The words you are searching are inside this book. To get more targeted content, please make full-text search by clicking here.
Discover the best professional documents and content resources in AnyFlip Document Base.
Search
Published by imstpuk, 2022-02-11 05:03:19

Companion to psychiatric studies Book 1

Companion to psychiatric studies Book 1

Companion to Psychiatric Studies

pound spent. Clinicians, however, are faced with a substantial Lastly, it does not play into the hands of those who wish to cut
problem: the research literature is chaotic and rapidly expand- the costs of healthcare, as the best available treatment is often
ing, and conventional approaches to keeping up-to-date are not the cheapest (although EBM may assist in the difficult
hopelessly inadequate. Evidence-based medicine (EBM), which process of balancing optimal practice against resource
consists of a coherent set of strategies based on developments in constraints).
information technology and clinical epidemiology, has been
proposed as a potential solution to this problem. EBM is a form Historical development
of ‘knowledge management’ that has been developed to meet
the needs of the practising clinician. Sackett et al (1997) have traced the philosophical origins of
EBM to the spirit of enquiry and call for external evidence in
EBM involves several key steps: 19th-century Paris. More specifically, the development of
the RCT as the clinical equivalent of a scientific experiment –
1. the recognition of uncertainty and the formulation of an first described by Daniels & Hill (1952) – has probably done
answerable question; more than anything else to place medical practice on secure
scientific foundations. The development and refinement of
2. the reliable and efficient identification of the best available techniques of meta-analysis (Smith & Glass 1977; Egger et al
evidence; 2001) has allowed the results of two or more studies to be
summarised quantitatively.
3. the critical appraisal (in terms of validity and usefulness) of
the evidence; EBM first became possible with developments in informa-
tion technology which allowed rapid bibliographic searching
4. the integration of the appraised best available evidence and retrieval, coupled with the application of biostatistical
with the clinician’s own experience and the patient’s principles, derived from population-based epidemiology, to
own preferences, biology and social circumstances; and the care of individual patients (Sackett et al 2005). At the
same time, the increasing demand from consumers of health-
5. evaluation and improvement of the clinician’s own care and others for more explicit use of treatments of proven
performance at each stage of the process. benefit increased the pressure on doctors to keep up-to-date
and to ensure that their practice was based on the best avail-
In this chapter we will briefly describe the main histori- able evidence (Cochrane 1972).
cal developments leading to EBM and critically examine its
potential clinical importance. We will then discuss in detail, Why EBM?
with examples, how each of the components of EBM
described above can be applied to the contemporary practice The central issue addressed by EBM is that keeping abreast
of psychiatry. Although EBM promises benefits for practi- of new medical developments is difficult. There are ever-
tioners (and their patients) throughout their career, we appre- increasing quantities of medical journals containing an ever-
ciate that most readers of this textbook will be studying for increasing literature to be sifted through. Relevant studies
postgraduate exams. We will therefore emphasise the pro- need to be read and assimilated and – when appropriate –
cesses of critical appraisal. This will be followed by an estimate new findings incorporated into clinical practice. Similarly, old
of how much of current psychiatric practice is evidence based, practices need to be discarded as evidence emerges of lack of
and how it compares with other medical disciplines. Finally, effectiveness or actual harm. For the average psychiatrist, as
we will discuss the main problems facing EBM in general and with most doctors, there are perhaps a handful of general
evidence-based mental health (EBMH) in particular. medical journals and a similar number of specialist journals
to be read, but time is limited and the journals are (clinically)
There has been a certain amount of misunderstanding about disorganised. Moreover, there is evidence that the further
EBM (Sackett et al 1997). Some have suggested that there is physicians are from graduation, the more out of date they
nothing new in EBM; certainly, the principle of basing health- become (Ramsey et al 1991). Not only are we likely to need
care on good empirical evidence has been generally accepted more time to study, but we are less likely to do so (and
for several decades, but the critical practical developments in less able to assimilate new information, due to age-related
information technology and epidemiology that make EBM fea- cognitive decline).
sible in the real world have only occurred recently. These same
developments mean that EBM is not impossible, as some have This is not merely an academic issue. Studies suggest that
suggested. EBM is not a purely academic pursuit – its whole the average physician, when asked, could usefully seek clini-
theory and practice is targeted directly at the needs of busy cally relevant information in two out of every three patients
clinicians and their patients. EBM does not ignore the results seen, and that such information would change their clinical
of research designs apart from randomised controlled trials decisions in one of every four (Covell et al 1985; Smith
(RCTs) or meta-analyses of RCTs; the most appropriate 1996). It has long been recognised that the mass of potentially
research design depends on the nature of the question being relevant information needs to be reliably reviewed and format-
asked, e.g. whether the question is to do with diagnostic tests, ted for easy access in everyday clinical situations. The problem
making valid prognostic statements or understanding patients’ is that traditional updating methods – books, review articles
experiences. EBM has been confused with the guidelines
‘industry’, or managed care, which are often seen as attempts
to restrict clinical practice inappropriately with a form of
‘cook-book’ medicine that ignores clinical expertise or patient
preferences. However, this is a willful misunderstanding of
EBM, because it has always been emphasised that the aim is
to integrate research evidence with other forms of knowledge.

190

Research methods, statistics and evidence-based practice CHAPTER 9

and even continuing medical education – are unreliable and do the situation, but productive questions tend to be composed
not change clinical practice. of four parts, known as the PICO format: the patient problem,
the type of clinical issue or intervention being considered (e.g.
Textbooks are almost inevitably out of date by the time they diagnostic, prognostic or therapeutic), the comparison interven-
are published. Most ‘expert’ review articles are unsystematic tion (if appropriate) and the clinical outcome(s) of interest.
(in the selection and interpretation of studies for review) and
often outdated. For example, conventional narrative review Treatment example: You are referred a 22-year-old female outpatient
methods delayed the use of thrombolysis for myocardial infarc- with a 5-year history of bulimia nervosa. She is not clinically depressed,
tion by about 10 years after a systematic review/meta-analysis but regularly abuses alcohol, and wishes for some sort of talking rather
would have shown convincing benefit (Antman et al 1992); than drug treatment. As a general adult psychiatrist, you are not sure
while the dangerous, sometimes lethal, use of lignocaine (lido- what the best sort of treatment would be. You ask her to begin to
caine) prophylaxis for ventricular fibrillation was extended by record a diary of her eating and bingeing and arrange to do a literature
about the same length of time after it had been shown to be search before you see her again in 2 weeks time.
counterproductive. By contrast, systematic reviews require a
comprehensive search for relevant studies using an explicit Question: In bulimia nervosa (problem), which psychotherapy
strategy, bias-free citations, accurate judgement of scientific (intervention) is most effective in securing recovery (outcome)?
quality of cited articles and appropriate synthesis of the articles’
conclusions. Unfortunately, but perhaps predictably, there is a There are of course several questions that one could ask in this
strong inverse relationship (r ¼ À0.52) between adherence to and any clinical situation – for example, how does drug treatment
these standards and self-professed expertise of the reviewer compare with psychotherapy? Are there alternative treatments?
(Oxman & Guyatt 1993). The question will, of course, depend on what is most useful to
you and your patient, what evidence you can actually find, and
Lastly, the processes of continuing medical education what might be of greatest benefit to know for similar common
(CME) are often ineffective. As with reading, those motivated situations in the future. Of course, the number of possible ques-
to attend CME courses are least likely to need to. It has been tions will always be greater than the time available to answer
shown that many commonly used methods of CME do not them, but once one question is answered (and ideally recorded
change doctors’ clinical behaviour and so are unlikely to bring somewhere for easy access), others can be dealt with as and when
about improvements in the quality of patient care (Davis the need and opportunity arises. Not only can several questions
et al 1995). What is needed is detailed audit and persona- be asked within a domain (e.g. treatment) but several domains
lised feedback on performance. The strategies of EBM offer (such as clinical findings, aetiology, differential diagnosis, pro-
a feasible alternative to haphazard reading and being at the gnosis, therapy, prevention and even self-improvement) may be
mercy of pharmaceutical company representatives for keeping pertinent in any one consultation.
up-to-date. Although EBM itself is still in need of rigorous
evaluation, there is some evidence that medical students at 2. Searching for evidence
institutions where the teaching has been problem-based rather
than knowledge-based for several years, find their undergradu- A really well-framed clinical question will make the literature
ate teaching more stimulating, satisfying and are better able to search parameters obvious, but this may sometimes require
keep up-to-date with medical advances than students taught in further thought (and perhaps slight modification of the ques-
traditional curricula (Sackett et al 1997). tion). In obtaining evidence, one must next consider where
and how to get it.
How to practise EBM
Electronic media for literature searching have several advan-
The five steps involved in EBM were outlined earlier. Below tages, in terms of access time and completeness, over hand
we will illustrate the process of EBM as applied to a single searching journals. The quickest and easiest way of identifying
treatment study, a prognostic study and a meta-analysis. It reliable information is to use evidence-based medical sum-
should be noted, however, that EBM also covers issues of maries generated by others, as mentioned above. However,
aetiology, audit, clinical decision analysis, clinical guidelines, if this does not answer the question, then a search of one of
diagnosis, economics, qualitative studies, quality improvement the computerised literature databases is often required.
and others which are beyond the scope of this brief discussion.
Those who are particularly interested or assiduous in preparing MEDLINE, BIDS (Bath Information Database Service) and
for exams are referred to Sackett et al (1997) and/or Lawrie PsychLit are all useful resources, but have different databases
et al (2000). stretching back to 1975, 1981 and 1990, respectively. It is
important to appreciate that even a well-conducted search on
1. Framing questions one of these databases will probably only detect 30–50% of
the relevant studies and that each will identify a different set
To be useful, such questions will be brief summaries of the clin- of publications, but in practice MEDLINE has the advantage
ical scenario and what information is required which are con- of being widely available, now with an improved user-friendly
structed in a way that makes searching for evidence likely to web interface.
succeed. Specific questions will of course vary according to
A MEDLINE search demands search terms, as textwords or
subject headings, to be typed in and allows certain operations

191

Companion to Psychiatric Studies

to be performed to identify publications of a particular type. Ide- Box 9.10
ally, when considering a treatment issue as in the brief example
above, one would be able to identify one or more systematic Critical appraisal for single treatment studies
reviews (indexed in MEDLINE under the heading ‘meta-
analysis’) of RCTs which monitored and compared relapse rates Is the research valid?
on drug and placebo. However, it is more likely that no meta-
analyses are available and one or more potentially relevant RCT Was the assignment of patients to treatments randomised?
must be identified and then evaluated. The secondary research Was the randomisation list concealed?
journal Evidence-Based Medicine (EBM) suggests search terms Were all subjects who entered the trial accounted for at its
to identify RCTs according to the year of publication – a ‘high-
quality yield search’ with one item would use ‘clinical trial conclusion?
(publication type [pt])’ for 1990 and after, as well as ‘random Were they analysed in the groups to which they were
(textword [tw])’ before 1990, although using more than one
term will increase the number of studies identified. PubMed randomised?
now includes a ‘clinical queries’ service that automatically uses Were subjects and clinicians blind to which treatment was being
filters for the most reliable design to answer a clinical question.
received?
If you cannot find any relevant studies, a number of possible Apart from the experimental treatment, were the groups treated
strategies are available. MESH (medical subject headings)
search terms can be ‘expanded’ into a broader category, alter- equally?
native words could be used or another database may be worth Were the groups similar at the start of the trial?
consulting. However, it is more common to get too much
rather than too little information from such a search. It is Is the research important?
often useful to get help in searching, or learning how to search,
from a trained librarian. Once the study or studies have been Absolute risk reduction (ARR, i.e. CER–EER) ARR ¼
identified, they must be evaluated in terms of their validity Number need to treat (NNT, i.e. 1/ARR) NNT ¼
and usefulness.
Can I apply it to my patient?
Literature search example: You remember seeing a structured
review of a paper on psychotherapy for bulimia in a recent edition of Is this patient so different from those in the trial that the results
Evidence-Based Medicine (on disk or in print). Searching the disk/ do not apply?
journal you find the one-page summary, although it was published
longer ago than you thought (EBM Jan/Feb 1996). You note that the How great would the benefit be for this particular patient?
summary contains some of the EBM summary data you require. What is the patient expected event rate (PEER) in my practice for

3. Appraising the evidence patients like this one?
What is the (adjusted) NNT for this patient? (PEER/CER ¼ F, or

estimate) NNT/F ¼

Is it consistent with my patient’s values and
preferences?

Do I have a clear assessment of the patient’s values and
preferences?

Are they met by this intervention and its potential
consequences?

Adapted from Sackett et al (1997/2000)
CER, control event rate; EER, experimental event rate.

The evidence needs to be critically appraised for its scientific At this point, it is worth briefly reviewing some of the mea-
validity and clinical importance. Validity criteria are essentially sures of clinical effectiveness and how to calculate them for
the same as the questions to be answered in critical appraisal treatment studies. We are primarily interested in comparing
(see Sackett et al 1997); while clinical importance can be the proportion of patients treated with a new treatment who
determined by some of the summary measures EBM practi- get the outcome of interest – or the experimental event rate
tioners find useful (particularly for treatment studies). (EER) – with the proportion of patients treated with an alter-
native (standard) treatment who get the outcome of interest
Appraisal example: Although you know that Evidence-Based Medicine — or control event rate (CER). The difference between these
only selects for inclusion treatment trials with random allocation, clini- two outcome rates is the absolute risk reduction (ARR), i.e.
cally important outcome measures and consistent data analysis, no CER À EER (for an undesired outcome) expressed as a per-
system is infallible, and therefore you evaluate the paper for yourself centage. This tells us the difference in the number of patients
(Fairburn et al 1995), following the checklist for treatment studies with a specific outcome for every 100 patients treated in
(Box 9.10). either way. The next term to introduce transforms this ARR
into a more clinically useful number – the number needed to
The paper compares the outcome after cognitive-behavioural ther- treat (NNT) – which is simply the reciprocal of the ARR and
apy (CBT), interpersonal therapy (IPT) and behaviour therapy. Treat- tells us how many such patients we would need to treat in a
ment allocation was random (although the paper does not mention particular way so as to avoid one outcome event. As a rough
whether or not the randomisation list was concealed); 90% of the rule of thumb, NNTs of less than 10 usually denote a powerful
patients were interviewed at follow-up and the groups were analysed and important treatment effect.
as randomised; the treatment was not blind (but outcome assessment
was); and the groups were treated equally other than with the interven-
tions of interest and did not differ significantly at the start of the trial.
You decide therefore that the study is valid.

192

Research methods, statistics and evidence-based practice CHAPTER 9

The results given in the paper are rates of still satisfying diagnostic asking each other for the evidence in support of some of their
criteria for bulimia at the end of the study: 37% for CBT, 28% for IPT statements. Audit of how ‘evidence-based’ your practice is and
and 86% for simple behavioural therapy. The ARR and NNT compared what changes we should aim for, preferably with individua-
with simple behavioural therapy are therefore 49% (86 À 37%) and lised feedback, will then be feasible. Similarly, it may be pos-
2 (95% confidence interval 1 to 4) for CBT, and 58% (86 À 28%) sible to begin to teach EBM principles to medical students or
and 2 (1 to 3) for IPT. There seems little to choose from between members of other disciplines. Existing structures, such as a
CBT and IPT, but the summary states that patients receiving CBT were journal club, can be reorganised along EBM lines (Sackett
less likely to have symptoms than those receiving either IPT or beha- et al 1997).
vioural therapy, and that CBT complete remission rates were highest.

4. Implementation Evaluation example: You ask the psychologist to inform you of the
treatment outcome and decide to audit the treatments (and their out-
We are here concerned with whether the results of valid, come) you have offered to patients with bulimia you have seen over
important studies can be applied to our particular patient or the past year. You discuss with your colleagues the possibility of a
group of patients. In essence, this depends on the similarity larger audit and whether all such patients are treated with CBT in the
and differences between the subjects in a paper and our own future (for which resources will need to be identified).
clinical population. Certain questions can be routinely asked
for therapeutic studies, as shown in Box 9.10. In practice, it Further examples of critical appraisal
is often quickest to answer these applicability questions first, and implementation
as this avoids the unnecessary evaluation of irrelevant papers,
as long as the other stages of critical appraisal are not Prognosis
forgotten.
Example: A 25-year-old male patient currently hospitalised for
One further term needs to be introduced here. This is sim- his first episode of schizophrenia is nearing discharge. His age-
ply an estimate of your own patients’ susceptibility to the out- ing parents ask to see you to discuss the likely outcome for
come of interest as compared with the average patient in the their son. At the meeting they are particularly interested in
trial – on the basis of age, sex, comorbidity, etc. This estimate whether he will be able to live independently and obtain paid
is called F (for fraction), as many patients will be less suscep- employment in the future. You tell them that the conventional
tible than those in the RCT (e.g. F ¼ 1/2), although some wisdom is that about half of all patients with schizophrenia
patients may be more liable to benefit and the F will be greater will return to their premorbid social situation. They thank
than 1 in such cases. The NNT for any particular patient can you for your time, but you are dissatisfied with the informa-
be simply calculated by dividing the NNT by F. tion you gave and sense that the parents were too. You resolve
to search the literature for a reliable prognosis study.
Implementation example: Your patient is clearly similar in age and sex
to those described in the study and would have been eligible for inclu- 1. Question
sion in the study. Interpersonal psychotherapy is not available locally,
but cognitive-behaviour therapy is (and it produces the best remission What percentage of patients in their first episode of schizo-
rates). In your team, the clinical psychologist provides the CBT, and so phrenia will be living independently and in paid employment
you refer the patient. You tell him that there is a good chance of a suc- at 5–10 years?
cessful outcome.

5. Evaluation 2. Literature search

The final stage of EBM is the continual evaluation and Using the MESH headings ‘schizophrenia’, ‘cohort’ and ‘first
improvement of the specific skills involved at each stage of episode’ on MEDLINE you identify several studies. However,
the process. It is useful to review periodically the clinical ques- one catches your eye as it describes the results of a 13-year
tions you have asked and your success in answering them. How follow-up in the UK and mentions disability and residence in
well can you critically appraise the scientific literature, and is the structured abstract (Mason et al 1995).
there any way that you can improve your skills (for example,
by using critical appraisal checklists prepared by others)? Are 3. Appraisal
you providing clinically useful summaries of the evidence,
and keeping them up-to-date? It is useful to share any pro- Using the critical appraisal checklist for prognosis studies
blems identified, by discussing them with other local practi- (Box 9.11), you decide that the Mason et al (1995) study
tioners or clinical epidemiologists. You may benefit from meets the validity criteria because it studied all first-episode
attendance at an EBM or critical appraisal skills workshop. psychosis patients in a defined catchment area, obtained
follow-up information in 94%, objective outcomes were deter-
One useful way of implementing EBM is to encourage a mined with good reliability and no subgroups with different
critical, but supportive, culture. Clinical colleagues should be prognoses were identified (making adjustment unnecessary),
even though there was no independent ‘test-set’ group of

193

Companion to Psychiatric Studies

Box 9.11 Treatment and systematic reviews and
meta-analysis
Critical appraisal for prognosis studies
Example: You want to improve the local implementation of
Is the research valid? case management for the severely mentally ill. You are aware
that there has been considerable uncertainty about the effec-
Was a defined, representative sample of patients assembled at a tiveness of case management and for which patients it is most
common (usually early) point in the course of their disease? useful.

Was patient follow-up sufficiently long and complete? 1. Question
Were objective outcome criteria applied in a ‘blind’ fashion?
If subgroups with different prognoses are identified, was there For patients with severe mental illness, what are the effects
of case management on the clinical state and service
adjustment for important prognostic factors? utilisation?
Was there validation in an independent group (‘test-set’) of
2. Searching for evidence
patients?
You search for a relevant systematic review of RCTs in the
Are the valid results of this prognosis study important? CDSR (Cochrane Database of Systematic Reviews) on the
Cochrane Library. You find a systematic review of case man-
How likely are the outcomes over time? agement for people with severe mental disorders which seems
How precise are the prognostic estimates? to address the issue (Marshall et al 2003).

Can you apply this valid, important evidence about 3. Appraisal
prognosis in caring for your patient?
Although the review has been conducted under the auspices
Were the study patients similar to your own? of the Cochrane Collaboration, it is still important to criti-
Will this evidence make a clinically important impact on your cally appraise it for validity and usefulness (Box 9.12). The
review includes 11 RCTs of case management versus stan-
conclusions about what to offer or tell your patient? dard care. The reviewers clearly describe the search strategy
for identifying the primary studies, as well as the inclusion
Adapted from Sackett et al (1997/2000) and exclusion criteria for including them in the review. All
the studies investigated a form of case management which
patients. The results also look important – 57 of the 59 sub- was broadly comparable; standard care was defined as the
jects (97%) had been living independently in the community usual level of psychiatric care provided in the area where
for most of the past 2 years, 16 (28%) alone and 22 (37%) each study was conducted. The studies examined a range of
had been employed for the past 2 years. clinical and health service utilisation outcomes. Although
there was some variation between the studies, the reviewers
You then calculate 95% confidence intervals for these out- investigated the reasons for this and concluded that it was
comes to ensure that the results given in the paper are suffi- mainly quantitative rather than qualitative. You decide that
ciently precise to be useful in practice. The 95% confidence the results of the overview are probably valid. As the primary
interval (CI) is the range of values in which you can be 95% studies were performed in the UK and USA, and included
sure that the true value lies. The approximate 95% CI for a patients with severe mental illness (however defined), you
proportion (expressed as a decimal) is the proportion plus or also think that the review can be applied to your own clinical
minus 1.96 times the square root of {[(the proportion) Â situation.
(1 À proportion)]/sample size}. For living independently, this
is 0.97 Æ 0.04, or 93–100%. The other confidence intervals are 4. Implementation
17–39% for living alone and 26–48% for employment. The
95% CI gives an estimate of the uncertainty of the proportion. The main findings of the review are that case management
assists community psychiatric teams to maintain contact with
4. Implementation patients. However, it also increases the rate of admission to
hospital. There was no significant effect on other outcome
Again using the checklist (Box 9.11), you decide that the variables (such as possible improvements in symptom severity
results can be applied to your patient because the study was or quality of life). Of 599 case-managed subjects, 150 were
of an unbiased inception (first-episode) cohort and your lost to follow-up (EER ¼ 25%), as compared with 195 of
patient is just recovering from his first episode. The study 611 standard care subjects (CER ¼ 32%). The absolute benefit
was in a UK secondary-care setting similar to your own service. increase is 7% and the NNT is 15 (rounded up from 14.3).
You also think that the information will be useful for you, your This means that 15 patients have to be treated with case man-
patient and his parents. You arrange to see them again briefly agement to prevent one less patient being lost to follow-up
to discuss the results of your endeavours. than would occur with standard care.

5. Evaluation

You note that the paper identified also gives information on
symptoms and treatment outcomes and file your critically
appraised summary of the paper under schizophrenia –
prognosis. The summary will need to be updated as new
evidence becomes available.

194

Research methods, statistics and evidence-based practice CHAPTER 9

Box 9.12 5. Evaluation

Critical appraisal for a systematic review You therefore decide not to use your scarce resources on
routine case management (NNT 49), but to concentrate on
Is the research valid? those with whom you have previously lost contact (NNT 6).
You audit the loss to follow-up in both groups over the next
Did the review address a clearly focused issue? (i.e. did the 6 months, both to ensure that your pre-case management audit
review describe: the population studied? the intervention given? figures are still applicable and to examine the effects of your
the outcome considered?) decision to focus your care programming.

Did the authors select the right sort of studies for review? How evidence-based are we?
(i.e. addressing the review question, with adequate study
design) An obvious place to begin to evaluate our individual and collec-
tive practice is to ask how much of contemporary medical and
Were the important, relevant studies included? (Look for which specifically psychiatric treatment is evidence based? We tend
bibliographic databases were used, personal contact with to assume that, because we are trying to do good, we are doing
experts, search for unpublished as well as published studies, good. However, few of our treatments are evidence-based in
search for non-English language studies) the sense that they are supported by high-quality studies that
would meet strict critical appraisal guidelines. Indeed, it has
Did the reviewers do enough to assess the quality of the been estimated that perhaps ‘only 15% of medical interven-
included studies? tions are supported by solid scientific evidence’ (Smith 1991).

Did they describe randomisation and/or use a rating scale? This debate raises the question: how can we deem a treat-
ment to be ‘evidence-based’? One way of answering this is to
What are the results? identify how well the intervention is supported by evidence
– using a hierarchy to rate the evidence according to the likeli-
Were the results similar from study to study? hood of it giving an unbiased estimate of the true effect. Most
Are the results of all studies clearly displayed? reliable are findings from systematic reviews of good-quality
If the results are not similar, are the reasons for the variations RCTs, followed by individual good-quality RCTs. However,
there are certain clinical situations where benefits could be
discussed? said to be obvious, or where RCTs would be difficult and,
What is the overall result of the review? arguably, even unethical (e.g. admission and observation for
Is there a clinical bottom line? suicidal ideation). Nonetheless, it might still be possible to
What is the numerical result? (ARR ¼ ?, NNT ¼ ?) conduct standard care versus alternative treatment RCTs in
How precise are the results? these areas. Alternatively, some useful information about
Is there a confidence interval? treatments may be available from naturalistic follow-up stud-
ies, although the information gained from cohort studies and
Can I apply the results to my patient(s)? case-control studies is most reliable for prognosis and diag-
nosis, respectively. Lastly, uncontrolled studies, case series
Is my patient so different from those in the trial that the results and individual case reports are too susceptible to bias to have
may not apply? general value, although they may sometimes identify a treat-
ment effect or therapeutic hazard that would merit further
Should I apply the results to my patient(s)? study.
How great would the benefit be?
Is the intervention consistent with my patient’s values and Using these criteria – where treatment is considered
evidence-based if it is based on results from randomised evi-
preferences? dence, either from systematic reviews or RCTs, or where
Were all the clinically important outcomes considered? benefit is obvious – how evidence based is contemporary med-
Are the benefits worth the harm and costs? icine and psychiatry? The answer crucially depends on whether
we talk about treatments or diseases. Chalmers et al (1989)
Adapted from Sackett et al (1997). examined the evidence for 226 obstetric procedures: only
50% had been evaluated in RCTs, with only 20% of procedures
The ratio of the odds of being lost to follow-up in the case having been shown to be beneficial; the other 30% were of
management group to the odds of the same in the standard dubious benefit or dangerous. On the other hand, taking
care group was 0.70 (95% CI 0.50 to 0.98). You know from patients as the denominator, Ellis et al (1995) reviewed the
an audit of your own service, carried out before the implemen- evidence for the main treatment of 121 consecutive admis-
tation of the care programme approach, that only 10% of your sions to a medical ward in Oxford over 1 month. One hundred
own patients were lost to follow-up each year. This is the and nine patients had an identified primary diagnosis, 58
patient expected event rate (PEER) and is about 30% (0.10/ (53%) of whom received evidence-based interventions (e.g.
0.32 ¼ 0.31) of the CER in the review. Assuming that case heparin and warfarin for deep venous thrombosis), a further
management has a fairly constant effect in all patient groups,
you can therefore adjust the NNT to apply to the loss of
follow-up rate of your service by dividing it by this proportion,
which is called the F value (Cook & Sackett 1995). This
revised NNT is approximately 49 (15/0.31). However, among
patients with a previous history of loss of contact, your drop-
out rate was 80%. Adjustment in a similar way produces a sec-
ond revised NNT of 6 for this ‘high risk of dropout’ patient
group.

