Optimizing Pharmacologic Treatment of Psychotic Disorders

I. MEDICATION SELECTION, DOSING, AND DOSE

EQUIVALENCE

Guideline 1: Selecting Initial Pharmacologic Treatment

Questions 1–3
for a Psychotic Disorder
1A. First-Episode Patient
For a first-episode patient with predominantly positive symptoms, the experts consider oral risperidone the treatment of
choice. Other recommended medications for this clinical situation are aripiprazole, olanzapine, ziprasidone, and queti-
apine (although the first two were rated first line and the second two high second line, these options clustered together
and all were rated first line by approximately two-thirds of the experts).
For a first-episode patient with predominantly negative symptoms, the experts recommend one of the newer oral atypi-
cal antipsychotics. Risperidone and aripiprazole received first line ratings, and the other three were rated high second
line; however, all the options clustered together with only small differences in their confidence intervals.
For a first-episode patient with both prominent positive and negative symptoms, the experts prefer oral risperidone.
Other recommended medications for this clinical situation are aripiprazole, ziprasidone, olanzapine, and quetiapine
(again these four options clustered together with only small differences in their confidence intervals).
The experts as a group varied in their ratings of using a long-acting injectable atypical antipsychotic for a first-episode
patient to such an extent that there was no consensus on this item (with approximately a quarter of the experts rating it
first line and approximately a third giving it third line ratings). The experts did not recommend the use of either oral or
depot conventional antipsychotics for a first-episode patient (conventional antipsychotics received third line ratings in
every case).
(bold italics = treatment of choice)

Presentation First Line* High Second Line Other Second Line

Predominantly positive Risperidone Ziprasidone Long-acting injectable atypical†

psychopathology
Aripiprazole Quetiapine
Olanzapine
Predominantly negative Risperidone Ziprasidone Long-acting injectable atypical
psychopathology
Aripiprazole Olanzapine
Quetiapine
Both prominent positive Risperidone Olanzapine Long-acting injectable atypical
and negative
Aripiprazole Quetiapine
symptomatology
Ziprasidone
*In this survey, we asked only about oral and long-acting injectable formulations of antipsychotics. Unless otherwise specified, all
medications listed in the tables refer to the oral formulation.
†At the time of this survey, a long-acting injectable atypical antipsychotic was not available in the United States, although it was
available in several other countries. In the survey, we asked the experts to rate how they would use such a formulation if it were
available.

J Clin Psychiatry 2003;64 (suppl 12) 21

Expert Consensus Guideline Series

1B. Multi-Episode Patient

For a multi-episode patient with predominantly positive symptoms, the experts consider oral risperidone the treatment
of choice. Other recommended first line medications for this clinical situation are aripiprazole, ziprasidone, olanzapine,
and quetiapine and a long-acting atypical antipsychotic. Clozapine was rated high second line. Other lower rated second
line options were a long-acting conventional antipsychotic (depot) and an oral high-potency conventional.
For a multi-episode patient with predominantly negative symptoms, risperidone, aripiprazole, and ziprasidone were
rated first line; high second line choices were olanzapine, quetiapine, a long-acting atypical antipsychotic, and clozapine.
(It should be noted that all these options tended to cluster together, with only small differences in their confidence inter-
vals.) A long-acting conventional antipsychotic was a lower rated second line option.
For a multi-episode patient with both prominent positive and negative symptoms, the experts preferred risperidone
followed by aripiprazole. Other first line options were ziprasidone and olanzapine. High second line choices were a long-
acting atypical antipsychotic, quetiapine, and clozapine. (Ratings for most of these options tended to cluster together
with only small differences in their confidence intervals.) Other lower rated second line options were a long-acting depot
conventional antipsychotic and an oral high-potency conventional.
The experts are clearly more willing to consider using clozapine or a long-acting injectable antipsychotic in a patient
with a history of previous psychotic episodes. The experts did not recommend the use of mid- or low-potency conven-
tional antipsychotics and gave only very limited support to the use of oral high-potency conventionals.
(bold italics = treatment of choice)

Presentation First Line High Second Line Other Second Line

Predominantly positive Risperidone Clozapine Long-acting conventional

(depot)
psychopathology
Aripiprazole Oral high-potency conventional
Ziprasidone
Olanzapine
Long-acting injectable
atypical
Quetiapine
Predominantly negative Risperidone Olanzapine Long-acting conventional
psychopathology
Aripiprazole Quetiapine
Ziprasidone Long-acting injectable
atypical
Clozapine
Both prominent positive Risperidone Long-acting injectable Long-acting conventional
and negative atypical Oral high-potency conventional
Aripiprazole
symptomatology
Quetiapine
Ziprasidone
Clozapine
Olanzapine

22 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 2: Adequate Dose of AntipsychoticsQuestions 4 & 6

We asked the experts to write-in doses of conventional and atypical antipsychotics that they would recommend in differ-
ent treatment situations. We used the mean and standard deviations of their responses to generate real-world doses
rounded to currently available pill strengths. The experts’ dosing recommendations generally agree closely with recom-
mended doses given in the package labeling. For olanzapine and quetiapine, their recommendations for highest acute
dose are somewhat higher than the highest doses for which safety data from clinical trials are available (20 mg of olan-
zapine and 800 mg of quetiapine). The panel would generally use higher doses for a patient who had had multiple epi-
sodes of psychosis than for a first-episode patient. The recommended dose ranges for maintenance treatment are also
slightly lower than for acute treatment.

First-episode patient Multi-episode patient

Acute Maintenance Acute Maintenance Highest final

treatment treatment treatment treatment acute dose
Medication (mg/day)* (mg/day) (mg/day)* (mg/day) (mg/day)

Atypicals
Aripiprazole 10–20 10–20 15–30 15–20 30
Clozapine 300–500 250–500 400–600 300–550 850
Olanzapine 10–20 10–20 15–25 12.5–22.5 40†
Quetiapine 350–700 300–600 500–800 400–750 950†
Risperidone 2.5–5.0 2.0–4.5 4.0–6.5 3.5–5.5 10.5
Ziprasidone 100–160 80–160 140–180 120–180 180
Conventionals
Chlorpromazine 200–650 150–600 400–800 250–750 950
Fluphenazine 2.5–15.0 2.5–12.5 5.0–22.5 5.0–15.0 25.0
Haloperidol 3.0–13.5 1.5–10.5 7.0–18.5 6.0–13.5 25.0
Perphenazine 8–38 6–36 16–48 12–42 56
Thioridazine‡ 225–550 150–500 350–650 250–550 650
Thiothixene 5–30 2–30 10–40 10–35 40
Trifluoperazine 5–30 2–20 10–35 10–30 40
Fluphenazine
decanoate
(mg/2–3 wk) 12.5–37.5 6.25–37.5 12.5–62.5 12.5–50.0 50.0
Haloperidol
decanoate
(mg/4 wk) 50–200 50–200 100–250 100–200 250
*In beginning treatment with an oral antipsychotic for which titration is not required or with a long-acting injectable antipsychotic, the
experts recommend either starting with a low dose and increasing the dose based on level of response and side effects, or starting with
Questions 10 & 11
a moderate dose. The experts do not recommend starting with a relatively high dose and then decreasing it if possible.

†Safety of doses of olanzapine > 20 mg/day and of quetiapine > 800 mg/day have not been evaluated in clinical trials.

‡The package labeling for thioridazine includes a black box warning stating that this agent “has been shown to prolong the QTc
interval in a dose related manner, and drugs with this potential, including thioridazine, have been associated with torsades de pointes-
type arrhythmias and sudden death. Due to its potential for significant, possibly life-threatening, proarrhythmic effects, thioridazine
should be reserved for use in the treatment of schizophrenic patients who fail to show an acceptable response to adequate courses of
treatment with other antipsychotic drugs.”

J Clin Psychiatry 2003;64 (suppl 12) 23

Expert Consensus Guideline Series

Guideline 3: Therapeutic Drug Monitoring (Using Plasma Levels)Question 5

Over 50% of the experts reported that plasma level assays were available to them only for clozapine, haloperidol, and
haloperidol decanoate. Clozapine was the agent for which the experts considered plasma levels most clinically useful.
Over half the experts use plasma levels of clozapine and haloperidol to monitor compliance; 88% of the experts use
clozapine levels to adjust dose, primarily if there has been an inadequate response or side effects are a problem. 50% of
the experts use plasma levels of haloperidol (oral or decanoate) to adjust dose levels if the patient has an inadequate
response or problematic side effects.

