Welcome to the Schizophrenia Resource Centre

Welcome, this website is intended for healthcare professionals in EMEA with an interest in the treatment of schizophrenia. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

The evolution of antipsychotic switch and polypharmacy in natural practice — A longitudinal perspective

Schizophrenia Research, 1-3, 130, pages 40 - 46



Most patients with schizophrenia first start with a single antipsychotic, and yet most finally end up ‘switching’ or using ‘polypharmacy’. The objective of this study was to examine the evolution of antipsychotic switch and polypharmacy in the real-world from a longitudinal perspective.


A systematic review of longitudinal antipsychotic prescriptions in 300 patients with schizophrenia (ICD-10) for up to 2 years after their first visit to one of the 4 participating psychiatric clinics in Tokyo, Japan between January, 2007 and June, 2008, was conducted. Reasons for prescription change were also examined. The evolution of switching and polypharmacy was studied, and prescribed doses were compared to suggested dose ranges by the Texas Medication Algorithm Project (TMAP).


208 patients started their antipsychotic treatment with monotherapy. 34.1% of the patients gave up monotherapy with an initial antipsychotic to move to antipsychotic switch (27.4%) and/or polypharmacy (17.8%) within 2 years. The main reason for antipsychotic switch was ‘ineffectiveness’; interestingly, this happened despite the fact that the monotherapy dose was below the recommended range in 47.4% of the antipsychotic switch. In a subgroup of 100 patients who started as antipsychotic-free, 2-year prevalence rates of switching and antipsychotic polypharmacy were 27.0% and 18.0%, respectively, and polypharmacy was resorted to after a median of 1 antipsychotic had been tried for 84 days (median).


These findings raise a concern that physicians may perform an antipsychotic switch without exploring the entire dose range and resort to antipsychotic polypharmacy without trying an adequate number of antipsychotics.

Keywords: Antipsychotic, Dose, Polypharmacy, Prescription, Schizophrenia.

1. Introduction

Antipsychotic drugs have been playing a pivotal role in the treatment of schizophrenia for more than 50 years ( Kapur and Mamo, 2003 ). Antipsychotic treatment significantly improves acute symptoms and reduces the risk of relapse; however, this also causes a variety of unwanted adverse events, including motor, metabolic, cognitive, and cardiovascular side effects (Uchida et al, 2003, Ray et al, 2009, Uchida et al, 2009b, Rajji et al, 2010, and Takeuchi et al, 2010), which contributes to poor adherence and unfavorable outcome ( Rettenbacher et al., 2004 ). To achieve optimal psychopharmacological treatment for schizophrenia, several treatment guidelines have been developed (Lehman et al, 2004, Argo et al, 2007, and Buchanan et al, 2009). However, in reality, daily clinical practice has often been reported to deviate from those guidelines (Chen et al, 2000, Sernyak et al, 2003, and Leslie and Rosenheck, 2004).

Deviated dosing of antipsychotic drugs from the guidelines and antipsychotic polypharmacy have been shown to be prevalent in previous cross-sectional prescription surveys (Chen et al, 2000, Sernyak et al, 2003, and Leslie and Rosenheck, 2004). Leslie et al. examined prescription drug records of 53,661 patients with schizophrenia in the US and found that the prescribed antipsychotic dose was outside the range recommended by the schizophrenia Patient Outcomes Research Team (PORT) at 40% ( Leslie and Rosenheck, 2004 ). Similarly, Chen et al. (2000) found that 47% of patients diagnosed with schizophrenia were not dosed within the PORT-recommended range ( Chen et al., 2000 ). Antipsychotic polypharmacy is also common in real-world clinical settings despite limited empirical support ( Pandurangi and Dalkilic, 2008 ) with prevalence rates varying widely (4% to 70%), depending on the setting and patient population (Ito et al, 2005, Stahl and Grady, 2006, Xiang et al, 2007, Koen et al, 2008, Procyshyn et al, 2010, and Santone et al, 2011). Antipsychotic polypharmacy has been linked to excessive dosing ( Procyshyn et al., 2010 ), which in turn is expected to lead to increased dose-dependent adverse events (Jeste et al, 1995, Lemmens et al, 1999, and Ray et al, 2009) — therefore, the use of this mode of therapy should be exceptional and justifiable.