195

Companion to Psychiatric Studies

32 (29%) received treatments supported by convincing non- used in multidisciplinary inpatient (and particularly outpatient)
experimental evidence (e.g. antibiotics for infections, resusci- settings.
tation for cardiac arrest) and the remaining 19 (18%) patients
were given treatments without any substantial supporting evi- Summers & Kehoe (1996) addressed some of these issues
dence (e.g. support for stroke or overdoses, treatment of pain). by examining 160 treatment decisions in 158 patients (56
Overall, therefore, 82% of the medical interventions were evi- inpatients, 29 day patients and 75 outpatients). They reported
dence based, taking patients rather than procedures as the that there was RCT evidence for 85 (53%) of the patient
focus of interest. This apparent discrepancy between proce- interventions (especially drug treatments), that 16 (10%) of
dures or patients can be explained by the simple fact that the interventions could not be ethically subject to RCT evalu-
there are many procedures which have never been evaluated, ation (e.g. observation levels in suicide) and that the remaining
whereas patients tend to have common diseases which are eas- 59 (37%) patients received supportive practical measures and
ier to study and therefore more likely to have been subject to psychotherapy. Thus, inpatient psychiatric treatment may be
treatment trials. We can take issue with the Ellis and collea- more evidence based than outpatient services, but, overall,
gues’ approach, or some of the deficiencies of the study – such psychiatry compares reasonably well with acute medicine.
as potential problems with generalisability, the accuracy of There is still a need, however, for further and more all-
diagnosis, the fact that few patients only have one problem encompassing assessments of these questions in psychiatry as
in clinical practice, and that self-evident treatments are not in other specialties.
necessarily effective – but the conclusion remains that most
medical patients receive treatments backed up by evidence Conclusions
from RCTs. How does psychiatry compare?
Archie Cochrane criticised psychiatry in 1972 for ‘using a large
Using a similar design, Geddes and colleagues (1996) eval- number of therapies whose effectiveness has not been proven’,
uated the treatments received by 40 consecutive admissions stated ‘it is basically inefficient’, and suggested increasing
to a general psychiatry ward in Oxford. Twenty-nine (65%) ‘grants for well-designed evaluatory research’. As he well
of these acute admissions received interventions supported appreciated, clinical research is required to improve the man-
by evidence from RCTs or meta-analysis. The main diagnoses agement of patients’ problems – otherwise new tests, treat-
and treatments they received are shown in Table 9.13. The ments and service initiatives risk being little more than
other 14 (35%) acute admissions received treatments (usually fashionable trends. Statistics are required to help make sense
combinations of medication) for which there was no good evi- of the data acquired, and to determine whether any apparent
dence of efficacy. This is a small study and subject to the same differences, e.g. between treatments, are likely to have
criticisms as apply to the study of Ellis et al (1995). Moreover, occurred by chance. EBM statistics (such as the NNT and
many of the primary trials were small, and the study only con- the likelihood ratio) then allow one to determine whether
sidered medical interventions: many other interventions are valid, statistically significant findings are likely to be clinically
significant.
Table 9.13 Evidence-based treatments received by 29 out of 40
psychiatry patients EBM also attempts to help doctors to stay abreast of devel-
opments in medical research by integrating clinical epidemiol-
Primary diagnosis (No. patients) Primary treatment ogy, medical informatics, biostatistics, pathophysiological
knowledge and clinical experience. Its underlying rationale is
Alcohol withdrawal (1) Individualised chlordiazepoxide that healthcare decisions should be based on the best available
regimen evidence. For each kind of clinical question it is usually possi-
Acute schizophrenia (12) Antipsychotic medication ble to identify the study design most likely to provide the most
Chronic schizophrenia (4) Oral/i.m. antipsychotic medication valid and useful information. Evidence is identified using the
Acute mania (3) Lithium most efficient available methods, critically appraised using
Bipolar affective disorder Lithium empirically derived criteria and integrated with clinical prac-
prophylaxis (2) tice using meaningful measures and indices.
Depressive disorder (4) Antidepressants
Treatment-resistant depression (3) Lithium augmentation Substantial progress has been made in understanding and
Adapted from Geddes et al (1996) managing psychiatric disorders in the past 30–40 years. The
slow but steady application of critical thinking, original
research, clinical epidemiology and audit is the only way of
reliably improving patient care into the future and offers one
viable means of job satisfaction throughout the average clinical
career. Who could argue with the assertion that our patients
deserve the rigour of a systematically gathered evidence base
for contemporary medical practice?

196

Research methods, statistics and evidence-based practice CHAPTER 9

Further reading

Carpenter Jr., W.T., Gold, J.M., Lahti, A.C., Gilbody, S.M., House, A.O., Sheldon, T.A., Straus, S.E., Scott Richardson, W.S.,
et al., 2000. Decisional capacity for informed 2002. Outcomes research in mental health. Glasziou, P., Haynes, R.B., 2005. Evidence-
consent in schizophrenia research. Arch. Br. J. Psychiatry 181, 8–16. based medicine, third ed. Churchill
Gen. Psychiatry 57, 533–538. Livingstone, Edinburgh.

References

Altman, D.G., 1991. Practical statistics for Davis, D.A., Thomson, M.A., Oxman, A.D., Hand, D.J., Taylor, C.C., 1987. Multivariate
medical research. CRC Press, Boca Raton, Haynes, R.B., 1995. Changing physician analysis of variance and repeated measures.
Florida. performance: a systematic review of the Chapman and Hall, London.
effect of continuing medical education
Altman, D.G., Bland, J.M., 2002. Time to event strategies. J. Am. Med. Assoc. 274, 700–705. Hart, A., 2001. Mann-Whitney test is not just a
(survival) data. Br. Med. J. 317, 468–469. test of medians: differences in spread can be
Department of Health, 2001. Treatment choice important. Br. Med. J. 323, 391–393.
Antman, E.M., Lau, J., Kupelnick, B., et al., in psychological therapies and counseling:
1992. A comparison of results of meta- evidence based clinical practical guideline. Hennekens, C.H., Buring, J.E., 1987.
analyses of randomized control trials and Department of Health, London. Epidemiology in medicine. Little, Brown,
recommendations of clinical experts: Boston.
treatments for myocardial infarction. J. Am. DerSimonian, R., Laird, N., 1986. Meta-analysis
Med. Assoc. 268, 240–248. in clinical trials. Control. Clin. Trials 7, Hotopf, M., Churchill, R., Lewis, G., 1999.
177–188. Pragmatic randomised controlled trials in
Bedi, N., Chilvers, C., Churchill, R., et al., 2000. psychiatry. Br. J. Psychiatry 175, 217–223.
Assessing effectiveness of treatment of Edwards, P., Roberts, I., Clarke, M., et al., 2002.
depression in primary care: partially Increasing response rates to postal Juni, P., Altman, D.G., Egger, M., 2001.
randomised preference trial. Br. J. Psychiatry questionnaires: systematic review. Br. Med. J. Systematic reviews in health care: assessing
177, 312–318. 324, 1183. the quality of controlled clinical trials.
Br. Med. J. 323, 42–46.
Bland, J.M., 2000. An introduction to medical Egger, M., Smith, G.D., Altman, D.G., 2001.
statistics. Oxford University Press, Oxford. Systematic reviews in health care: meta- Kline, P., 1994. An easy guide to factor analysis.
analysis in context. BMJ Books, London. Routledge, London.
Bland, J.M., Altman, D.G., 1986. Statistical
methods for assessing agreement between Ellis, J., Mulligan, I., Rowe, J., Sackett, D.L., Kraemer, H.C., Stice, E., Kazdin, A., et al.,
two methods of clinical measurement. Lancet 1995. Inpatient general medicine is evidence- 2001. How do risk factors work together?
i, 307–310. based. Lancet 346, 407–410. Mediators, moderators, and independent,
overlapping, and proxy risk factors. Am. J.
Bland, J.M., Altman, D.G., 1994. Statistics Even, C., Siobud-Dorocant, E., Dardennes, R. Psychiatry 158, 848–856.
notes: one and two sides tests of significance. M., 2000. Critical approach to antidepressant
Br. Med. J. 309, 248. trials: blindness protection is necessary, Last, J.M., 1995. A dictionary of epidemiology.
feasible and measurable. Br. J. Psychiatry Oxford University Press, Oxford.
Bland, J.M., Altman, D.G., 1997. Statistics 177, 47–51.
notes: Cronbach’s alpha. Br. Med. J. 314, Lawrie, S.M., McIntosh, A.M., Rao, S., 2000.
572. Everitt, B.S., Wessely, S., 2008. Clinical trials in Critical appraisal for psychiatry. Churchill
psychiatry, second ed. Wiley Blackwell, Livingstone, Edinburgh.
Bland, J.M., Altman, D.G., 1998. Survival Chichester.
probabilities (the Kaplan-Meier method). Lewis, G., Pelosi, A.J., 1990. The case-control
Br. Med. J. 317, 1572. Fairburn, C.G., Norman, P.A., Welch, S.L., study in psychiatry. Br. J. Psychiatry 157,
et al., 1995. A prospective study of outcome 197–207.
Bland, J.M., Altman, D.G., 2002. Validating in bulimia nervosa and the long-term effects
scales and indexes. Br. Med. J. 324, 606–607. of three psychological treatments. Arch. Gen. Louis, T.A., Lavori, P.W., Bailar 3rd, J.C.,
Psychiatry 52, 304–312 (summarised in Polansky, M., 1984. Crossover and self-
Carpenter Jr., W.T., Sadler, J.H., Light, P.D., Evidence-Based Medicine 1:48). controlled designs in clinical research. N.
et al., 1983. The therapeutic efficacy of Engl. J. Med. 310, 24–31.
hemodialysis in schizophrenia. N. Engl. Farmer, A., 1999. The demise of the published
J. Med. 308, 669–675. case report – is resuscitation necessary? Br. J. McMahon, A.D., 2002. Study control, violators,
Psychiatry 174, 93–94. inclusion criteria and defining explanatory
Chalmers, T.C., Celano, P., Sacks, H.S., and pragmatic trials. Statistical Medicine 21,
Smith Jr., H., 1983. Bias in treatment Feinstein, A.R., 1985. The architecture of 1365–1376.
assignment in controlled clinical trials. clinical research. Saunders, Philadelphia.
N. Engl. J. Med. 309, 1358–1361. Marshall, M., Gray, A., Lockwood, A.,
Geddes, J.R., Game, D., Jenkins, N.E., et al., Green, R., 2003. Case management for
Chalmers, I., Enkin, M., Keirse, M.J.N.C., 1989. 1996. What proportion of primary people with severe mental disorders.
Effective care in pregnancy and childbirth. psychiatric interventions are based on Available in The Cochrane Library. Update
Oxford University Press, Oxford. randomised evidence. Qual. Health Care 5, Software, Oxford.
215–217.
Cochrane, A., 1972. Effectiveness and efficiency: Mason, P., Harrison, G., Glazebrook, C., et al.,
random reflections on health services. Gilbody, S.M., Song, F., 2000. Publication bias 1995. Characteristics of outcome in
Cambridge University Press, Cambridge and the integrity of psychiatry research. schizophrenia at 13 years. Br. J. Psychiatry
(reprinted 1989 by BMJ Publishing, London). Psychol. Med. 30, 253–258. 167, 596–603.

Cook, R.J., Sackett, D.L., 1995. The number Gilbody, S., Whitty, P., 2002. Improving the Moher, D., Cook, D.J., Eastwood, S., et al.,
needed to treat: a clinically useful measure of delivery and organisation of mental health 1999. Improving the quality of reports of
treatment effect. Br. Med. J. 310, 452–454. services: beyond the conventional randomised meta-analyses of randomised controlled
controlled trial. Br. J. Psychiatry 180, 13–18. trials: the QUOROM statement. Lancet 354,
Covell, D.G., Uman, G.C., Manning, P.R., 1985. 1896–1900.
Information needs in office practice: are they Gore, S.M., Altman, D.G., 1982. Statistics in
being met? Ann. Intern. Med. 103, 596–599. practice. BMJ Books, Wiley-Blackwell. Moncrieff, J., Wessely, S., Hardy, R., 1998.
Meta-analysis of trials comparing
Daniels, M., Hill, A.B., 1952. Chemotherapy of Grimes, D.A., Schulz, K.F., 2002. Descriptive antidepressants with active placebos. Br. J.
pulmonary tuberculosis in young adults: an studies: what they can and cannot do. Lancet Psychiatry 172, 227–231; discussion 232–234.
analysis of the combined results of three 359, 145–149.
Medical Research Council trials. Br. Med. J. Norman, G.R., Streiner, D.L., 1999. PDQ
1, 1162. statistics. BC Decker, Hamilton, Ontario.

197

Companion to Psychiatric Studies

Offord, D.R., Kraemer, H.C., 2000. Risk factors practice and problems. Statistical Medicine effects in controlled trials. J. Am. Med.
and prevention. Evid. Based Ment. Health 70, 21, 2917–2930. Assoc. 273, 408–412.
70–71.
Ramsey, P.G., Carline, J.D., Inui, T.S., et al., Siegel, S., 1998. Nonparametric statistics.
Oxman, A., Guyatt, G.H., 1993. The science of 1991. Changes over time in the knowledge McGraw-Hill, Singapore.
reviewing research. Ann. N. Y. Acad. Sci. base of practicing internists. J. Am. Med.
703, 125–134. Assoc. 266, 1103–1107. Smith, M.L., Glass, G.V., 1977. Meta-analysis of
psychotherapy outcome studies. Am.
Oxman, A.D., Cook, D.J., Guyatt, G.H., Sackett, D.L., 1979. Bias in analytic research. Psychol. (September), 752–760.
Evidence-Based Medicine Working Group, J. Chronic. Dis. 32, 51–63.
1994. Users’ guides to the medical literature. Smith, R., 1991. Where is the wisdom? Br. Med.
VI. How to use an overview. JAMA 272, Sackett, D.L., Haynes, R.B., Guyatt, G.H., J. 303, 798–799.
1367–1371. Tugwell, P., 2005. Clinical epidemiology:
how to do clinical practice research, third ed. Smith, R., 1996. What clinical information do
Parmar, M.K.B., Machin, D., 1995. Survival Lippincott Williams & Wilkins, Philadelphia, doctors need? Br. Med. J. 313, 1062–1068.
analysis: a practical approach. Wiley, PA.
Chichester. Streiner, D.L., Norman, G.R., 1996. Health
Sackett, D.L., Richardson, S., Rosenberg, W., measurement scales: a practical guide to their
Perneger, T.V., 1998. What’s wrong with Haynes, R.B., 1997. Evidence-based development and use. Oxford University
Bonferroni adjustments. Br. Med. J. 316, medicine: how to practise and teach EBM. Press, Oxford.
1236–1238. Churchill Livingstone, London.
Stroup, D.F., Berlin, J.A., Morton, S.C., et al.,
Peto, R., Collins, R., Gray, R., 1993. Large-scale Schulz, K.F., Grimes, D.A., 2002a. Case-control 2000. Meta-analysis of observational studies
randomized evidence: large, simple trials and studies: research in reverse. Lancet 359, in epidemiology. J. Am. Med. Assoc. 283,
overviews of trials. Ann. N. Y. Acad. Sci. 703, 431–434. 2008–2012.
314–340.
Schulz, K.F., Grimes, D.A., 2002b. Unequal Summers, A., Kehoe, R.F., 1996. Is psychiatric
Pocock, S.J., 1983. Clinical trials: a practical group sizes in randomised trials: guarding treatment evidence-based? Lancet 347,
approach. John Wiley, Chichester. against guessing. Lancet 359, 966–970. 409–410.

Pocock, S.J., Assmann, S.E., Enos, L.E., Schulz, K.F., Chalmers, I., Hayes, R.J., Thornley, B., Adams, C., 1998. Content and
Kasten, L.E., 2002. Subgroup analysis, Altman, D.G., 1995. Empirical evidence of quality of 2000 controlled trials in
covariate adjustment and baseline bias: dimensions of methodological quality schizophrenia over 50 years. Br. Med. J. 317,
comparisons in clinical trial reporting: current associated with estimates of treatment 1181–1184.

198

Clinical assessment: interviewing 10
and examination

David Cunningham Owens Peter J McKenna Richard Davenport

The psychiatric interview ability to be effective in promoting a good clinical outcome
can be clearly linked to the competence of our communication
It is often said that medicine is as much art as science but this skills. It goes without saying therefore that being ‘nice’ and
overlooks an additional dimension particularly pertinent to ‘well-intentioned’ is not enough. Psychiatric interviewing
psychiatry – that of craft. Much of what comprises good is not social chit-chat but a professional interaction geared
psychiatric practice takes what is inherently ‘good’ about an towards engaging the patient in a therapeutic relationship.
individual’s ability to relate to others and hones it, under the As such, there are certain factors which the interviewer must
influence of a mentor of experience, into an effective set of be aware of from his/her own, as well as the interviewees,
professional skills. As a result, learning the tools of one’s trade point of view – factors in the way the interaction is conducted,
should be viewed as largely an apprenticeship, best learned in the circumstances in which it occurs and external factors
the shadow of someone who does it well. It is a source of which act to promote or inhibit it.
regret, if not alarm, to educators that service pressures and
those who now structure training programmes seem to con- Starting points : empathy and engagement
spire in unison to squeeze the element of apprenticeship
learning from specialist training but trainees themselves should It has been shown that patients gauge a ‘good’ doctor on a
remain aware of its importance and seek it out wherever simple measure: ‘Is this someone I can trust?’ Obtaining that
possible. trust is the doctor’s most important goal, and while no-one will
be able to achieve this with every patient, one will end up
The present chapter cannot ‘teach’ interview skills nor diminished therapeutically if one cannot go some way to
substitute for specific reading on psychopathology. Its aim is achieving it with most.
modest – to encourage an awareness of firstly, the process
elements of the psychiatric interview and secondly, the impor- The key to developing trust is the demonstration of
tance of structure in organising clinical material. Its orientation empathy. Although this word has entered the vernacular, the
is pragmatic and does not emanate from any specific theoreti- concept itself is frequently misunderstood. Empathy refers to
cal framework. Those interested in theoretical aspects are the ability to place oneself in the emotional perspective of
referred to the suggestions for further reading. Its target is another – while maintaining one’s own emotional perspective.
predominantly trainees working in general adult services. This latter point is crucial. Joining the patient in their emo-
Specialist practice will require new skills, though these will tional environment – full stop – is identification, the last thing
rest firmly on the principles relevant to general psychiatric a distressed patient requires of their doctor!
practice.
Empathy is not so much a single event but a cycle of inter-
Process action – one in which the doctor, via phrase, gesture, action,
etc., plays the key initiating role. To maintain the cycle, the
There is no technology yet devised to replace the fundamental patient requires to register this initiating element for what it
medical skill of interpersonal communication. For psychiatry, is and respond accordingly. This in turn must be picked up
it is not just a core skill – it is the skill, competence in which and responded to appropriately by the doctor – and so on.
forms the basis of any claim to professional expertise. Our Empathy is therefore an active process which should be ongo-
ing throughout any interview. Work in the time-restricted
settings of primary care has shown that the empathic cycle

ã 2010, Elsevier Ltd.
DOI: 10.1016/B978-0-7020-3137-3.00010-3

Companion to Psychiatric Studies

can be initiated simply, for example by expressions of personal Box 10.2
concern or interest (e.g. about the patient’s family), or raising
matters of common interest (the weather!) – i.e. beginning with Some barriers to engagement
things unrelated to the purpose of the visit. Without an aware-
ness of the central role and the dynamic of empathy, any inter- Patient
view may fail to start or is soon likely to grind to a halt.
Personality insufficiently sophisticated, naturally diffident
Empathy may be seen as one of the principle process mechan- Untrusting disposition, e.g. ‘negative’ past experience, rejection
isms whereby one achieves engagement of the patient in the
interview and its purposes, though it is not the only one. Engage- of authority
ment is clearly helped by being ‘engaging’ – by projecting a warm, Insufficiently ‘verbal’/articulate including first language not
friendly and interested demeanour from the start, but there are
risks in misjudging the means of doing this (see below). Some English
‘rules of engagement’ are outlined in Box 10.1. Resentment, e.g. ‘negative’ perceptions of psychiatry
Circumstances, e.g. ‘forced’ interview (e.g. prison)
Barriers to engagement emanate not only from the patient Mental state, e.g. fear/perplexity/suspicion/ preoccupation/
but also from factors inherent to the doctor or the circum-
stances of the interview (Box 10.2). Those relating to why retardation/cognitive impairment/intoxication
the consultation is taking place are obvious. However, other
factors to do with the patient’s personality, their lack of ease Doctor
in a verbal medium, social or cultural alienation, past experi-
ences of similar or other circumstances or frank resentment Personality aloof/socially awkward/diffident
at the whole exercise, are easily overlooked. Barriers presented Disposition condescending/paternalistic/patronising/‘rushed’
by doctors may also relate to personality characteristics. Those Verbal ‘pitch’ too sophisticated/ unsophisticated, too much
in whom interest alone is the attraction to psychiatry are not
immune from social awkwardness, lack of verbal fluency, ‘jargon’, says too much/too little
superficiality, etc. Negative influences may also emerge from Non-verbal communication absent/inappropriate, open to
one’s manner – be it aloof, supercilious, patronising, off-hand.
One of the biggest problems for trainees in busy clinical con- misinterpretation
texts is to avoid the impression of being rushed, with the Questions – weak techniques (process/structure)
patient just one more ‘task’ for the day. It can also be difficult
for trainees to aim the ‘pitch’ of an exchange correctly, adopt-
ing a flexible presentational style appropriate to patients from
16 to 65 of all social and educational backgrounds. Undoubt-
edly the most important doctor-related barriers emerge from
the style, structure and delivery of questions themselves.

Box 10.1

‘Rules of engagement’: some characteristics of a ‘good’ interviewer

Disposition (cf. the much quoted words of Truax: ‘non-possessive particular expertise. Clinically it is helpful to project that expertise
warmth’) – the natural qualities of character, or personality, one to patients – that one can be effective in addressing their
projects. There are two elements to this: concerns. However, expertise is as much about having the
confidence to accept alternative opinion as it is about standing
—an open and friendly approach – which is not the same as trying firm in splendid isolation. The ‘good’ psychiatrist will take on board
to be seen as the patient’s ‘friend’. Friendship is a personal, not only alternative information, but alternative explanation, in
intimate and possessive relationship, open to exploitation, and constructing a picture that can be meaningfully viewed from many
may represent an area of difficulty that brought the patient to angles.
psychiatric care in the first place.
A non-judgemental attitude. Many patients feel they have been
—respect for the patient as an individual, their beliefs and judged and found wanting. This perception may extend beyond
aspirations, often described under the general heading of the obvious moral and ethical issues which may bring them into
‘autonomy’. This raises complex issues, especially when one’s conflict with social norms. Those with psychotic symptomatology
duty of care sets one against the patient’s expressed wishes, may also feel that the validity of their experience is open to
but it is necessary for the effective psychiatrist to convey a judgement and rejection. Maintaining a non-judgemental attitude
respect for the patient as an individual and for the views they is not the same as being complicit. Boundaries exist in all human
express, despite arriving at different conclusions. relationships and it is not helpful to create in the patient’s mind the
idea that this, for them, might not be so. It is, however, important
Awareness of the process elements of the interview. In busy that one is able to convey a sensitivity to their dilemma and
clinical situations, it is easy to concentrate on the content of the provide alternative models against which their situation may be
interview – the factual material to be acquired – to the exclusion of appraised.
the methods one is employing. Psychiatric interviewing must
always have a formality that separates it from social exchanges Sensitivity to the subtleties of human interaction, both verbal and
and those who practice it must have a constant awareness of not non-verbal. This is one aspect of psychiatric expertise that cannot
just ‘what’ they are attempting, but ‘how’ they are attempting it. be taught. You either have it – or you haven’t! For those not gifted
No question, comment or gesture should be without purpose and in this regard, the best that can be recommended is to maintain
none should slip through unnoticed or without its place in the awareness of the deficit – and read copiously from quality novels,
process clearly understood. for more will be found there about what they are missing than in
the pages of any textbook.
A demeanour of confidence without dogmatism. Psychiatrists
enjoy respect based on the understanding that they possess

200

Clinical assessment: interviewing and examination CHAPTER 10

Preliminaries patients to be referred by colleagues in other disciplines
without it being made clear that referral is to a psychiatrist.
Before proceeding to aspects of questioning, it is worth raising The sudden realisation of your occupation may stimulate
some preliminary considerations. powerful emotions, including fear and anger, especially if you
appear complicit in a deception. Secrets do not make for
The room sound clinical relationships and while there may be exceptions,
it is generally better to make clear from the start in such situa-
This is not something that doctors, especially trainees, may tions not only who, but what, you are. One can, should there
have a lot of say in, but arranging – or being seen to arrange – be concerns or objections, then work through them. Following
the most conducive environment can be taken as an empathic from that, it is also reasonable to inform the patient what it is
gesture, so a quick rearrangement of the seating as the patient you are going to do, which might include an indication of the
enters, whether necessary or not, is to be commended. It is time the interview is likely to take.
unhelpful to interview across large obstructing desks, which
in addition to creating physical barriers, creates psychological A hand-shake is a widespread sign of introduction and
ones by reinforcing hierarchies. Seats should be high, comfort- welcome but there are limits. For example, a male doctor
able and of comparable height for doctor and patient. The low, offering a hand to a traditional Muslim lady may create an
soft chairs so beloved of those who furnish out-patient depart- uncomfortable dilemma for the patient. The same may pertain
ments should be avoided. They impose an uncomfortable to those who exude an obvious suspicion and anxiety. Related
posture and create a barrier to elegant exit for the elderly, to this is touch. A pat on the back, shoulder or forearm can be a
the overweight and those with even mild parkinsonism. In potent and genuine signal of support or approbation, but
arranging seating, it is best to avoid a direct face-to-face align- because of its potency, touch can be all too easily misinter-
ment, which can feel inquisitorial. Seating should be slightly preted as invasive or worse, sexual in its connotations. In gen-
off-centre, allowing the patient the opportunity to avoid the eral, and especially with opposite-sexed patients, it should be
examiner, which may facilitate their comfort, without them avoided.
having to turn away, which they might construe as rude. Seats
should be sufficiently close to foster ease in the sharing of When on ‘your’ territory – hospital or clinic – the patient
confidences, but not so close that feet might touch. occupies in effect the status of guest. It should go without
saying that on entering the room they should be invited to take
Interview rooms should in general be spartan to avoid the a seat and not left to make awkward social assumptions. Some
attractive ornament becoming a missile. Telephone cables statement that they are welcome would also be appropriate:
should be tightly fixed and as short as possible. Interviews e.g. ‘Good afternoon, Mrs X. My name is doctor Z. Do have
should never be conducted with the patient seated between a seat - and make yourself comfortable.’
the doctor and the door. Alarms are a further consideration
nowadays. Such simple safety points will seem trivial – until Mode of address is a further preliminary open to misinter-
the day they aren’t! pretation. In increasingly informal times, the use of first names
is taken as a sign of informality and friendliness, especially by
While privacy is a basic expectation of anyone attending a the young but equally may be seen as a reflection of superfici-
psychiatric interview, psychiatrists must increasingly consider ality or lack of respect. Doctors should make no assumptions
the value of a chaperone by appraising the risks of aggression in this regard and patients should always be addressed as
and accusations of exploitation. ‘Mr’, ‘Mrs’ or ‘Miss’ until given leave by the patient to use a
Christian name. It is usually best not to seek permission to
Introductions use first names immediately, but to wait until the formal
exploratory part of the interview – i.e. after the ‘Presenting
First impressions count – especially to the anxious, the be- Complaint’. There is an argument that those considerably in
wildered and those with negative preconceptions. Thus, it is advance of oneself in years deserve the respect of a formal
essential that the ‘obvious’ – so easy to overlook – become mode of address throughout.
integral and formal parts of every introduction.
An additional common ‘negative’ on first contact is the
The attire appropriate to a medical practitioner is a fraught impression of lack of preparation, which in terms of patient
topic and one on which there is unlikely to be a ready consen- expectations, is disrespectful and rude. It must be frustrating
sus. It furthermore may be one that is open to variability, to have to sit in silence while an ill-prepared and discourteous
depending on the doctor’s ‘target’ population – those working doctor ignores one to read a referral letter or past case note
with adolescents, for example, may find casual attire more material. Any such preparation, no matter how brief, should
conducive to engagement than a three piece suit. Suffice to be done in private, before first introductions.
say that while ‘casual’ may have its place, ‘scruffy’ never does!
To write or not to write
Informing the patient of who you are is common courtesy.
No matter a doctor’s eminence within their institution or pro- This is a question on which there can be little dogmatism. The
fession, he/she is likely to be a stranger to the patient at first rights and wrongs of taking a contemporaneous record will
meeting. Furthermore, patients may believe they are coming depend on the attributes of the interviewer (whether they
to see a specific individual, especially the consultant to whose can maintain engagement overall while intermittently disenga-
clinic they have been referred. Sometimes patients may also ging to write), the interviewee (whether they feel diminished
have no awareness of what you are. It is not uncommon for by, or suspicious of, note-keeping) and the nature of the

201

Companion to Psychiatric Studies

interview (see below). There are however strong arguments to • Semistructured interviews comprise the usual format in
justify the recording of contemporaneous notes, especially of routine practice. Such a format need not imply adherence
initial assessment/diagnostic interviews. to any particular school or specified set of questions, but
merely suggests that the interviewer accepts a proactive
• No-one can remember all information conveyed over role in establishing a core body of factual information and
a 45–50 minute exchange, not only factual material but, the structure within which such information can be
as importantly, how much structure had to be imposed, sensitively acquired and evaluated.
how much dissemblance was detectable, at what points
particular emotions were shown, etc. Psychiatric interviews can also be considered in terms of their
purpose. Initial interviews are usually diagnostic assessments.
• There is no record so valuable as that liberally peppered They are likely to be the most structured, demanding as they
with verbatim quotations. This can be helpful in later do, the acquisition of basic levels of information. Diagnostic
consideration of the issues; it can provide invaluable interviews are conventionally 45–50 minutes in duration. This
information for successors who become involved in the is not just tradition but reflects the difficulty that anyone –
case; and, crucially, it provides unassailable evidence of doctor or patient – has in maintaining active attention for
what was actually said should your opinions come under longer!
subsequent challenge, be these legal or from patients who
later deny their initial symptomatology. Initial assessments however are not just about diagnosis.
At a clinical level, diagnosis is the key to treatment planning
• It takes great skill to avoid an unbroken 50-minute and prognostication. In psychiatry, an initial interview may
interview evolving into an interrogation or degenerating into not provide sufficient information to formulate a diagnosis
chit-chat and the breaks offered by a pause to write can with adequate probability of accuracy. Thus, in practice, diag-
provide useful oases of peace and reflection. nostic assessments are geared to provide a differential set of
possibilities, ranked on a hierarchy of probability. Unlike in
Engagement is generally facilitated by an initial period of general medicine, psychiatry should not, as a rule, distinguish
attention to the patient’s expressed concerns without the dis- between ‘the’ diagnosis and the group of ‘differentials’ which
engagement inherent to writing. Starting to write is something follow on. For us, first assessments should provide only
else best left till after one has clarified the ‘Presenting Com- ‘differentials’, albeit ranked by probability.
plaint’. A skilled interviewer will also cultivate the art of
recording while maintaining the flow of the interview with A modification of this type of interview is the problem-
simple social exchanges, facilitating comments or run-ins to orientated interview, most pertinent to emergency and risk
the next question. assessment situations, in which evaluation of a specific pre-
senting problem (usually a behaviour or set of behaviours) is
Types of interview of greater importance than a particular diagnosis.