Guideline 4: Duration of an Adequate TrialQuestion 13

If a patient is having little or no response to the initial or to the second antipsychotic that is tried, the experts recommend
waiting a minimum of 3 weeks and a maximum of 6 weeks before making a major change in treatment regimen. If the
patient is showing a partial response to treatment, the experts would extend the duration of the trial somewhat, waiting
4–10 weeks before making a change for the initial antipsychotic and 5–11 weeks for the second antipsychotic. A major
change in treatment regimen could mean either a significant dose increase or switching to a different agent. Note that the
experts would wait longer if the patient is having a partial response, especially in the second trial. Although the differ-
ences were not dramatic, they are interesting, particularly given the lack of data from controlled trials addressing these
issues. These results are similar to those from the 1996 Expert Consensus Guidelines on the Treatment of Schizophre-
nia,* which recommended waiting 3–8 weeks if there is no response and 5–12 weeks if there is a partial response before
switching to another pharmacologic strategy.

4A. Inadequate Response to Initial Antipsychotic

Minimum number of Maximum number of
weeks to wait weeks to wait

Little or no response to treatment 3 6

Partial response to treatment 4 10

4B. Inadequate Response to Second Antipsychotic

Minimum number of Maximum number of
weeks to wait weeks to wait

Little or no response to treatment 3 6

Partial response to treatment 5 11
* McEvoy JP, Weiden PJ, Smith TE, et al. The expert consensus guideline series: treatment of schizophrenia. J Clin Psychiatry
1996;57(suppl 12b):1–58

24 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 5: Dose Equivalency

Question 7
5A. To Haloperidol
We asked the experts to write-in doses of conventional and atypical antipsychotics that they would consider equivalent
to a range of haloperidol doses. We used the mean and standard deviations of their responses to generate real-world
doses rounded to currently available pill strengths. The goal was to obtain a better sense of the equivalency between the
older conventional antipsychotics and the new generation of atypical antipsychotics. In general, the experts’ responses
followed a very linear pattern, indicating that it would probably be possible to use linear formulas to calculate dose
equivalency. It is interesting to note that, in every case, the dose the experts consider equivalent to 30 mg of haloperidol
is higher than the highest acute dose the experts indicated they would usually use (see Guideline 2).

Haloperidol 1 mg 5 mg 10 mg 20 mg 30 mg

Atypicals

Aripiprazole 5 10 20 30 35
Clozapine 75 250 425 675 900
Olanzapine 2.5 10 20 30 45
Quetiapine 100 325 600 900 1200
Risperidone 1.0 3.0 5.5 10.5 15.0
Ziprasidone 40 100 140 180 240

Conventionals

Chlorpromazine 60 250 500 900 1300

Fluphenazine 1 5 10 20 30
Perphenazine 4 16 32 64 88
Thioridazine 50 200 450 750 1000
Thiothixene 3 12 25 40 60
Trifluoperazine 3 12 25 40 55
Fluphenazine 6.25 12.5 25 50 75
decanoate*
(mg/2–3 wk)
Haloperidol 25 100 150 250 300
decanoate*
(mg/4 wk)
*For fluphenazine decanoate and haloperidol decanoate, the experts were asked to indicate the dosage they consider equivalent to that
dose of oral haloperidol being given daily on an ongoing basis.

J Clin Psychiatry 2003;64 (suppl 12) 25

Expert Consensus Guideline Series

Question 8
5B. To Risperidone
We asked the experts to write-in doses of conventional and atypical antipsychotics that they would consider equivalent
to a range of risperidone doses. We used the mean ± the standard deviation of their responses to generate real-world
doses rounded to currently available pill strengths. The goal here was to obtain a better sense of the equivalency of doses
among the new generation of atypical antipsychotics. Again, the experts’ responses generally followed a very linear
pattern, indicating that it would probably be possible to use linear formulas to calculate dose equivalency. It is interesting
to note that the doses the experts consider equivalent to 10 mg of risperidone are closest to those they consider equiva-
lent to 20 mg of haloperidol (as would be expected since they indicated that they considered 10.5 mg of risperidone to be
equivalent to 20 mg of haloperidol, see Guideline 5A).

Risperidone 1 mg 2 mg 4 mg 6 mg 10 mg

Atypicals

Aripiprazole 5 10 15 25 30
Clozapine 75 175 350 500 700
Olanzapine 5 7.5 15 20 30
Quetiapine 100 225 450 600 825
Ziprasidone 40 60 120 160 200

Conventionals

Chlorpromazine 80 175 350 550 800

Fluphenazine 1 5 7.5 12.5 15
Haloperidol 1.5 3.5 7.5 11.5 17
Perphenazine 6 12 24 40 54
Thioridazine 75 150 300 475 650
Thiothixene 4 8 17 25 35
Trifluoperazine 4 10 15 25 35
Fluphenazine 6.25 12.5 25 37.5 50
decanoate*
(mg/2–3 wk)
Haloperidol 25 50 100 150 225
decanoate*
(mg/4 wk)
*For fluphenazine decanoate and haloperidol decanoate, the experts were asked to indicate the dosage that they consider equivalent to
that dose of oral risperidone being given daily on an ongoing basis.

26 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 6: Dose Adjustment

Question 9
6A. Factors to Consider in Dose Adjustment
The experts considered the use of concomitant medications, the patient’s age, and the presence of hepatic disease the
most important factors to consider in adjusting the acute antipsychotic dose. The priority given to the use of concomitant
medications reflects our expanding knowledge of drug-drug interactions and their potential consequences. Other impor-
tant factors to consider are the presence of cardiovascular or renal disease, whether or not the patient smokes, and the
patient’s weight. There was no consensus about the importance of the patient’s sex, with 30% of the experts saying they
would nearly always consider the patient’s sex in dose adjustment and 23% saying they would rarely or never consider
it. It is surprising that many of the experts (45%) would only sometimes consider the patient’s weight in adjusting the
dose. This is consistent with the observation that the determination of psychiatric drug dosage is infrequently influenced
by the patient’s weight, despite the fact that (given the highly lipophilic nature of these compounds) blood levels may
ultimately be influenced by body mass. It may also reflect the pharmaceutical industry’s desire to simplify dosage de-
termination in the treatment of psychiatric disorders.

Always consider Sometimes consider

Use of concomitant medications Presence of cardiovascular disease*

Patient’s age Presence of renal disease
Presence of hepatic disease Whether or not the patient smokes
Patient’s weight
Patient’s sex
*Very high second line

J Clin Psychiatry 2003;64 (suppl 12) 27

Expert Consensus Guideline Series

Question 12
6B. Dose Selection for Special Populations
Dose Selection for Children and Adolescents. A majority of the experts would not generally use the following medica-
tions in children with a psychotic disorder who are 12 years of age or younger: aripiprazole, clozapine, chlorpromazine,
fluphenazine, perphenazine, thioridazine, thiothixene, trifluoperazine, fluphenazine decanoate, and haloperidol de-
canoate. A majority of the experts would not generally use the following medications in an adolescent (13–18 years old)
with a psychotic disorder: chlorpromazine, perphenazine, thioridazine, thiothixene, trifluoperazine. The doses recom-
mended for pediatric patients are generally much lower than those given for adult patients (see Guideline 2), while the
doses recommended for adolescents are only somewhat lower than those recommended for adults. These results under-
score the need for more data on optimum dosing for children and adolescents.
Dose Selection for Elderly Patients. The experts generally recommend using lower doses in elderly patients than in
younger adults. This probably reflects concerns about slower metabolism and greater sensitivity to adverse effects in
older patients. Older patients are also more likely to have comorbid medical conditions and to be taking multiple medi-
cations, increasing the risk for adverse effects and drug-drug interactions. The experts generally recommend using much
lower doses in elderly patients with dementia than in those with a psychotic disorder. The majority of the experts would
not generally use the following medications in an elderly patient with a psychotic disorder or with dementia: chlorpro-
mazine, thioridazine, thiothixene, trifluoperazine; 70% would also avoid haloperidol or fluphenazine decanoate in eld-
erly patients with dementia.