Due to a nature of cross-sectional design in previous prescription surveys, they could not address the dynamic process of prescribing behaviors in the context of time course. For example, previous cross-sectional studies (Chen et al, 2000, Sernyak et al, 2003, Leslie and Rosenheck, 2004, Ito et al, 2005, Stahl and Grady, 2006, and Procyshyn et al, 2010) failed to address the maximum antipsychotic doses patients had received for each antipsychotic used, but just examined the doses prescribed at one time point. Given that antipsychotic dose is usually increased to achieve desirable therapeutic effects, a lack of this perspective greatly limits interpretation of the data. Moreover, reasons for changes in antipsychotic regimen have not systematically been investigated in the context of prescribed dosage. Likewise, how antipsychotic polypharmacy is developed has not yet been systematically studied; in other words, clinical data have been lacking on how many and how long antipsychotic drugs have been tried prior to resorting to antipsychotic polypharmacy. Examination of prescriptions in a longitudinal fashion in the same patients could help elucidate the above issues and therefore provide clinically relevant data with which educational interventions and future clinical trials can be developed.

To examine the evolution of antipsychotic switch and polypharmacy in the real-world, we conducted a systematic chart review that assessed antipsychotic medications prescribed for patients with schizophrenia over 2 years after their first visit.

2. Materials and methods

2.1. Subjects

A systematic chart review was conducted at Keio University Hospital, Ohizumi Hospital, Asakadai Mental Clinic, Ohizumi Mental Clinic, and Ongata Hospital in Tokyo, Japan. Outpatients were included if they visited one of the participating sites for the first time between January 1, 2007 and June 30, 2008, and were diagnosed with schizophrenia (F20 according to the International Classification of Diseases, 10th edition) ( World Health Organization, 1992 ). Outpatients who were hospitalized immediately after their first visit or only sought a second opinion were excluded.

The study was approved by the institutional review board or equivalent committee at all participating sites and exempted from the requirement for informed consent because the study involved de-identified data acquired during routine care.

2.2. Collected information

The collected information included baseline age, sex, antipsychotic medications prescribed for up to 2 years since their first visit, reasons for prescription change, duration of psychosis (DOP), and the dates of their first visit and dropout, if applicable. When patients did not receive any antipsychotic treatment within 2 months of their first visit to one of the participating sites during the target period, they were regarded as antipsychotic-free. Details of prescribed antipsychotic medications included drug names and daily doses of antipsychotics, and start and stop dates of these prescriptions.

When patients did not complete a 2-year outpatient treatment, reasons for dropout were examined and then sorted to one of the followings: “loss of follow-up”, “referral to another clinic/hospital”, and “hospitalization”. Reasons for a regimen change were also examined, based on descriptions in medical charts; they were sorted to one of the following categories: “ineffectiveness”, “side effects”, “symptomatic improvement”, and “unknown”. Prescribing physicians were routinely requested to document reasons for medication changes in participating sites that were tertiary care, university based hospitals. When reasons for a regimen change were not documented, corresponding physicians were contacted for further information by one of the investigators (CT). Onset of illness was identified as the time of the first appearance of at least one of the psychotic symptoms listed in the F20 criteria (a–i). Maximum prescribed doses of antipsychotics for individual episodes of antipsychotic monotherapy were compared with recommended dose ranges by the two major treatment guidelines: the Schizophrenia Patient Outcomes Research Team (PORT) guidelines (Buchanan et al.) and the Texas Medication Algorithm Project (TMAP) ( Argo et al., 2007 ). Dose ranges recommended by these guidelines are summarized in Table 3 .

2.3. Statistical analyses

Statistical analyses were carried out, using the Statistical Package for Social Science (SPSS) version 18.0 for Windows. Differences between groups were tested using χ2 test. A p-value < 0.05 was considered statistically significant and all tests were two-tailed.

3. Results

3.1. Study sample

After an initial review of 4464 patients, 586 patients with schizophrenia who visited one of the participating sites for the first time between January 1, 2007 and June 30, 2008 were identified. Of these, 286 patients were excluded since they were hospitalized immediately after their first visit or only sought a second opinion. Thus, the remaining 300 patients were included in the analysis, among whom 100 patients were regarded as antipsychotic-free. Demographic and clinical characteristics of these patients are summarized in Table 1 . Prescriptions were provided by a total of 48 psychiatrists.

Table 1 Demographic and clinical characteristics and 2-year outcomes of patients.