There are various ways of looking at psychiatric interviews but Follow-up interviews are geared to monitoring, which
two readily recognisable classifications relate to the form of the includes not only symptom remission and quality of life/
interview and its purpose. psychosocial functioning, but issues relating to the treatment
plan, including tolerability (side-effects) and patient satisfac-
Three main formal types can be identified: tion. In addition, these interviews should incorporate a con-
scious element of psychological intervention, whether formal
• Structured interviews have their origins in research and their or informal, covering supportive issues, practical problem
form is determined by the requirements of the recording solving, adherence and so on. In general adult practice, such
instrument, which may relate to just the areas that require interviews are often conducted without structure in an
to be covered or may in addition specify the precise ad hoc way and can readily develop into chit-chat. It is impor-
questions. For example, the Present State Examination tant to keep in mind the purpose of the interview and to have
(Wing et al 1974), which comprises a mixed set of 140 at one’s disposal a structure within which to address the key
symptom and behavioural items, bases its reliability largely issues.
on the fixed questioning presented in the manual and
precise adherence to predefined anchor points. Such It is not uncommon, especially in specialist practice, that
methods require training to establish, and retraining to interviews must be geared towards the needs of third parties,
maintain, reliability and have not been widely applied in the most frequent example of which is legal reports. In this
routine practice but knowledge of them can provide a situation, diagnostic issues may be of less import, having
comprehensive structure for routine use. They also offer a already been established, and interviewing may have to be
valuable insight into the levels of evidence necessary before modified to address the specific questions raised by those
the presence of psychopathology can be accepted with requesting the report.
reliability.
Conduct of the interview
• Unstructured interviews are traditionally associated with
dynamically orientated practice, in which the way This is the most important component of professional inter-
information emerges and its symbolic significance (i.e. the viewing and concerns the verbal and non-verbal interactions
narrative) is considered as important as factual material. of both interviewer and interviewee. The ‘patient’ part
The prolonged or open-ended contracts that go with such will be the focus of the subsequent section on mental state
styles of interviewing are incompatible with routine practice.

202

Clinical assessment: interviewing and examination CHAPTER 10

examination. Here, we will concentrate on the interviewer’s While hopelessness is usually associated with depression, the
contribution. opposite is not necessarily the case. A compound question
(a) Formal questions come in two main varieties. Open questions can often be identified by the presence of a conjunction,
require some opinion or judgement from the respondent and though an ill-chosen adjective can produce a similar effect, e.g.
allow for answers which are open to variability and debate, e.g.
‘Are you troubled by voices?’
‘How have you been sleeping recently?’
The patient may have voices but not be in the least ‘troubled’
‘How have you been feeling in your spirits recently?’ by them – or may feel troubled in themselves without halluci-
nations as a feature.
Both these would leave it open to the patient to provide a (d) Avoid leading questions. This is perhaps the commonest
qualitative account of the relevant area and lend themselves mistake trainees make, particularly when losing control of an
to further probing with minimal intervention. Closed questions, interview or simply getting stuck. At best, leading questions
on the other hand, are usually geared towards some factual place interviewees in a corner out of which they may not feel
response about which there is little room for debate, e.g. able to manoeuvre tactfully. At worst, they present a challenge
to disagree, an offer most will feel unable to accept. Thus, they
‘Have you been sleeping poorly recently?’ – ‘Yes’/’No’ encourage the responses the patient thinks the doctor wants
to hear. Leading questions come in two forms, positive and
‘Have you been feeling depressed recently?’ – ‘Yes’/’No’ negative, of which the latter are the more powerful, e.g.

In general, ‘open’ questions open up an interview, allowing a ‘You were depressed at that time (?)’
freer exchange of information, while ‘closed’ questions trim
things down to basic, largely factual, exchanges. Over-reliance ‘You were not happy at that time (?)’
on either can result in major control problems – under-control
with too much of the former because of lack of constraint Trickiest of all, is to compound the error of a leading question
and inadequate direction; over-control with the latter from a by finishing it off with ‘were you?’, e.g.
failure to facilitate free expression.
‘You were depressed at that time, were you?’
All interviews should start with an open question, e.g.
Leading questions are frequently prefaced by that sinister little
‘What have been the difficulties that have brought you to see me?’ word, ‘So’, e.g.

Thereafter, they should comprise a fluid mix of ‘open’ and ‘So, you were depressed at that time, were you?’
‘closed’ formats, with the balance deliberately chosen to facili-
tate the often competing requirements of; (1) acquiring suffi- This is a classic way of presenting the patient with a ‘chal-
cient information; and (2) allowing the patient free expression lenge’. Merely raising one’s tone at the end of such sentences
of concerns. As a rule, ‘open’ questions should predominate in in an attempt to transform statements of opinion into ‘ques-
the earlier part of the interview, with ‘closed’ questions coming tions’ does not instil them with merit.
in later to fill in factual blanks. The important point is
that whereas in social contexts the interaction itself determines While as a general style, leading questions remain examples
the balance, in formal settings the interviewer must organise the of unsophisticated interviewing, they can have a place in the
balance in order to maintain a reign on the interview. Thus, an psychiatric interview as part of a process of clarification where
interviewer must always be aware of where the interview clarity remains elusive (see below).
is ‘at’, what type of question is being utilised and whether the (e) Utilise recapitulation and summarising statements. Direct
best way to take things forward involves ‘open’ or ‘closed’ statements used for recapitulation and summarising are an
formats. important part of the structure of psychiatric interviews.
(b) Questions should be kept short. Verbose language has an They allow for clarification while providing a natural break
enervating impact on the listener and lengthy questions bewil- for ‘air’. They also allow the examiner to facilitate
der. If possible, questions should be delivered in a single sen- the interview by offering approbation for the efforts so far
tence. The only alternative is a succinct introductory (or (see below). And, they can very effectively begin with
‘lead’) statement to preface a question that requires to be ‘So’, e.g.
placed in some context, e.g.
‘So, am I correct in concluding. . . . . . . . . ’
‘Depression is a word with many meanings. What does it mean to you?’ ‘So, is it the case that. . . . . . . . . .’

c) Questions should be simple (as opposed to ‘compound’) – the (f) Be aware of ‘pitch’. This refers to the intellectual level at
interviewer should only be expecting the patient to address a which you approach the patient (not one’s voice register!).
single point with each answer, and that point should be clear. Even with intelligent individuals, vocabulary should be kept
With compound questions it is not possible to be certain which simple and devoid of technical jargon. This can be difficult
part of the question the answer refers to, leaving it open for the in a specialty such as psychiatry, which shares an extensive
interviewer to jump to false conclusions. Compound questions technical language with lay usage.
can come in remarkably straightforward guises, e.g. (g) Avoid direct questions, which are the key element of inter-
rogation. It is easy when conducting an interview to latch on to
‘Do you feel sad and hopeless about the future?’

203

Companion to Psychiatric Studies

a particular piece of information that strikes one as important ‘I am sorry to hear that.’
and at the first opportunity, dive in with a frontal assault! One
can still come across trainees enquiring along the lines of: ‘Do ‘You’re doing very well.’
you ever feel like killing yourself?’ Direct questioning should
be resisted, largely because it will be interpreted as traumatis- ‘That must have been very difficult for you.’
ing or, quite simply, rude. Sensitive areas may indeed have to
be explored, but ideally with a run in, in which the approach One eminent neurologist, after exploring the mysteries of the
can be ‘softened’, e.g. bizarre symptomatology that was his field, would conclude
each section authoritatively with: ‘Thank you, Mr X. That’s
‘May I ask you,. . . ’ clear to me’ – whether it was or not!

Even the addition of words that generalise potentially embar- It should also be remembered that statements of profes-
rassing topics can improve acceptability, e.g. sional opinion can act as powerful facilitators. The classic
way of putting this is: ‘It seems to me . . . ’. Although super-
‘Have you noticed any change in your interest in matters of sex?’ ficially presented as a statement of opinion – ‘It seems to me
you are very angry about this’ – such comments can be similar
(h) Be prudent in the use of the imperative. The imperative is in their effects to open questions, urging a reaction or opinion,
the case of command, something that is hard to avoid even and can act as powerful facilitators.
with the most empathic delivery. For most, the distinc- (k) Remember the power of intonation. The natural rhythm,
tion between ‘Tell me what happened?’ and ‘Tell me what variability and inflection of the human voice is one of the most
happened!’ is easily blurred. One should similarly be wary of important elements in our communications, the subtleties of
the word ‘why’, which in splendid isolation, at best infers which can be powerful tools in the psychiatric interview. The
non-understanding, at worst censure or disapproval, e.g. ‘way’ a sentence is delivered can be as important as the words
used. Written text can give little instruction on the ‘goods’ and
‘Why did you do that?’ ‘bads’ in this regard and all that can be offered is further gen-
eral encouragement to give as much attention to the ‘how’ of
Both these styles suggest justification or accountability, neither verbal exchange as to the ‘what’.
of which should be expected of those participating in a psy- (l) Be sensitive to ‘flow’. Human communication has a natural
chiatric interview. A more empathic approach is to seek leave tempo, periods of intense exchange giving way to less emotive
to explore the relevant issue, e.g. interactions when the ‘static’ is notably less evident; periods
when interaction flows readily interspersed with spells when
‘Would you tell me what happened?’ communication is less spontaneous. This natural ‘ebb and flow’
is to be encouraged, but always within a framework of control.
‘Can you tell me why you did that?’ Trainees often have difficulty with this aspect of the interview
reflecting their own insecurities in their professional role and
(i) Ban ‘just’ – as in, ‘I would just like to ask you. . . . . . .’ It lack of confidence in their process skills. As a result they can
often seems that doctors view this word as a technical medical end up being either overwhelmed by a verbal deluge or
term, as it crops up so often: ‘If I could just have a listen to excluded by a barrier of silence. In fostering the ‘flow’, it is
your heart?’; or ‘Just a little prick in the skin’, etc., etc. No important to be ready to intervene in the face of verbal
doubt, those who rely on it do so in the belief it is a ‘softening’ onslaught; to recognise and respect natural pauses; and to pro-
word, a way of making a request less like a command, or vide lubrication to points when the process seizes up.
making something inherently unpleasant or embarrassing, less
so. However, its effect on the receiver is invariably negative, Rambling verbosity is often a consequence of a timid,
conveying hesitancy or a general interpersonal awkwardness – unconfident interviewer, unable to take hold of natural breaks
or worst of all, an impression that the context in which is it (however short) to intervene effectively, and/or an inter-
used is actually not that important. In fact for most doctors viewee who has little understanding of what is expected. Such
it becomes a ‘comfort’ word, whose function is to relieve their interviews are best rescued by resolve – clear interventions on
anxiety about communicating effectively. the part of the examiner, clear statements of the purpose of
(j) Use prompts and facilitators judiciously. These are techni- the questioning and short, simple closed questions. Interviews
ques, frequent in social interactions, which make it easier, by which fail to get off the launch-pad are usually influenced by
providing encouragement or approval, for the patient to con- barriers to engagement on the part of the patient, as noted
tinue. Prompts are usually preverbal – Mmmm?’s or Uh-huh’s, above, while those which start, however tenuously, but come
delivered with an empathic non-verbal gesture or expression, to an immovable halt, are usually inhibited by barriers from
such as a raising of the eyebrows or a tilt or nod of the head. the interviewer. ‘Freezing’, the crushing silence that ensues
They convey curiosity, interest, understanding and act as a from either party unable to find a way forward, is particularly
lubricant to promote the patient’s participation. Facilitators destructive when it afflicts the interviewer. These situations
are words, or more usually short phrases or sentences, which arise especially when interviewers are emotionally over-
make it easier for the patient to respond, by conveying involve- whelmed or embarrassed, usually with overt or powerful displays
ment or providing reinforcement of their efforts so far. They of emotion, or when the interview takes on an unexpectedly
are demonstrations of concern and appreciation and are a key disinhibited or inappropriately direct or aggressive quality.
part of the examiner’s ‘returns’ in maintaining the empathic
cycle, e.g. A ‘freeze’ is different from a planned pause, which eases inter-
personal tension and reinforces engagement. There are few more
awkward atmospheres than those created by psychiatrists who

204

Clinical assessment: interviewing and examination CHAPTER 10

‘freeze’ mid-interview. ‘Freezing’ is the ultimate demonstration adopt the postures, and possibly even the gestures, of their
of loss of control and can readily turn one’s best efforts at a examiner. When the doctor subconsciously comes to ‘mirror’
therapeutic relationship into something distinctly counter- the patient, however, a serious control problem is evident!
therapeutic. Such situations require a liberal use of empathic
statements and gestures, carefully planted prompts and facilita- Structure
tors and an open questioning style.
Any separation of the process and structural elements of the
Sometimes, patient ‘freezes’ can be helped by providing a psychiatric interview is somewhat artificial. However, in this
statement to which they can respond that is ‘one degree’ section we shall refer briefly to the conventional way in which
removed from painful emotional areas. For example, rather the material from a psychiatric interview is organised for
than asking directly about fearfulness or anxiety per se, one presentational purposes (emphasising predominantly initial
might comment on observed behaviour, e.g. interviews). Psychiatric history-taking is a complicated and
extensive process, partly due to the need to obtain consider-
‘You seem somewhat anxious at the moment.’ able amounts of information in order to place the patient’s
symptomatic material in its personal and social (including
Agreement on the presence of psychopathology may provide its hereditary) context. This makes for schemes with many
the necessary bridge to allow its further exploration. headings and subheadings. A number of presentational
(m) Listen to what the patient says! How obvious this may seem, schemes are in existence, and while the one illustrated here
but the ability to convey a listening attitude is one of the most (and summarised in Box 10.3) is only one of many, it contains
empathic, and engaging, of gestures. In the anxiety of trying to the conventional core elements. The paperwork comprising
do everything else, however, it is not uncommon to find trainees the patient’s record is traditionally referred to as the ‘Case
paying greater attention to the next question than to the answer Notes’ – emphasising that it does not have to be written in per-
they are getting to the one they have just asked. Most damaging fect English. It is more important that all relevant information
of all is repeating a question already asked. is included than that the prose is perfect.
(n) While focusing on the verbal, do not ignore the non-verbal
components of communication which ethologists tell us com- Preliminary statement
prise the major contribution to human communication. Be this
as it may, evaluation of this will comprise an important part of In writing up a case, and in formal presentations (e.g. examina-
your assessment of the patient – and a sizable component of tions), it is useful to begin with a general introductory state-
their assessment of you! This element is crucial in setting the ment which ‘sets the scene’:
tone within which formal verbal elements are evaluated and
is key in initiating and maintaining the empathic cycle that is Preliminary (including identifying) statement.
such an important part of the ‘work’ of a psychiatric interview.
It is, however, a vast area whose insights and hypotheses are Mrs Margaret Smith, 53 years of age.
too numerous to be presented here. Routine referral by General Practitioner (see letter).

However, be wary of: Presenting complaint

• Facial expressions. These can betray disinterest, This should be the starting point for every psychiatric inter-
disengagement, amusement, bewilderment, scepticism and view and most commonly is taken from the patient’s own
frank incredulity even in the presence of soothing and words in response to a general enquiry. This should, where
contrary words. While a controlled smile, judiciously used, possible, be an ‘open’ question which is neither too pointed –
can be a powerful signal of warmth, laughter – one of the ‘Why are you here?’ – nor too vague, allowing the patient to
most potent behaviours we share in social bonding – is as a free associate.
rule too potent for the psychiatric interview, and hence too
open to misinterpretation. Examples might be:

• Interactive posture and gesture. Engaging communication is ‘Can you tell me the difficulties you’ve been experiencing recently?’
fertilised by body posture that is open and gesture that is ‘What are the major problems that have been troubling you recently?’
inviting. To lean slightly towards a speaker suggests
confidentiality and interest; to slouch in one’s chair, the A common introductory medical question – ‘What brought
opposite. For speakers, continuous eye contact is not you here today?’ – should be avoided, in view of the common
essential – for listeners, it is (except when writing). answer – ‘An ambulance.’!
Controlled use of hand movements can be a powerful sign
of emphasis or encouragement; uncontrolled, they can Different standardised psychiatric assessments adopt dif-
become comedic, while ‘professional bradykinesia’ conveys ferent criteria for ‘recently’ – i.e. what ‘present’ means in the
a flatness interpretable as indifference. ‘present mental state’. For the Positive and Negative Syndromes
Scale (PANSS) this is the past 7 days, while for Research
• Incidental, stereotyped movements. To jiggle one’s legs, or
fiddle with pens, etc. is to convey boredom, impatience or
social anxiety. To sit with arms crossed over one’s chest
creates barriers, while resting one’s chin in a heavy hand
wreaks of fatigue. ‘Mirror’ attitudes not infrequently develop
in the course of therapeutic dyads, where patients come to

205

Companion to Psychiatric Studies

Box 10.3

Summary of psychiatric assessment – history and mental state examination

1. Reason for referral/contact – co-habitation
– marriage
Presenting complaint – separation/divorce
— quality of adult relations
Patient and/or third party sources — children
– age(s)
2. Symptomatic context – difficulties
– relationship
History of present episode/illness Interests/hobbies/recreations
Past psychiatric history*{ (including treatments: satisfaction/
Establishing boundaries – normal syndromal dissatisfaction)
Medical history*{ (including known allergies)
3. Social context Substance use*
Forensic history*
Family and personal Subjective appraisal of personality
Family
Parents: 4. Formal examination

— alive (a) Mental state
– ages/occupations
– physical health Appearance
Behaviour (/manner)
— dead Speech
– age at death Thought:
– cause of death Form
– occupations Content (including ‘possession’ – e.g. ruminations)
Perception:
— separated/divorced Distortions
Siblings: Deceptions
(Other abnormal mental contents)
— number (incl. patient’s place) Affect/mood{:
– alive/dead Subjective
– ages Objective
– marital status Risk (e.g. suicide)
– occupation(s) Cognition:
Orientation
Relationships: past/present Concentration/attention
Medical histories Memory:
Psychiatric history – extended family (incl. suicide/alcohol)
— short-term
Personal — long-term (antero/retrograde)
Date/place of birth Executive function
Known obstetric problems (Abstract thinking)
Developmental milestones (if known) (Insight)
Early behavioural problems
Schooling: (b) Neurological assessment

— elementary Observation – at rest: sitting/lying
– special educational input Abnormal postures : e.g. dystonia
– friendships Restlessness: e.g. akathisia
Additional movements:
— secondary
– special educational input — tremors
– friendships — dyskinesias – choreoathetoid (sometimes mistaken for
– age of leaving
– academic attainment fidgetiness)
— tics
— higher (if relevant) – — jerks
– institution/course Lack (poverty) of movement: e.g. immobile facies/reduced blink
– attainments
rate
Employment record
– attainments/failures
– (e.g. promotion/demotion/dismissals)

Psychosexual development
— puberty
— sexual orientation
— relationships

Continued

206

Clinical assessment: interviewing and examination CHAPTER 10

Box 10.3—cont’d

Examination: gross cognitive assessment (see above) — stooped posture, e.g. parkinsonism
Observation — reduced/absent arm swing, e.g. parkinsonism
Walking: walk the patient 10 m along a corridor, watch them turn — instability on turning
Examination: sit patient on edge of couch
and return Eye movements
Gait: e.g. ‘waddling’ in proximal muscle weakness Arms above head
Arms outstretched with eyes open and closed
— ‘marche a petit pas’ (reduced length/height of step) in ‘Piano-playing’/other fine repetitive movements with hands
parkinsonism Examination: lie patient semirecumbent
Tendon reflexes (brisk or normal)
— ataxia (broad-based, unsteady): e.g. cerebellar/Huntington’s Plantar responses
— stiff-legged ‘scissors’ gait (spastic paraparesis)
— ‘functional’ gait disorder: e.g. dragging apparently useless limb

behind one/exaggerated reliance on people or objects for
support

*May be presented as separate headings.

{Often presented immediately after 2. ‘History or presenting illness’.
{Often placed after ‘Speech’.
()Optional.

Diagnostic Criteria (RDC) it is the past 2 weeks. This can seem History of present illness
confusing but partly reflects the different purposes for which
different instruments were devised. However, no system This covers firstly, the symptomatic context (how the lead
restricts the ‘present mental state’ to ‘right now’. We would symptomatology identified in the ‘presenting complaint’
recommend, in line with the Present State Examination blends with normality at one end or with the range of possible
(PSE), that for routine clinical use the previous 4 weeks additional phenomena to form an identifiable syndrome at the
provides a representative period for consideration. other); and secondly, the temporal context. In establishing the
symptomatic context, a patient complaining of anxiety for
Acutely disturbed patients may fail to acknowledge such example, would be asked about the nature of their symptom-
enquiries with a simple statement – or any problem statement – atology, whether it is associated with panic attacks, autonomic
so the ‘presenting complaint’ can be taken from third party and other concomitant symptoms, whether it is free-floating
sources and can thus comprise either subjective or objective or confined to specific situations, the extent of any fluctua-
information. tions and how severe it gets, whether they have always been
an anxious person and how long the features they complain
In secondary care settings one will usually have a GP’s of have been present.
letter, or some other referral information which can provide
a smooth introduction, e.g. It is in establishing the symptomatic context that all those
process elements noted above come into play – plus one not
‘Your GP has given me an outline of your difficulties but it would be mentioned: curiosity. Curiosity is one of the surest qualities
helpful to me if you could summarise these in your own words.’ to guide a comprehensive psychiatric assessment – curiosity
not only about the presence of mental state disorder, but
The ‘Presenting Complaint’ should be brief – it can literally about its unique and intricate detail, its boundaries, and its
be a single word, or short phrase. Only in exceptional cir- impact. Demonstration of curiosity about the patient’s experi-
cumstances, which usually relate to disturbed patients brought ences can also provide an invaluable fillip to the empathic
to medical attention by others, should it extend to more than a cycle.
sentence.
Temporal context bears on one of the most valuable pieces
It can, where possible, be helpful to spend a few moments of information in mental state assessment – a history of change
longer in introductory remarks, by explaining to the patient (i.e. the extent to which the patient’s symptomatology repre-
that to start with you would like to hear how they would sents a change from the norms and in what way it has evolved
summarise their difficulties – ‘and by that I mean your main over time). While patients may chart their path to psychiatric
symptoms’, then explaining that you will go over the context care over many years, it is the identification of change in
in which these have developed subsequently. some aspect of mental state and/or psychosocial functioning,
without spontaneous resolution, that is the strongest pointer
Most of the rest of the interview is about placing to illness development.
the features of the ‘Presenting Complaint’ in various
contexts. Whatever the initial complaint, it is always important to
specifically explore mood, firstly, because of the prevalence
Presenting complaint of mood disturbance and its significance in differential diag-
nosis (‘depression’ features as a possible primary diagnosis in
‘I feel so miserable all the time.’

207

Companion to Psychiatric Studies

patients presenting with everything from anxiety disorder to presenting symptomatology tends to open up a range of possibi-
psychosis), and secondly, because of the key role it plays in lities, which are then explored by further questioning, exami-
risk assessment. It is also a useful domain to resort to at times nation, special investigations, etc. In this process, refutation is
when interviews are closing down or becoming difficult, as few as important as confirmation and the solution (i.e. the diagno-
people have reservations about discussing how they ‘feel’. sis) comes from the balanced play of the two. It has been
shown that psychiatrists have a tendency to take the present-
While some presentational systems recommend that so- ing complaint as a feature of the diagnosis, with subsequent
called ‘biological’ or ‘vegetative’ features of affective disor- questioning closing down the options to confirm an initial
ders – sleep, appetite, weight, etc. – should be presented in impression. Thus, diagnosis can become largely a process
the mental state examination, we would recommend they are of construction built by confirmation, with refutation playing
considered under ‘history of present illness’, primarily because little part. Clearly, medicine tends to have wider diagnostic
they are subjective phenomena, appropriately considered with horizons than psychiatry but the danger of not ‘testing’
other subjective symptomatologies. The wide range of other assumptions as the examination proceeds is fulfilling one’s
accessory depressive symptoms, such as anergia, poor con- own prophesy – which requires no expertise whatsoever!
centration and memory, irritability, lack of interest, anhedonia
and ideas of hopelessness, self-depreciation, self-blame should In assimilating the patient’s history, it is important that
also be routinely explored and noted. By contrast, while neither deficient process elements in the interview nor initial
psychotic phenomena are naturally explored at this point, bias limit a full and objective exploration of all potential
it is patient descriptions that should be recorded here symptomatology.
with MSE reserved for recording findings using technical
terminology, e.g. History of presenting illness

‘history of presenting complaint: ‘talks of strangers in the street looking Lead symptom(s)
at him threateningly’ Feeling ‘depressed’ ¼ ‘sad all the time’, ‘miserable’, ‘awful’.

mental state examination: ‘describes delusions of reference’ Elucidation of lead symptom(s)

Risk assessment is a further area that should be specifically Never felt like this before – ‘I’ve had my moments, but this is different’ –
addressed in this section, and is now of such potential impor- ‘can’t seem to shake myself out of it’. Feels depressed all the time –
tance as to justify a separate heading. This includes not only ‘every day the same’ – but worse in the mornings – ‘that terrible feeling
assessment of suicide risk, but also of the possibility of harm of dread’ on waking. Eases as the day goes on but never feels like her
to others. Planned (as opposed to impulsive) acts of self-harm old self. No matter how much better she becomes in the evening,
tend to progress in significance along a continuum, and while ‘I can never carry that over to the next day’. Nothing can buck her
establishing any patient’s ‘place’ on the cascade does not allow up – ‘I just can’t see the joy in anything’. Doesn’t know when she last
for complacency, it does provide a measure of reassurance, or enjoyed herself – ‘last summer, probably’ (i.e. 1 year ago).
alarm, as the case may be. This ‘pathway to suicide’ is illu-
strated in Fig. 10.1. Associated symptoms

In forming diagnostic inferences, psychiatrists tend to oper- Gets no pleasure from anything, even her grandson: ‘He’s become
ate in a significantly different way to physicians. For the latter, such a burden – that’s terrible, isn’t it’. Feels tearful sometimes but
‘I can’t cry’ – ‘Maybe I’d feel better if I did’. Feels guilty that she’s

Depressed mood Hopelessness Fig. 10.1 Pathway to suicide
(severe and/or
Pointlessness ‘I can’t see any future’
consistent) of life
Thoughts ‘I’d be better off dead’
Passive rejection
Plans of life ‘I wish I were dead’

Active rejection ‘I could take an overdose’
of life transient
recurrent
Passive thoughts
of self-harm ‘I will take an overdose’
not acted on
Active thoughts acted on
of self-harm

Act

208

Clinical assessment: interviewing and examination CHAPTER 10

letting everyone down – ‘They all try so hard – but it’s me’. Feels she’s depression’. Frequently forgotten is the importance of enquir-
‘no good to anyone’ – ‘they (the family) would be better off without ing about suicide within the extended family, and also
me’. Wonders if this is all a punishment for taking on too much – ‘I tried alcoholism.
to be superwoman – but I’m not’.
Having established a factual base, one may then explore
Has no energy – everything’s ‘an effort’. Even getting dressed is ‘an superficially the patient’s perception of the family’s dynamics.
uphill struggle’. Hasn’t been out the house in weeks – ‘I get so anxious In the 1960s and 70s social theory and family dynamics
– it’s ridiculous’. No appetite – ‘The sight of food makes me feel sick’. assumed a major role in paradigms of psychiatric disorder
Has lost about a stone or more in weight in the past couple of months. (for the most part in a far from positive way) and in recent
Sleep ‘dreadful’ – feels ‘exhausted’ in the evening and gets off to sleep years the issue of childhood abuse has become prominent.
without difficulty. Wakens at 4.00 a.m. – ‘I could set my alarm by it’. Such areas, so dependent on inference and recollection in
Can’t get over again. ‘That’s the worst time – that terrible churning situations of charged emotion, must be explored not only sen-
in my stomach’. Getting about 5 hours sleep per night. Feels tired sitively, but with sophisticated, non-directive and ultimately
in the day but can’t cat-nap. Concentration poor – ‘I look at the non-collusive interview techniques which may be beyond the
television – but it goes right over my head’. Used to be an avid reader, average trainee. The routine out-patient clinic is not, as a rule,
but now ‘can’t be bothered’. Forgetful of everyday things which she the forum to explore other than descriptive accounts of family
finds embarrassing. dynamics perceived as distorted or factual statements of early
abusive events.
No other symptoms including misperceptions
Personal history
Risk assessment (suicide/homicide)
This should comprise a chronological account of the patient’s
Feels hopeless about the future – ‘I can’t see anything to look forward personal experience, starting with their date of birth and
to – just more of the same’.’ Frequently wishes she was ‘out of it’. Has birth details, and covering early development, schooling, rela-
thought of ending her life – ‘It would be a release for everyone’ – e.g. tionships, academic achievements, occupational history, adult
by taking an overdose or drowning herself, but says she would not relationships (psychosexual history, partner(s), children), and
do it – ‘I’m too big a coward’. ‘If only I could go to sleep and never present social circumstances.
wake up – I long for that’.
While separate headings are often given to ‘past medical
Evolution/associated factors history’, ‘forensic history’, ‘substance use’ and so on, these
are really subheadings of ‘personal history’, and it is reasonable
She relates the onset of her difficulties to being made redundant from to exercise judgement about whether to record pertinent
her job 6 months ago. This was unexpected and came as ‘a terrible information separately or as part of ‘personal history’.
blow’. ‘I was told I wasn’t up to it’. Admits that there had been
‘problems’ latterly but was ‘shocked’ when she was, as she sees it,
‘fired’. Since then has brooded a lot on this and feels she was ill done
by. Problems have progressively worsened since then.