Elderly Patients with

Children with a Adolescents with a Psychotic Dementia with behavioral

Medication psychotic disorder psychotic disorder disorder disturbance and/or psychosis

Atypicals

Aripiprazole (10–15)* 10–20 10–15 10–15

Clozapine (100–350)* 225–450 175–375 50–175
Olanzapine 5–10 10–15 5–15 5–10
Quetiapine 150–400 250–550 225–450 75–300
Risperidone 1.0–2.0 2.5–4.0 1.5–3.5 1.0–3.0
Ziprasidone 40–100 80–140 80–140 40–100

Conventionals

Chlorpromazine (150–200)* (225–375)* (150–300)* (75–150)*

Fluphenazine (1.5–5.0)* 2.5–10.0 2.5–7.5 1.0–5.0
Haloperidol 1.0–4.0 2.0–9.0 2.0–6.0 1.0–3.5
Perphenazine (6–12)* (12–22)* 6–24 2–14
Thioridazine (100–250)* (225–325)* (150–300)* (50–125)*
Thiothixene (4–7)* (4–20)* (2–20)* (1–11)*
Trifluoperazine (2–10)* (6–15)* (4–15)* (3–10)*
Fluphenazine (6.25–12.5)* 12.5–25.0 6.25–25.0 (6.25–12.5)*
decanoate
,† ,
Haloperidol (15–50)* 50–150 25–100 (15–100)* †
decanoate
*A majority of the experts would not generally use this medication in this population.

†Although with current formulations it would be difficult to administer 15 mg of haloperidol decanoate, this low mean suggests that
the experts would be very cautious in dosing if it is decided to use this medication in children or elderly patients with dementia.

28 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 7: Strategies When There Is an Inadequate Response

Question 14
7A. When to Switch Antipsychotics
For each antipsychotic, we asked the experts whether they would increase the dose or switch to another agent if a multi-
episode patient was having an inadequate response to the average target dose of the medication (see Guideline 2 for
recommended target doses). Over 90% of the experts would first increase the dose of clozapine and olanzapine before
switching, going as high as 850 mg/day of clozapine and 40 mg/day of olanzapine. Over 80% would increase the dose of
quetiapine and risperidone before switching, going as high as 1100 mg/day of quetiapine and 10 mg/day of risperidone.
Approximately 60% or more of the experts would also increase the dose of aripiprazole, ziprasidone, and the decanoate
formulations of fluphenazine and haloperidol. The experts are divided fairly evenly as to whether increasing the dose or
switching is the best strategy if a patient is having an inadequate response to the recommended target dose of one of the
conventional oral antipsychotics, except for thioridazine, where 67% would switch to another agent. The experts may be
less willing to increase the dose of the conventional oral medications because of concern about side effects, especially
EPS and TD, at higher doses.

Inadequate response
to adequate dose of Strategy

Atypicals Increase dose Target dose Switch medications

(% of experts) (mg/day) (% of experts)

Aripiprazole 68% 30–35 32%

Clozapine 93% 600–850 7%
Olanzapine 93% 25–40 7%
Quetiapine 84% 650–1100 16%
Risperidone 84% 6–10 16%
Ziprasidone 57% 160–220 43%

Conventionals

Chlorpromazine 56% 550–1300 44%

Fluphenazine 55% 10–30 45%
Haloperidol 52% 10–30 48%
Perphenazine 51% 24–64 49%
Thioridazine 33% 500–800 67%
Thiothixene 49% 25–50 51%
Trifluoperazine 53% 20–55 47%
Fluphenazine
64% 37.5–62.5 mg/2–3 wk 36%
decanoate
Haloperidol decanoate 64% 125–325 mg/4 wk 36%

J Clin Psychiatry 2003;64 (suppl 12) 29

Expert Consensus Guideline Series

Question 15
7B. Switching Antipsychotics: Selecting the Next Agent
We asked the experts to indicate the first and second antipsychotics they would try after an inadequate response to the
initial medication. The table lists those agents written in by 10% or more of the experts in Question 15. Note that, after
trials of two atypical antipsychotics, 30% or more of the experts would switch to clozapine; this was recommended as a
first line strategy in this situation by 70% of the experts in Question 18. The discrepancy between the responses in
Questions 15 and 18 probably reflects differences in the way the question was posed as well as lack of certainty in the
field as to the most appropriate place for clozapine in the treatment algorithm. The editors would endorse the response
given in question 18, where approximately three quarters of the experts recommend switching to clozapine after inade-
quate response to two atypical antipsychotics (see Guideline 7G). For patients who had started with a conventional
antipsychotic, the experts are more likely to try two other atypical antipsychotics before moving on to clozapine.

First medication you Second medication you would

Inadequate response to: would switch to* (%) switch to (%)

Aripiprazole Risperidone (54%) Clozapine (39%)

Olanzapine (19%) Olanzapine (25%)
Ziprasidone (16%) Risperidone (19%)
Clozapine Risperidone (34%) Olanzapine (23%)
Aripiprazole (25%) Quetiapine (17%)
Aripiprazole (13%)
Risperidone (13%)
Ziprasidone (10%)
Olanzapine Risperidone (60%) Clozapine (43%)
Aripiprazole (12%) Aripiprazole (21%)
Ziprasidone (12%) Quetiapine (12%)
Risperidone (10%)
Quetiapine Risperidone (64%) Olanzapine (38%)
Olanzapine (14%) Clozapine (31%)
Aripiprazole (12%) Aripiprazole (14%)
Risperidone Olanzapine (50%) Clozapine (35%)
Aripiprazole (19%) Aripiprazole (25%)
Clozapine (12%) Quetiapine (13%)
Quetiapine (10%)
Ziprasidone (10%)
Ziprasidone Risperidone (44%) Clozapine (34%)
Aripiprazole (21%) Olanzapine (29%)
Olanzapine (21%) Aripiprazole (16%)
Quetiapine (10%) Risperidone (13%)
Chlorpromazine Risperidone (64%) Olanzapine (35%)
Olanzapine (18%) Clozapine (19%)
Quetiapine (14%)
Aripiprazole (11%)
Risperidone (11%)
Ziprasidone (11%)
Fluphenazine Risperidone (62%) Olanzapine (29%)
Olanzapine (16%) Clozapine (18%)
Aripiprazole (11%) Quetiapine (15%)
Risperidone (15%)
Aripiprazole (12%)
Ziprasidone (12%)

30 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

7B. continued

First medication you Second medication you would

Inadequate response to: would switch to* (%) switch to (%)

Haloperidol Risperidone (59%) Olanzapine (28%)

Olanzapine (18%) Clozapine (19%)
Aripiprazole (13%) Quetiapine (14%)
Risperidone (14%)
Ziprasidone (14%)
Aripiprazole (11%)
Perphenazine Risperidone (62%) Olanzapine (29%)
Olanzapine (14%) Clozapine (18%)
Aripiprazole (11%) Quetiapine (15%)
Ziprasidone (11%) Risperidone (15%)
Aripiprazole (12%)
Ziprasidone (12%)
Thioridazine Risperidone (68%) Olanzapine (29%)
Olanzapine (14%) Clozapine (18%)
Aripiprazole (15%)
Risperidone (15%)
Quetiapine (12%)
Ziprasidone (12%)
Thiothixene Risperidone (64%) Olanzapine (30%)
Olanzapine (14%) Clozapine (18%)
Aripiprazole (11%) Risperidone (15%)
Aripiprazole (12%)
Quetiapine (12%)
Ziprasidone (12%)
Trifluoperazine Risperidone (61%) Olanzapine (27%)
Olanzapine (17%) Clozapine (18%)
Aripiprazole (11%) Risperidone (15%)
Ziprasidone (15%)
Aripiprazole (12%)
Quetiapine (12%)
Long-acting injectable Clozapine (27%) Clozapine (40%)
atypical Risperidone (24%) Olanzapine (17%)
Haloperidol decanoate (15%) Aripiprazole (10%)
Ziprasidone (10%)
Injectable fluphenazine Long-acting injectable atypical (38%) Clozapine (41%)
decanoate Risperidone (24%) Olanzapine (21%)
Injectable haloperidol Long-acting injectable atypical (39%) Clozapine (45%)
decanoate Risperidone (22%) Olanzapine (15%)

*If the patient did not respond to the initial antipsychotic you tried and you have switched to another antipsychotic, the experts recom-
mend waiting 3–6 weeks before making a major change in treatment regimen (e.g., switching to yet another antipsychotic) if the patient is
Question 13
having little or no response to treatment, and waiting 5–11 weeks if the patient is having a partial response to treatment.