  Total sample Antipsychotic-free patients
(N = 300) (N = 100)
Age (mean ± SD), years 38.5 ± 15.1 38.2 ± 15.6
Sex: male, N (%) 126 (42) 43 (43)
DOP [mean ± SD (range)], years 9.2 ± 10.7 6.2 ± 9.2
(0.1–60.0) (0.1–50)
Duration of follow-up (mean ± SD), days 383.2 ± 316.0 280.8 ± 308.5
Successful follow-up completion, N (%) 123 (41.0) 32 (32.0)
Loss of follow-up, N (%) 81 (27.0) 39 (39.0)
Referral to another clinic/hospital, N (%) 52 (17.3) 13 (13.0)
Hospitalization, N (%) 44 (14.7) 16 (16.0)

DOP = duration of psychosis.

3.2. Two-year disposition of patients

As shown in Table 1 , approximately 40% of the patients successfully completed the 2-year follow-up period. The most frequent reason for dropout was “loss of follow-up”, followed by “referral to another clinic/hospital” and “hospitalization”. The dropout rate for a subgroup of antipsychotic-free patients was numerically higher at 68%, compared to the whole sample (59%) although the difference was not statistically significant (χ2 = 2.6, p = 0.068).

3.3. Dosing of antipsychotic drugs

A total of 327 episodes of antipsychotic monotherapy were identified. Of these, in 262 episodes, any antipsychotic drugs used were listed up in at least one of the two treatment guidelines (i.e., the PORT and the TMAP), and then maximum prescribed doses of the antipsychotic drugs in these episodes were compared to the suggested dose ranges of those guidelines. As shown in Table 2 , about 58% of the episodes were associated with the recommended dose ranges while the antipsychotic drugs were under-dosed at approximately 40% and excessively dosed in 1–2%. No statistically significant differences in these rates among individual antipsychotic drugs were found ( Table 3 ). In the antipsychotic-free patients, 108 episodes of antipsychotic monotherapy were associated with antipsychotic drugs listed in at least one of the two guidelines. In slightly over half the episodes, the maximum prescribed doses were within the suggested dose range. The rest was lower than the range and there was no single episode of excessive dosing in this specific subgroup.

Table 2 Comparison of maximum prescribed antipsychotic doses in antipsychotic monotherapy episodes with suggested dose ranges of the two treatment guidelines.

  Total sample Antipsychotic-free patients
(N = 256) (N = 262) (N = 107) (N = 108)
Lower than the range, N (%) 105 (41.0) 103 (39.3) 50 (46.7) 50 (46.3)
Within the range, N (%) 148 (57.8) 153 (58.4) 57 (53.3) 58 (53.7)
Higher than the range, N (%) 3 (1.2) 6 (2.3) 0 (0.0) 0 (0.0)

PORT = Schizophrenia Patient Outcomes Research Team guidelines; TMAP = Texas Medication Algorithm Project.

Table 3 Comparison of maximum prescribed doses of individual antipsychotics in antipsychotic monotherapy episodes with suggested dose ranges of the two treatment guidelines.

  Risperidone (N = 83) Olanzapine (N = 62) Aripiprazole (N = 70) Quetiapine (N = 26) Haloperidol (N = 10) Chlorpromazine (N = 3) Fluphenazine (N = 2) Haloperidol depot (N = 2) Fluphenazine depot (N = 4)
Maximum prescribed dose (mean ± SD) 3.0 ± 2.2 mg/day 10.5 ± 6.5 mg/day 12.4 ± 7.0 mg/day 197.6 ± 191.2 mg/day 3.8 ± 2.7 mg/day 83.3 ± 11.8 mg/day 1.0 ± 0.0 mg/day 100.0 ± 0.0 mg/2 weeks 28.0 ± 15.2 mg/2 weeks
Duration of episodes (mean ± SD) 187.8 ± 256.4 days 259.0 ± 285.3 days 266.0 ± 287.0 days 153.9 ± 179.0 days 153.7 ± 224.0 days 145.7 ± 104.3 days 351.5 ± 341.5 days 379.5 ± 350.5 days 551.5 ± 292.7 days
Recommended dose range 2–8 mg/day 2–6 mg/day 10–20 mg/day 10–20 mg/day 10–30 mg/day 10–30 mg/day 300–750 mg/day 300–800 mg/day 6–12 mg/day 2–20 mg/day 300–600 mg/day 300–1000 mg/day 6–12 mg/day 5–20 mg/day N.A. 50–200 mg/2–4 weeks N.A. 6.25–50 mg/2–4 weeks
Lower than the range, N (%) 22 (26.5) 22 (26.5) 27 (43.5) 27(43.5) 27 (38.6) 27 (38.6) 18 (69.2) 18 (69.2) 6 (60.0) 4 (40.0) 3 (100) 3 (100) 2 (100) 2 (100) N.A. 0 (0.0) N.A. 0 (0.0)
Within the range, N (%) 58 (69.9) 55 (66.3) 35 (56.5) 35 (56.5) 43 (61.4) 43 (61.4) 8 (30.8) 8 (30.8) 4 (40.0) 6 60.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) N.A. 2 (100) N.A. 4 (100)
Higher than the range, N (%) 3 (3.6) 6 (7.2) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) N.A. 0 (0.0) N.A. 0 (0.0)