Personal and social history Past psychiatric history

Although the ‘History’ conventionally comprises a series of There is merit in specifically presenting details of past psy-
further headings, additional information is essentially about chiatric disorder as this often has a major bearing on the current
placing the symptomatic material in a social context. The presentation and on treatment options. Previous periods of
order of presentation is not fixed – some may see it as more identifiable illness should be summarised, which will often con-
logical to progress from details about the present episode to sist of episodes of diagnosed and treated illness. However, it is
enquiry about the ‘past psychiatric history’, while for others important to include any episode that comes to light, even if
this is part of the patient’s wider personal life more logically it did not result in medical attention at the time. For example,
placed with ‘personal history’. if a patient complaining of depression in the setting of consider-
able life stress describes a previous similar but milder episode
Family history some years previously which occurred without obvious precipi-
tant, or they give an account of a circumscribed period of appar-
After obtaining the symptomatic information, it is reasonable ent mild hypomania, this will have implications for diagnosis.
to explore the family relationships from which the patient
has emerged and to which they may still belong. It is important to record the patient’s past treatment his-
tory, though this can be a surprisingly difficult area in which
It is traditional to begin with factual information: whether to gain accurate information. The importance of past drug
parents are alive and if so their ages and occupations (which history extends beyond the obvious – such as avoiding reuse
can provide a useful gauge of advancement or decline), and of preparations to which the patient may have developed bad
whether they suffer from any physical illness; if they are reactions/allergies in the past, etc. The patient will have
deceased, their age at death and the cause. This is repeated expectations of their consultation which are likely to extend
for siblings, then followed by enquiry about psychiatric disor- beyond merely a re-prescription of a previously failed regime.
der in all blood relatives, not just immediate family. If positive If there are reasons to recommend a compound tried before,
responses are elicited it is useful to supplement the enquiry it is important to be able to justify this clearly, as a positive
with pointers to a likely diagnosis, if known – e.g. ‘he/she gets decision – in terms, for example, of its previous use being
an injection every month’: ‘he/she saw a counsellor for inadequate. This cannot be done if prior regimes have not been
elucidated in detail.

209

Companion to Psychiatric Studies

Personal and social history chart the whole evolution of the patient’s symptoms under
‘history of present illness’, even if this spans a number of years:
Family history on another, it may be sufficient merely to state that the
patient has had ‘N’ previous admissions with similar symptom-
Father died 5 years ago, aged 81. Had Alzheimer’s disease. Retired atology beginning at age ‘X’, describing only those episodes
engineer. which may not appear to fit the stereotype. In presenting
‘long’ cases for examination purposes, it is important to
Mother alive and well aged 78. Housewife. Lives independently. explain at the start of the presentation which approach you
She sees her regularly. have opted for, and why.

Eldest of three siblings – Sister aged 49, married with three chil- Premorbid personality
dren. Housewife.
Some schemes advocate specific assessment of ‘premorbid
Brother aged 44, married, one son. Lives in USA. Advertising personality’. This is a complex area that requires expertise to
executive. attain reliability and the routine interview only allows for a
‘stab’. One should begin with general enquiries, e.g. ‘Would
Describes a happy family life – remains close to her family and you say you were a shy person or an outgoing person?’:and ‘Is
keeps in touch regularly. your mood generally stable?’, though one must always remem-
ber to be diplomatic – no-one likes to admit they have no
Maternal uncle committed suicide in early adult life – ‘alcohol pro- friends! It is better to approach this sort of area gently – e.g.
blems, I think’. ‘Do you have lots of friends or just one or two close friends?’.
One can then move on to more specific questions designed to
One cousin on mother’s side treated for depression in the past – elicit evidence of various different abnormal personality traits
thinks she is still on antidepressants. No medical history of note in (Box 10.4). Questioning about premorbid personality provides
family. little or no useful information about some personality traits,
such as dissocial, histrionic and dependent.
Personal history
Mental state examination
Born 14/10/53, Northern General Hospital. Normal full-term delivery.
No early developmental or behavioural problems. Started school at To some extent, examination of the mental state follows the
5 years old – enjoyed school and had many friends. High school from conventions of general medicine, but with some important dif-
12 – was popular and made friends easily. Academically successful. ferences. Firstly, the distinction between symptoms (which
Left at 18 with 5 ‘Higher’ grades (2 ‘A’s’/3 ‘B’s). the patient describes) and signs (abnormalities the clinician eli-
cits on examination) becomes blurred in psychiatry (Sims
Went to Edinburgh University – honours course (MA) in politics/ 1988) – many of the phenomena of interest to psychiatrists,
economics. Graduated 22 years. Joined Bank : ‘fast-track’ manage- whether described spontaneously or elicited by questioning,
ment programme. Achieved steady promotion till resignation with first exist only as subjective descriptions of experience. Because
pregnancy at 29. Worked from home as freelance financial advisor of this, it is common practice to use the term ‘symptom’ to
for 6 years before returning to work as financial analyst for an invest- cover both types of phenomena. However, it is important for
ment company. Promoted to team leader 4 years ago. Work ‘stressful’ the clinician to keep an awareness of the distinction in mind.
but enjoyed and always felt capable. Achieved regular bonuses. Made
redundant 6 months ago – was a reorganisation 18 months ago, Secondly, the identification and categorisation of mental
though her role changed little. Told by senior colleague she was no symptoms depends on the discipline of descriptive psychopa-
longer able to do the job. Accepted a ‘generous’ settlement but felt thology, or phenomenology. Nowadays, these two terms tend
‘devastated’. to be viewed synonymously but traditionally in clinical psy-
chiatry the latter referred to a method of evaluating patient
Menarche aged 11 years. Regular periods. No problems. One experience embracing a broader, more individualised, less cat-
steady boyfriend at school (not physical). Met husband at University. egorical (and hence dogmatic) approach to mental assessment
Lived together for 3 years before marriage. Describes good relation- than the kind that has tended to follow the introduction of
ship with husband, now partner in civil engineering firm. ‘He’s very operationalised diagnostic techniques, where the interview
understanding’. He travels extensively. She’s always accepted this tends to address the criteria rather than the criteria being
but recently ‘I can’t stand it when he’s away’. Feels ‘terrified’ on her applied to the products of the interview. The reader is
own, and stays with mother. referred to Mullen (2007) for a brief but articulate exposition
of the essentials and a passionate plea for its wider utilisation
Two children – son of 24 (married) works with airline, daughter of the authors would endorse. In essence, the phenomenological
19 (single), first-year medical student in London. Children are ‘wonder- method attempts to recognise and categorise mental abnormal-
ful’. One grandson (18 months old). ities without embellishing them with explanation, but at the
same time trying to understand the nature of the experience
Owns own home. No money worries – ‘life should be very as far as possible.
comfortable’.

Normally enjoys hill walking, country dancing and reading (lost
interest in all in past few months).

Describes herself as diligent, ‘slightly obsessive’, gregarious. Has
high expectations of others but feels she was ‘a good manager’ –
was popular with staff.

Drinks moderately: 12–15 units per week. Mainly wine. Non-
smoker. Denies illicit drug use.

No medical history of note.
Was on contraceptive pill for some years. Now takes HRT. Nil else.
No forensic history.

Sometimes, in patients with extensive histories and chronic
symptomatology, separation of ‘history of present illness’ from
‘past psychiatric history’ can be difficult, and no hard and fast
rules can be applied. On one occasion, where no clear breaks of
well-being can be established, the best approach may be to

210

Clinical assessment: interviewing and examination CHAPTER 10

Box 10.4 appearance and behaviour, speech, mood and affect, thought
content and perception, concluding with cognition and insight.
Useful themes of enquiry for probing premorbid The exact ordering of these is, however, determined largely
personality by convention. While some will place ‘mood/affect’ immedi-
ately after the general descriptive components, others will con-
Schizoid sider it more logical to progress to ‘thought’ immediately after
Does he* have solitary or eccentric hobbies? ‘speech’, as speech is the medium from which thought
Does he have few friends? processes and contents are inferred.
Is he shy?
Paranoid Appearance
Has he ever had serious rows with people so that they have
This initial category comprises a brief descriptive note of
never spoken to him again? observations at first contact and throughout the interview.
Has he ever taken anyone to court? It refers not only to matters of dress and grooming, but would
Would other people regard him as a touchy person, i.e. quick to cover physical characteristics worthy of note, such as posture,
build, age-appropriateness of appearance and so on. Facial
take offence expression is an evident part of one’s ‘appearance’, but unless
Does he tend to take remarks the wrong way? clearly associated with neurologically mediated disorder, for
Does he have any beliefs which he holds very strongly? example parkinsonism, is best covered elsewhere.
Anankastic (obsessional)
Is he very tidy/fussy about the way things are arranged? Despite the wide range of dress currently acceptable in west-
Would other people regard her as being very houseproud? ern society, it is generally possible to judge, often ‘at a glance’,
Is he the type of person who ‘goes by the book’/believes in the whether acceptable codes are being followed, and whether to
an acceptable level – i.e. whether the patient is unkempt
motto ‘a place for everything and everything in its place’? or not. If a patient follows the fashion of some identifiable
Does he find it difficult to make decisions about trivial things, subculture – for example, new age travellers, punks, bikers,
etc. – it is appropriate to record this, even though it is not in
such as what to wear, which restaurant to go to? itself indicative of clinical abnormality.
Does he have to check things that he knows he has already
Any degree of self-neglect may be seen in schizophrenia, and
done, such as doors being locked, appliances being switched may be subtle or isolated – e.g. odd socks, flies undone, etc.
off? Mild–moderate dental decay is a fairly non-specific sign in
Hyperthymic psychiatric patients but in severe degree, especially in the
Is he very extravert? young, can be a pointer to substance misuse, particularly
Would people regard him as the life and soul of the party? metamphetamine (‘crystal meth’). In females, letting appear-
Does he need less sleep than most people? ance go, as indicated by lack of make-up or dishevelled hair,
Does he suffer from brief spells of depression lasting a few days can be a telling sign of depression. General decline in self-care
every so often? may also be a marker of emerging dementia. While cognitive
Histrionic impairment may be evident in dress and grooming – e.g. buttons
Is he the kind of person who likes to be the centre of attention at done up incorrectly – it is worth remembering that some
social gatherings? adverse treatment effects, especially parkinsonism, can impair
Can he be on top of the world one minute and then in the depths dexterity, resulting in neglect which is more apparent than real.
of despair the next?
Dissocial Excessively loud (or, in females, revealing) clothes can be a
Does he lose his temper easily? feature of hypomania, as can garish or bizarre make-up, exces-
Dependent sive jewellery, ornaments or inappropriate dress (e.g. evening
At what age did he leave home (for the first time)? clothes). Highly unfashionable and/or clashing garments may
reflect long-standing schizoid personality traits or Asperger’s
*Male personal pronouns are used to imply either male or female, unless a female syndrome, while an excessively neat appearance, sometimes
form is more appropriate in the context. disguised by studied casualness or even affectation, has been
said to be a sign of obsessional personality (Enoch & Ball
Avoidance of leading and direct questions as an interview 2001). Judgement must, however, be exercised with such
‘style’ was noted above but in determining whether a particu- alleged ‘signs’. While important to note, they can rarely
lar symptom is present, what its specific characteristics are provide more than adjunctive information.
and how it can be differentiated from other similar symptoms,
some degree of ‘cross-examination’ is inevitable. As the pref- Behaviour
ace to the PSE notes: ‘. . . an interview designed to discover
whether defined symptoms are present must be based to some Of necessity, the main focus of this part of the examination is
extent upon the technique of ‘cross-examination’. Patients behaviour during the interview. However, it is perfectly
find this completely acceptable and, to the extent that inter- acceptable to include accounts of noteworthy behaviour in
viewer and patient are together successful in producing an
exact description of the symptoms, it can be a rewarding and
therapeutic experience in itself’. (Wing et al 1974).

As with the history, categorising and recording mental state
has a fairly standard structure, covering the domains of

211

Companion to Psychiatric Studies

prior contexts from reliable third party sources (e.g. family, (Rogers 1985; Lund et al 1991), and can also be seen in both
general practitioner, police, etc.). phases of affective psychosis, learning disabilities, autism and
some forms of organic brain disease (Rogers 1992; Wing &
Any automatic motor behaviours should be noted, such as Shah 2000). Some of these wide-ranging phenomena are
restlessness, fidgetiness and nervousness, as well as involuntary summarised in Box 10.5.
movements such as dyskinesias. Any loss of expected motor
activity in relation to, for example, facial expression, expres- Lack of volition is one of the main negative features of
sive posture and gesture, can be recorded here or above, as schizophrenia. As its name suggests, it refers to reduced motor
noted. It is traditional to comment on patients’ eye contact behaviour which is based on a failure of motivation and inter-
though poor eye contact is of such low diagnostic significance est. It is probably identical to, or at least shades into, apathy,
as to be largely without value. It is more informative to note although this term tends to be reserved for generally more
gaze avoidance if present, as this can be a sign of hostility. severe states associated with marked emotional indifference
A useful additional consideration is the patient’s manner seen in organic disorders.
towards the interviewer: What was their approach to you as
an individual? Were they co-operative, forthcoming, engaged? In affective disorders the main abnormalities are retarda-
Did they observe the social proprieties of politeness, etc? tion and agitation in depression and a rather less well-defined
set of inverse abnormalities in mania. Retardation consists of
The major abnormalities of movement and behaviour are delay in initiating and slowness in executing motor acts which
those associated with schizophrenia and affective disorder. may be subtle and only noticeable as invariant facial expres-
Acutely psychotic patients may act in response to delusions sion and paucity of expressive gestures, or more pronounced,
(searching for bombs in cupboards, etc.), respond to abnormal manifesting itself in the classical bowed posture, slow hesitant
perceptions (‘behaving as if hallucinated’), or break social actions and soft monotonous speech. The most extreme mani-
taboos without compunction (e.g. masturbate, make sexual festation is depressive stupor, where the patient lies nearly
advances or, in the tactful language of the PSE, ‘pass loud fla- motionless, struggles to whisper a few words and has, as
tus’!). They may be readily distractible by incidental noises, or Kraepelin put it (1913a), a characteristic ‘peculiar vacant,
so suspicious or bewildered as to be totally unforthcoming. strained, disturbed, facial expression’.
Catatonic features are nowadays uncommon in acute
schizophrenic patients but remain common in chronic patients Agitation is a much misused term which, strictly, describes
patients who are in a state of severe motor unrest in the

Box 10.5

Catatonic symptoms

Disorders of movement Positivism (Lohr & Wisniewski 1987): includes echopraxia; mitgehen
(passive movement overcomplied with, limbs elevate themselves
Stereotypies: purposeless motor acts which are carried out at the slightest touch like an anglepoise lamp); and automatic
repetitively and with a high degree of uniformity, e.g. rocking, obedience, where any suggestion (even one that is merely implied)
rubbing hands and tapping objects, more complicated, is complied with in an exaggerated way
‘gymnastic’ or ‘contortionist’ movements
Other: ambitendence — the patient appears to simultaneously wish to
Mannerisms: everyday goal-directed acts like washing, dressing and and not wish to carry out some action, e.g. walking through a
eating which are executed in idiosyncratic ways, e.g. keeping one doorway
arm tucked under armpit. (NB: In practice, it is difficult to separate
stereotypies and mannerisms from one another; many patients Disorders of behaviour
show motor behaviour which is abnormal by virtue of varying
degrees of repetitiveness, purposelessness, and stiltedness of Catatonic stupor: the patient sits or lies motionless, mute, often in a
execution.) contorted posture; waxy flexibility and gegenhalten may be evident.
Although unresponsive, the patient is aware of his surroundings
Manneristic gaits: e.g. over-precise or overelaborate, walking on the
toes, with interpolated sidesteps and bowing Catatonic excitement: aimless overactivity, destructiveness and
violence, often associated with manneristic and stereotyped
Posturing: hunched and constrained, hugging sides, sitting perched actions, e.g. moving around striking an endless series of quasi-
on edge of chair, adopting statuesque, ‘pharaonic’ poses (Sims symbolic poses
1988)
Catatonic impulsiveness: sudden, incomprehensible and often violent
Other: grimacing, blocking/freezing, waxy flexibility (catalepsy), acts for which the patient is unable to give any more than a facile
psychological pillow (where the patient lies with his head 2–3 explanation. May interrupt stupor
inches off the bed)
Catatonic speech disorders
Disorders of volition
Aprosodic speech: speaks in monotone, peculiar scanning,
Negativism: patients do the reverse of whatever is asked of them; e.g. automaton-like, telegrammatic sing-song, or affected intonations
hold their breath when asked to breathe deeply, resist attempts to
get them to stand up, then refuse to lie down. Also gegenhalten Speech stereotypies and mannerisms: speaking in affected accents,
(opposition): ‘springy’ resistance to passive movement which adding -ism or -io to every word
increases with the force exerted
Other: mutism, echolalia and palilalia (repeating last word or last few
words of a sentence over and over again)

After Lund et al (1991) and McKenna (2007).

212

Clinical assessment: interviewing and examination CHAPTER 10

context of obvious distress. Agitated patients cannot remain on the other hand, is the one point in the MSE where empathy
seated, pace back and forth, wring their hands, pull their hair must be used to gauge the impact of the patient’s presentation.
and clothes and make despairing gestures. Many psychiatrists Depressed patients often display a heightened depth of emo-
also apply the term to lesser degrees of motor unrest accompa- tion and distress.
nied by anxiety but it should not be used merely to describe
restlessness. Retardation and agitation can be present in the The classic objectively observed affective abnormality in
same patient. schizophrenia is flattening of affect in which both the range
and depth of emotion are restricted with loss of subtlety and
Manic and hypomanic patients may appear infused with nuance in moment-to-moment emotional interaction. Beyond
energy, drum their fingers impatiently, make emphatic and this, affective flattening is not easily defined and, it is fair to
exaggerated gestures, smile broadly and display other emotions say, not altogether understood. Psychophysiological studies
with animation. Restlessness, if present, usually reflects impa- suggest that such individuals with affective flattening are often
tience, is without distress, and is indicated by fidgeting, walk- highly emotionally ‘aroused’.
ing around the room, leaving the room and coming back, etc.
In more severe states patients may be continuously ‘on the Even experienced clinicians can disagree over the presence
go’, interfering and partaking in pointless acts which they are of milder degrees of affective flattening. Problems arise from
rapidly distracted from, sometimes referred to as a press of the wide variation in the display of emotion, both culturally
activity. and individually – some cultures valuing emotional reserve;
some individuals always maintaining an unemotive, ‘dead-pan’
Affect (mood) demeanour. Many normal people will show (and feel) no emo-
tion when bereaved or on receiving news of other traumatic
The concepts of ‘affect’ and ‘mood’ are so confusing for psy- events. Some individuals with dissocial personality disorder
chiatry that a general consensus appears to have been reached habitually describe horrifying crimes with indifference or
that they should be considered interchangeably. This con- amusement (Hamilton 1984) Finally, the distinction between
fusion was cemented with the designation of major mood dis- affective ‘flattening’ and drug-induced bradykinesia can be dif-
orders as ‘affective disorders’, attributed to Manfred Bleuler ficult, especially if no steps have been taken to examine for
in the 1930s. In English-language psychiatry, many clinicians other parkinsonian features (Owens 1999). Because of such
use the term ‘affect’ to refer to emotionality objectively variability and context, assessment of ‘flattening’ should never
observed during the interview, and ‘mood’ to describe the be a ‘spot’ diagnosis but should come from observation of the
patient’s subjective account of their emotional state. This is patient’s interaction throughout the interview. Furthermore,
inaccurate. ‘Affect’ refers to that short-term component of because of its specific diagnostic inference (schizophrenia),
emotionality which is responsive to circumstance and environ- the term should not be used loosely.
ment, and comprises a multitude of generally short-lasting
feeling states, such as fear, anxiety, contentment, anger, jeal- Blunting of affect is commonly used synonymously with
ousy, etc. (i.e. emotional ‘waves’). ‘Mood’ on the other hand, affective ‘flattening’, but the two are separate phenomena.
refers to one’s longer-term emotional predisposition, as gauged ‘Blunting’ refers to coarsening of emotions and an insensitivity
along the dimension of happiness–sadness (i.e. emotional to social context – what Kraepelin referred to as a ‘loss of the
‘currents’). delicacy of emotion’ (Kraepelin 1913b). For example, Sims
(1988) described a schizophrenic woman who, with obvious
Affective symptomatology, strictly defined, usually appears relish for the sensational effect, took her visitors upstairs to
at different points in the presentation of mental state mate- view the corpse of her mother, deceased for 2 days.
rial. Many affective features, for example, are best deter-
mined by questioning and hence will appear mainly in the Inappropriate or incongruous affect is the appearance of sud-
‘History of the Presenting Complaint’, while those which den emotional states which are out of keeping with events. In
present overtly will usually be described under ‘Behaviour’, practice, this is invariably evidenced by patients laughing when
as noted above. However, any affective symptomatology describing distressing events. For example, when asked why
should be briefly brought together here – e.g. noting anxiety, there was a bandage on her wrist, one patient replied that
anger, irritability, etc. she had been trying to kill herself that morning, and burst
out laughing! It is important to remember, however, that inap-
Subjective mood state should be explored and described in propriate laughter may be a sign of social awkwardness or
the history but it is important to summarise the key findings embarrassment and assessment of affective incongruity should
at this point in the MSE. This would include the patient’s always be circumspect. Rightly or wrongly, sudden causeless
impression of severity and quality of mood change, the pres- laughter, essentially seen only in patients with severe chronic
ence of anhedonia (inability to experience pleasure) and main- schizophrenia, is considered by some a manifestation of inap-
tenance of reactivity (the patient can temporarily cheer up in propriate affect.
the right circumstances). The objective expression of mood
should also be noted and can be considered along two axes – In mania, mood is characteristically one of increased vital-
breadth and depth. ‘Breadth’ affect refers to the variability ity, gaiety and pleasure, with an infectious quality, rather than
of emotion exhibited within the interview seen in facial simple happiness. In mild states, this may only be evident in
expression, interactive posture, gesture, intonation, flow of the persistence of the mood state regardless of context, or in
speech, etc. Assessment is on the basis of observation. ‘Depth’, its ready emergence in response to trivial or incidental stimuli.
Alternatively (and perhaps increasingly) irritability and exces-
sive response to frustration may predominate. Euphoria – a
heightened sense of happiness – is distinct from the typical

213

Companion to Psychiatric Studies

mood change. Similarly, patients with major depression not individuals may also communicate in a ‘pressured’ fashion).
infrequently describe their mood as different from normal When speech is rapid and difficult to interrupt the term
unhappiness, communicating a peculiarly painful quality. pressure of speech is applied. Catatonia can also affect speech
(see Box 10.5).
Additional important affective abnormalities include suspi-
cion, in which the palpebral fissures are narrowed, the mouth Speech latency refers to the time between a question being
tightened and the eyes disproportionately alert and active in posed and an answer being given. Delays in providing response
otherwise fixed facial contours; and perplexity, characterised can be the result of an intense individual taking time to ensure
by lowered brows, slowed gaze and partially open mouth. the adequacy of their reply or can be found in the presence of
Perplexity is seen par excellence in puerperal psychosis, but anxiety intruding into the normal organisation of responses.
can be present in any acute psychotic state, including organic Striking increases in speech latency can be seen in pro-
ones. The so-called omega sign, in which the inner third of found depression but are also to be found in psychotically
the eyebrows are depressed, the outer third elevated and ver- preoccupied patients.
tical ridging is evident over the glabellar eminence, is said to be
characteristic of major depression though is more likely to Poverty of speech is a characteristic negative symptom of
be a general expression of extreme worry. Lability of mood, schizophrenia, which can be regarded as the same disorder as
where the emotional state shifts from cheerfulness to tears lack of volition, affecting the domain of speech. Patients
to irritability and back again over brief periods is common in answer questions readily enough but use only the minimum
mania, may be found in dementias or other organic states necessary number of words. They do not elaborate what they
and is also a rare symptom of acute schizophrenia (Bleuler say, and make few spontaneous comments. Encouragement
1911). Facile euphoria is said to be characteristic of the frontal with open questions meets with little success.
lobe syndrome and Korsakoff’s syndrome. Ecstasy is a state of
intense tranquil euphoria which usually has a religious colour- Speech is, as noted, the principal means by which we
ing, often attracting descriptions like ‘exalted’ and ‘trans- express thought. Formal thought disorder refers to abnormality
figured’ (Hamilton 1984; Sims 1988). Patients may become in the coherence and logical structure of thought – the ‘how’ or
so absorbed in their inner state as to be unresponsive to ‘mechanics’ of thinking – which makes conversation difficult
the external world, and objectively appear to be in a state of to follow or at times completely incomprehensible. Many
so-called euphoric (or manic) stupor. different abnormalities have been proposed to account for
the loss of coherence of thought seen in schizophrenia, and
Other emotional abnormalities said to be characteristic of in some cases the same phenomenon has acquired different
particular disorders include the catastrophic reaction in early terminologies – e.g. ‘derailment’ and ‘knight’s move thinking’.
dementia, where the patient reacts excessively to realisation Current approaches, such as DSM IV and structured inter-
of failure on a cognitive task (the poor performance is abnor- views like the PSE, place the emphasis on just a few types of
mal but so is the response to it) and so-called obsessional affect disorder:
(‘warmth, with something held back’), described in individuals
with anankastic (obsessional) personality. Histrionicity is the • derailment (loosening of associations, tangential thinking,
exaggerated, melodramatic, but at the same time shallow, ‘knight’s move’ thinking): the ‘train’ of thought slips off the
expression of emotions. ‘track’. Individual derailments may be slight, so the speaker
only gradually, but progressively, digresses from the original
Speech and thought form point, though in some cases the links between ideas are
clearly obscure and difficult to decipher.
Thought is a purely subjective process and the primary means
open to us to make inferences regarding its form is via speech. • incoherence (word salad, schizophasia): speech is
As a result, the separation demanded by the traditional MSE completely incomprehensible. Typically, individual
into ‘speech’ and ‘thought’ is somewhat artificial. Strictly sentences themselves have no meaning, in contrast to
speaking, ‘speech’ should be limited to the motor component derailment, where the meaning is lost over several
of verbal expression – everything from the central speech areas sentences.
to the lips. Abnormalities include dysarthria, dysphasias and
disorders of articulation. However, there are other compo- • neologisms: often found in conjunction with incoherence.
nents to speech including pitch, power and intonation (or This refers to the invention of new words (e.g. ‘tarn-harn’;
‘prosody’– the natural musicality of word usage), which are ‘bathroot’). The term has also been applied to the use of
the preserve of psychiatry, as are a number of so-called psy- existing words in individualistic ways without any clear
chomotor qualities of speech, such as rate, quantity, flow and meaning, though this is more precisely idiosyncratic word
so on. These should also be recorded under ‘speech’ even if usage (e.g. ‘frequenting of clairvoyance’). A further variant
the origins of the change lie in other domains (e.g. mood). is word approximations, where imprecise words or phrases
are substituted for familiar ones – e.g. one patient spoke of
Many abnormalities of speech are fairly obvious counter- having ‘a menu three times a day’ instead of three meals a
parts of the disorders of behaviour described above, applied day (Cameron 1938).
to the particular motor act of speaking. Thus, speech may be
soft, delayed and prone to tail off in depression or, conversely, • poverty of content (of speech): speech which is spontaneous
loud and increased in quantity in mania (socially anxious and normal in quantity (i.e. as opposed to ‘poverty of
speech’, the number of words used is normal/appropriate)
but which fails to communicate any meaningful
information. If ‘incoherence’ is the most severe expression
of formal thought disorder, then poverty of content is the

214

Clinical assessment: interviewing and examination CHAPTER 10

mild end of the spectrum. Bear in mind, however, that ‘this In depression, thought may become narrowed and repeti-
symptom may appear to be readily recognisable in some of tive and in severe cases can be reduced to a small circle of
one’s colleagues, therefore only rate it when it really is painful ruminations: in response to questioning on any topic,
pathological’ (Wing et al 1974)! one patient would almost immediately revert to stereotyped
repetitions of depressive delusions – ‘my husband has been
The two classical disorders of thought form in mania are: taken away. . . the house is boarded up. . . the children are on
flight of ideas (or distractible speech), in which the train of the street’.
thought is distracted by irrelevant associations, either internal
or external, which capture the patient’s attention; and clang Andreasen (1979) has developed a detailed classification of
associations, where irrelevant connections are made on the these and the many other abnormalities of thought form seen
basis of assonance (similarity of sound), rhymes or puns. in psychiatry (Box 10.6). This is broad and includes