J Clin Psychiatry 2003;64 (suppl 12) 31

Expert Consensus Guideline Series

Question 15
7C. Switching Antipsychotics: Target Doses
The recommended target doses for the second and third antipsychotics the experts would try are, for the most part,
consistent with the acute target doses shown in Guideline 2, although there is a tendency to consider using doses at the
higher end of the range, especially for the third medication tried.

Dosing of first switch Dosing of second switch

(mg/day) (mg/day)

Atypicals

Aripiprazole 20–30 15–30

Clozapine 350–450 350–500
Olanzapine 15–30 15–25
Quetiapine 550–750 500–800
Risperidone 3.5–7 4.5–8
Ziprasidone 120–160 120–180
Long-acting injectable
37.5–50 mg/2 wk 50 mg/2 wk*
atypical (risperidone)

Conventionals

Fluphenazine — 50*
Haloperidol 10* 10–20
Fluphenazine decanoate 6.25–62.5 mg/2–3 wk 75 mg/2–3 wk*
Haloperidol decanoate 100–250 mg/4 wk 100–450 mg/4 wk
*Only one write in.

32 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

7D. Preferred Switching Strategies for Oral AntipsychoticsQuestion 16

We asked the experts what strategy they would use in switching to each of the oral atypical antipsychotics, assuming that
the first antipsychotic does not require tapering before discontinuation. In switching to any of the oral atypicals except
clozapine, the experts recommend using cross-titration (gradually tapering the dose of the first antipsychotic while
gradually increasing the dose of the second) or overlap and taper (continuing the same dose of the first antipsychotic
while gradually increasing the second to a therapeutic level and then tapering the first). For each drug, a larger percent-
age of the experts considered cross-titration first line. In switching to clozapine, the experts recommend using cross-
titration, probably reflecting the need to institute clozapine treatment gradually and not to withdraw the previous medi-
cation abruptly or prematurely. They would also consider using overlap and taper in switching to clozapine (rated high
second line). The experts do not recommend strategies that involve stopping the first antipsychotic before beginning the
second.

When switching to: First Line High Second Line

Oral atypical antipsychotic Cross-titration

other than clozapine
Overlap and taper
Clozapine Cross-titration Overlap and taper

Question 17
7E. Preferred Switching Strategies for Injectable Antipsychotics
In switching to a depot conventional antipsychotic, the experts recommend either continuing the oral antipsychotic at the
same dose until therapeutic drug levels of the injectable antipsychotic are achieved and then gradually tapering the oral
antipsychotic or else beginning to taper the oral antipsychotic gradually after giving the first injection, with a larger
percentage of the experts favoring the first strategy. Some experts would consider discontinuing the oral antipsychotic
immediately once therapeutic levels of the injectable antipsychotic are achieved.
The experts’ recommendations for switching to a long-acting atypical antipsychotic are similar, except that there is
stronger support for continuing the oral antipsychotic at the same dose until therapeutic drug levels of the injectable
antipsychotic are achieved and then gradually tapering the oral antipsychotic compared with the other options.
It should be noted that the experts definitely do not recommend stopping the oral antipsychotic when the first long-acting
injection is given, since this would leave the patient without adequate antipsychotic coverage during the switchover and
potentially increase the risk of relapse.

When switching to: First Line High Second Line Other Second Line

Depot conventional Continue oral antipsychotic at Continue oral antipsychotic at same dose
until patient achieves therapeutic blood
same dose until patient achieves levels of the injectable antipsychotic and
therapeutic blood levels of the then immediately discontinue the oral
injectable antipsychotic and then antipsychotic
gradually taper the oral
antipsychotic
Taper the oral antipsychotic
gradually after giving the first
long-acting injection
Long-acting Continue oral antipsychotic at Taper the oral antipsychotic gradually after
injectable atypical same dose until patient achieves giving the first long-acting injection
therapeutic blood levels of the Continue oral antipsychotic at same dose
injectable antipsychotic and then until patient achieves therapeutic blood
levels of the injectable antipsychotic and
gradually taper the oral then immediately discontinue the oral
antipsychotic antipsychotic

J Clin Psychiatry 2003;64 (suppl 12) 33

Expert Consensus Guideline Series

Question 19
7F. Strategies When There Is a Partial Response
We asked the experts about the appropriateness of a number of strategies to try to improve response in a patient who is
having a partial but still inadequate response (e.g., a patient with some persisting positive symptoms). The experts gave
only limited support to any of the options and rated many of them third line, probably reflecting the lack of empirical
data concerning these strategies.

If partial response to: First Line High Second Line Other Second Line

Oral conventional Add a long-acting injectable atypical antipsychotic

Add an oral atypical antipsychotic
Add valproate
Add a benzodiazepine

Oral atypical Add a long-acting injectable atypical antipsychotic

Add valproate
Add an oral atypical antipsychotic
Add a benzodiazepine
Add lithium
Add ECT

Depot conventional Add an oral atypical antipsychotic

Add valproate

Question 18
7G. When to Switch to Clozapine
Clozapine is indicated for treatment-refractory schizophrenia. However, clinicians vary in how they define treatment-
refractory illness and there are no universally accepted criteria for treatment-refractoriness in schizophrenia. We there-
fore asked the experts in what clinical situations they would be most likely to consider a switch to clozapine. The experts
consider a trial of clozapine a strategy of choice for a patient who has failed to respond to adequate trials of one or more
conventional antipsychotics and two atypical antipsychotics. They would also consider it a strategy of choice for a
patient who had failed to respond to trials of one or more conventionals and all the atypicals. However, 13% of the
experts rated this option third line, probably because there would be no advantage in trying all the other five atypical
antipsychotics before going to clozapine. The experts also consider a trial of clozapine a first line option for patients who
have failed to respond to trials of two or three atypicals or trials of one or more conventionals and one atypical. Although
some experts would consider clozapine for patients who have not responded to two conventionals or one atypical, there
was much less support for these options. When it is most appropriate to switch to clozapine remains an area of contro-
versy with few data to inform clinical practice. We may in fact be doing our patients a disservice by trying multiple
drugs before going to clozapine.
(bold italics = indications receiving the highest rating from at least 50% of the experts)

First Line High Second Line Other Second Line

Trials of one or more conventional antipsychotics Trials of two conventional antipsychotics

and two atypical antipsychotics Trial of one atypical antipsychotic
Trials of one or more conventional antipsychotics
and all of the other atypical antipsychotics
Trials of three atypical antipsychotics
Trials of two atypical antipsychotics
Trials of one or more conventional antipsychotics
and one atypical antipsychotic

34 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 8: Pharmacologic Strategies for Managing Relapse

Questions 20–22
8A. Relapse When Taking an Oral Antipsychotic
If a patient relapses whom the clinician believes is compliant with medication based on all available evidence (e.g.,
family report, plasma levels), the experts recommend (high second line ratings) either switching to a different oral
antipsychotic or increasing the dose of the current medication. Another second line option the experts would consider is
switching to a long-acting injectable antipsychotic. This probably reflects concerns that the patient may not actually be
compliant, since studies have found that clinicians are often incorrect in their assessment of patients’ compliance. It may
also reflect concerns about absorption problems with the oral formulations.
When the clinician is unsure of the level of compliance or there is clear evidence of noncompliance, the experts’ first
line recommendation is to switch to a long-acting injectable atypical. They would also consider a long-acting conven-
tional depot antipsychotic (high second line). If the clinician is unsure of the level of compliance, the experts would also
consider adding a long-acting atypical to the oral antipsychotic.