PORT = Schizophrenia Patient Outcomes Research Team guidelines; TMAP = Texas Medication Algorithm Project; N.A. = not available.

Ninety-nine episodes of antipsychotic monotherapy out of the 262 episodes were terminated during the 2-year follow-up period because of an antipsychotic switch. Reasons for the regimen change are shown in Table 4 , the most frequent one being “ineffectiveness” (55%) followed by “side effects”. Also when a subgroup of episodes associated with lower doses (45 and 46 episodes for the PORT and the TMAP, respectively) were separately analyzed, approximately half the antipsychotic regimen change was conducted because of “ineffectiveness”. No statistically significant difference in the reasons among individual antipsychotic drugs was observed.

Table 4 Maximum antipsychotic dosage and reasons for terminating the regimen in 99 antipsychotic monotherapy episodes.

  PORT (N = 97) TMAP (N = 99)
Maximum prescribed antipsychotic dose   Duration of episodes   Duration of episodes
(mean ± SD), days (mean ± SD), days
Lower than the range, N (%) 46 (47.4) 55.5 ± 66.4 45 (45.5) 56.4 ± 66.8
Within the range, N (%) 51 (52.6) 121.5 ± 138.0 52 (52.5) 114.1 ± 130.9
Higher than the range, N (%) 0 (0.0) N.A. 2 (2.0) 224.0 ± 210.0
Reasons for a total of 97 and 99 episodes, respectively
Ineffectiveness, N (%) 53 (54.6) 101.3 ± 128.8 54 (54.5) 100.0 ± 128.0
Side effect, N (%) 40 (41.2) 79.7 ± 96.6 41 (41.4) 78.8 ± 95.5
Unknown, N (%) 4 (4.1) 48 ± 46.7 4 (4.0) 48.0 ± 46.7
Reasons for a subgroup of 46 and 45 episodes sorted to “lower than the range”, respectively
Ineffectiveness 21 (45.7) 59.3 ± 83.1 21 (46.7) 59.3 ± 83.1
Side effect 22 (47.8) 52.3 ± 47.7 21 (46.7) 54.4 ± 47.2
Unknown 3 (6.5) 50.0 ± 53.8 3 (6.7) 50.0 ± 53.8

PORT = Schizophrenia Patient Outcomes Research Team guidelines; TMAP = Texas Medication Algorithm Project; N.A. = not available

3.4. Antipsychotic polypharmacy

One-hundred forty-two patients had received, at least once, more than one antipsychotic drug at the same time during the 2-year period. Of these, 11 patients transiently received antipsychotic polypharmacy during a switch to another antipsychotic; thus, 131 patients (43.7%) received treatment with “true” antipsychotic polypharmacy. A total of 285 episodes of antipsychotic polypharmacy were identified in the 131 patients. A wide variety of combinations were found to be used; the most frequent actual combination was risperidone plus chlorpromazine (N = 16, 5.6%), followed by aripiprazole plus quetiapine (N = 13, 4.6%), risperidone plus quetiapine (N = 12, 4.2%), risperidone plus olanzapine (N = 10, 3.5%), and risperidone plus aripiprazole (N = 10, 3.5%).

All episodes of increase in the number of prescribed antipsychotic drugs (e.g., antipsychotic monotherapy to two antipsychotics, two antipsychotics to three) were reviewed. A total number of such episodes was 124; of these, the increment in 29 episodes was transient due to an antipsychotic switch. Consequently, the remaining 95 episodes were examined. The most common reason for the increment in the number of prescribed antipsychotic drugs was “ineffectiveness” (N = 76, 80%), followed by “side effects” (N = 14, 14.7%), and “unknown” (N = 5, 5.3%).