Box 10.6

Abnormalities of thought, language and communication

Poverty of speech. Restriction in the amount of spontaneous speech, that is unwarranted or illogical. It may take the form of faulty
so that replies to questions tend to be brief, concrete and inductive inferences. It may also take the form of reaching
unelaborated. Unprompted additional information is rarely conclusions based on faulty premises without any actual
provided delusional thinking

Poverty of content of speech. Although replies are long enough so Clanging. A pattern of speech in which sounds rather than meaningful
that speech is adequate in amount, it conveys little information. relationships appear to govern word choice, so that the
Language tends to be vague, often over-abstract or over- intelligibility of the speech is impaired and redundant words are
concrete, repetitive and stereotyped. The interviewer may introduced. In addition to rhyming relationships, this pattern of
recognise this finding by observing that the patient has spoken at speech may also include punning associations, so that a word
some length but has not given adequate information to answer the similar in sound brings in a new thought
question. Sometimes characterised as ‘empty philosophising’
Word approximations. Old words that are used in a new and
Pressure of speech. An increase in the amount of spontaneous unconventional way, or new words that are developed by
speech as compared with what is considered ordinary or socially conventional rules of word formation. Often the meaning will be
customary. The patient talks rapidly and is difficult to interrupt. evident even though the usage seems peculiar or bizarre
Some sentences may be left uncompleted because of eagerness
to get on to a new idea. Even when interrupted, the speaker often Circumstantiality. A pattern of speech that is very indirect and delayed
continues to talk. Speech tends to be loud and emphatic in reaching its goal idea. In the process of explaining something,
the speaker brings in many tedious details and sometimes makes
Distractible speech. During the course of a discussion or interview, parenthetical remarks. Circumstantial replies or statements may
the patient repeatedly stops talking in the middle of a sentence or last for many minutes if the speaker is not interrupted and urged to
idea and changes the subject in response to a nearby stimulus, get to the point. When not called circumstantial, these people are
such as an object on a desk, the interviewer’s clothing or often referred to as ‘long-winded’
appearance, etc.
Loss of goal. Failure to follow a chain of thought through to its natural
Tangentiality. Replying to a question in an oblique, tangential or even conclusion. This is usually manifested in speech that begins with a
irrelevant manner. The reply may be related to the question in particular subject, wanders away from the subject and never
some distant way. Or the reply may be unrelated and seem totally returns to it. This often occurs in association with derailment
irrelevant
Perseveration. Persistent repetition of words, ideas, or subjects so
Derailment. A pattern of spontaneous speech in which the ideas slip that, once a patient begins a particular subject or uses a particular
off the track onto another one that is clearly but obliquely related, word, he continually returns to it in the process of speaking
or onto one that is completely unrelated. Things may be said in
juxtaposition that lack a meaningful relationship, or the patient Echolalia. A pattern of speech in which the patient echoes words or
may shift idiosyncratically from one frame of reference to another. phrases of the interviewer. Typical echolalia tends to be repetitive
At times, there may be a vague connection between the ideas; and persistent. The echo is often uttered with a mocking,
at others, none will be apparent mumbling or staccato intonation

Incoherence. A pattern of speech that is essentially incomprehensible Stilted speech. Speech that has an excessively formal quality. It may
at times. The incoherence is due to several different mechanisms, seem rather quaint or outdated, or may appear pompous, distant,
which may sometimes all occur simultaneously. Sometimes the or overpolite. The stilted quality is usually achieved through use of
rules of grammar and syntax are ignored, and a series of words particular word choices (multisyllabic when monosyllabic
or phrases seem to be joined together arbitrarily and at random. alternatives are available and equally appropriate), extremely
Sometimes the disturbance appears to be at a semantic level, polite phraseology (‘Excuse me, madam, may I request a
so that words are substituted in a phrase or sentence so that the conference in your office at your convenience’), or stiff and formal
meaning seems to be distorted or destroyed. Sometimes syntax (‘Whereas the attorney comported himself indecorously,
‘cementing words’ (conjunctions such as ‘and’ and ‘although’ and the physician behaved as is customary for a born gentleman’)
articles such as ‘the’ ‘a’ and ‘an’) are deleted
Self-reference. A disorder in which the patient repeatedly refers the
Illogicality. A pattern of speech in which conclusions are reached that subject under discussion back to himself when someone else is
do not follow logically. This may take the form of non sequiturs, in talking and also refers apparently neutral subjects to himself when
which the patient makes a logical inference between two clauses he himself is talking

After Andreasen (1979).

215

Companion to Psychiatric Studies

abnormalities such as echolalia, a catatonic symptom, persever- Latin – ‘de mentia’. There was no specific inference of persecu-
ation, the inference of which is most frequently an organic dis- tion, as has become the case in popular usage today. Thus, ‘para-
order, and circumstantiality which is not uncommon in normal noid’ schizophrenia was simply schizophrenia dominated by
individuals. delusions (as opposed to motor or other abnormalities). Techni-
cally, to talk of ‘paranoid delusions’ (i.e. ‘delusional’ delusions)
Trainees tend to absorb descriptively based classifications of is clearly a nonsense!
communication disorders in psychoses as if in clinical practice
they have strong discriminative power, whereas in reality they Delusions of reference can be viewed as one end of a ‘spec-
often do not. When a patient is considered to show evidence trum’ of psychopathology. With reference, the disorder is
of formal thought disorder it is essential to quote an example broad, as the events described, while distressing the patient,
(or preferably examples) of the speech on which this con- may be happening to others and usually have no clearly defined
clusion is based. It is surprisingly easy to jot down a few con- source. The world in general is devoid of coincidence with, as
secutive sentences while the patient is talking, the example the term implies, every incidental occurrence referred directly
often appearing more disordered when read back outside the to the patient, assuming a personal significance, usually imply-
interview. ing threat. They find hints and double meanings in the per-
fectly ordinary behaviour, gestures and statements of others,
Thought content are the victims of gossip or are followed by strangers checking
up on them. Further along the ‘spectrum’, the ‘evidence’
Disorders of content relate to the ‘what’, as opposed to the becomes more sophisticated, while the focus narrows more
‘how’, of thinking – to the ideas and beliefs one holds. The to the patient. There may be references to him/her, or mes-
prototypical abnormality is delusions. However, it is also sages planted for them alone, in newspapers, magazines or on
appropriate to record here any other abnormal ideas and television or they may see themselves as the subject of some
beliefs, including disorders of the possession of thought, over- experiment set up specifically to test them out – an extension
valued ideas, obsessional ruminations, depressive ideas of self- the PSE calls delusions of misinterpretation/misidentification.
depreciation, self-blame and guilt, simple ideas of reference With delusions of persecution, symptomatology is ‘crystallised’
and so on. into the morbid belief that the patient alone is the focus of
some conspiracy or plot on the part of defined individuals or
In English-language psychiatry, delusions are traditionally organisations, such as the Freemasons, the Government, etc.,
defined as: which is reaping adversity in their life that may end in dire
consequences, including death.
false, unshakable beliefs, out of keeping with the individual’s
educational, social and cultural background, which are held with utter Whether persecution is always preceded by a prior phase of
conviction and are impervious to counter-argument (i.e. are reference and/or misinterpretation is open to debate but the
‘incorrigible’). value of considering this type of symptomatology this way is
firstly, that ‘reference’ is much commoner than, and not the
While this provides a yardstick that can be clinically useful, it same as, ‘persecution’; and secondly, that despite the often
is flawed in each of its elements. For example, all of us hold to well-preserved exterior, ‘persecution’ can be viewed as reflect-
beliefs that are, technically, ‘false’ (i.e. without, or contrary to, ing more profound disorder.
fact or evidence) while in other instances the alleged ‘falsity’
is a matter of opinion. The most noticeable omission is that Occasionally, those same individuals/organisations that
conventional definitions fail to emphasise that the belief is ‘persecute’ can be seen as working surreptitiously to promote
‘morbid’ – i.e. that they in some way reflect pathology. As or help, a phenomenon known as delusions of assistance. This
Jaspers (1963) pointed out, phenomenologically, a delusion is one of a group of phenomena the PSE refers to as ‘expansive
cannot be defined – it is simply a ‘belief’, but one that repre- delusions’, which includes grandiose ability and identity and
sents a ‘pathologically falsified judgement’. Patients may also illustrates that ‘persecution’ and ‘grandiosity’ can be consid-
experience partial delusions, beliefs which are expressed with ered opposite sides of the same coin – both reflecting a dis-
doubt, as a possibility but not a certainty (Wing et al 1974). turbance of the patient’s perception of ‘self-in-the-world’
(‘inferior’ versus ‘superior’). In practical terms, persecutory
In practice, descriptive classifications of delusions can be ideation is not infrequently found with grandiose delusions
useful as clinical ‘shorthand’ for targeting questions. Although and grandiosity is often detectable in the ‘victims’ of
there are various schemes for classifying delusions descrip- persecution.
tively, the most comprehensive is that used in the Present
State Examination (Wing et al 1974), outlined in Box 10.7. Delusions of reference must be distinguished from simple
ideas of reference, which are an exaggerated form of self-
Patients often refer to themselves as feeling ‘paranoid’ consciousness, usually driven by social anxiety, comprising an
when experiencing probably the commonest type of delusion, uncomfortable feeling of undue attention from others. While
delusions of reference, but psychiatrists should be more sophis- the person cannot help thinking that others are taking notice
ticated with terminology. In the 19th century as now, a ver- of them, the feeling is recognised as an ‘as if’ experience origi-
nacular term for ‘mad’ was ‘out of the mind’, which in nating from within themselves. Almost everyone has experi-
Greek translates as ‘para noia’. This was the term adopted by enced this symptom at some time (e.g. on entering a room
early German-language psychiatry for patients whose mental and noticing that conversation seems to stop) though insight
state was dominated by delusions, the age-old marker for into the lack of reality is retained. Simple ideas of reference
‘madness’. The French, on the other hand, preferred the can, however, become pervasive and socially incapacitating

216

Clinical assessment: interviewing and examination CHAPTER 10

Box 10.7

The present state examination classification of delusions

Delusional mood. The subject feels that his familiar surroundings have including hypnotism, telepathy, magic, witchcraft, etc. Or it may
changed in a puzzling way which he may be unable to describe, be in terms of physical processes such as electricity, X-rays,
but which seems to be especially significant for him. The state television, radio or machines of various kinds. (This term largely
often accompanies the development of full delusions replaces the older term ‘secondary delusions’.)

Delusions of reference and misinterpretation. What is said has a Delusions of control. The subject experiences his will as replaced by
double meaning, someone makes a gesture which is construed as that of some other force or agency. The basic experience may be
a deliberate message, the whole neighbourhood may be gossiping elaborated in various ways — the subject believes that someone
about the patient. He may see references to himself on the else’s words are coming out using his voice, or that what he writes
television, radio and in newspapers, or feel he is being followed, is not his own, or that he is the victim of possession — a zombie or
that his movements are observed, and that what he says is tape a robot controlled by someone else’s will, even his bodily
recorded. Circumstances appear to the patient to be arranged to movements being willed by some other power. This is one form of
test him out, objects are placed in particular positions to convey a passivity experience (‘somatic passivity’)
meaning to him, whole armies of people are deployed to discover
what he is doing or to convey some information to him Sexual delusions. Delusions of having a fantasy lover, that one’s sex
is changing. Also delusions of pregnancy
Delusions of persecution. The subject believes that someone, or
some organisation, or some force or power, is trying to harm him Delusional memories and delusional confabulation. Experiences of
in some way, to damage his reputation, to cause him bodily injury, past events which clearly did not occur but which the subject
to drive him mad or to bring about his death equally clearly remembers, e.g. ‘I came to Earth on a silver star’.
Delusional confabulations are beliefs which the subject either
Delusions of assistance. The subject believes that someone, or some elaborates on or appears to make up on the spot
organisation, or some force or power, is trying to help him
Fantastic delusions. Delusions which violate elementary common
Delusions of grandiose ability. The subject thinks he is chosen by sense and logic, e.g. England’s coast is melting, she has given
some power, or by destiny, because of his unusual talents. He birth to thousands of children
thinks he is able to read people’s thoughts, or that he is
particularly good at helping them, that he is much cleverer than Simple delusions concerning appearance. The subject believes that
anyone else, that he has invented machines, composed music, something is wrong with his appearance. He looks ugly or old or
solved mathematical problems, etc. beyond most people’s dead, his skin is cracked, his teeth misshapen, his nose too large
comprehension or his body crooked

Delusions of grandiose identity. The subject believes he is famous, Delusions of depersonalisation or nihilism. The subject has a
rich, titled or related to prominent people. He may believe that he conviction that he has no head, that he cannot see himself in the
is a changeling and that his real parents are royalty, etc. mirror, that he has a shadow but no body or that he does not exist
at all
Religious delusions. The patient believes he is a saint, an angel or
even God, or has special spiritual powers, or a divine purpose Hypochondriacal delusions. The subject feels that his body is
unhealthy, rotten or diseased; he has incurable cancer, his bowels
Delusional explanations. Explanation or elaboration of other abnormal are stopped up or rotting away. (In schizophrenia there may be
experiences. This may be in terms of paranormal phenomena, bizarre beliefs about bodily change or malfunction.)

After Wing et al (1974).

in some conditions, such as anxiety and depressive states, or in their head – the term delusional explanation is now
occasionally following some event perceived as embarrassing preferred.
or shameful, or sometimes just in times of severe stress.
Subjective alterations in thought, or disorders of the posses-
Some other aspects of delusions deserve comment. While sion of thought, include a heterogeneous group of phenomena
most delusional beliefs develop over time, they can emerge covering core psychotic symptomatology, such as thought
without detectable antecedents in an ‘autochthonous’, or insertion/withdrawal and broadcasting. They also cover obses-
‘eureka’ fashion like a ‘brainwave’. Such beliefs have tradition- sional patterns of thinking. Obsessional thoughts, or rumina-
ally been designated ‘primary delusions’. In practice there is tions, have three characteristics: they are recognised by the
confusion surrounding this concept but there is merit in patient as their own thoughts (i.e. despite having an imposed
unequivocally identifying delusions as ‘primary’ on the rare quality, they are ‘non-alien’), their content is acknowledged
occasions in which they are to be found because of the strong as absurd and there is an attempt to resist them, though the
diagnostic inference of schizophrenia they allow. In the older latter two qualities may be less evident in severe or chronic
literature, primary delusions were distinguished from ‘second- disorder. Other subjective alterations of thought include slow-
ary’ delusions, where the abnormal belief seems to be based ness in depression, and crowding or acceleration of thought in
on, grow understandably out of or represent an elaboration of mania, in which ideas (and associated images) flash rapidly
some other element of psychopathology – as in depressed through the mind, each suggesting others.
patients who develop beliefs that they have sinned greatly, lost
everything, died already, etc. – though this term is less often Overvalued ideas are beliefs which, because of the excess of
used nowadays. Where a delusion seems obviously secondary emotional tone invested in them, come to dominate to an
to another psychotic symptom – e.g. a patient with auditory abnormal degree. They do not have obsessional characteristics
hallucinations who believes he or she has a radio transmitter (i.e. they are not recognised as absurd or resisted) but rather
are qualitatively akin to ‘preoccupations’ that any of us can,

217

Companion to Psychiatric Studies

from time to time, develop. Examples include querulous inside their heads, which are typically as real and compelling
paranoid states, in which individuals pursue redress for some as those emerging from external space and, in line with PSE
real or imagined wrong with extraordinary tenacity, and some conventions, these should be accorded equal diagnostic
forms of morbid jealousy and hypochondriasis. Another significance to ‘true’ hallucinations.
example of an overvalued idea, which also illustrates Jasper’s
argument that not all fixed, incorrigible beliefs are delusions, Visual hallucinations can be formless (shapeless images,
is the core belief of anorexia nervosa. lights, shadows) or formed objects (fiery crosses, faces, peo-
ple). Scenic or panoramic hallucinations are visions of whole
Abnormal perceptions scenes such as battles, the crucifixion, etc. Visual hallucina-
tions are perhaps most commonly described in delirium,
The most important type of misperception in clinical practice where they may take all these forms. In the elderly,
is hallucinations, which are most simply defined as ‘percepts they may take the characteristic form of ‘silent boarders’ –
without objects’ (i.e. they represent perceptual deceptions). full-sized figures seen around the house who do not speak.
There are, however, other definitions (Sims 1988), several of Visual hallucinations are less common in schizophrenia than
which emphasise the fact that, unlike tinnitus, the phantom auditory hallucinations, but the view that they are so uncom-
limb syndrome, etc., patients with hallucinations usually mon that their presence should point to an organic disorder
consider their experiences to be real. (Hamilton 1984) is probably extreme. They are certainly seen
in psychotic depression. Kraepelin (1913a) described patients
While hallucinatory experiences may occur in any sensory as seeing evil spirits, corpses, crowds of monsters and much
modality, in clinical practice hallucinations of hearing (audi- else. As with hearing, visual deceptions can take the form of
tory hallucinations) are the most frequent and important. ‘pseudohallucinations’, where the images, although intense,
‘Hearing voices’ in the alert state have been reported in indivi- are seen ‘in the mind’s eye’, or sometimes projected onto
duals who do not demonstrate other features of psychotic external – but subjective – space in an unreal way which the
illness (Dhossche et al 2002) and are not uncommon experi- patient often finds difficult to describe.
ences in normal individuals during sleep induction (hypnagogic
hallucinations), but the importance of auditory hallucinations Somatic hallucinations can be classified on the basis of the
is as a diagnostic pointer to psychotic illness, especially specific sensation they replicate – e.g. haptic (touch, tickling,
schizophrenia. pricking); thermic (heat and cold); hygric (wetness); kinaes-
thetic (movement and joint position) (Sims 1988). Patients
The range and variety of auditory hallucinations in schizo- may complain of pain and sexual sensations, as well as more
phrenia is remarkable. They may be elementary (or rudimen- incomprehensible experiences of movement, vibrations, etc.
tary), comprising tapping, banging, music or preverbal inside the body. Somatic hallucinations are frequently inextri-
whispers, mutterings and mumblings where individual words cably bound up with delusional elaboration, to the extent that
cannot be made out (but onto which the patient may still graft the relative contributions of hallucination and delusion are
content). They may comprise a single voice heard only sporad- impossible to disentangle (Berrios 1982). One patient, for
ically, through conversations involving several parties to example, stated she had a Coca-Cola bottle in her stomach,
thousands of voices babbling constantly (mass hallucinations). while another felt semen travelling up his vertebral column
The voice(s) may be recognisable as family, friends, neighbours – to his brain, where it was laid out in sheets (Sims 1988).
even the patient’s own voice – or may be total strangers.
Their source may be localised in space to ‘next door’, ‘the Like somatic hallucinations, olfactory and gustatory halluci-
walls’, ‘behind my left shoulder’ or they may seem to come nations may be simple or elaborated in delusional ways.
from all around. They may arise from locations outside the While olfactory hallucinations are not an uncommon symptom
normal perceptual sphere, e.g. ‘from Italy’ (extracampine of psychosis, their occurrence in organic brain disorders
hallucinations). They may occur spontaneously or occasion- (especially temporal lobe epilepsy) must be borne in mind.
ally are superimposed on a normal auditory stimulus such
as traffic noise or running water (functional auditory halluci- Perception of real objects may also be changed (i.e. per-
nations) or be precipitated by a perceptual stimulus in ceptual distortions). The commonest distortions are the non-
another modality (reflex hallucinations). pathological illusions, such as pareidolia (elaboration of faces
or human forms from complex backgrounds such as flames,
Traditional phenomenology distinguishes between ‘true’ carpets, wallpaper, etc.), and affect illusions, in which a height-
hallucinations and ‘pseudohallucinations’ (where the per- ened affective state leads to figures or forms being elaborated
ception is located to inner subjective space, ‘the mind’s eye’, from poorly visualised, ambiguous stimuli (e.g. seeing a noctur-
or heard ‘inside the head’). The basis of this distinction has nal stalker in wind-blown trees). Analogous phenomena also
never been well understood in English-language psychiatry occur in pathological states, but here the term ‘illusion’ tends
(Taylor 1981) ‘Pseudohallucination’ has also been erroneously to be avoided, with perceptual misinterpretation preferred,
applied to genuine hallucinations that are recognised by the though the key elements of heightened affect and diminished
patient as being not ‘real’ or which are under some degree of attention are common to both. Perhaps the most familiar
voluntary control (Hare 1973), which probably accounts for examples of this are seen in delirium, where, for example,
the mistaken view that voices ‘inside the head’ cannot be patients may misperceive the window panes as bars in a cell
‘proper’ hallucinations and hence carry a lesser diagnostic (Lishman 1998). Patients may also experience objects
inference. Many schizophrenic patients describe hearing voices changing shape, size or colour or complain that people appear
‘different’, ‘ugly’, etc (changed perception). One patient saw a
chair take on a ‘demonic’ appearance.

218

Clinical assessment: interviewing and examination CHAPTER 10

Perception may in addition be ‘heightened’ or ‘dulled’ Box 10.8
(Wing et al 1974). In heightened perception, all sensation
is experienced vividly – sounds seem unnaturally clear, loud Cognitive state examination
or intense; colours appear more brilliant or beautiful, details
of the environment seem to stand out in a particular way. In Alertness Level of wakefulness and reactivity
dulled perception, a rare symptom of depression, things seem Orientation Time: day of week, time of day, month, year
dark, grey or colourless, with uniform, flat, uninteresting Place: Building, town, county, country
texture. Taste and appetite is blunted, colours may appear Attention/ Name, age, date of birth
muddy or dirty, and sounds ugly or impure. concentration Months of year backwards or
Short-term serial 7s
Other abnormal mental contents memory Digit span
Long-term Immediate recall of name and address
A number of other phenomena cannot readily be placed within memory Delayed recall of name and address
standard MSE headings and it is reasonable to keep a section Recall of conversation, journey to hospital,
for noting features such as depersonalisation/derealisation, Executive recent news, famous events
obsessional images and impulses, d´ej`a vu, out of body experi- function Proverbs, similarities, differences
ences, bodily experiences not obviously hallucinatory in nature, Verbal fluency
including burning sensations, numbness, parasthesiae and Cognitive estimates
impulses, whether obsessional or otherwise.
After Hodges (1994).

Cognitive state examination estimate the time to within half-an-hour and any answer that
is wrong by more than this should be regarded as suspicious.
The purpose of this is primarily to confirm the presence or Mildly disorientated patients often over- or underestimate
absence of one of the two forms of global cognitive impairment, how long they have been in hospital, so as well as the standard
delirium (the acute confusional state) or dementia, though it questions, a question about this should be included. Also, it is
can also establish specific areas of neuropsychological deficit, surprising how often patients are unaware they are in hospital
such as amnesia or frontal lobe syndrome. In practice, however, (Hodges 1994), and it is also worth asking ‘What is the name
such examination represents only a ‘broad brush’ assessment of this place (building etc.)?’. In delirium, the degree of
and is not a substitute for formal neuropsychological evaluation. impairment characteristically fluctuates, and it is quite possi-
ble for orientation to be intact at the particular time the
Apart from the requirements of exams, there may in prac- patient is being examined. When an organic condition is sus-
tice be little to be gained by a formal cognitive assessment pected, it is important to continue to check this aspect of
in young individuals with an established non-organic psy- mental state, for example by re-examining later in the day
chiatric diagnosis, as it is only likely to confirm the obvious – when fatigue may make subtle deficit more evident, or the
diminished attention and impaired concentration. However, next day, when the previous day’s visit may have been
such screening should routinely be carried out in middle-aged forgotten.
and elderly patients, and in patients of any age in whom there
is diagnostic uncertainty or physical illness that can affect brain Attention and concentration
function. In these contexts, cognitive evaluation should be
combined with neurological examination and, often neglected, The suspicion of significant deficits in these areas is often
examination of the cardiovascular system. raised by the patient’s behaviour and responses during earlier
parts of the interview. For formal assessment, the ‘serial-7’s
A typical scheme for cognitive examination (after Hodges test’ is conventional. However, normal elderly people may
1994) is shown in Box 10.8. Examination begins with general make errors on this and even younger individuals require a
observations concerning behaviour and alertness, going on to certain level of numerical accomplishment to perform it
cover orientation, attention and concentration, memory and accurately and with reasonable speed. Errors must therefore
executive function. This can be complemented by carrying be interpreted. A simple alternative is asking the patient to
out the Mini-Mental State Examination (Folstein et al 1975), recite the months of the year backwards.
a ubiquitous practice in psychogeriatrics, which provides a
crude but useful measure of the severity of cognitive Memory
impairment. It is important to bear in mind that the MMSE
can be relatively unimpaired in the presence of specific types ‘Digit span’ is a standard test of short-term memory, the store
of cognitive disorder, especially executive dysfunction. which holds a limited amount of information for periods up
to 30 seconds. ‘Immediate recall’ of a name and address may
Orientation also reflect partly the operation of short-term memory.
Delayed recall of the same name and address tests long-term
Orientation tends to be compromised progressively, with only memory, the store of all information that needs to be held
‘time’ affected in mild disorder, followed by ‘place’, and ‘per- for more than 30 seconds.
son’ only disrupted in severe cases. Most people are able to

219

Companion to Psychiatric Studies

A useful test of the ability to retain new information Box 10.9
(anterograde long-term memory) is to find a topic of interest,
such as the patient’s family or a recent holiday, and then tell The Cognitive Estimates Test
them about one’s own family or recent holiday. The patient
can then be asked about this later. Retrograde long-term mem- 1. What is the height of the Post Office Tower?
ory can be assessed by beginning with open-ended questions,
such as ‘What important events have been in the news >1500 ft 3 <60 ft 3
recently?’, then asking about a standard list of famous events. 2
Although there is no psychological support for the time- ¼ 1500 ft 2 ¼ 60 ft 1
honoured clinical distinction between ‘recent’ and ‘remote’ 3
long-term memory, patients with retrograde amnesia usually >800 ft 1 <100 ft 2
show a temporal gradient of recall, with more impairment of
recent and less impairment of remote events. 2. How fast do race horses gallop? 3
2
Executive function >50 mph 3 <9 mph 1
3
Most of the commonly advocated ‘bedside’ tests of executive ¼ 50 mph 2 <15 mph 1
function are insensitive and will only pick up abnormality in 3
gross cases. Furthermore, patients with frontal lobe lesions >40 mph 1 1
are notorious for performing normally on some, or even most, 3
executive tests, while disastrously failing others. 3. What is the best paid job or occupation in Britain today? 1
3
One formal test that can be easily applied in the clinical Manual workers 2
setting is ‘verbal fluency’ – asking the patient to generate as 1
many items as possible over a minute in a set category, such Car workers (or other special groups of well-paid blue- 3
as animals, words beginning with ‘S’, or items that can be 1
bought in a supermarket. Another is the Cognitive Estimates collar workers)
Test (Shallice et al 1978). This requires the subject to give 1
educated guesses to questions they are unlikely to know the Professional (up to and including Prime Minister)
exact answers to – like, ‘How fast do race horses gallop?’,
and ‘What is the largest object normally found in a house?’. 4. What is the age of the oldest person in Britain today?
The full version of this is shown in Box 10.9. Sometimes the
answers can be revealing: when asked what the population of >115 3 <103
Britain was, one patient with suspected frontal dementia
replied ‘About a thousand’. When the interviewer pointed ¼ 115 2
out that there were probably a thousand people working in
the hospital they were in, she responded ‘All right – two ¼ 114 1 ¼ 103
thousand’ (R. Dolan, personal communication).
5. What is the length of an average man’s spine?
Tests of abstract thinking – ‘proverb interpretation’/
’similarities–differences’ – are popular with psychiatrists and >5 ft 3 <1 ft 6 in
can provide clues to cognitive difficulty, such as early demen-
tia, before orientation and memory impairment can be >4 ft 2
detected by routine bedside methods. However, proverb inter-
pretation is highly dependent on educational and cultural ¼ 4 ft 1 ¼ 1 ft 6 in
factors and the choice should be restricted to simple, well-
known examples. Many people (including some doctors!) do 6. How tall is the average English woman?
not know the meaning of: ‘It’s an ill wind that blows nobody
any good’. Assessment should always be prefaced by: ‘Have >6 ft 3 <5 ft 2 in
you heard the old saying. . . ?’, as ignorance is the surest way
to ‘fail’ proverb interpretation! ¼ 5 ft 11 in; 6 ft 2

Perseveration – the inappropriate repetition of words, ideas ¼ 5 ft 9 in; 5 ft 10 in 1 ¼ 5 ft 2 in
or themes – is a presumptive dysexecutive sign which has been
considered to be a feature of schizophrenia, and was included 7. What is the population of Britain?
by Andreasen (1979) as one of her abnormalities of thought,
language and communication. Whether or not it is genuinely >1000 million 3 <2 million
a symptom of schizophrenia or merely a reflection of the
cognitive impairment found in this disorder, it is certainly a >500 million 2 <3 million
feature of organic disorders, particularly frontal lobe disease.
Two striking examples of perseveration in patients with frontal ¼ 500 million 1 <10 million
dementia are shown in Fig. 10.2.
8. How heavy is a full pint bottle of milk?