Relapse First Line High Second Line Other Second Line

Despite compliance Switch to a different oral Switch to long-acting injectable

atypical antipsychotic
antipsychotic
Add an adjunctive agent
Increase the dose of the
current antipsychotic Add a long-acting injectable
atypical antipsychotic
Add another oral antipsychotic
Switch to long-acting
conventional depot

When unsure of level of Switch to long-acting Switch to long-acting Switch to a different oral
antipsychotic
compliance injectable atypical conventional depot
antipsychotic* Add a long-acting conventional
Add a long-acting depot
injectable atypical
Add an adjunctive agent
antipsychotic
When noncompliant Switch to long-acting Switch to long-acting Switch to a different oral
antipsychotic
injectable atypical conventional depot
antipsychotic
*At the time of this survey, a long-acting injectable atypical antipsychotic was not available in the United States, although it was
available in several other countries. In the survey we asked the experts to rate how they would use such a formulation if it were
available.

J Clin Psychiatry 2003;64 (suppl 12) 35

Expert Consensus Guideline Series

Questions 23, 54
8B. Relapse on a Long-Acting Injectable Antipsychotic
If a patient relapses when receiving a long-acting conventional antipsychotic (depot), the experts’ first line recommen-
dation is to switch to a long-acting injectable atypical antipsychotic. They would also consider increasing the dose or the
frequency of injections of the long-acting conventional (high second line options).
If a patient relapses when receiving a long-acting injectable atypical antipsychotic, the experts’ first line recommenda-
tion is to increase the dose of the injectable antipsychotic. They would also strongly consider adding the oral form of the
injectable antipsychotic to try to boost response (very high second line). The experts do not recommend switching to a
conventional depot antipsychotic (third line rating).

Current Treatment First Line High Second Line Other Second Line

Long-acting depot Switch to long-acting Increase the dose of the Add an oral antipsychotic
conventional injectable atypical long-acting Obtain plasma levels
antipsychotic antipsychotic* conventional
Add an adjunctive agent
antipsychotic
Switch to a different oral
Increase the frequency of antipsychotic
injections of the long-
Switch to a different
acting conventional conventional depot agent if
antipsychotic not previously tried

Long-acting injectable Increase the dose of the Add the oral form of the Add an adjunctive agent
atypical antipsychotic long-acting injectable long-acting injectable Obtain plasma levels
atypical atypical
Add a different oral antipsychotic
Switch to a different oral
antipsychotic

*At the time of this survey, a long-acting injectable atypical antipsychotic was not available in the United States, although it was
available in several other countries. In the survey we asked the experts to rate how they would use such a formulation if it were
available.

36 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 9: Dose Adjustment in Stable PatientsQuestion 24

If the patient is being treated with an atypical antipsychotics or with fluphenazine or haloperidol decanoate, the majority
of the experts would continue maintenance treatment with the same dose that was effective acutely, although over 40%
would lower the dose of olanzapine or risperidone. A majority of the experts said they would lower the dose of an oral
conventional antipsychotic for maintenance treatment; however, the percentages are very close, with 40% or more of the
experts recommending continuing the acute dose of the conventional antipsychotic. The uncertainties shown in this area
are consistent with a lack of information concerning optimum doses for maintenance treatment with both conventional
and atypical antipsychotics.

Medications to continue at acute dose during Target maintenance

maintenance treatment dose if it is decided to
% of experts Medications lower dose*
endorsing this strategy
Atypicals
Aripiprazole 78%
Aripiprazole (10–15)†
Clozapine 66%
Clozapine (225–375)†
Olanzapine 59%
Olanzapine (7.5–15.0)†
Quetiapine 71%
Quetiapine (250–500)†
Risperidone 51%
Risperidone (2.5–4.0)†
Ziprasidone 72%
Ziprasidone (60–120)†
Fluphenazine decanoate 59%
Haloperidol decanoate 58% Conventionals

Chlorpromazine 175–425
Fluphenazine 3.5–10
Haloperidol 3–8
Perphenazine 8–24
Thioridazine 150–350
Thiothixene 7–20
Trifluoperazine 5–20
Fluphenazine decanoate (6.25–25)†
Haloperidol decanoate (50–125)†
*The experts recommend waiting at least 6 months and prefera-
bly a year after a patient has become stable before lowering the
Question 25
dose of the antipsychotic.

†The majority of the experts would not lower the dose of this
medication during maintenance treatment.

J Clin Psychiatry 2003;64 (suppl 12) 37

Expert Consensus Guideline Series

Guideline 10: Managing Complicating Problems

Question 26
10A. Selecting Antipsychotics for Patients With Complicating Problems
The experts consider clozapine the treatment of choice for patients who present with suicidal behavior. This is consistent
with a new indication for clozapine for “reducing the risk of recurrent suicidal behavior.” Clozapine is also the top
choice for aggression and violence. Other highly rated options for aggression and violence are risperidone (rated first
line), olanzapine, and a long-acting injectable atypical (both rated high second line). There were no first line recommen-
dations for the other problems we asked about—dysphoria/depression, cognitive problems, and substance abuse—for
which all of the oral atypical antipsychotics as well as a long-acting injectable atypical received second line ratings. The
experts would also consider a long-acting depot conventional for a patient with substance abuse. The lack of first line
consensus on these items probably reflects the fact that, although an increasing number of studies have looked at the
effects of atypical antipsychotics on mood, cognition, and substance use, the data are not yet sufficiently consistent or
dramatic to influence clinical practice. It is interesting that the experts would not recommend oral conventional antipsy-
chotics for patients with any of the problems that we asked about, except aggression/violence, for which conventional
orals were second line options. It is possible that these complicating problems may be caused or exacerbated by non-
compliance. Therefore, it is not surprising that a long-acting atypical antipsychotic was a prominent alternative, espe-
cially for aggression/violence and substance-abuse problems.
(bold italics = treatments of choice)

Complicating problem First Line* High Second Line Other Second Line

Aggression/violence Clozapine Olanzapine Quetiapine

Ziprasidone
Risperidone Long-acting injectable
atypical Aripiprazole
Long-acting depot conventional
Conventional
Suicidal behavior Clozapine Risperidone Aripiprazole
Quetiapine
Olanzapine
Long-acting injectable atypical
Ziprasidone
Long-acting depot conventional
Dysphoria/depression Olanzapine Quetiapine
Long-acting injectable atypical
Clozapine
Aripiprazole
Risperidone
Ziprasidone
Cognitive problems Risperidone Quetiapine
Long-acting injectable atypical
Aripiprazole
Olanzapine
Ziprasidone
Clozapine
Substance abuse Clozapine Quetiapine
Ziprasidone
Risperidone
Long-acting depot conventional
Long-acting injectable
atypical
Aripiprazole
Olanzapine
*In this survey, we asked only about oral and long-acting injectable formulations of antipsychotics. Unless otherwise specified, all
medications listed in the tables refer to the oral formulation.

38 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Questions 27–30
10B. Selecting Adjunctive Treatments for Patients With Complicating Problems
When we asked about a number of adjunctive medications that are commonly used in clinical practice to treat a variety
of complicating problems in patients with schizophrenia, the experts as a group had few strong recommendations, proba-
bly reflecting the lack of decisive empirical data in this area. The only first line recommendation was a selective seroto-
nin reuptake inhibitor (SSRI) for dysphoria/depression, reflecting studies showing that antidepressants can be helpful for
patients with comorbid depression. Venlafaxine was a very high second line for dysphoria/depression. For aggression
and violence, valproate and lithium received high second line ratings. For suicidal behavior, the same two antidepres-
sants recommended for dysphoria/depression received high second line ratings, with ECT another high second line
option. The question of how to treat persisting negative symptoms has long been a difficult issue in the field. Although
there was no consensus on any of the adjunctive treatments which were rated second line for negative symptoms, it
should be noted that approximately a quarter of the experts or more rated the following options first line: a glutaminergic
agent, an SSRI, another antipsychotic, or venlafaxine.