An additional analysis in a subgroup of 100 antipsychotic-free patients found the number of patients who had received antipsychotic polypharmacy was 22 (22.0%) during the 2-year period. The rate of true polypharmacy was 18 (18.0%) after excluding transient polypharmacy in a process of antipsychotic switch. These patients received antipsychotic monotherapy for 66 ± 101 (mean ± SD) (median, 8.5 days; range, 0–365) and tried 1.1 ± 1.4 antipsychotic drugs (mean ± SD) (median, 1; range, 0–5) before starting antipsychotic polypharmacy ( Fig. 1 ). When the seven patients who started antipsychotic treatment with more than one antipsychotics on their first visit were excluded, antipsychotic polypharmacy was resorted to after a median of 1 antipsychotic (mean ± SD, 1.8 ± 1.4) had been tried for 84 days (median) (mean ± SD, 108.1 ± 116.0). In this subgroup, a 2-year prevalence rate of antipsychotic switch was 27.0% (N = 27).


Fig. 1 Disposition of 100 antipsychotic-free patients.

4. Discussion

We conducted a longitudinal chart review that assessed antipsychotic drugs prescribed for schizophrenia in a naturalistic setting in Japan and observed that physicians sometimes judged antipsychotics as being ‘ineffective’ without sufficiently exploring the entire dose range. Another interesting finding is that the antipsychotic polypharmacy was employed in as short as 2 months and with as few as a single antipsychotic trial, suggesting that physicians may resort to antipsychotic polypharmacy without trying an adequate number of antipsychotics in the treatment of schizophrenia.

Antipsychotic dosage prescribed during the follow-up period was found to be modest or somewhat low in almost all the patients, irrespective of whether they were antipsychotic-free or not at baseline. Intriguingly, antipsychotics were switched to another in approximately half the antipsychotic monotherapy episodes even though the maximum dose of the previous antipsychotic drugs did not reach even the lower limit of the recommended dose range. This would be understandable if patients experienced side effects from these medications; however, the rates of antipsychotic switch due to side effects and insufficient effectiveness were similar. These data suggest that physicians may tend to switch a current antipsychotic drug to another one without sufficiently increasing the dose in spite of a lack of side effects. Still, one clinical question still remains — should we further increase the dose when the dose is already within the recommended dosage range and side effects are not observed? There is little evidence to support to go beyond the recommended doses; however, to our knowledge, only one study by Kinon et al. (1993) systematically addressed this question ( Kinon et al., 1993 ). Thus, a paucity of the data on this clinically relevant dosing issue still limits our dosing strategies for the treatment of schizophrenia ( Uchida et al., 2009d ), emphasizing the need of pragmatic clinical trials to further investigate this topic.

In the present study, approximately one-third of the patients received antipsychotic polypharmacy, which is within the range of rates reported in the previous surveys (Ito et al, 2005, Koen et al, 2008, Procyshyn et al, 2010, Santone et al, 2011, Stahl and Grady, 2006, and Xiang et al, 2007) although direct comparison is difficult in light of differences in clinical settings and patient populations. As expected, when a subgroup of antipsychotic-free patients was separately analyzed, the use rate was lower (18.0%). However, it should be noted that the mean number of antipsychotic trials was as small as one and the duration was as short as approximately two months before antipsychotic polypharmacy was employed. Furthermore, 7 patients (7.0%) were started with two antipsychotic drugs on their first visit; this trend may not always be unique only in Japan given a comparable rate (3.7%) in the California's Medicaid claims data ( Lee and Walker, 2008 ). Although unavailability of clozapine until 2010 may contribute to this hasty resort to antipsychotic polypharmacy to some extent, these findings suggest that it may indeed be employed without sufficient trials of antipsychotic monotherapy in the real-world. The evidence generally does not support the use of antipsychotic polypharmacy; for example, one recent randomized, double-blind, placebo-controlled trial failed to show any added efficacy of the addition of aripiprazole to risperidone or quetiapine ( Kane et al., 2009 ). A possibility of effectiveness of treatment with combined antipsychotic treatment cannot entirely be rejected in difficult-to-treat patients (Suzuki et al, 2008 and Correll et al, 2009). Previous surveys that interviewed physicians who prescribed antipsychotic polypharmacy demonstrated that polypharmacy was utilized because of a failure of antipsychotic monotherapy and/or physicians' disbelief in treatment guidelines (Sernyak and Rosenheck, 2004 and Ito et al, 2005). Still, in light of increased risks of serious dose-dependent adverse effects, including cardiac sudden death ( Ray et al., 2009 ), antipsychotic polypharmacy should be utilized with a great degree of caution.