>3 lb 3 <1 lb

¼ 3 lb 1 ¼ 1 lb

9. What is the largest object normally found in a house?

<Carpet 3

Carpet 2

Piano, sideboard, settee 1

10. How many camels are there in Holland?

Very large number 3 None

From Shallice & Evans (1978), with permission.
Responses are scored in terms of extremeness from 0 (normal) to 3 (very extreme).
The signs < (less than) and > (greater than) should be interpreted strictly, not as
approximate values. 5th percentile cutoff for abnormality in 84 normal adults aged
17–68 with a wide range of estimated IQ ¼ 12.75.

Insight

Traditionally the mental state examination ends with a state-
ment about insight, usually gauged simply on the patient’s
ability to accept they are ill. Recent research has demonstrated
that insight is a considerably more complex construct than
this (David et al 1995) and what meaningful information
rough bed-side appraisals can provide must be questionable.
Nonetheless, it is useful to explore at this point the patient’s
willingness to accept medical recommendations, response to
which may in some circumstances have major implications
for immediate management.

220

Clinical assessment: interviewing and examination CHAPTER 10
B

A

Fig. 10.2 Perseveration in fronto-temporal dementia. (A) The patient was asked to write a sentence about the weather, and then to write
something about her stay in hospital. She continued to reproduce the original sentence despite repeated instructions to write something
different. (B) The patient was asked to draw as many different designs as possible made up of four lines. Reproduced with permission from
Snowden et al (1996).

Formulation/summary a pr´ecis of significant phenomenology elicited during the inter-
view (only those features present, not absent ones), its evolu-
A traditional part of diagnostic assessments, especially in tion and context; the third, key mental state disorders
exams, has been the formulation. The ‘exam formulation’ has evident from formal examination. Differential diagnoses are
been seen as a way of assessing a candidate’s ability to focus too important to be ‘summarised’ and any proposed treatment
thought on the essentials and prioritise issues. plan deserves full exploration, so these should not be included.
Unlike a ‘formulation’, a ‘summary’ should not encroach on
Unfortunately, ‘formulation’ means different things to speculative dynamics.
different people and comes with heavy theoretical baggage.
Its origins lie in psychoanalytical practice, where its functions Proposals from some examination boards might now require
are more explanatory than descriptive. This approach is illu- trainees to present a ‘psychodynamic formulation’ or a ‘beha-
strated in the views of the American analyst, Karl Menninger, vioural formulation’ of a case. This is to be regretted. Such
who considered individuals and their predicaments unique and skills are for those in specialist training to acquire and have
incapable of classification. He, amongst others, rejected formal as much relevance to the generalist as a formulation based on
taxonomy in favour of individualised formulations that encom- the patient’s serotonergic receptor status!
passed every facet of the patient’s being and life experience
that had brought them to a symptomatic state. This concept Summary
goes far beyond what is expected of the average examinee –
or clinician! A 51-year-old happily married professional lady with a family history of affec-
tive disorder, referred by GP with a 6 month þ history of progressive, persis-
The concept of ‘formulation’ is too muddled for the exami- tent depression of mood, associated with anhedonia, loss of emotional
nation room. If the skills of ‘extraction’ and ‘prioritisation’ resonance, diurnal variation, early waking with daytime fatigue, poor appe-
are deemed important, the same aim can be achieved by con- tite and weight loss, and impaired concentration. She presents as mildly
sidering a ‘summary’ statement, which is entirely descriptive agitated and objectively depressed with ideas (but not delusions) of guilt,
and has no theoretical implications. This should contain no low self-esteem and hopelessness associated with passive ideas of self-
more than three (occasionally four) brief sentences – the first, harm. She relates the onset of her illness to enforced redundancy but
comprising basic demographic information, mode of referral admits to declining performance in the preceding months.
and presenting complaint, including its duration; the second,

221

Companion to Psychiatric Studies

Conclusions disorders, it should be offered. However, many patients, espe-
cially those presenting with psychotic disorders, may not wish
A formal psychiatric interview should have a formal con- a diagnosis at an early stage. This is perhaps as well, because all
clusion. This is not so much a statement or question – and is that one may strictly be able to offer is a set of differential pos-
certainly not merely the passing on of a prescription! This is sibilities, which may be interpreted as lack of competence!
a section in itself, in which the doctor will, for initial contacts, Relatives may of course have profound interest in the outcome
provide some discussion of the problems as he/she sees them, of the interview, and while every effort should be made to
including therapeutic recommendations, or for follow-up con- inform them – including the limitations imposed by cross-
tacts, some assessment of progress and recommendations sectional evaluations – this must be within the bounds of
regarding on-going management. It is therefore important patient confidentiality.
in busy out-patient settings that information gathering allows
sufficient time for this crucial exchange. In appraising material from psychiatric interviews there are
two ‘rules’ worth applying. The first, to paraphrase Kendell
Ultimately, the examiner determines the structure and (1975), is what might be called the ‘Rule of Intersubjective
duration of most psychiatric interviews, but with skill and a Certifiability’. While the central role of empathy in the con-
competent exercise of control over process elements, they duct of the psychiatric interview has been emphasised, this
should nonetheless have allowed the patient ample opportu- must be rejected when it comes to evaluating the material
nity to present the issues of importance to them. This cannot comprising the MSE (with the exception of depth of emo-
however be guaranteed, and especially if the patient has pre- tion). As much of the mental state exam as possible must be
conceptions that have not been fulfilled, they may still feel submitted to objective appraisal, such that other examiners
there are significant areas that remain unaddressed. It is there- would reach similar conclusions to oneself. Not only must
fore often useful after concluding the above to give them the empathy be banned from the diagnostic process, so too should
opportunity to note this, for example: intuition, identification and any of the interminable dynamic
‘insights’ so beloved of experts with Sunday supplement
‘We’ve covered a lot of issues during the interview. Are there any expertise. Possible roads to the disorder should be addressed
other issues we haven’t covered that you feel it would be important after the disorder has been diagnosed, and that may include
for me to know about?’ psychodynamic ones, but these principles should have no place
in the diagnostic process itself.
Alternatively, a useful signal to the patient that the informa-
tion part of the exercise has been completed is to say: The second rule might be called the ‘Rule of Counter-
Intuition’. It is everyday experience that, in all of us,
‘I’ve asked you a lot of questions. Are there any questions you would unpleasant life circumstances produce unpleasant emotional
like to ask me?’ consequences. However, what may be intuitive in relation to
everyday life may not be so in relation to psychiatric disorder.
Most individuals will not take this as an opportunity to ask It is well established that disordered mental states can them-
about what’s to come – i.e. treatment issues – but will relate selves produce adverse life events – that is, that life events
it back to what has been before, to whether or not they are may bear a ‘dependent’ or an ‘independent’ relationship to
satisfied with what you have asked about. Thus, it can act as psychiatric illness. The situation where psychiatrists have satis-
a useful way of linking what has been established with what fied themselves that an event bears a causative relationship to
you will then be recommending. the symptomatology they have elicited is one in which the
principle of refutation should be invoked. Rather than accept-
A final and useful question relates to patient expectations. ing that ‘The event caused the illness’, the question ‘Could the
One of the most powerful determinants of efficacy for any illness have caused the event?’ must be put forward for consid-
psychiatric intervention is the patient’s belief in it. One will eration. It is surprising how often ‘intuitive’ impressions fail to
experience opposition, if not hostility, if one’s recommenda- be sustained!
tions are predominantly pharmacological when the patient’s
expectations were psychotherapeutic – and vice versa. Aware- The reader will have noticed the recurrent use of ‘recom-
ness of the patient’s expectations is important in helping to mendation’ in this chapter and when it comes to proposing
frame recommendations sensitively, especially when they treatments, this word cannot be ‘recommended’ too strongly!
may not gel with what was anticipated. One might ask: While in psychiatric practice, ‘recommendations’ must some-
times be imposed, in the majority of instances the patient
‘In what way do you feel I might help you?’ must be the willing recipient of professional advice. With its
connotations of professional authority yet personal choice,
or it is the essence of what clinical practice is about.

‘What do you feel might be the best way forward?’ Basic neurological examination

In concluding the interview, the psychiatrist’s task is to pro- Patients with neurological symptoms are often viewed with
vide an ‘assessment’, not a ‘formulation’ (too theoretically bur- trepidation by doctors, with the neurological examination an
dened) nor a ‘summary’ (too stark), a task relevant to both impenetrable ritual understood only by neurologists, an aura
initial and follow-up interviews, though slightly different for of mystery they sometimes encourage. The nervous system is
each. The patient may wish a diagnosis, and if available with
reasonable probability of accuracy, as with many non-psychotic

222

Clinical assessment: interviewing and examination CHAPTER 10

probably the most complex and poorly understood aspect of as noted above: the presenting complaint and its evolution; the
human function, producing symptomatologies that can be past medical history; current and previous medications (pre-
as vague as they are varied. Many neurological diseases are scribed and otherwise); and social and family history.
relentlessly disabling and untreatable, adding fear and anxiety
to their presentations and worry in doctors that they may be As many neurological and psychiatric disorders are chronic
missing something. Yet many neurological symptoms are benign or relapsing/remitting in nature, past histories are crucial. This
and do not reflect organic pathology. can be problematic, as patients often forget or dismiss previous
symptoms, particularly if they resolved and seemingly have lit-
The division between neurology and psychiatry is in many tle relevance to current complaints. Even if recalled, previous
ways artificial but likely to be maintained so long as training symptoms are often inaccurately placed in their timing or in
schemes are kept separate and it remains the case that many the sequence of events. Sometimes they may be actively con-
psychiatrists, when confronted with potential neurological cealed. These pitfalls may be averted by reviewing contempo-
disease and the complexities of neurological examination, feel raneous medical records and although hospital practitioners
uneasy. The purpose of this section is not to turn psychiatrists rarely have easy access to primary care records, obtaining these
into neurologists but simply to try and demystify the approach can at times be invaluable, even diagnostic. Simply knowing
to neurological assessment and diagnosis. A secondary aim is that the patient in front of you possesses a bulky set of primary
to allow psychiatrists to feel comfortable about the limited care records reflecting referral to several different depart-
expectations possible to them from a brief, standard neuro- ments and hospitals with a variety of ill-defined symptoms
logical examination. can immediately alert one to the possibility of functional
or factitious disorder. Alternatively, one may identify long-
History taking forgotten symptoms, such as a brief episode of reduced visual
acuity in one eye in someone presenting years later with gait dis-
While diagnosis is the key aim of neurological assessment, neu- turbance, indicating a probable diagnosis of multiple sclerosis.
rologists – like psychiatrists – appreciate that many symptoms
do not indicate disease and that some patients defy accurate By the end of the history taking, you should have a differen-
diagnosis, certainly initially and perhaps indefinitely. tial diagnostic list in your head. If you do not, the examination
is unlikely to help.
One should make no apology for emphasising an ‘obvious’
fact – in neurological assessment, there is no substitute for an The examination
accurate and well-interpreted history. Without an accurate
history a neurological diagnosis is usually impossible, and many Professional anxiety about the neurological examination can
neurological diseases can be diagnosed on the history alone. lead doctors to either omit it or, worse still, perform it incom-
This will be familiar to psychiatrists but as in psychiatry, his- petently but a basic neurological examination should be within
tory taking skills (as opposed to examination techniques) are the grasp of all practising clinicians. There is, however, little
rarely taught and published material is scarce (Thrush 2002). point in non-neurologists attempting to learn a ‘full’ neuro-
logical examination: they are likely to get little opportunity
There is no difference in taking a history from patients with to perform this regularly and when they do, inexperience is
neurological symptomatology than from those with any other likely to throw up false positive as well as negative findings.
types of presentation, though nervous system symptoms, ‘psy- Examination findings are subject to such variability that even
chiatric’ or ‘neurological’, often require more patience to pur- experienced neurologists find themselves frequently disagree-
sue. All history taking is an active process and while ing over ‘hard’ signs, like plantar responses.
checklists or protocols may be helpful for students or in diag-
nosing certain easily identified and specific disease processes, Many practitioners place more faith in their abilities in
such as asthma attacks or stroke, they are hopelessly inade- undertaking an accurate neurological examination than in their
quate when pursuing less well-defined histories of the sort history-taking skills, with the result that they are dispropor-
not uncommonly associated with neurological presentations. tionately influenced by isolated neurological ‘signs’, regardless
No two histories will be the same. One must be constantly lis- of what the history is telling them. A reasonable neurological
tening, thinking and updating the potential differential diag- view would be that if you suspect neurological disease at the
noses as one elicits the story, as this will open up new end of the history, you should refer the patient to a neuro-
avenues of enquiry. In a further similarity with psychiatry, logist, regardless of what you may or may not have found on
information from witnesses or third parties is often invaluable, examination. Similarly, you can afford to ignore an isolated,
though is something that may be best obtained privately. With potentially unreliable ‘sign’ if it makes no sense in the clinical
neurological presentations, family members or other witnesses setting. As a result, psychiatrists need only be competent in a
may be embarrassed about discussing symptomatology in front very basic neurological examination (see Box 10.3).
of the patient, for example where cognitive problems pre-
dominate, or where relevant family histories have remained A logical approach
concealed or unspoken.
A basic neurological examination does not require detailed
A neurological history can usefully begin with three stan- neuroanatomical knowledge. It is sufficient to consider the
dard questions – age, handedness (a peculiar neurological nervous system in three sections: cognitive apparatus, cranial
obsession) and occupation (past and present) – and thereafter nerve function and limb function. Most neuropsychiatric syn-
can be pursued within the same sort of traditional framework dromes, or neurological diseases presenting to psychiatrists

223

Companion to Psychiatric Studies

will, with few exceptions, involve the central nervous system Nystagmus should be tested for in both the horizontal and the
(i.e. brain and spinal cord), so one might particularly con- vertical planes of gaze. In interpreting eye signs one must be
centrate on cognitive disturbance and signs of upper motor aware of nystagmoid jerks which are common if the eyes
neuron pathology. are forced beyond their natural sweep and indicate a faulty
examiner, not a faulty patient!
1. Observation
The interpretation of an eye movement disorder can be
Careful observation from the ‘end of the bed’ is the first stage extremely difficult, even for specialists. Simple observation
of any examination, followed by observation of the patient of the face is likely to be more instructive than further detailed
walking an adequate distance down an uncluttered corridor. testing of facial sensation or movements – the immobile face
It is surprising how much information this can reveal (see of parkinsonism with reduced blinking and drooling; ptosis
Box 10.3). Testing tandem gait (heel–toe walking) and for suggestive of myopathic or mitochondrial disorder; tics, bleph-
Romberg’s sign (a test of proprioception) are common sources arospasm or other dyskinesias. Lower cranial nerve dysfunction
of false-positive errors and cannot be recommended for non- is most commonly manifest as dysarthria and may be accompa-
specialists. If the patient’s standard gait appears normal, this nied by emotional lability. In general, however, interpretation
can be accepted as a reasonable indication that legs and cerebel- of cranial nerve deficits is best left to experts and for the
lum are grossly intact. Finally, observing movements when non-specialist it is sufficient to take observed abnormality as
undressing/dressing can also be instructive, though the impor- an indication for specialist referral.
tance of having a chaperone present cannot be overemphasised.
4. The limbs
2. Cognition
Gross motor power in the legs will already have been assessed
This has already been dealt with and only a couple of issues by observing the patient walking. A common functional sign is
of importance in the neurological examination will be re- the ability of the patient to walk (albeit oddly) yet to have
presented, especially from the perspective of ‘bedside’ testing virtually no power in the legs when examined on the bed,
as opposed to formal neuropsychological evaluation. Although e.g. they appear unable to lift either leg off the bed despite
clinicians enjoy eliciting the ‘primitive’ reflexes – e.g. the pout, having somehow got there in the first place!
snout, rooting and palmar-mental responses – in practice, these
reflexes only infrequently add significantly to cognitive testing. The upper limbs may be grossly assessed in similar fashion
by asking the patient to raise both arms above their head (gross
• Consciousness. This is a more complex issue than most motor power) then stretch them out in front with palms
clinicians appreciate or bedside testing allows for, but when uppermost – initially with eyes open, then with eyes closed.
required the Glasgow Coma Scale is a widely utilised One is looking for any ‘drift’ of the outstretched limb, often
international standard. It should be remembered that it is a a subtle early sign of pyramidal disease. Lastly, with eyes open,
three–domain scale and findings should be presented as the patient can be asked to wiggle their fingers in ‘piano-
such, rather than as a composite score, which means little. playing’ fashion, another subtle measure of pyramidal or extra-
pyramidal dysfunction, when fine finger movements are
• Orientation. affected first. Interpretation of findings must bear in mind
• Attention. As noted above. that the dominant hand will always be more dextrous than
• Memory. the non-dominant one. Other fine motor tasks that can be
• Executive function. In addition to points noted above, usefully observed include hand-writing and doing up/undoing
buttons.
testing motor sequences with the Luria ‘three hand
position’ test can be useful. Conventional teaching would dictate that the standard neu-
• Perception. In a neurological context, this has a slightly rological should include a detailed assessment of resting tone
different meaning than in psychiatry. Here, what’s being and power in individual muscles. However, it must be doubtful
tested is neurological substrate, not percepts themselves. how useful these exercises actually are to the non-specialist.
This can be assessed with clock drawing tasks or line Assessing tone is difficult unless it is markedly increased, in
bisection. which case there will be other, more obvious signs evident
• Praxis. This is difficult to assess unless marked, but can be (such as exaggerated tendon reflexes and extensor plantars),
roughly evaluated by testing the patient’s ability to mime and unless one possesses a working knowledge of the individ-
actions, such as lighting a cigarette, brushing their teeth or ual nerves and roots which supply individual muscles there
combing their hair. would seem to be little point in testing them. It is far more
important to decide whether or not the patient is weak (as
3. Cranial nerves opposed to stiff, apraxic, in pain or subject to any of the other
impairments which they may misinterpret as weakness), and if
Detailed assessment of all 12 pairs of cranial nerves is unneces- so, whether or not that weakness appears consistent. Simply
sary in a psychiatric context, but certain signs are more com- observing gait, the ability to lift their arms above the head
mon than others and worth looking for, particularly disorders and fine finger movements will usually identify significant
of eye movements. Eye movement disorders occur in a number weakness.
of neuropsychiatric disorders and asking the patient to follow
your finger in an ‘H’ pattern with the head fixed will identify Testing upper (biceps, supinator, triceps) and lower (knee,
gross abnormalities. The eye movements should be smooth ankle) limb reflexes should however be done, simply to see if
and coordinated, without extra movements such as nystagmus. they are exaggerated (which would suggest an upper motor

224

Clinical assessment: interviewing and examination CHAPTER 10

neuron lesion) or not. But it should be remembered that disorder, refer the patient to a neurologist, indicating why you
absence of an isolated reflex in non-specialist hands is as often think this is the case. In making a referral, remember that the
due to poor technique as to any underlying pathology. Inter- neurologist (a good one, at least) will be more interested in your
pretation should also be informed by the fact that deep tendon history than your examination findings. Keep your own exami-
reflexes may appear brisk without indicating pathology, espe- nation simple for it is from simple observation and testing that
cially in anxious and tense patients. The plantar response the most reliable information will emerge; and be prepared to
should only be interpreted as abnormal (i.e. extensor) when disregard signs which you find but which make no sense, as they
it is unmistakable and repeatable and should not be confused are more than likely false positives. And remember that plenty
with a withdrawal response, commonly found in people who of neurological disorders produce normal examinations.
dislike having their feet ‘interferred with’!
Epilogue
As with any part of the examination, it is important not to
fall into the trap of overinterpreting isolated signs. A patient Most of the aforegoing has been focused on the doctor–patient
with normal gait and otherwise entirely normal legs apart from interaction. It is of course axiomatic that the quality of mate-
an isolated extensor plantar probably has an intact motor sys- rial relevant to any patient’s case may be greatly enhanced by
tem, while in the patient with a limp where it is suspected that the addition of third party information. Where possible – and
the reflexes are brisk in the limping leg, a ‘probable’ extensor with the patient’s agreement – third party sources should be
plantar on that side is altogether more suspicious of central considered and if relevant sought, though the interview princi-
nervous system pathology (which may or may not be relevant ples will be the same as those noted above.
to the presenting complaint).
Despite the many advances in practice in the post-war
Summary period, it remains the case that one of the most potent thera-
peutic tools available to the psychiatrist is the ‘Dr.’ in front of
The concluding message is simple – dedicate your time to history his or her name. This still embodies powerful expectations on
and pursue all the witnesses available, including from medical the part of most patients and those who carry the title would
records, especially in difficult cases. If you suspect a neurological do well to respect it.

Further reading Hawkes, C.H., 1991. How to perform a rapid Memon, A., Bull, R. (Eds.), 2000. Handbook of
neurological examination. Hospital Update the Psychology of Interviewing. John Wiley
Carlat, D.J., 2004. The Psychiatric Interview: 17, 125–131. and Sons, Chichester.
A Practical Guide, second ed. Lippincott
Williams and Wilkins, Philadelphia. Hawkes, C.H., 1997. Diagnosis of functional Myersough, P.R., Ford, M.J. (Eds.), 1996. Talking
neurological disease. Br. J. Hosp. Med. 57, with patients: keys to good communication,
Fish, F.J., Casey, P.R., Kelly, B., 2007. Fish’s 373–377. third ed. Oxford University Press, Oxford.
Clinical Psychopathology: Signs and
Symptoms in Psychiatry. RCPsych Kendell, R.E., 1975. The Role of Diagnosis in Shea, S.C., 1998. Psychiatric Interviewing,
Publications. Psychiatry. Blackwell Science, Oxford. second ed. Saunders, Philadelphia.

Frith, C.D., 1992. The Cognitive Neuro- McKenna, P.J., 2007. Schizophrenia and Related Tasman, A., Kay, J., Lieberman, J.A., First, M.B.,
psychology of Schizophrenia. Erlbaum, Syndromes, second ed. Routledge. Maj, M. (Eds.), 2008. Psychiatry, third ed.
Hillsdale, New Jersey. John Wiley and Sons, Chichester.

References

Andreasen, N.C., 1979. Thought, language and psychotic illness: cross-sectional and Hare, E.H., 1973. A short note on pseudo-
communication disorders: 1. Clinical longitudinal associations. Br. J. Psychiatry hallucinations. Br. J. Psychiatry 122, 469–476.
assessment, definition of terms and 167, 621–628.
evaluation of their reliability. Arch. Gen. Hodges, J.R., 1994. Cognitive Assessment
Psychiatry 36, 1315–1321. Dhossche, E., Ferdinand, R., van der End, J., for Clinicians. Oxford University Press Oxford.
Hofstra, M.D., Verhulst, F., 2002. Diagnostic
Berrios, G.E., 1982. Tactile hallucinations: outcome of self-reported hallucinations in a Jaspers, K., 1963. General Psychopathology
conceptual and historical aspects. J. Neurol. community sample of adolescents. Psychol. (J. Hoenig and M.W. Hamilton, Trans.).
Neurosurg. Psychiatry 45, 285–293. Med. 32, 619–627. Manchester University Press, Manchester.

Bleuler, E., 1911. Dementia Praecox or the Enoch, M.D., Ball, H.N., 2001. Uncommon Kraepelin, E., 1913a. Manic-Depressive Insanity
Group of Schizophrenias (J. Zinkin, 1950, Psychiatric Syndromes. Arnold, London. and Paranoia (R.M. Barclay, 1921, Trans.).
Trans.). International Universities Press, Livingstone, Edinburgh.
New York. Folstein, M.F., Folstein, S.E., McHugh, P.R.,
1975. Mini-Mental State’: a practical method Kraepelin, E., 1913b. Dementia Praecox and
Cameron, N., 1938. Reasoning, regression and for grading the cognitive state of patients Paraphrenia (R.M. Barclay, 1919, Trans.).
communication in schizophrenics. Psychol. for the clinician. J. Psychiatr. Res. 12, Livingstone, Edinburgh.
Monogr. 50, 1–34. 189–198.
Lishman, W.A., 1998. Organic Psychiatry: the
David, A., van Os, J., Jones, P., Harvey, I., Hamilton, M., 1984. Fish’s Schizophrenia, Psychological Consequences of Cerebral
Foerster, A., Fahy, T., 1995. Insight in third ed. Wright, Bristol. Disorder. Blackwell Scientific Publications,
Oxford.

225

Companion to Psychiatric Studies

Lund, C.E., Mortimer, A.M., Rogers, D., Rogers, D., 1992. Motor Disorders in Psychiatry. Taylor, F.K., 1981. On pseudohallucinations.
McKenna, P.J., 1991. Motor, volitional and John Wiley and Sons, Chichester. Psychol. Med. 11, 265–271.
behavioural disorders in schizophrenia. I:
Assessment using the modified Rogers scale. Rogers, D., 1985. The motor disorder of severe Thrush, D., 2002. How to do it: take a good
Br. J. Psychiatry 158, 323–327. psychiatric illness: a conflict of paradigms. history. Pract. Neurol. 2, 113–116.
Br. J. Psychiatry 47, 221–232.
McKenna, P.J., 2007. Disorders with overvalued Wing, L., Shah, A., 2000. Catatonia in autistic
ideas. Br. J. Psychiatry 145, 579–585. Shallice, T., Evans, M.E., 1978. The involvement spectrum disorders. Br. J. Psychiatry 176,
of frontal lobes in cognitive estimation. 357–362.
Mullen, P.E., 2007. A modest proposal for Cortex 14, 294–303.
another phenomenological approach to Wing, J.K., Cooper, J.E., Sartorius, N., 1974.
psychopathology. Schizophr. Bull. 33, Sims, A., 1988. Symptoms in the Mind. Balliere The Measurement and Classification of
113–121. Tindall, London. Psychiatric Symptoms. Cambridge
University Press, Cambridge.
Owens, D.G.C., 1999. A Guide to the Snowden, J.S.S., 1996. Fronto-temporal Lobar
Extrapyramidal Side-Effects of Antipsychotic Degeneration. Churchill Livingstone,
Drugs. Cambridge University Press, New York.
Cambridge.

226

Clinical psychopharmacology 11

David Cunningham Owens

Introduction However, 1949 and 1960 form the ‘bookends’ to a noteworthy
era where almost every year was marked by some new advance.
Ancient societies knew more about ‘mind-altering’ substances In 1950, Paul Charpentier synthesised chloropromazine (as it
than we give them credit for. From this perspective, the dis- was originally called), its place as ‘foundation drug’ of the new
cipline we call ‘psychopharmacology’ is an infant, the term era confirmed by Delay and Deniker in 1952. The same year,
attributable to the American pharmacologist, David Macht, Selikoff noted the mood elevating effects of iproniazid, a
in 1920. While there is inevitably an overlap with neurophar- synthetic analogue of the antituberculous agent, isoniazid, while
macology, the aim of drug treatments in clinical practice is in 1954 the rauwalfia alkaloid, reserpine, long a tool in Ayurvedic
to bring about beneficial changes in disordered mental states medicine, was introduced into Western practice as an anti-
and the prime responsibility for clinicians is acquisition of depressant by Nathan Kline. The same year saw the synthesis
expertise in drug use in patients. Hence, there is merit in of methylphenidate and 1955 brought the first ‘tranquilliser’,
maintaining the distinction between neuro- and clinical meprobamate, within 2 years the most prescribed drug in
psychopharmacology. America. Also in 1955, the iminodibenzyl derivative of the
renamed chlorpromazine was evaluated for antipsychotic effi-
The whole undertaking of producing, studying and adminis- cacy though, disastrously, seemed to produce only mania.
tering ‘psychotropic’ drugs (i.e. compounds whose primary However, in 1957 the Swiss psychiatrist Roland Kuhn, demon-
target is, via modifications to brain function, mental state) strated its efficacy in ‘vital depression’ and imipramine became
is viewed with suspicion in some quarters. It is curious how the first tricyclic antidepressant with its launch in 1959. Also
substances of which we have scientific knowledge are so fre- in 1957, Randall discovered the behavioural effects of the 1,4-
quently rejected in the clinic while those of which we know benzodiazepines while Klein introduced what were technically
little are accepted without question at the bus-stop outside! the first antidepressants, based on the earlier observations with
Doctors must be vigilant to the compliance issues that are antituberculous agents. In 1958, Zeller demonstrated that these
the most frequent causes of non- or poor response. Further- mood-elevating effects were due to monoamine oxidase inhibi-
more, even the most ardent psychopharmacologist must admit tion, thereby providing the basis for the most durable biological
that the boundaries of their ‘therapies’ are often narrower than theory of mood disorders. Also in 1958 Paul Janssen, working
they like to think. Psychotropic drugs are rarely ‘insulins’ to (in a rented garage) on the effects of heat on organic molecules,
psychiatric ‘diabetes’. Undoubted benefits are usually relative converted pethidine (meperidine) to norpethidine to haloperi-
and do not absolve the psychiatrist from devising comprehen- dol, while Petersen in Denmark synthesised the thioxanthenes.
sive treatment plans that reflect the complex interplay of fac- The Swiss company Wander, collaborating with the Munich
tors contributing to the development of psychiatric disorders. group under Hans Hippius in search of new antidepressants,
synthesised clozapine in 1959, with the first year of the new
The introduction of psychotropics is often described as ‘seren- decade, and Cohen and Tobin’s demonstration of the anxiolytic
dipitous’ (i.e. a ‘fluke’!) but a lineage of quality science preceded actions of chlordiazepoxide, bringing to an end this remarkable
many and clinical application was sometimes suggested by exam- ‘Golden Age’.
ples of remarkable observational skill. The term ‘empirical’ is
more accurate. It began in 1949, when John Cade published It is not only ‘Golden Ages’ that are limited. So too is
the first account of the antimanic and mood-stabilising proper- empiricism and much of what followed was largely derivative
ties of lithium salts, though it was some years and some distance until the introduction of compounds developed from theory,
away before these observations were brought to clinical fruition.