Complicating problem First Line High Second Line Other Second Line

Aggression/violence Valproate Carbamazepine

Beta-blocker
Lithium
Benzodiazepine
Gabapentin
ECT
Lamotrigine
Topiramate

Suicidal behavior Selective serotonin Mirtazapine

reuptake inhibitor Lithium
(SSRI)
Valproate
Electroconvulsive therapy Bupropion
(ECT)
Nefazodone
Venlafaxine
Lamotrigine

Dysphoria/depression SSRI Venlafaxine ECT

Mirtazapine
Bupropion
Nefazodone
Lithium
Tricyclic antidepressant
Valproate
Lamotrigine
Trazodone

Persisting negative A glutamatergic agent (e.g.,

glycine, cyclo-serine)
symptoms
SSRI
Another antipsychotic
Venlafaxine
A stimulant

J Clin Psychiatry 2003;64 (suppl 12) 39

Expert Consensus Guideline Series

Questions 31, 32
10C. Strategies for a Patient With Clinically Significant Obesity
There is increasing concern about long-term medical problems in patients with schizophrenia, especially obesity and its
complications. Many antipsychotics can contribute to weight gain and clinicians face difficult clinical dilemmas when a
patient with clinically significant obesity (BMI ≥ 30) responds well to a medication that is likely to be contributing to the
patient’s weight problem. If a patient with clinically significant obesity has responded to an antipsychotic other than
clozapine, the experts recommend a trial of a different antipsychotic with less weight gain liability combined with nutri-
tional and exercise counseling if possible. They would also consider (high second line) continuing the same antipsy-
chotic and providing nutritional and exercise counseling to try to help the patient lose weight. However, reflecting the
fact that most patients receiving clozapine have already failed to respond to other agents, the experts would continue
clozapine in this situation and try to address the weight problem with nutritional and exercise counseling. Although the
experts gave a high second line rating to lowering the dose of clozapine in this situation, clinical studies have found that
weight gain does not appear to be a dose-related effect. It is interesting that the experts gave second line ratings to the
addition of topiramate. Although there have been case reports of weight loss with this agent in schizophrenia, there are
no controlled studies supporting this practice. The experts did not recommend the use of weight loss medications
(orlistat, sibutramine) or surgical treatment of obesity in this population.

Clinical presentation First Line High Second Line Other Second Line

Patient who has responded Switch to a different Switch to a different Lower the dose of the current
antipsychotic and provide
well to an antipsychotic antipsychotic with less antipsychotic with less nutritional and exercise
other than clozapine weight gain liability and weight gain liability counseling
provide nutritional and
Continue treatment with Add topiramate and provide
exercise counseling nutritional and exercise
the same antipsychotic
counseling
at the same dose and
provide nutritional and
exercise counseling
Patient with treatment Continue treatment with Lower the clozapine dose Switch to a different
antipsychotic with less weight
resistant illness who has clozapine at the same and provide nutritional gain liability and provide
responded well to dose and provide and exercise counseling nutritional and exercise
clozapine nutritional and exercise counseling
counseling Add topiramate and provide
nutritional and exercise
counseling

40 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Question 33
10D. Monitoring for Comorbid Conditions and Risk Factors
Many patients with schizophrenia rely on their psychiatric care provider for general medical care. With the improving
outcomes being achieved with the newer atypical antipsychotics, more attention is being focused on short- and long-term
health and wellness in this population. We asked the experts which conditions and risk factors they felt it was most
important to monitor. We also asked which ones it was feasible to monitor in a psychiatric treatment setting. The experts
felt that it was important to monitor for all the conditions we asked about, with obesity and diabetes considered the most
important (rated 9 by 60% and 56% of the experts, respectively). The experts’ ratings of feasibility reflect the relative
difficulty of the assessments involved (e.g., it is relatively simple to monitor weight and blood pressure, but much harder
to evaluate osteoporosis). Although we did not ask about obtaining lipid profiles, the editors note that clinicians should
also obtain lipid levels on a regular basis, because some antipsychotics are associated with hyperlipidemia. A recent
expert conference concluded that, as part of routine care, a lipid panel should be obtained if one is not available. Given
that individuals with schizophrenia, as a group, are considered to be at high risk for coronary heart disease, lipid screen-
ing should be carried out at least once every 5 years and more often when there is evidence of lipid levels that approach
those that would lead to treatment.* The same conference also recommended that clinicians should be aware of, and
monitor regularly for, symptoms of increased prolactin. If clinically indicated, prolactin should be measured, and, if
elevated, a work-up for the cause of the elevation should be initiated. Consideration should also be given to switching to
a prolactin-sparing medication—if the symptoms disappear and prolactin levels fall to normal, an endocrine work-up can
then be avoided. Recommendations on other complicating conditions, such as cardiac problems (QTc prolongation and
myocarditis), cataracts, and EPS will also be included in the Mount Sinai guideline when it is published.
(bold italics = conditions receiving the highest rating from at least 50% of the experts)

Conditions and risk First Line Second Line

factors to monitor for

Most important Obesity Galactorrhea

Diabetes Osteoporosis
Cardiovascular problems
HIV risk behavior
Medical complications of substance
abuse
Heavy smoking
Hypertension
Amenorrhea
Most feasible for Obesity HIV risk behavior
psychiatric treatment
Hypertension Medical complications of substance
team to monitor
abuse
Amenorrhea
Osteoporosis
Diabetes
Heavy smoking
Galactorrhea
Cardiovascular problems
*Marder SR, Essock SM, Miller AL, et al. The Mount Sinai Conference on the Health Monitoring of Patients with Schizophrenia. Am
J Psychiatry (submitted)

J Clin Psychiatry 2003;64 (suppl 12) 41

Expert Consensus Guideline Series

II. COMPLIANCE (ADHERENCE)

Guideline 11: Levels of Compliance

Question 36
11A. Defining Levels of Compliance
We provided the experts with the definitions of compliance given below to use as benchmarks in answering a series of
questions about the assessment and management of compliance problems. We also asked them to tell us how they would
define levels of compliance. On average, the expert panel would set a higher threshold for compliance, as shown below,
and would consider a patient who missed more than 65% of his or her medication noncompliant.

Level of compliance Definitions provided in the survey Average of experts’ preferred definitions

Compliant Misses < 20% of medication Misses < 25% of medication

Partially compliant Misses 20%–80% of medication Misses 25%–65% of medication
Noncompliant Misses > 80% of medication Misses > 65% of medication

Questions 34 & 35
11B. Reported Extent of Compliance
Not surprisingly, the experts report that their patients show higher levels of compliance than are generally reported in the
literature.

Level of compliance Levels reported in Experts’ estimate of compliance

the literature levels in their patients

Compliant (misses < 20% of medication) 28% 43%

Partially compliant (misses 20%–80% of medication) 46% 38%
Noncompliant (misses > 80% of medication) 26% 19%

Guideline 12: Assessing ComplianceQuestion 37

The experts consider asking the caregiver or patient first line strategies for assessing compliance; they would also con-
sider pill counts, obtaining blood levels, and using self-rating scales. They did not consider routine use of urine tests
appropriate.

Preferred strategies Also consider

Asking relative or caregiver Pill counts

Asking patient Blood levels
Self-rating scale for compliance

42 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 13: When to Intervene for Compliance ProblemsQuestion 38

The experts were unanimous about the need to intervene if a patient is missing more than 80% of medication. They
would usually intervene if a patient is missing approximately 50% of prescribed medication (91% would usually inter-
vene). The majority of the experts (52%) would also usually intervene when a patient is missing approximately 20% of
medication. There was less agreement about whether to intervene if a patient is only missing occasional doses (13%
would usually intervene, 39% would sometimes intervene, and 48% would generally not intervene).
(bold italics = over 50% of the experts gave the highest rating to intervention)

Usually intervene Sometimes intervene

Patient missing more than 80% Patient missing occasional doses

of medication doses or has
stopped medication completely
Patient missing approximately
50% of medication
Patient missing approximately
20% of medication*
*High second line

Guideline 14: Strategies for Addressing Compliance Problems

Questions 39 & 40
14A. Selecting Initial Strategies
We asked the experts about the appropriateness of three different types of strategies that have been used to address
compliance problems:
•Pharmacologic interventions (e.g., switching to a long-acting medication)
•Psychosocial interventions (e.g., patient education, compliance therapy [focused cognitive-behavioral therapy tar-
geting compliance issues])
•Programmatic interventions (e.g., intensive case management, assertive community treatment)
The experts gave first line ratings to all three types of interventions. The editors note that clinicians should generally
employ a combination of strategies tailored to the specific needs of the patient. The experts gave the highest ratings to
psychosocial interventions for patients who are partially compliant, probably reflecting findings that such interventions
can improve compliance levels. Psychopharmacologic interventions received the highest ratings for noncompliant pa-
tients, probably reflecting the fact that patients who are not taking their medication are at the highest risk for relapse and
it is especially important to try to get the patient back on medication as quickly as possible.
(bold italics = intervention of choice)