This study gathered the data on patients treated in naturalistic settings, obviating any bias and reflecting the actual clinical needs of the patients with respect to antipsychotic dosing. The fact that the rates of antipsychotic switch due to side effects and insufficient effectiveness were similar could implicate different sensitivity to antipsychotics on an individual basis (Uchida et al, 2009a, Uchida et al, 2009b, and Uchida and Mamo, 2009). In addition, the larger SDs than means (i.e., variability) in the days on antipsychotic monotherapy and in the number of antipsychotics previously tried in antipsychotic-free patients may be suggestive of wide interindividual differences in physicians' preferences in antipsychotic treatment as well as patients' responsivity to antipsychotic drugs.

The results of our study must be interpreted in light of a number of limitations. First, the total number of patients included in the present study was relatively small despite an initial review of 4464 charts. Second, due to the nature of the retrospective study design, reasons for a regimen change were not always identified although the rate of such episodes was as small as 4%. Still, no systematic usage of clinical rating scales limits the interpretation of the data. Furthermore, it would have been ideal to gather other collateral information or to examine detailed reasoning behind physicians' prescribing decisions in a prospective manner. On the other hand, reasons for treatment decisions need to be shared with other treatment staff, including physicians and nurses. A lack of such documentation is seriously problematic since a treatment failure, especially side effects, could be repeated by succeeding physicians who are not aware of the past history. Third, prescribing behaviors may reflect a bias in training and direct and indirect influence of current and local standard of care (Uchida et al, 2008 and Uchida et al, 2009c) as well as the dissemination of the recommendation. In addition, all patients included in this survey were Asian, which may limit the extrapolation and direct comparison of our results to those in other ethnic groups. In fact, differences in the activity of cytochrome P450 among ethnicities could have effects on antipsychotic doses needed for therapeutic/side effects. For examples, Asians and Pacific islanders have a high frequency of approximately 40% of a reduced function allele, CYP2D6*10 ( Bradford, 2002 ), which results in slower metabolism and in turn may lead to lower dose requirement for antipsychotics metabolized by CYP2D6, such as risperidone and aripiprazole. The Japanese Society of Psychiatry and Neurology updated the treatment guidelines for schizophrenia in 2008 where they suggested therapeutic dose ranges of several atypical antipsychotics for Japanese patients ( Japanese Society of Psychiatry and Neurology, 2008 ). They suggest the same dose ranges for risperidone (2–6 mg/day) and olanzapine (10–20 mg/day) as those recommended by the TMAP while aripiprazole and quetiapine are recommended to be given at lower doses of 6–24 mg/day and 150–600 mg/day, respectively, compared to the TMAP. We have employed an additional analysis to apply these lower dose ranges for quetiapine and aripiprazole to the present dataset. In this additional analysis, we found that 10.0% and 57.7% of monotherapy episodes of aripiprazole and quetiapine were sorted to “lower than the recommended dose ranges”, respectively. Although a majority of subjects were receiving risperidone and olanzapine whose recommended dose ranges for Japanese were the same as those in the TMAP, these potential genetic effects should be taken into consideration when the data are interpreted. Fourth, sex differences in the P450 3A4 activity should be paid attention to; indeed, women would be expected to require lower doses of olanzapine to achieve given clinical effects ( Bigos et al., 2008 ). We therefore performed an additional analysis in patients who were treated with olanzapine with a focus on sex and found that the rates of episodes sorted to “lower than the recommended dose range” were comparable between men and women (38.1% vs. 46.3%; n.s. by chi-squared test or Fisher's exact test). Therefore, the potential impact of sex on our results may be limited in the present study. Finally, adherence to the regimen could not be assessed in this chart review because medication adherence was not routinely assessed in participating clinics or hospitals. In addition, the presence of substance use was not examined, either, which could confound the results. Furthermore, if information on p.r.n. medications had been included, the rate of polypharmacy would have increased. Moreover, it would have been ideal to collect the information on smoking status that could have impacts on the clearance of several antipsychotic drugs, including olanzapine ( Bigos et al., 2008 ).