ã 2010, Elsevier Ltd.
DOI: 10.1016/B978-0-7020-3137-3.00011-5

Companion to Psychiatric Studies

beginning with new antidepressants in the 1980s, then antipsy- pharmacological agents classified on the basis of an adverse
chotics the following decade. Increasing clinical experience action and in recent years it has fallen from use.
however is forcing us to question the accuracy of the theories
and the assumptions drawn from them! It may, however, be premature to see things ‘in this light’.
The unique psychotropic actions of chlorpromazine were not
This chapter aims to provide an overview of the major drug spotted by psychiatrists (who saw only sedation) but by a sur-
treatments currently used in psychiatric practice, though the geon, Henri Laborit, who noted an essentially affective change
frame of reference is necessarily general adult psychiatry. (‘detachment’). And he noted it after single – and low – doses!
The emphasis is on clinical application – but from a foundation Also in 1955, Fabing and Cameron suggested an alternative
of pharmacology, as this should provide the starting point for class nomenclature – ‘ataraxics’: drugs producing a state
therapeutics, not an ‘additional extra’. It is no longer accept- of ‘equanimity’. Opposition to chlorpromazine from a largely
able for psychiatrists to ‘pick up’ psychopharmacology ‘on analytically orientated profession was intense, something now
the hoof’, or from the British National Formulary – or worst forgotten, so it was necessary to present a strictly medical
of all, from the marketing literature for commercially available theory of action to cement their place. ‘Ataraxy’ lost out,
products. Reference to ‘practice standards’ is necessary nowa- only surviving in the term ‘specific sedation’ still utilised in
days. However, it is important to be sensitive to the empiri- Continental Europe. Ataraxy however defined the class in
cal element behind much of what comprises present terms of a unique mental state property and it is unfortunate
‘guidelines’. Psychopharmacological research remains plagued it was lost for its mechanism – in particular whether it is dopa-
by methodological limitations too infrequently acknowledged minergically mediated – remains unknown. Could it be how-
by those who undertake it, raising qualitative issues readily ever that after even single doses these drugs do indeed
masked by considering solely quantitative ‘bottom lines’. ‘grasp’ (if not ‘seize’) the nervous system in what is their
‘Guidelines’ are just that – advice to provide boundaries fundamental class action? This must await future elucidation,
against the excesses of ill-informed or maverick practice but the practical point being that these drugs can produce
that still leave much room for the application of true expertise, clinically subtle subjective effects from earliest exposure.
something that is the clinician’s responsibility to acquire. If the
following strays from guidelines to which the reader is bound, The actions of the class are pharmacologically specific, not
one is encouraged to view this as part of an on-going debate, merely ‘barbiturate-like’ and secondary to sedation, so any
legitimate and necessary. Most importantly, what follows reference to them being ‘tranquillisers’, major or otherwise,
should not be seen as something for passing exams to there- is erroneous. They are however clinically (i.e. syndromally)
after lie neglected but as something which defines the terrain non-specific effects in that they can be demonstrated on psy-
over which one will become a life-long traveller, guided by chotic symptomatology regardless of diagnosis. It is therefore
primary sources. equally inaccurate to refer to them as ‘antischizophrenics’.
The current view is that drugs clinically effective against
Since the last edition of this book, an EU directive has psychotic symptomatology should be classified descriptively,
substituted British Approved Names (BAN) with the ‘recom- on the basis of this primary action – i.e. as ‘antipsychotics’.
mended International Non-proprietary Name’ (rINN). This
has led to a number of minor and a few major changes in drug The battle for class terminology was heated half a century
nomenclature. The rINNs will be used here but if the author ago. The impending one over group terminology is just warming
slips, difficulty in erasing 30 years of practice is to blame! up! It was evident from the start that clozapine was ‘different’ –
‘atypical’ in its clinical effects and, by implication, ‘novel’ in its
It goes without saying that in providing an overview, we are mode of action. In the early 1970s a number of drugs were
here using an atlas, not ‘sat-nav’. It is possible that some of the put forward to share these accolades (Owens 2008) but as each
following will be superseded or qualified before the present re-joined the pack clozapine was left alone, its isolation empha-
volume comes to print. sised by a dangerous reputation and limited use. Tremendous
impetus was given to the quest for ‘atypicality’ by the USA
Antipsychotics Multicenter Study showing that clozapine had, in specific
circumstances (‘treatment-resistant schizophrenia’), advan-
Terminology tages over a standard regime (Kane et al 1988) – namely : (1)
enhanced efficacy on ‘positive’ symptoms; (2) enhanced effi-
Two years after the introduction of chlorpromazine, Steck cacy on ‘negative’ symptoms; and (3) enhanced neurological
and Thiebaux reported parkinsonism with its use, something (i.e. extrapyramidal) tolerability. The new ‘atypical’ age was
evident from its earliest administration. This was soon viewed ushered in with the launch of risperidone in 1993.
as integral – a ‘marker’ of therapeutic action. Accordingly,
in 1955 the pharmacologist Jean Delay proposed the term The notion that there are ‘two dichotomous groups of
‘neuroleptics’ for the class – literally, drugs which ‘grasp’ antipsychotic’ (Kinon and Lieberman 1996), ‘typical’ and
or ‘seize’ the nervous system. By the 1960s, the view that ‘atypical’, has embedded itself in the professional vernacular
parkinsonism was an inevitable part of the therapeutic action despite increasing evidence that ‘atypicality’ is no more valid
fell into disrepute on the basis of both clinical and laboratory a concept now than four decades ago. There is no pharmaco-
evidence, though the term ‘neuroleptic’ stuck. Seen in this logical characteristic that links all ‘new’ antipsychotics and
light, ‘neuroleptic’ is redundant, making these the only the concept continues to rest on the flimsy foundation of
reduced parkinsonian liability, a mirage as likely to reflect
practice inequities between study arms as pharmacological
difference (for details see Owens 2008).

228

Clinical psychopharmacology CHAPTER 11

Notwithstanding on-going debate about the interpretation Box 11.1
of these efficacy studies, the distinction between efficacy
(results in ideal situations) and effectiveness (results in ‘real Classification of antipsychotic drugs
life’ situations) is also crucial. The Clinical Antipsychotic
Trials of Intervention Effectiveness (CATIE) study (Lieberman Chemical classes Aliphatic
et al 2005) is important not simply because it challenges us to Piperidine
learn new facts but because it challenges in the far more Phenothiazines Piperazine
demanding task of unlearning erroneous ones. A sound
knowledge of this study that only comes from source material Thioxanthenes Diphenylbutylpiperidines
is essential for all psychiatrists so only outlines will be pre- Butyrophenones
sented here. Essentially, this pragmatic USA study could Benzaides Dibenzoxazepine
not identify clinically significant differences between new Dibenzazepines Dibenzodiazepine
drugs and a standard comparitor (the intermediate potency, Dibenzothiazepine
perphenazine). Importantly, time to develop parkinsonism Benzisoxazoles Thienobenzodiazepine
in those without it at entry did not differ between com- Indoles
pounds (Miller et al 2008) (Fig. 11.1). CATIE is part of an Quinolinones
expanding literature, including CUtLASS (Jones et al
2006), EUFEST (Kahn et al 2008) and AGATE (Fischer- Clinical pharmacology High
Barnicol et al 2008), all of different aims and design but all Intermediate
pointing in the same direction – that ‘atypical’ antipsychotics Potency Low
offer no fundamental advantage over older drugs when older D2 selectivity
preparations are used sparingly. Receptor binding profile Non-dopaminergic binding
Tolerability Non-extrapyramidal
Previous editions of the present volume adopted a ‘circum- Receptor binding ‘mechanics’ Extrapyramidal
spect’ but conventional view of ‘atypicality’ but current evi- ‘tight’
dence is sufficiently robust to support a dogmatic stance. In ‘loose’
line with recommendations elsewhere (Fischer-Barnicol et al
2008; Owens 2008), this chapter asserts that the term ‘atypi- Chemical structures
cal’ must now be dropped from teaching and clinical practice
as a misleading diversion from the real task for clinicians – Phenothiazines
developing sound expertise in the use of the entire class of
antipsychotics, something that now allows for truly individua- S 4
lised treatment planning. ‘New’ in relation to antipsychotics is 5 3
used here in a purely temporal sense, i.e. drugs launched more 76
recently than ‘old’ ones.
89 10 2
Classification N 1 R1

Conventional classifications of antipsychotics on the basis of R2
chemical structure are of limited value to clinicians though
can make a general contribution to treatment planning. This large group sprang from the commercial dye industry
Greater utility comes from considering clinical pharmacology (Owens 1999). Phenothiazine lies at the heart of methylene
(Box 11.1). blue, one of the most commercially prized of the substances
synthesised in the wake of William Perkin’s extraction of
mauve (‘Perkins’ purple’) from coal tar in 1856. As the name
suggests, it comprises two benzene rings (‘pheno’), linked by

Fig. 11.1 Extrapyramidal side-effects of Proportion of patients without the event 1
antipsychotics (CATIE): time to parkinsonism in those 0.8
with no parkinsonism at baseline (after Miller et al 2008).

0.6

Olanzapine * * No significant differences
0.4 Quetiapine *

Risperadone *

0.2 Ziprasidone *
Perphenazine *

0

0 100 200 300 400 500

Time to event (days)

229

Companion to Psychiatric Studies

a central ring structure incorporating sulphur (‘thio’) and nitro- used representative. In practice, these drugs have high levels of
gen (‘azo’) atoms. Two side-chain substitutions are possible – general tolerability and parenteral use is long established in the
the R1 at position 2, comprising electronegative moieties management of psychiatric emergencies. They have however
(e.g. Cl or S-CH3) largely responsible for potency but inde- very low extrapyramidal tolerability, with the suggestion that hal-
pendently contributing to efficacy; and the R2 at position 10, operidol is uniquely prone to promote EPS (Owens 1999).
the major contributor to potency but which must comprise
at least a 3-carbon chain for antipsychotic action. This termi- The addition of an extra fluorinated benzene ring to the
nates in a tertiary amine or cyclic analogue of a tertiary amine. butyrophenone side-chain created diphenylbutylpiperidines,
with the longest half-lives of any antipsychotic. Only pimozide
R2 substitutions form the basis of subgrouping: has sustained a clinical profile, though its use has declined.

• aliphatic (or aminoalkyl), such as chlorpromazine, have Substituted benzamides
straight chains with methyl or alkyl substituents;
Sulpiride
• piperidine (following the withdrawal of thioridazine, only H OCH3
represented in the UK by pericyazine though mesoridazine
is available in some markets) in which the amino nitrogen is CONH CH2 N
incorporated into a cyclic structure; and H H CH2 CH3

• piperazine (e.g. trifluoperazine) in which the cyclic ring NH2 SO2
comprises two nitrogens and is consequently extended.
Substitutions to the benzene ring have applications in many
Aliphatic compounds tend to be of low potency with broad branches of medicine. In the 1960s, modifications to the meto-
receptor binding profiles, endowing them with relatively high clopramide molecule (itself a modification of the antiarrhythmic,
levels of general adverse effects but relatively low extrapyra- procainamide) produced sulpiride, one of the first drugs termed
midal liability. Piperazines, on the other hand, tend to be of ‘atypical’/‘novel’. Although a number of substituted benzamides
high potency, with lower general adverse effects but higher are centrally acting (e.g tiapride, raclopride, etc.), with the
extrapyramidal side-effects (EPS) liability. demise of remoxipride in 1993 over concerns about aplastic anae-
mia, only sulpiride and amisulpride remain in widespread use.
Thioxanthenes

S 4
5 3
76
Dibenzazepines
89 10 2
1 Dibenzodiazepine

R1 Cl N

R2 N N CH3

Chemically, these represent only a minor shift from the phe- N Clozapine
nothiazines though one that introduces considerable pharma- H
cological difference. Position 10 is taken by carbon rather
than nitrogen, allowing the R2 substitution to attach via a Modifications to the dibenzazepine ring structure have proved
double bond. These compounds therefore exhibit stereoisom- fruitful paths for psychotropic drug development (see tricyclics,
erism (side-chains can attach in mirror-image fashion). The below). Clozapine is the antipsychotic template, the two nitro-
clinical consequence is that isomers differ markedly in affinity gens of the seven-membered central ring making it a dibenzo-
for dopamine binding sites, from potent to weak. The major diazepine. From the same generation came another compound
representatives are the high potency flupentixol (the thio- that illustrates the limitations of chemical classifications to clini-
xanthene analogue of fluphenazine) and zuclopentixol (the cians. Loxapine, a debenzoxazepine substituted at position 11, is
analogue of perphenazine), a lower potency compound. Some remarkably similar in structure to clozapine yet is not clozapine-
markets also have chlorprothixene and thiothixene. like. The same can be said of the dibenzothiapine, zotepine, a
child of the 1970s that failed to find a niche until the ‘atypical’
Butyrophenones explosion 25 years later. Commercially, the most successful
modifications have been the thienobenzodiazepine, olanzapine
O and the dibenzothiazepine, quetiapine.

F C CH2 CH2 CH2 R

Indoles

F CH CH2 CH2 CH2 R HH

H

F H N¨
H H
Structurally, these are ‘phenylbutylpiperidines’ and amongst
the most potent and D2-selective antipsychotics. Since the with- Indole, a naturally occurring aromatic heterocyclic compound,
drawal of droperidol in 2000, haloperidol remains the only widely is a constituent of both orange blossom and human faeces!

230

Clinical psychopharmacology CHAPTER 11

It was also exploited as a dye (indigo) but as the basic structure could have real impact on prescribing choices. The author is
lies at the heart of tryptophan and serotonin, it had obvious psy- aware of only two quality studies on the clinical pharmacology
chopharmacological interest. For many years the sole indole of antipsychotics in normal volunteers (Ramaekers et al 1999;
antipsychotic in the UK was the infrequently used and now Artaloytia et al 2006), work that in some countries (e.g. USA)
withdrawn oxypertine (though molindone has long had a mar- would still have to confront major ethical hurdles. It can only
ket in the US) but recent years have seen the arrival of sertin- be hoped that future funding can fill this black hole.
dole and ziprasidone (technically, a benzisothiazoylpiperazine).
Pharmacokinetics
Miscellaneous
Oral formulations
Risperidone and paliperidone belong to a novel class, the ben-
zisoxazoles, though the latter will be recalled as 9-hydroxyris- Following oral ingestion most antipsychotics appear to be rap-
peridone, the major metabolite of risperidone, so represents idly and completely absorbed from the proximal small bowel,
little that is new. Aripiprazole is technically a ‘dihydrocarbos- regardless of gastric status though time to peak may be delayed
tyril’, a derivative of quinoline. Both groups are under intensive in the absence of dietary fat (e.g aripiprazole). Times to peak
investigation with further compounds likely. blood levels (Tmax) are in the range of 2–5 hours, though
for reasons that are unclear wide discrepancies (0.5–6 hours)
Clinical pharmacology have been reported for haloperidol (Jorgensen 1986).

A more pragmatic, if less precise, way for clinicians to view Orally administered antipsychotics entering the stomach
antispsychotics is on the basis of clinical pharmacology (see are subject to presystemic extraction or first-pass effects
Box 11.1). As a diverse class, their pharmacology offers multiple with passage through the liver. The extent of these effects is
‘dimensions’ along which they may be considered, though it must dependent on a drug’s clearance which, for antipsychotics,
be emphasised that the following cannot be substantiated by is flow-limited, not capacity-limited (Greenblatt 1993). Thus,
‘evidence’. Study of the clinical pharmacology of antipsychotics clearance depends only on the ability of the portal system to
has been woefully inadequate and widespread comparative deliver drug to the liver and not on intermediate metabolism.
explorations have not been undertaken, especially with older The consequence is that the bulk of an orally administered
compounds. However, clinical utility can be found in: antipsychotic dose does not reach the systemic circulation.
Of that which does, only free drug is available for end-receptor
• potency; activity. However, all antipsychotics are highly membrane
• D2 selectivity; bound and protein bound, especially to albumen though also
• general pharmacological tolerability; to other proteins such as alpha-1-glycoprotein. The bound frac-
• EPS liability; and tion varies from around 90% to more than 99% (ziprasidone
• receptor binding ‘mechanics’. and aripiprazole). Free drug is rapidly and widely distributed
due to the physicochemical property crucial to the primary
We have already touched on the general rule (cf. phenothia- class action – lipophilicity. This not only means that unbound
zines) that high potency compounds tend to raise fewer general drug readily crosses the blood–brain barrier and is widely avail-
tolerability issues but more extrapyramidal ones. The same able to brain sites but also that uptake into peripheral organs is
general point holds for butyrophenones compared overall to extensive. Antipsychotics therefore have a large apparent vol-
phenothiazines and to new drugs also – e.g. risperidone pro- ume of distribution (Vd). Drug is reversibly and dynamically
duces more dose-dependent EPS than, say, olanzapine or que- bound to peripheral sites especially in lung and other organs
tiapine. D2 selectivity is also useful. One would predict that with rich blood supplies, as well as adipose tissue, from where
risperidone, with its broad receptor binding profile, would it is readily released back into the systemic circulation as excre-
dose-for-dose have lower EPS liability than, say, the equally tion progresses, a factor underlying the persistence of efficacy,
potent but more selective fluphenazine. A further consider- adverse actions and active moieties in plasma long after cessa-
ation might be the ‘emphasis’ of a drug’s general tolerability tion of treatment.
profile. Olanzapine’s high proclivity to produce metabolic dis-
turbance places it in a less favourable position than, say, the As a result of these kinetic properties, standard antipsycho-
pharmacologically comparable perphenazine and while both tics have low bioavailability, which for chlorpromazine is in the
pericyazine and olanzapine have strong affinity for H1 sites, region of 30%, though for haloperidol is higher at around 60%
pericyazine is powerfully sedative compared to olanzapine at (Dahl and Strandjord, 1977). By contrast, aripiprazole in
similar doses. If diminishing EPS liability is a priority, one tablet form has very high bioavailability ( 90%).
might, in addition to keeping doses low, consider a broad spec-
trum drug of any sort as opposed to a potent and selective one Elimination half-lives (t1/2) of antipsychotics, as with
and theoretically, the ‘loose-binding’ quetiapine (see below) all drugs, bear a mathematical relationship to their hepatic
would have advantage over, say, chlorpromazine at similar clearance and volume of distribution. Despite their extensive
dosage. clearance and wide distributions, the half-life of most anti-
psychotics is in the intermediate range at around 20 hours.
It is depressing that after half a century there is still inade- Exceptions are sulpiride, quetiapine and ziprasidone, all in
quate evidence to assess those issues that at a clinical level the 6–8 hour range, clozapine in the intermediate 12–14 hour
range and aripiprazole at approximately 75 hours. On this
parameter, therefore, most antipsychotics are suitable for once

231

Companion to Psychiatric Studies

daily dosing, though half-life alone may bear only a tenuous amounts of sulpiride can be recovered unchanged from faeces
relationship to receptor activity in drugs so widely distributed. implying that part of its low bioavailability ( 27%) is a conse-
However, wide distribution also means that following cessa- quence of poor absorption (as opposed to wide distribution).
tion of treatment drug can seep back into the systemic circula- This, plus difficulty in crossing the blood–brain barrier, result
tion over long periods. Thus, the terminal phase half-life of from the fact that, unusually for an antipsychotic, sulpiride is
some of these compounds is very long (in the case of chlor- predominantly water soluble. In addition, it is presented as a
promazine up to 60 days) resulting in persistence of detectable 50:50 racemic mix (‘l-’ and ‘d-’ isomers), only l-sulpiride being
drug moieties for many months – if not years – after exposure active. Sulpiride is a hard drug to use in ‘dose equivalence’,
(Curry 1986). switches most commonly representing effective dose reduc-
tions. It is therefore unsurprising that its EPS liability is per-
CSF levels have not been extensively studied but appear ceived as low and that negative symptoms appear to improve
to represent only about 3–4% of plasma levels for chlorproma- following transfer! Amisulpride has better absorption charac-
zine and haloperidol, which correspond approximately to free teristics than sulpiride but relatively poor kinetic parameters
fractions in plasma, though they do seem to vary considerably may still be a factor in any advantages this drug is perceived
in different individuals and with different preparations to have.
(Jorgensen 1986).
Sulpiride is subject to little first-pass effects and is substan-
Metabolism is mainly hepatic, for the simple reason that tially less protein bound than others of the class (approx. 40–
lipophilic compounds must be transformed into water soluble 50%). Estimated half-life is widely variable (Jorgensen 1986)
ones in order to be excreted via the kidney. Antipsychotic but, with an average of approximately 8 hours, is shorter than
metabolism is generally extensive with little parent drug those of most other drugs. Thus, sulpiride is not suitable for
eliminated. The processes involved are for the most part unso- once daily dosing, perhaps a more valid assumption for a drug
phisticated, including oxidation, N-dealkylation and conjuga- of poor distribution than for those with large distributions.
tion with glucuronic acid. Aripiprazole is mainly metabolised Benzamides as a group are subject to a number of metabolic steps
by dehydrogenation to dehydro-aripiprazole. Some drug which produce, in general, inactive products. Most benzamide
metabolites are excreted in bile. metabolites are excreted largely unchanged in the urine.

While some antipsychotics, such as flupentixol and pimo- The conventional view is that the specific therapeutic (i.e.
zide, do not appear to produce active metabolites, the majority ‘antipsychotic’) action is delayed till about the third week of
do and some produce several which have clinically significant treatment. However, analysis of placebo-controlled trials shows
antipsychotic actions (e.g risperidone/9-hydroxyrisperidone). that improvement in positive symptoms is evident in the first
Dehydro-aripiprazole contributes up to 40% of drug in plasma week (Agid et al 2003). The question is whether this undoubted
and has equal affinity for D2 receptors as the parent. This has improvement represents the specific antipsychotic action or
complicated the field of therapeutic blood monitoring, as the non-specific effects of, for example, sedation or diminished
percentage of active metabolite varies considerably across indi- ‘arousal’. Studying the isomers of flupentixol, one of which is
viduals and these products have their own pharmacokinetic a D2 antagonist and antipsychotic, the other of which is not,
properties. This latter principle lies behind the marketing Johnstone (1979) found that one of the most striking changes
of metabolically and isomerically refined products (e.g. 9- within the first 1 – 2 weeks was reduction in anxiety. Practically,
hydroxyrisperidone or ‘paliperidone’) in recent years though it is important not to take early improvements as indications to
whether kinetic parameters are sufficiently different to justify reduce or significantly modify antipsychotic regimes as this is
these being considered ‘different’ drugs is doubtful. likely to result in symptom exacerbation.

Although therapeutic monitoring has been a recurrent Delayed onset of ‘antipsychosis’ is compatible with longer
theme of the American literature, because of these complex term receptor and intracellular modifications that underlie
kinetic issues it has led to little that is clinically useful. An drug action though may also reflect the time to steady state.
exception may be clozapine, measurement of which, along This is achieved when the overall mean concentration of
with its major and active metabolite, N-desmethyl-clozapine parent drug and active metabolites does not alter, so long as
(as a ratio), has been advocated in assessing compliance, ‘opti- the daily dose and factors influencing clearance do not change.
mising’ therapy and minimizing toxicity. Nomograms are avail- It is described for all drugs by a constant relationship – i.e. 4–5
able to aid dose adjustment in populations in whom kinetic half-lives are required to achieve within 10% of the steady-
variables may influence plasma levels, such as males/females, state condition (Greenblatt 1993; Richelson 2001). Clearly,
smokers/non-smokers (Rostami-Hodjegan et al 2004). Thera- drugs with very long half-lives, such as pimozide, will require
peutic drug monitoring depends on acceptance that clear rela- longer to achieve steady state than the likes of sulpiride, which
tionships have been established between blood levels and has a much shorter half-life. The figure of 4–5 half-lives also
clinical response, a proposition still contentious even with applies to the time to achieve 90% elimination of a drug from
clozapine and the author remains phlegmatic about the value, plasma. When drugs are administered at dose-intervals shorter
and especially the cost-effectiveness, of clozapine monitoring. than 4–5 times their half-lives accumulation (or ‘cumulation’)
It should be viewed as an aid to decision-making and not a occurs resulting in blood levels at steady state that are substan-
substitute for the exercise of clinical skills. tially higher than peak doses after single ingestion, as the
administration interval is less than that required to eliminate
An exception to the general kinetic principles above is sul- most of the previous dose (Richelson 2001). This is theoreti-
piride, which has somewhat different – and less favourable – cally relevant to most psychotropics though is usually only of
pharmacokinetic parameters than other antipsychotics. Herein
may lie the roots of its perception as ‘atypical’. Considerable

232

Clinical psychopharmacology CHAPTER 11

clinical relevance in physically compromised patients (e.g. These manufacturing principles radically alter the kinetics
major hepatic or renal disease). As can be deduced, in theory of depots. This is because pharmacokinetics are limited by
at least, drugs of short half-life may be associated with a more the rate of drug absorption, not the rate of metabolism as is
rapid onset of action but a swifter relapse rate on discon- the case with other pharmaceuticals. Traditional pharmacoki-
tinuation. Those with half-lives greater than approximately netic models do not apply in such situations and a so-called
36 hours probably do not require daily dosing though for ‘flip-flop’ model, where the absorption rate constant is less
compliance reasons, longer ingestion intervals are usually not than the elimination rate constant (Ereshefsky et al 1983; Jann
favoured in practice. et al 1985), has been proposed. Interpreting plasma curves in
this situation is difficult, as declines in blood levels reflect
Recently, so-called ‘orodispersible’ (technically, ‘orally dis- not only metabolism but also protracted absorption.
integrating’) formulations have been introduced (e.g risperi-
done, olanzapine, aripiprazole). These not only facilitate ease Most available long-acting depots (excluding zuclopentixol
of administration and allow greater certainty regarding inges- acetate) share similar kinetic properties. Most achieve maxi-
tion and hence compliance, but because absorption is pre- mal plasma levels gradually over approximately 4–7 days
gastric, they avoid first-pass effects and, theoretically, increase followed by a gentle decline in the subsequent few weeks.
bioavailability. Different technologies are used which (despite The exception is fluphenazine decanoate which, uniquely,
the theory) appear to result in similar kinetics, including achieves peak levels rapidly within about 12–24 hours of
bioavailability, to tablet formulations. Both orodispersibles of administration, with an equally rapid decline to about one
risperidone and aripiprazole contain aspartame, a source of third peak, followed by a more gradual reduction over the
phenylalanine, and should not be given to those with phenylke- subsequent few weeks (Jann et al 1985). The reason for this
tonuria, while the bovine origin of the gelatin in olanzapine rapid postinjection peak is not understood but may be related
‘Velotabs’ may be objectionable to some. to the drug’s vulnerability to muscle hydrolases. It probably
underlies some difference in the pattern of early neurological
Orally disintegrating formulations dissolve rapidly in normal adverse effects reported with depots (Owens 1999). Half-lives
saliva on the tongue and can be swallowed if necessary. How- in the range of 5–7 days following single doses of depots
ever, in those with dry mouth, a prior drink is a prudent aid. appear to increase with multiple dosing to something in the
range of 14–21 days (Jann et al 1985) though in a few indivi-
Depot formulations duals the half-life of flupentixol decanoate has been found to
be greatly extended (up to 112 days).
Traditionally, the term ‘depot’ refers to a specific drug deliv-
ery system involving chemical alteration of the parent com- Risperidone was the first new antipsychotic to be presented
pound, the use of an oil-based storage vehicle and enzymatic in long-acting injectable form. Risperdal Consta is an aqueous
release from intramuscular sites. The molecular structure of suspension of microspheres comprising drug contained within
most newer drugs does not permit the first step in this pro- a glycolic acid–lactate copolymer matrix. Following injec-
cess, so novel pharmaceutical methods have had to be tion, the copolymer undergoes gradual, sustained hydrolysis,
employed. Technically, these newer formulations are not resulting in steady release of active substance. The slow nature
‘depot’ per se but aqueous solutions more correctly termed of this process means that, allowing for distribution, it takes
‘long-acting injectables’. approximately 3 weeks for blood levels to rise, the peak being
attained at about 4–5 weeks and lasting to approximately
The kinetics of traditional depots are highly complex and week 7. With repeated treatment, steady state (risperidone
markedly different from oral formulations. The chemical and 9-hydroxyrisperidone) is usually reached after the fourth
change is esterification of a terminal side-chain hydroxyl group injection on a 2-weekly regime. On cessation, concentrations
(the required structural component) to a long-chain fatty are maintained for some 4–5 weeks before declining rapidly
acid with the ester dissolved in an inert oil base. The most (Gefvert et al 2005). Again, this process results in complex
commonly utilised base is sesame oil, though the Lundbeck and different pharmacokinetics from oral administration but
products (Depixol, Clopixol, Acuphase) use a synthetic from a practical point of view, bioequivalence between the
triglyceride based on coconut oil (‘Viscoleo’). Following intra- two has been established (Eerdekens et al 2004).
muscular injection, esters slowly diffuse from the oil base
(release may also be determined by partial metabolism of the At time of writing, a long-acting injectable formulation of
oil) and are thereafter rapidly hydrolysed by plasma, and olazapine is not available but is likely to be licensed immi-
possibly muscle, esterases to release active drug. nently. This exploits the physical properties of olanzapine
pamoate, the sustained-release mechanism being simply slow
Both the oily base and the fatty acid utilised in esterifi- dissolution from an aqueous solution of a salt of inherently
cation are important in determining the kinetics of depots. low solubility.
Viscoleo may be degraded more rapidly than sesame oil, one
possible reason why the decanoate of flupentixol is detectable The pharmacokinetics of depots and long-acting injectables
for a less protracted period after last injection than that of have clear clinical implications. These formulations avoid
fluphenazine. The role of the fatty acid in determining the rate first-pass effects, so the desired actions can be achieved with
of drug release can be seen with zuclopentixol which, when much lower absolute doses. The lower steady-state peak con-
esterified with acetic acid (Acuphase), gives plasma concentra- centrations with Consta compared to an equivalent oral regime
tion curves closer to those of an aqueous solution than the of risperidone has been, along with diminished fluctua-
characteristic depot curves found when the drug is esterified tions, particularly suggested as advantageous (Eerdekens et al
with decanoic acid (Clopixol). 2004), but although intuitive, supportive evidence is lacking.