Clinical presentation Preferred interventions to improve compliance

Partially compliant Psychosocial interventions

Pharmacologic interventions
Programmatic interventions
Noncompliant Pharmacologic interventions
Programmatic interventions
Psychosocial interventions

J Clin Psychiatry 2003;64 (suppl 12) 43

Expert Consensus Guideline Series

Questions 41 & 42
14B. Psychosocial and Programmatic Interventions to Improve Compliance
Among psychosocial interventions for improving compliance, the experts gave the highest ratings to patient/family
education, medication monitoring, and compliance therapy. Their ratings agree with research findings concerning the
efficacy of these strategies in improving compliance. Findings concerning the efficacy of group and individual psycho-
therapy in improving compliance are equivocal, as shown by the lower ratings given to these options.
Among programmatic interventions the experts recommend assertive community treatment (ACT), ensuring continuity
of treatment provider across treatment settings, and intensive case management services. These recommendations reflect
findings in the literature that intensive case management, in particular the kind of assistance provided by ACT programs,
can significantly improve compliance levels. Lack of continuity in care providers can lead to serious compliance prob-
lems, since patients may be continued on an ineffective or difficult-to-tolerate treatment regimen or may not receive
continuing medication coverage after discharge. The experts also considered supervised residential services, partial
hospitalization, rehabilitation services, and involuntary outpatient commitment useful options for improving compliance.

Psychosocial interventions Programmatic interventions

Preferred Also consider Preferred Also consider

Patient education Symptom and side effect Assertive community Supervised residential
monitoring treatment (ACT) services
Family education and
support Individual or group Continuity of primary Partial hospitalization
psychotherapy clinician across services
Medication monitoring
treatment modalities
Rehabilitation services
Compliance therapy (e.g., inpatient,
(focused cognitive- outpatient, and Involuntary outpatient
behavioral therapy residential programs) commitment
targeting compliance
Intensive services (e.g.,
issues)
contact 1–5 times
weekly or more
frequently as needed)

44 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Questions 43 & 44
14C. Pharmacologic Strategies for Addressing Compliance Problems
There was strong agreement among the experts that the first line pharmacologic strategy for addressing compliance
problems is to switch the patient to a long-acting injectable atypical antipsychotic once this option is available (rated first
line for partially compliant patients and treatment of choice for noncompliant patients). High second line options are to
switch to a long-acting depot conventional or add a long-acting injectable atypical. Another high second line option for a
patient who is partially compliant is to continue the same pharmacotherapy and intensify psychosocial interventions to
improve compliance. However, the experts do not recommend this strategy for a patient who is noncompliant.
(bold italics = treatment of choice)

Clinical presentation First Line High Second Line Other Second Line

Partially compliant Switch to a long-acting Switch to a long-acting Switch to a different oral

antipsychotic that has not
atypical antipsychotic* conventional depot previously been used
antipsychotic
Regular monitoring of plasma
Add a long-acting injectable levels
atypical antipsychotic Add a long-acting conventional
depot antipsychotic
No change in
pharmacotherapy; intensify
psychosocial treatment
Noncompliant Switch to a long-acting Switch to a long-acting Add a long-acting conventional
depot antipsychotic
atypical antipsychotic conventional depot
antipsychotic Regular monitoring of plasma
levels
Add a long-acting injectable
Switch to a different oral
atypical antipsychotic antipsychotic that has not
previously been used

*At the time of this survey, a long-acting injectable atypical antipsychotic was not available in the United States, although it was
available in several other countries. In the survey we asked the experts to rate how they would use such a formulation if it were
available.

J Clin Psychiatry 2003;64 (suppl 12) 45

Expert Consensus Guideline Series

III. LONG-ACTING INJECTABLE ANTIPSYCHOTICS

Guideline 15: Benefits of Long-Acting Injectable AntipsychoticsQuestion 45

The experts consider the greatest benefit of a long-acting injectable antipsychotic to be assured medication delivery.
Other important advantages are the ability to know immediately when a patient misses medication and the fact that the
patient continues to have some medication in his or her system even after a missed dose. Additional advantages are the
reduced risk of relapse associated with continuous medication, and the ability to know that relapse, if it occurs, is not the
result of compliance problems.
(bold italics = benefits receiving the highest rating from at least 50% of the experts)

Most important Somewhat important

Assured medication delivery Regular contact with patient

Knowing immediately when medication is missed Convenience for patient
Reduced risk of relapse Ability to use lower effective dose
Some continuing medication coverage after a missed dose
Knowing that relapse has occurred despite adequate pharmacotherapy

Guideline 16: Potential Disadvantages of Long-Acting Injectable

Question 46
Antipsychotics
The experts consider lack of patient acceptance the most important potential disadvantage of long-acting injectable
antipsychotics. To some extent, this response probably reflects an assumption that patients will not accept the idea of
continuing injections. However, once they try a long-acting medication, many patients are surprised to find how easy it
is to tolerate receiving medication in this way. Although lack of patient autonomy is another potential concern that is
sometimes mentioned, patient surveys do not support this as being a major factor. Although the experts said that they
considered inability to stop medication immediately should side effects become a problem somewhat important as a
potential disadvantage, the editors were hard pressed to find examples of situations in which immediate discontinuation
of an antipsychotic in a long-acting formulation was a medical necessity. Even in neuroleptic malignant syndrome, there
is no evidence that mortality rates are higher among patients receiving a long-acting injectable antipsychotic than in
those receiving an oral medication (assuming that the condition is identified and appropriately treated).

Most important Somewhat important Not too important

Lack of patient acceptance Logistical issues Reimbursement issues

Inability to stop medication immediately should Inadequately established benefit
side effects become a problem
Negative physician perceptions
Stigma associated with injections or depot
clinics
Inadequately appreciated benefit
Local effects of repeated injections

46 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 17: Factors Favoring Use of Long-Acting Injectable

Question 47
Antipsychotics
In deciding whether to use a long-acting injectable antipsychotic, 96% of the experts consider the availability of an
atypical antipsychotic in such a formulation very important. This probably reflects concerns about side effects associated
with the conventional depot antipsychotics. Other factors that the experts consider very important in deciding to use a
long-acting injectable are good patient acceptance of the injection, evidence that the rate of relapses and side effects will
be lower than with oral equivalents, better quality of life for their patients, and ease of administration.
(bold italics = factors receiving the highest rating from at least 50% of the experts)

Most important Somewhat important

Availability of an atypical antipsychotic in a long- Longer interval between injections

acting injectable formulation
Demonstrated superior efficacy to oral equivalent
Good patient acceptance of injection
Easy preparation of injection
Demonstrated fewer relapses/hospital admissions than
Little dose titration required with long-acting injectable
oral equivalent
formulation
Fewer side effects than oral medications
Easy dose conversion from oral equivalent
Better quality of life/patients say they feel better
Easy dose conversion from other oral antipsychotic
Easy administration of injection agent

J Clin Psychiatry 2003;64 (suppl 12) 47

Expert Consensus Guideline Series

Guideline 18: Indications for Switching From an Oral Antipsychotic to a Long-

Questions 48 & 49
Acting Injectable Atypical
We asked the experts about the appropriateness of using a long-acting injectable atypical antipsychotic in a variety of
clinical situations. The experts consider a long-acting atypical antipsychotic the treatment of choice for a patient who is
taking an oral atypical and requests the long-acting formulation, for a patient who relapses because of noncompliance
with an oral atypical antipsychotic, and for a patient who is experiencing EPS on a depot conventional antipsychotic. The
experts consider a long-acting injectable atypical first line for a patient in involuntary outpatient commitment, for a
patient who is chronically relapsing on an oral conventional, for a patient with lack of insight or denial of illness, for a
patient taking an oral atypical antipsychotic who is relapsing for reasons that are unclear, and for a patient with a history
of aggressive or violent behavior. It is interesting that the experts perceive a role for the use of long-acting injectable
atypicals that goes well beyond treatment of patients with compliance problems (see the many other second line indica-
tions listed below). Of all the situations that we asked about, the only ones in which the experts would not generally
consider a long-acting injectable atypical are a patient taking an oral atypical or conventional who is stable and not
experiencing EPS or a patient who has been newly diagnosed with schizophrenia and has had no previous antipsychotic
treatment.

