Evidence-based guidelines for

the pharmacological treatment of

schizophrenia: recommendations from the
British Association for Psychopharmacology
Evidence categories
Categories of evidence for causal relationships
.and treatment
(I) Evidence from meta-analysis of RCTs, at least one
large, good quality, RCT or replicated, smaller, RCTs.
(II) Evidence from small, non-replicated, RCTs, at least one
controlled study without randomization or evidence from
at least one other type of quasi-experimental study. RCTs
must have an appropriate control treatment arm; for primary
efficacy this should include a placebo condition.
(III) Evidence from non-experimental descriptive studies,
such as uncontrolled, comparative, correlation and
casecontrol studies.
(IV) Evidence from expert committee reports or opinions
and/or clinical experience of respected authorities.

Proposed categories of evidence for non-causal

r.elationships
(I) Evidence from large representative population samples.
(II) Evidence from small, well-designed, but not necessarily
representative samples.
(III) Evidence from non-representative surveys, case reports.
(IV) Evidence from expert committee reports or opinions
and/or clinical experience of respected authorities.

Strength of recommendations

The strength of our recommendations are graded A to D, as

described below. Where there was a need to extrapolate from
limited available evidence or clinical opinion, the recommendations
have a weaker grading (B, C or D), although they
may still cover key areas of practice.
Where
recommendations were predominantly derived from a consensus
view, in the absence of valid, systematic evidence, they are
graded as S (standard of good practice).
(A) Directly based on category I evidence.
(B) Directly based on category II evidence or extrapolated
recommendation from category I evidence.
(C) Directly based on category III evidence or extrapolated
recommendation from category I or II evidence.
(D) Directly based on category IV evidence or extrapolated
recommendation from category I, II or III evidence.
(E) Standard of good practice.

Scope and target of the guidelines

The content of these guidelines is relevant for all prescribers

treating patients with schizophrenia. We expect that in most
cases these will be doctors who are specialists in psychiatry.

However, the guidelines were written with an eye also to

informing general practitioners, patients and their families,
and other healthcare providers with an interest in the care
of people with schizophrenia.

Early intervention: Prodromal/at-risk

mental states
Incidence

The incidence of schizophrenia, relatively low at around

15 per 100,000 population a year (Saha et al., 2005) has
long been held to be relatively similar worldwide, reflecting
an apparent genetic aetiology.
the
bulk of which was non-affective psychosis within the schizophrenia
definition. In contrast to the commonly accepted
malefemale ratio of 1.4:1
There is also evidence that risk factors for later illness may
be evident in childhood. For example, in a study of a normal
population of 6000 young boys (Jones et al., 1994), the boys
were classified on the basis of whether they could toddle without
support at 9 months, 10 months, 11 months or 12 months.
The cumulative incidence of schizophrenia in adult life in
these four groups was found to increase progressively with
the age of learning to stand without support.
The findings
suggest that the less efficient the development of motor coordination,
the greater the risk that, as an adult, an individual
may develop schizophrenia.
One interpretation is that there
are common pathophysiological mechanisms between early
developmental processes and adult cognitive function and
schizophrenia.

Classification of the prodromal/at-risk mental state

The prodrome is a classic area of uncertainty, partly because
it has only been the subject of careful research in the last
10 years or so. Definitions vary, but there are two major
forms of categorization. The PACE (Personal Assessment
and Crisis Evaluation Clinic) (Yung et al., 1998) and COPS
(Criteria of Prodromal Syndromes) (Miller et al., 2002) criteria,
commonly used in English-speaking countries, focus on
what are called attenuated positive symptoms, essentially the
positive symptoms that are seen in frank psychotic illness, but
less severe. These criteria can also be fulfilled if there is a
major recent decline in function in someone who has a schizotypal
personality disorder, or a family history of psychosis.
People with this syndrome are said to have an at-risk mental
state for psychosis, or to be at Ultra High Risk of psychosis.
The German research field has developed a different
approach, focusing more on subjective cognitive disturbances
(termed Basic symptoms) than positive symptoms (Haefner
et al., 2004; Klosterko tter et al., 2001). This work suggests
that subjective disturbances in thinking, language and attention
are predictive of later psychosis, but that this is not such
an imminent risk as with other criteria. Early studies showed

that both the PACE and COPS, and the Basic symptoms
criteria, are associated with a high risk (2040%) of progression
to frank psychosis within 2 years