233

Companion to Psychiatric Studies

They also take much longer than oral preparations to reach well-received drugs showing a similar predilection. However,
steady state and longer to clear following discontinuation. the pattern of preferences for dopamine subtypes is so varied
Most ‘classical’ depots achieve steady state after about among equally efficacious drugs that early hope this might pro-
3 months, though fluphenazine decanoate may again be differ- vide a more detailed means of understanding pharmacodynam-
ent at about 6 weeks (Jann et al 1985). This latter preparation ics has not been realised.
can however be detected in plasma up to 12 weeks or longer
following cessation of treatment, while the decanoate of A further, potentially interesting way of modifying D2 func-
flupentixol is, in general, present for only 6 weeks. Elimination tion comes from partial agonist actions. A drug’s biological
of risperidone on stopping Consta is complete after 7–8 weeks effectiveness (technically referred to as ‘intrinsic activity’) is
(Harrison & Goa 2004). The practical implications are firstly, independent of its affinity for receptors. Partial agonists are
that depots and long-acting injectables are not, as a rule, compounds with low intrinsic activity but high affinity, as a
suitable for acute phase management. This is not always result of which they can act to inhibit the actions of other ago-
acknowledged with depots but recommendations for Risper- nist molecules with high intrinsic activity – i.e they can antag-
dal Consta, suggesting maintenance of oral risperidone for onize them. They are called ‘partial’ agonists because they
the first 3 weeks, enshrine this principle. Secondly, after ces- possess some limited intrinsic activity but their net effect
sation, relapse with these formulations may be substantially overall is to diminish the actions of the more powerful agonist.
delayed by the actions of residual medication, especially with Furthermore, they do this to a degree that relates to the
‘classical’ depots – and with Modecate more than Depixol. amount of agonist available – e.g. D2 partial agonists reduce
Furthermore, dose regimes must be flexible and empirically high dopaminergic transmission states but increase low dopa-
derived in order to avoid detrimental cumulative effects in minergic transmission. Hence, these drugs have been referred
some patients. to as ‘dopamine stabilisers’, a catchy term that diminishes a
subtle concept. Aripiprazole is the first D2 partial agonist to
Pharmacodynamics reach the market though others are likely to follow.

Therapeutic actions represent the desired clinical effect(s) No antipsychotic acts only on D2 receptors, most interact-
reflecting a (usually) specific pharmacological mechanism. ing with a broad range of other receptor types. This was for
While all other actions are conventionally referred to as many years felt to represent contamination contributing to
‘adverse’, resulting clinically in ‘side-effects’, there is value adverse, rather than therapeutic, effects, a view that springs from
for psychiatrists adopting a more flexible view – namely, that the inferences of the Dopamine Hypothesis of Schizophrenia
of ‘target’ and ‘non-target’ actions. The reason is that in psy- (Box 11.2) which has held strong sway.
chiatric practice not all ‘non-target’ actions are ‘adverse’ (e.g.
sedation). From this perspective, the early thrust was to produce
increasingly selective D2 antagonists targeted on mesolimbic,
Therapeutic (target) actions as opposed to nigrostriatal, dopamine. The benzamides are to
some extent the realisation of this – sulpiride has minimal
In 1963, Carlsson and Lundquist suggested that antipsychotics 5HT and a-adrenergic activity while amisulpride has negligible
act as postsynaptic dopamine antagonists, something now affinity for all receptors except D2 and D3. However, this
clearly established. In 1979, Calne and Kebabian classified approach was turned on its head by the demonstration that
central dopamine receptors into two types (D1 and D2) on clozapine, a highly unselective transmitter antagonist (see
the basis that the former stimulates synthesis of adenylate Table 11.1), was superior to a standard drug (chlorpromazine)
cyclase while the latter does not. Whatever else individual in, at least, treatment-resistant schizophrenia (Kane et al
compounds may do in relation to other receptor systems, all 1988). Drugs previously considered pharmacologically ‘dirty’
current antipsychotics of proven efficacy block central dopamine came to be seen as having ‘rich and challenging’ pharmacolo-
D2 receptors at postsynaptic sites, though with varying affi- gies and in the search for new treatments, the highly selective
nities. Other targets have been explored, and may ultimately approach gave way to the highly unselective one.
hold potential, but to date no strategies invoking other than
postsynaptic D2 antagonism have translated into effective anti- The model from clozapine that has been pursued most
psychotic treatments. focuses on serotonergic mechanisms, and in particular 5HT2a.
Many new antipsychotics seek to combine D2 antagonism with
We now know that the D1/D2 classification represents at a greater degree of 5HT2a antagonism and have been marketed
least two ‘families’ of receptor subtypes – the D1 ‘family’ heavily on this basis, with some commentators suggesting that
(comprising D1 and D5 isomorphs), and the D2 ‘family’ (com- by ‘enhancing’ efficacy, 5HT2a mechanisms (conventionally
prising the so-called ‘long’ and ‘short’ varieties of D2, plus the seen as ‘non-target’) themselves contribute to the target
D3 and D4 isomorphs). Antipsychotics differ in their affinities action. This must be rejected. While many new antipsychotics
for different isomorphs of the D2 receptor (Table 11.1), which have relatively high 5HT2a:D2 affinity ratios (Box 11.3) this
has been a focus for exploring the detailed basis of the antipsy- if anything is due to low affinity for dopamine, not higher
chotic effect and, especially, for seeking out differences affinity for 5HT (Kapur & Seeman 2000). What is more, some
between drugs. There was great interest in the demonstration new drugs do not have high ratios while these can be found
that clozapine exerted a preference for D4 sites, with other with a number of long-established compounds, including
chlorpromazine. This aspect of clozapine’s profile revivified
antipsychotic psychopharmacology by breaking away from
the confines of the classical Dopamine Hypothesis but in
terms of results has turned out to be a damp squib!

234

Clinical psychopharmacology CHAPTER 11

Table 11.1 Relative in vitro receptor binding potencies of some antipsychotic drugs (Ki values – nM)

Receptor CPZ HALO SUL AMISUL RIS OLAN QUET CLO

D1 þþþ þþþ i i þþþ þþþ þ þþ

(56) (25) (–) (–) (75) (31) (455) (85)

D2 þþþ þþþþþ þþþ þþþþ þþþþ þþþþ þ þ

(19) (1) (34) (2.8) (3) (11) (160) (125)

D3 þþþ þþþþ i þþþþ þþþ þþþ þ þ

(29) (2.7) (–) (3.2) (14) (49) (340) (280)

D4 þþþ þþþþ i i þþþþ þþþ i þþþ

(37) (5) (–) (–) (7) (27) (–) (21)

5HT1A þ i iiþ i ii
(642) (–) (–) (–) (490) (–) (–) (–)

5HT2A þþþþ þþþ þ/i i þþþþþ þþþþ þ þþþ
(3.3) (78) (>1000) (–) (0.6) (4) (220) (12)

5HT2C þþþ i i i þþþ þþþ þ þþþþ
(27) (–)
(–) (–) (26) (11) (615) (8)

a1 þþþþþ þþþ þ/i i þþþþ þþþ þþþþ þþþþ

(1.7) (46) (>1000) (–) (2) (19) (7) (7)

H1 þþþþ i i i þ þþþþ þþþþ þþþþ

(9.1) (–) (–) (–) (155) (7) (11) (6)

M1 þþþ i ii i þþþþþ þ þþþþþ
(–) (–) (–) (1.9) (120) (1.9)
(60) (–)

Average values: the lower the value, the greater the potency.
AMISUL, amisulpride; CLO, clozapine; CPZ, chlorpromazine; HALO, haloperidol; i, insignificant (>1000); OLAN, olanzapine; QUET, quetiapine; RIS, risperidone; SUL, sulpiride.
Data from Buckley (2007) and other sources.

However, if clozapine is our model, there is another factor at a threshold of 65%, while no patients with occupancy below
of interest – receptor binding ‘mechanics’. In vivo imaging sug- 78% developed EPS. This ‘rule’ – occupancy below 65% does
gested that therapeutic doses of older drugs were associated not achieve efficacy; occupancy over 80% and EPS emerge –
with 70–80% blockade of dopamine D2 receptors (Farde is now a generally accepted principle for all antipsychotics.
et al 1992) (Fig. 11.2). Using haloperidol, Kapur et al (2000) However, early studies also suggested that clozapine obtained
confirmed that D2 occupancy significantly predicted response efficacy with lower occupancy levels ( 50%) which was used
to explain the drug’s difference (see Fig. 11.2). It seems this
Box 11.2 was an artefact of methodology and an assumption that
drug–receptor binding is a static process. It has been cogently
argued that as a dynamic process, binding reflects an interplay
of drug association with and dissociation from its site and that

The dopamine hypothesis Box 11.3

Mark 1 5HT2a:D2 affinity ratios of some antipsychotic drugs

Schizophrenia is the result of increased dopaminergic function in Olanzapine 50:1
certain brain areas:
• from disruption at predominantly postsynaptic sites, Clozapine 30:1

e.g. increased postsynaptic receptor numbers (supersensitivity) Chlorpromazine 10:1
• from disruption at predominantly presynaptic sites,
Risperidone 8:1
e.g. increased absolute levels/turnover (?mesolimbic system)
Loxapine 7:1
Mark 2
Thioridazine 5:1
Antipsychotic drug efficacy is mediated via postsynaptic dopamine
(D2) blockade (?mesolimbic system)

235

Companion to Psychiatric Studies

Flupenthixol
Haloperidol
Chlorpromazine
Pimozide
Trifluoperazine
Raclopride
Risperidone
Ziprasidone
Sertindole
Olanzapine
Loxapine
Sulpiride
Amoxapine
Remoxipride
Clozapine
Quetiapine

D2 occupancy % 100
90
80
70
60
50
40
30
20
10

0.01 1 10 100
Antipsychotic binding constant K (nM)

Fig. 11.2 D2 receptor occupancy with maintenance doses of antipsychotics in schizophrenia
(PET & SPET) (after Seeman and Tallerico 1999).

what is referred to as ‘low affinity’ results from the property Pharmacological: central or peripheral actions By
of fast dissociation (Kapur & Seeman 2001). From this which can be anticipated by body
perspective, clozapine appears different in that although effi- a drug’s known pharmacology system
cacy is achieved with similar occupancy levels to other drugs
(65–80%), its ‘loose binding’ characteristics means that it readily Non-pharmacological: actions not predictable from
dissociates from dopamine sites in response to surges of endoge- known pharmacology
nous dopamine. This elegant proposal may have profound impli- chemical properties
cations in explaining not so much a unique pattern of efficacy but (toxic effects)
of neurological tolerability (see below) and its ramifications. idiosyncratic response
Quetiapine is the only other available drug that, to some extent, (allergic effects)
shares this characteristic (Seeman & Tallerico 1999).

Adverse (non-target) actions Fig. 11.3 Classification of adverse (non-target) drug action.

In terms of serious adverse events, antipsychotics are, in gen- Mediated via general pharmacology
eral, safe. They have however a wide range of side-effects that
while not medically significant, can be subjectively unpleasant Gastrointestinal actions of antipsychotics provide some of the
and major factors in non-compliance. Thus, if safety is con- best examples of medically ‘minor’ but subjectively unpleasant
sidered, antipsychotics have a wide ‘therapeutic index’ but non-target actions. Dry mouth may not be disconcerting to doc-
substitute this with tolerability and a different picture tors but is invariably so to patients. It predisposes to stomatitis,
emerges. The problem with the concept of ‘therapeutic index’ especially with dentures, as well as oral candidiasis (which may
is that it encourages the idea that ‘safe’ use is the same as extend the length of the oesophagus) and long-term dental
‘easy’ use, which is not the case. caries and gum disease (Lucas 1993). Dry mouth can be found
with clozapine, a powerful anticholinergic (see Table 11.1),
The ideal way to classify non-target actions would be in but excessive saliva and drooling are also common, a paradox
terms of receptor binding profiles but in vitro binding does probably resulting from a parallel and greater cholinergic reup-
not reliably translate to in vivo predictions. The most prag- take inhibition. As this hypersalivation represents true excess
matic classification of non-target actions is simply those phar- production, it is rightly distinguished from the drooling of
macologically predictable and those not, on a system basis parkinsonism, where the disorder is less one of excess pro-
(Fig. 11.3). Because of the importance of extrapyramidal duction than impaired clearance as a result of bradykinesia
actions, these are considered separately from what are called affecting the swallowing musculature. Hypersalivation is better
here ‘general’ non-target actions.

236

Clinical psychopharmacology CHAPTER 11

avoided – utilising low starting and slow incremental doses – With these electrophysiological actions comes lowering
than treated. It may respond to antimuscarinics such as hyo- of the seizure threshold. With old antipsychotics, there is
scine (e.g as transdermal patches) or low-dose tricyclic but once approximately a 1% risk of precipitating fits. The risk with
established treatment is often unsatisfactory. low-dose clozapine (<300 mg per day) does not appear to be
increased, though above 300 mg prevalences as high as 14%
GI hypomotility is common with antipsychotics, especially have been reported, amounting to an average risk of 4% over-
potent anticholinergic drugs (see Table 11.1). Constipation all (Baldessarini & Frankenburg 1991). Predisposing factors
is upsetting for patients but predisposes to paralytic ileus, include a past or family history of epilepsy or pre-existing
a serious condition especially in the elderly. Review of cerebral pathology but in the absence of such predisposition,
clozapine-related GI hypomotility suggests a mortality of treatment variables such as polypharmacy, rapid increments
27.5% (Palmer et al 2008), emphasising that constipation in and high-dose regimes are relevant. A rank order of risk for
antipsychotic-treated patients should be vigorously managed. older drugs places chlorpromazine at the top, with pimozide,
Impaired bowel motility may also underlie the suggestion, as haloperidol and fluphenazine at the bottom, behind trifluoper-
yet unconfirmed, that antipsychotics may be associated with azine which appears intermediate (Cold et al 1990). Seizures
an increased risk of the rare but potentially fatal haemorrhagic in themselves are a potentially serious complication but as they
enterocolitis, in which commensal flora, especially claustridia, can usually be managed by a gradual switch to another prepa-
invade bowel mucosa. This condition is also associated with ration and/or the introduction of anticonvulsants, they do not
the generally poor nutrition and life-style known to afflict represent a contraindication.
chronic psychotic patients, but sudden onset of bloody diar-
rhoea in treated, at-risk individuals should trigger concern. Any drug with potent anticholinergic actions can impede
temperature control by impairing peripheral sweating, though
Impaired accommodation is distressing but can promote this is not clinically significant with antipsychotics (see Anti-
angle closure glaucoma in the predisposed, while dry eyes are cholinergics). In addition, however, phenothiazines, and in
not only uncomfortable but may lead to corneal ulceration, particular chlorpromazine, are poikilothermic, i.e. they act to
especially with contact lens use. Urinary problems (retention, lower core body temperature. This is a clinically significant
frequency, urgency, nocturia) are all more common in males but overlooked action and was the property surgeons built on
and often must be sought as patients infrequently bring them to promote chlorpromazine’s development as an operative pre-
to attention. These lower urinary tract symptoms, often med. This usually presents practical problems only when these
mimicking prostatism, are more common with clozapine than drugs are combined with another that increases heat loss
other antipsychotics (Jeong et al 2008). through peripheral vasodilation, such as alcohol, a combination
that in cold environments can lead to fatal hypothermia. The
The best treatment for these types of problem is dose opposite problem – hyperthermia and neuroleptic malignant
reduction or change to an alternative preparation of lower syndrome (NMS) – is discussed below.
anticholinergic potential.
Antipsychotics can promote a number of other endocrine
Psychiatrists are roundly criticised for poor attention to the changes, though these are either of no or unknown significance.
sexual dysfunction antipsychotics (and other psychotropics) Oedema may rarely result from inappropriate secretion of anti-
can promote. Any drug with anticholinergic actions can impair diuretic hormone, usually with phenothiazines, while suppres-
performance directly, especially in males, resulting in difficulty sion of corticotropin, growth hormone, thyroid-stimulating
in obtaining/sustaining an erection and premature, retrograde hormone (TSH), follicle-stimulating hormone and luteinising
or failed ejaculation. In females, impaired arousal response hormone is probably of negligible clinical import (Edwards
and anorgasmia may be the consequence. Performance pro- 1986).
blems are not infrequently volunteered whereas loss of libido
usually emerges from enquiry. This will be discussed below Specific issues
(hyperprolactinaemia).
Three areas of non-target antipsychotic activity are worth
In classificatory terms, antipsychotics are not ‘sedatives’ but singling out for detailed consideration. These relate to cardio-
can have sedative effects that are helpful to management of vascular effects, metabolic actions and hyperprolactinaemia.
disturbed behaviour. This particularly applies to low potency
compounds, though risperidone is also sedating. This is most Cardiovascular effects
commonly a consequence of antihistamine (H1) activity but can
be found in drugs with little inherent antihistaminic activity, Elementary Antiautonomic actions result in an increase in rest-
such as haloperidol in higher doses. It is usually a phenomenon ing heart rate in the order of 10 beats/minute or less and are not
that habituates over a few days on a steady regime. clinically significant. The more powerfully anti-adrenergic drugs
will also produce a significant fall in blood pressure, mainly pos-
Almost all antipsychotics alter the EEG, particularly cloza- tural, which can itself increase heart rate via reflex tachycardia.
pine, with up to one third of patients demonstrating striking Psychiatrists tend to consider postural hypotension a minor
changes. The effect is towards general slowing of waveforms medical issue but it is a potentially serious development that
with a decrease in alpha waves and an increase in theta and can precipitate myocardial ischaemia/infarction or cerebrovas-
delta waves. Chronic exposure results in increasing synchro- cular accident especially in the elderly and should always be
nization, with increasing slow wave activity and increasing greeted with concern. It can occur with new generation antipsy-
amplitude. Spike and sharp waves can be superimposed and chotics as well as low potency phenothiazines. For this reason,
paroxysmal discharges, similar to epileptiform activity, may the latter should never be administered intravenously, and
be seen. Hence the need for a detailed drug history to caution should always be exercised with intramuscular use.
accurately interpret the EEG.

237

Companion to Psychiatric Studies

Complex Box 11.4

QTc Prolonged QTc interval: factors influencing its length

Up to 25% of antipsychotic recipients have ECG changes, Age Coronary heart disease
including ST depression, T wave flattening and occasional Gender Multiple medications
emergence of U waves. Such changes seem to relate to the Time of day Pharmacokinetic/dynamic changes
chemical structure of the drug and probably to potency as they Electrolyte imbalance > in females
are more associated with phenothiazines, and with chlorprom- Phase of menstrual cycle
azine more than trifluoperazine (approximately 50% versus Family history > during the night
16%, respectively) (Lipscomb 1980). Cardiovascular disease Hypokalaemia
Diuretic use
The major ECG issue is delayed ventricular repolarisation Drug treatment Hypomagnesaemia
seen in prolongation of the QT interval. Prolonged QTc (i.e. Vomiting/diarrhoea
duration of the QT interval corrected for heart rate) has gone CYP 450 polymorphisms Long QT syndrome
from marginal to central concern in clinical pharmacology in e.g. CYP 1A2 1F Sudden death syndrome
general, because of its association with serious tachyarrhymias, Sinus bradycardia Ischaemic heart disease
such as the polymorphic ventricular tachycardia torsade de Psychological Acute/chronic ischaemia
pointes, which can be fatal (see below). Reilly et al (2000) Method of calculation Predisposition to arrhythmia
found that predictors of QTc prolongation in a psychiatric pop- Cardiomyopathy
ulation included age, exposure to tricyclic antidepressants and QT prolonging drugs
especially to thioridazine (OR 5.4: 2.0–13.7) and droperidol Dose-dependent action on ion
(OR 6.7: 1.8–24.8), findings that led to the withdrawal of channels governing QT length
both antipsychotics. There is therefore sound reason for psy- Therapeutic doses of drugs of known
chiatrists to be aware of this issue – but perhaps not alarmed. enhanced risk
Overdose of drugs of known
QTc is a complex phenomenon with many causes and sev- enhanced risk
eral methods of measurement (Box 11.4), all of which impact Pharmacokinetic interactions,
on assessment. While ‘norms’ are suggested (e.g. females especially with inhibition of CYP 450
<450 ms; males <430 ms), as ‘boundaries of normality’ these 3A3/4
are very general. In addition, while blockade of rapidly activat-
ing delayed rectified potassium channels (Ikr) is a likely mech- ‘hyperarousal’: excessive
anism, this is not the only cause of serious tachyarrhythmias. sympathetic drive
It is fair to assume that all antipsychotics have the potential > with manual vs. automated
to prolong QTc though the greatest amount of evidence exists Bazett method (easiest but greater
for haloperidol and pimozide with very little for loxapine, variance)
perphenazine and pipothiazine, with others intermediate, but Fridericia (most accurate with fast
this undoubtedly reflects usage rather than inherent risk. heart rate)
Prolonged QTc is uncommon, affecting no more than 2–3%
of exposures and in only a minority of these will clinical conse- þ
quences be significant. Thus, when it comes to prescribing Hodges
choices, clinical need and not QTc issues should predominate. Framingham
In Reilly et al’s study, prolonged QTc was dose dependent and Nomogram method of Karjalainen
polypharmacy has been suggested as a predisposing factor,
both of which emphasise the importance of ‘minimal effective the risk remains ‘potential’ then monitoring is probably suffi-
treatment’. Dose relationships are, however, complicated as cient. Routine ECG monitoring is widely recommended both
prolonged QTc has been reported with depots, even high doses at baseline and through treatment though one must be sceptical
of which attain only nanomolar plasma concentrations. of its value. It is certainly not cost-effective. It has, however, a
place in high-risk groups: the elderly; established heart disease;
QTc prolongation should be avoided rather than being family history of prolonged QTc syndrome/sudden death;
allowed to develop then treated and again calls for cautious uti- administration of other QT-prolonging drugs; and electrolyte dis-
lisation of antipsychotics and avoidance of complex regimes, turbance (e.g. hypokalaemia, hypomagnesaemia, hypocalcaemia).
including polypharmacy of all sorts – including electolyte-
lowering diuretics and other drugs that prolong QTc (e.g. antiar- The concern surrounding a regime’s dysrhythmogenic
rhymics such as disopyramide and sotalol; antimalarials such as potential resides with sudden death syndrome. Although very
artemether with lumefantrine). Because of specific concerns rare, this particularly involves apparently fit young males, so
(which may or may not be justified) ceilings on recommended concern is justified. SDS is usually associated with high-dose
doses are set for pimozide (20mg per day), while ECG moni- regimes (Levinson & Simpson 1987) but more than 100mg
toring is required with pimozide and sertindole (relaunched thioridazine or its equivalent increases the risk of sudden death
in the UK under limited license). If QTc is long when identified approximately 2.5 fold (Ray et al 2001). Thus, individual vul-
(e.g. >500ms), medication should be stopped and con- nerability may be more relevant than simple dose, emphasising
sideration given to alternatives (in lower dose) when values fall, the importance of exploring family history of cardiac disorders
which usually takes only a few days. If in the 450–500ms range, and sudden premature death where possible in all potential

238

Clinical psychopharmacology CHAPTER 11

recipients of antipsychotics. This is likely to be more valuable Table 11.2 Stroke risk in the elderly receiving antipsychotics
than a poor-quality screening ECG. of different types

Perhaps the most common context for sudden death has been Risk ratio
in relation to the treatment of psychiatric emergencies. Extreme
muscular overactivity can adversely alter the pharmacokinetics Unadjusted Adjusted
of certain agents (Jusic & Lader 1994) resulting in exceptionally
high blood levels, possibly as a result of the loss of muscle storage Relative to non-exposed controls
sites. Management of psychiatric emergencies should always be
viewed as potentially dangerous medically, requiring care and Phenothiazines 5.26 5.79
skill. Where risk can be appraised from, for example, family 3.55
history, it may be worth considering emergency treatment as a Butyrophenones 3.32 2.20
‘two-step’ process, where ‘pretreatment’ with intravenous 2.46
benzodiazepine might act against the kinetic effects of extreme Substituted benzamides 1.58
muscular overactivity, making subsequently administered anti- 2.34
psychotic less risky. The compound effects of sedation in com- ‘Atypical’ antipsychotics 2.07
plex strategies such as this would require close monitoring.
‘Standards’ relative to ‘atypical’ controls

Phenothiazines 2.57

Cerebrovascular incidents in the elderly Butyrophenones 1.55 1.44

In 2004, data emerged suggesting that elderly patients treated Substituted benzamides 1.08 0.89
with risperidone for the behavioural and psychological symp-
toms of dementia (BPSD) experienced a three-fold increase Non-users 0.65 0.40
in cerebrovascular incidents, including both transient ischae-
mic attacks (TIA) and stroke, compared to those on placebo. Correlates of stroke
Data for olanzapine and quetiapine were weak but suggestive,
reflecting the relative absence of studies. The European regula- Age Crude incidence doubles over 80 years
tor issued a warning. Subsequently, three studies of aripipra-
zole in psychosis associated with dementia found incidents Gender Higher in males
doubled over placebo and a voluntary restriction was applied
to its use in psychosis in dementia. No clear predisposing fac- ‘Chronic disease score’ Reflecting number of different medications
tors emerged from these studies. taken

If ‘typical’/‘atypical’ does not represent a valid pharmacolog- Comorbidities Parkinson’s disease
ical distinction, these data force the question: is this a class
effect, not restricted to those with dementia? Reviewing the Concurrent medications Anticoagulants
data, the European Pharmacovigilance Working Party found that
data ‘do not provide strong evidence that the risk of stroke. . . . . . Data from Saccheti et al (2008).
does not extend to other. . . . Antipsychotics’ (Pharmacovigi-
lance Working Party Assessment Report 2005). Data on stroke all possible alternatives have been exhausted. Where antipsy-
from an epidemiological cohort study in the elderly (Sacchetti chotics still hold the balance of benefit, the decision-making
et al 2008) strongly suggest a class effect not specifically relevant process should be clearly documented.
to dementia (Table 11.2). The greatest risk seems to lie with
phenothiazines (the low relative risk with substituted benza- Metabolic effects
mides probably reflects small numbers). This translates into
approximately a six-fold increase in crude stroke incidence for Early studies of antipsychotics suggested that up to 30% of
those on phenothiazines and a two to four-fold increase for those patients experienced significant weight gain, defined as 3 kg
on other types of antipsychotic. As with the original data, no over a 6-week exposure interval. While it is doubtful if any
clear associations with conventional cerebrovascular risk factors antipsychotic can be viewed as an angel in this regard (zipraza-
emerged but in contradistinction to these data, dementia per se done and aripiprazole are the most likely to be ‘weight neu-
was not a significant factor (see Table 11.2). This study suggests tral’; Consensus Statement, 2004), it is clear that some are
that age and the frailties accompanying it impart a particular vul- more villainous than others, especially clozapine and olanza-
nerability to cerebrovascular events in those exposed to antipsy- pine. The mechanism of antipsychotic-induced weight gain is
chotics, especially phenothiazines. not understood but in recent years the issue has become linked
to the wider one of metabolic dysfunction.
Despite inherent weaknesses, these data support other
analyses and RCTs with specific compounds in suggesting An increased risk of diabetes in schizophrenic patients was
there is an issue of risk with the use of antipsychotics in the suggested by Kooy as far back as 1919, and was noted with chlor-
elderly and a new and weighty consideration for risk:benefit promazine in 1953, immediately after its introduction. More
appraisal. The use of antipsychotics for treatment of non- recently it has been suggested that schizophrenia itself may rep-
specific symptoms, such as anxiety and insomnia, and first-line resent a significant independent risk for both diabetes and
in BPSD, should be avoided and only enter consideration when impaired glucose tolerance (Bushe & Holt 2004). Good data on
the prevalence of diabetes in untreated psychosis are lacking
but reasonable estimates put it at twice as common as in the nor-
mal population ( 14% vs. 7%; Dixon et al 2000).

239


Click to View FlipBook Version