Further recommendations: We asked the experts how concern about the potential for TD would affect their decision to
switch to an injectable atypical antipsychotic. The majority of the experts would definitely switch if there is concern
about TD in a patient who is experiencing EPS on a depot or oral conventional antipsychotic (96% and 73% first line,
respectively). Even if the patient is not experiencing EPS, many of the experts would consider switching from a depot or
oral conventional if there is concern about TD (49% and 38% first line, respectively). The editors were unsure on what
Question 50
basis a clinician would decide that there was in fact no or minimal risk of TD.
We asked the experts about the appropriateness of beginning treatment with a long-acting injectable atypical while
the patient is hospitalized, given shorter lengths of hospital stays. This strategy was rated high second line by the expert
panel, in order to ensure continuing medication coverage when the patient is discharged and to facilitate acceptance of an
injectable medication in outpatient treatment. The experts also noted that this strategy may be helpful because patients
Questions 52 & 53
are most vulnerable to relapse soon after discharge.
(bold italics = indications receiving the highest rating from at least 50% of the experts)

First Line High Second Line Other Second Line

Patient taking an oral atypical History of or potential for suicidal Other severe psychosocial stressor
antipsychotic who requests a long-acting behavior Early episode schizophrenia
antipsychotic
Homelessness Patient taking a depot conventional
Patient taking an oral atypical antipsychotic who is stable and is
Comorbid substance abuse not experiencing serous EPS
antipsychotic who is experiencing relapse problems
because he or she stopped taking Bipolar mania with psychosis
medication Lack of social supports Dementia with psychosis

Patient taking a depot conventional Elderly patient taking an oral Elderly patient taking an oral
conventional antipsychotic who conventional antipsychotic who is
antipsychotic who is stable but having troublesome side effects
experiencing EPS forgets to take medication
A patient with treatment-refractory
Involuntary outpatient commitment Patient taking an oral illness who is taking clozapine
conventional antipsychotic who and having troublesome side
Patient taking an oral conventional is stable but experiencing EPS effects
antipsychotic who is chronically relapsing
Persistent lack of insight/denial of illness
Patient taking an oral atypical antipsychotic
who is experiencing relapse for reasons
that are unclear
History of or potential for aggressive or
violent behavior

48 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 19: Factors Motivating Patients to Return for Repeat InjectionsQuestion 51

The experts consider the influence of family/caregivers and physician/treatment team to be most important in motivating
patients to return for repeat injections.

Most important Somewhat important

Urging/insistence of family or caregivers Involuntary outpatient commitment

Urging of physician/treatment team Contact with treatment team
Decreased risk of relapse
Not having to remember to take oral medication
Convenience
Better efficacy

J Clin Psychiatry 2003;64 (suppl 12) 49

Expert Consensus Guideline Series

IV. DEFINING REMISSION AND RECOVERY

Guideline 20: Indicators of Remission and RecoveryQuestion 55

With improving outcomes, research studies are now trying to evaluate the effectiveness of different antipsychotics not
only in producing remission of symptoms but in promoting long-term recovery in patients with schizophrenia. However,
as yet there is no general consensus on how best to define these terms. We therefore asked the experts to rate the appro-
priateness of a number of factors as indicators of remission and recovery. There was strong agreement that the level of
positive symptoms is the single most important indicator of remission. High second line indicators are levels of cogni-
tive/disorganized, negative, and depressive symptoms, reflecting studies that show that these associated symptoms
contribute in a substantial way to the functional disability associated with schizophrenia. In defining recovery, however,
the experts gave almost equal weight to all of the indicators that we asked about, indicating that recovery is a concept
involving improvement in multiple domains.
Rank ordering of symptomatic indicators: When the experts were asked to rank order four key indicators of remission
and recovery, their responses agreed very closely with those presented in the table below: 89% considered level of
positive symptoms the most important indicator of remission, followed by cognitive/disorganized symptoms, negative
symptoms, and depressive symptoms, all three of which were ranked similarly. However, there was less agreement on
the most important indicator of recovery, with 41% considering level of positive symptoms most important, 33% giving
the highest ranking to level of cognitive/disorganized symptoms, and 28% ranking level of negative symptoms as most
Question 56
important.
Rank ordering of functional outcomes. When asked to rank order three functional outcomes as indicators of remission,
the experts were divided, with 45% considering independent living, 32% occupational/education functioning, and 20%
peer relationships the most important functional indicator of remission. This division among the panel may reflect the
fact that one is unlikely to see major changes in any of these areas in the shorter time frame that is usually used to meas-
ure remission (see Guideline 21). However, when asked about the same functional outcomes as indicators of recovery,
the majority (64%) felt that occupational/educational functioning was the most important functional outcome in recov-
ery, followed by peer relationships (considered most important by 20%) and independent living (considered most im-
portant by 18%). When asked about the most appropriate way of defining functional improvement in their patients, 86%
Questions 57 & 58
of the experts considered relative rather than absolute change in the patient the most appropriate indicator.
(bold italics = indicators receiving the highest rating from at least 50% of the experts)

Remission Recovery

First Line High second line Other second line First line

Level of positive Level of cognitive/ Meaningful peer Occupational/educational functioning

relationships
symptoms disorganized
Meaningful peer relationships
symptoms Ability to live
independently Level of negative symptoms
Level of negative
Occupational/educational
symptoms functioning Ability to live independently
Level of depressive Level of positive symptoms
symptoms
Level of cognitive/disorganized symptoms
Level of depressive symptoms

50 J Clin Psychiatry 2003;64 (suppl 12)

 Optimizing Pharmacologic Treatment of Psychotic Disorders

Guideline 21: Severity and Duration of Symptoms as Indicators of Remission

Questions 59 & 60
and Recovery
We asked the experts what levels of symptom severity were most appropriate to use in defining remission and recovery.
Their ratings are presented in the bar charts below. The majority of the experts would consider a patient in remission
who had mild levels of positive, cognitive/disorganized, negative, and depressive symptoms (62%, 69%, 62%, and 73%
of the experts, respectively). However, a third of the experts felt that no positive symptoms should be present for a
patient to be considered in remission.
The experts’ ratings shifted to the left when asked about indicators for recovery, with a majority (62%) saying that there
should be no positive symptoms for a patient to be considered in recovery. In terms of negative symptoms, 62% of the
panel would consider a patient in recovery who had mild negative symptoms while 33% would look for no negative
symptoms. The panel was more evenly split as to whether a patient could have mild cognitive or depressive symptoms
and still be considered in recovery.
Duration of symptoms. The expert panel said that the improvement in symptomatic indicators should be maintained for
at least 3 months for a patient to be considered in remission and for a year or more for a patient to be considered in
recovery. The experts believe that improvement in functional indicators (occupational/vocational functioning, independ-
ent living, peer relationships) needs to be maintained for somewhat longer, 15–17 months, for the patient to be consid-
ered in recovery.

Severity of symptoms as indicators of remission and recovery

Remission Recovery
None (33%)

Mild (62%)

Moderate (4%)

None (13%)

Mild (69%)

Moderate (18%)

None (7%)

Mild (62%)

Moderate (31%)

None (18%)

Mild (73%)
Moderate (9%)

None (62%)
Mild (33%)

Moderate (4%)

None (44%)

Mild (51%)

Moderate (4%)

None (33%)

Mild (62%)

Moderate (4%)

None (42%)

Mild (51%)

Moderate (7%)

Positive Cognitive/ Negative Depressive Positive Cognitive/ Negative Depressive

symptoms disorganized symptoms symptoms symptoms disorganized symptoms symptoms
symptoms symptoms

J Clin Psychiatry 2003;64 (suppl 12) 51