Antipsychotic medication
The potential objectives of pharmacotherapy in cases that
come under the care of clinical services at this early stage
are threefold.
First, for those individuals who are seeking
help for their presenting symptoms (such as attenuated
psychotic symptoms), very low-dose antipsychotic medication
can be considered for short-term symptom relief, although
such a prescription would be off-label in terms of indication.
In general, the dosages of antipsychotics prescribed for
people with an At-risk Mental State are even lower than
those used in first-episode psychosis, as affected individuals
tend to be exquisitely sensitive to both the therapeutic effects
and adverse effects of such medication.
However, individuals with an At-risk Mental State are often
reluctant to take medication, and frequently express a preference
for psychological intervention (Broome et al., 2005).
There is preliminary evidence that both low-dose antipsychotics
and cognitive behavioural therapy (CBT) can improve
presenting symptoms
The second potential objective is to delay, prevent or
reduce the severity of the onset of a psychotic illness. The
findings of a few clinical trials (Larson et al., 2010;
McGlashan et al., 2006; McGorry et al., 2002; Morrison
et al., 2004) suggest that this may be possible with either
low-dose antipsychotic drugs or CBT.
The third objective for treatment is to intervene as soon as
psychosis develops, in order to improve the subsequent outcome.
If subjects at high risk have already been engaged by
mental health services before the onset of illness, the delay
between the onset of frank psychosis and the initiation of
treatment can be substantially reduced. For example, in
South London the mean duration of untreated psychosis
(DUP) in patients who developed psychosis after presenting
to a service for people with an At-risk Mental State
was 10 days, as compared with 12 months in patients
whose first contact with mental health services was after the
onset of illness (Valmaggia et al., 2009). This may account
for the lower rates of hospital admission and compulsory
treatment in the former group (Valmaggia et al., 2009).

Antidepressants
Data are also available as to whether antidepressants in
the At-risk Mental State are effective, although to date
these have been derived from clinical audits rather than
clinical trials. Cornblatt et al. (2007) reported a very low
risk of transition to psychosis in individuals with high-risk
features who had been treated with antidepressants, as
opposed to antipsychotics. Similar findings emerged from
an audit of treatment in at-risk subjects in the UK (FusarPoli et al., 2007). However, it is unclear whether these low
rates of transition are attributable to an effect of the drug
treatment or reflect factors that might lead a clinician to
choose an antidepressant as opposed to an antipsychotic or

psychological treatment for someone with an At-risk Mental

State.
Antidepressants and antipsychotics may also play a role in
the treatment of comorbid depression and anxiety in this
group, which is common (Broome et al., 2005). In clinical
trials, antipsychotic treatment has been associated with an
improvement in both depressive and anxiety symptoms in
the At-risk Mental State (Ruhrmann et al., 2007; Woods
et al., 2007).

Recommendations for the At-risk Mental State

. Encourage a therapeutic relationship to allow for further
assessment, review, watchful waiting and monitoring of
symptoms. (S)
. Assess the nature and impact of any substance use (see the
section below on Pharmacological strategies for comorbid
substance misuse). (S) Substance use in this group is not
common and rarely leads to diagnostic confusion. There is
no evidence that it has any effect on the risk of transition
to psychosis.
. If antipsychotic medication is considered for symptom
relief in the prodromal phase of the illness:
_ This should be treated as off-label prescribing. (S)
_ The prescription should be treated as a short-term,
individual trial. (D)
_ Very low doses should be used. (D)
_ Symptom response should be monitored. (D)
_ Side effects should be carefully monitored. (D)
_ It should be prescribed by specialist psychiatric
services, such as an early intervention team. (D)
. Individual CBT can be considered to be an acceptable
alternative to drug treatment on the preliminary evidence
available. (D)

First-episode psychosis
Diagnosis

Early intervention services are generally designed for firstepisode

psychosis, as opposed to first-episode schizophrenia
or bipolar disorder, as there is often a blurring of affective
and non-affective psychotic features in the early stages.

First-episode psychosis
Diagnosis

Early intervention services are generally designed for firstepisode

psychosis, as opposed to first-episode schizophrenia
or bipolar disorder, as there is often a blurring of affective
and non-affective psychotic features in the early stages.
A definitive diagnosis of the type of psychotic disorder is
frequently postponed until 12 months or so after initial presentation,
by which time differences in psychopathology and
course will have emerged. Typically, within a first-episode
psychosis sample, around 25% have bipolar disorder or
psychotic depression, and only 3040% will meet criteria
for schizophrenia at presentation, although this latter proportion
will increase over time (Singh et al., 2004; Yung et al.,
2003). Schwartz et al. (2000) noted that diagnostic changes
in psychosis over the 2 years after first presentation
may reflect the evolution of an illness, the emergence of new
information, or unreliability of measurement.

DSM-IV diagnostic
criteria for schizophrenia include continuous signs of the
disturbance for at least 6 months, including at least 1 month
of symptoms, and for ICD-10 diagnosis, symptoms must be
clearly present for most of the time during a period of
1 month or more.

Response to antipsychotic medication

The ethos of early intervention services for first-episode
psychosis is to reduce DUP, and to provide high-quality pharmacotherapeutic,
psychological and psychosocial interventions
in the critical early phase of the disorder

Response to antipsychotic medication

The ethos of early intervention services for first-episode
psychosis is to reduce DUP, and to provide high-quality pharmacotherapeutic,
psychological and psychosocial interventions
in the critical early phase of the disorder
(International Early Psychosis Association Writing Group,
2005). The rationale for such an approach is that people are
accessed at a relatively treatment-responsive stage of illness,
and the possible adverse consequences of a putative active
morbid process associated with untreated psychosis may be
minimized by early treatment, thus improving symptomatic
and functional outcomes (Marshall et al., 2005; Perkins et al.,
2005). Whether treatment with antipsychotic medication at
first episode can prevent the progressive structural brain
changes associated with psychotic illness remains uncertain,
despite some early positive findings
These results challenge the almost exclusive use of SGAs
for the treatment of first-onset schizophrenia and schizoaffective
disorder. The safety findings related to weight gain and
metabolic problems (Alvarez-Jiminez et al., 2008a; Tschoner
et al., 2007) raise important public health concerns, given the
widespread use of SGAs in youth for non-psychotic disorders.
Monitoring for metabolic disturbance may be important
within the first 8 weeks of treatment, as such changes can
occur early in antipsychotic treatment (Kelly et al., 2008;
Zhang et al., 2004).

Dosage
There is evidence that first-episode psychosis responds to
lower doses of antipsychotic medication than those required
for the treatment of established schizophrenia, even when
stringent criteria for response are applied
There is a biological sensitivity to such medication in the
early stages of the illness which applies to both the therapeutic
effects and the adverse effects. Thus, there is a consensus that
clinicians should use the lowest recommended dosage of an
antipsychotic when initiating medication in an individual
presenting with their first episode of psychosis.
As little difference has emerged between individual antipsychotic
drugs used for first-episode psychosis, the selection
of an antipsychotic drug will be more dependent on the side
effect profiles as far as these are known, and the perceived
susceptibility to, and tolerability of, particular side effects in
the individual to be treated. Minimizing adverse effects, such

as extrapyramidal and aversive subjective side effects, is

particularly important at this stage given that they can be
a short-term disincentive for medication adherence and have
an impact on attitudes to drug treatment and mental health
care over the longer term

Continuation of antipsychotic medication

A key clinical question is how long antipsychotic medication
should be maintained after the first episode, when the illness is
in remission. Given the absence of reliable predictors of prognosis
or drug response, consensus guidelines recommend
continued antipsychotic medication for every patient diagnosed
with schizophrenia for 12 years

Recommendations for first-episode schizophrenia

. If the onset of psychosis is suspected in primary care, the
person should be referred to specialist mental health services,
ideally an early intervention in psychosis service if
this is available. (S)
. Assess the nature and impact of any substance use (see the
section below on Pharmacological strategies for comorbid
substance misuse). (S)
. Choice of first-line antipsychotic drug should be based on:
_ The evidence for relative liability for side effects,
particularly considering common and serious effects
such as extrapyramidal motor syndromes and metabolic
problems. (B)
_ Individual patient preference. (S)
_ Individual patient risk factors for side effects. (B)
_ Relevant medical history. (S)
. Antipsychotic medication should be initiated at the lower
end of the licensed dosage range. (A)
. An individual trial of the antipsychotic of choice should be
conducted:
_ The indications for oral antipsychotic medication,
the expected benefits and risks, and the anticipated
timeframe for improvement of symptoms and emergence
of side effects should be considered and documented.
(S)
_ Dosage titration should be within the dose range
identified in the British National Formulary (BNF)
or Summary of Product Characteristics (SmPC), and
conducted gradually, based on the response of symptoms
or behaviour and the nature and tolerability of
side effects. (S)
_ The results of symptom and side effect review
should be documented in the clinical records, with
the rationale for any change in medication or its continuation.
(S)
_ Aim to achieve an adequate trial: optimum dosage
with good adherence for 4 weeks. (A)
_ If an FGA is selected, this probably should be a
medium- or low-potency drug rather than a highpotency
drug. (S)
. Following antipsychotic drug initiation, side effects
should be closely monitored with regular, systematic
and comprehensive assessment. Consideration should be
given to the use of formal side effect checklists or rating
scales. (B)