In conclusion, the results from the present survey raise a concern that physicians may perform an antipsychotic switch without exploring the entire dose range suggested by treatment guidelines for schizophrenia. Moreover, physicians may resort to antipsychotic polypharmacy without trying an adequate number of antipsychotics. To either advocate or criticize these prescribing behaviors is beyond the purpose of this study; rather, in light of a lack of robust findings in the literature, the findings highlight the need of further ‘real world’ investigations on the dosing issues. Future pragmatic clinical trials seeking any pros and cons of these prescribing habits that we often see in our daily clinical practice are critically needed to understand and optimize what really happens in the real world of psychopharmacological intervention for schizophrenia.

Role of funding source

This work was funded by Japan Research Foundation for Clinical Pharmacology, Tokyo, Japan (HU), Research Group for Schizophrenia, Tokyo, Japan (HU), Grant-in-Aid for Young Scientists-B from the Ministry of Education, Culture, Sports, Science and Technology, Tokyo, Japan (HU), Japanese Society of Clinical Neuropsychopharmacology, Tokyo, Japan (TS), Government of Canada Post-Doctoral Research Fellowships, Ottawa, Canada (TS), Kanae Foundation, Tokyo, Japan (TS) and Mochida Memorial Foundation, Tokyo, Japan (TS).


All authors contributed to and have approved the design and the protocol of the study and the literature searches. Drs. Tsutsumi and Uchida managed the collection of the data and analyses. Dr. Tsutsumi wrote the first draft of the manuscript, and all authors contributed to and have approved the final manuscript.

Conflicts of interest

Dr. Uchida has received grants, speaker's honoraria, or manuscript fees from Pfizer Health Research Foundation, GlaxoSmithKline, Otsuka Pharmaceutical, Dainippon Sumitomo Pharma, Janssen Pharmaceutical, and Pfizer within the past 5 years. Dr. Suzuki has received manuscript fees from Dainippon Sumitomo Pharma and Kyowa Hakko Kirin within the past 5 years. Dr. Takeuchi has received speaker's honoraria or manuscript fees from Dainippon Sumitomo Pharma, Eli Lilly, GlaxoSmithKline, Janssen Pharmaceutical, and Otsuka Pharmaceutical within the past 5 years. Dr. Nakajima has received grants from Pfizer and speaker's honoraria or manuscript fees from Dainippon Sumitomo Pharma, GlaxoSmithKline, Jannsen Pharmaceutical and Pfizer within the past 5 years. Dr. Imasaka has received speaker's honoraria from Eli Lilly, Pfizer, Meiji, GlaxoSmithKline, Janssen Pharmaceutical, and Otsuka within the past 5 years. Dr. Tsutsumi has received consultant fees or speaker's honoraria from Astellas Pharma, Eli Lilly, Otsuka Pharmaceutical, Dainippon Sumitomo Pharma, Janssen Pharmaceutical, GlaxoSmithKline, and Schering-Plough within the past 5 years. Dr. Watanabe has received grants, or consultant fees from Dainippon Sumitomo Pharma, Eli Lilly, GlaxoSmithKline, Janssen Pharmaceutical, and Pfizer, and received speaker's honoraria from Astellas Pharma, Dainippon Sumitomo Pharma, Eli Lilly, GlaxoSmithKline, Janssen Pharmaceutical, Meiji, Otsuka Pharmaceutical, Pfizer, and Yoshitomiyakuhin within the past 5 years. Dr. Kapur has received grant support from AstraZeneca and GSK; and has served as consultant and/or speaker for AstraZeneca, Bioline, BMS-Otsuka, Eli Lilly, Janssen (J&J), Lundbeck, NeuroSearch, Pfizer, Roche, Servier and Solvay Wyeth. Other authors have nothing to disclose.




  • Argo et al., 2007 T.R. Argo, M.L. Crismon, A.L. Miller, T.A. Moore, S.D. Bendele, B. Suehs. Texas Medication Algorithm Project Procedural Manual: Schizophrenia Algorithm. (The Texas Department of State Health services, 2007)
  • Bigos et al., 2008 K.L. Bigos, B.G. Pollock, K.C. Coley, D.D. Miller, S.R. Marder, M. Aravagiri, M.A. Kirshner, L.S. Schneider, R.R. Bies. Sex, race, and smoking impact olanzapine exposure. J Clin Pharmacol. 2008;48:157-165 Crossref
  • Bradford, 2002 L.D. Bradford. CYP2D6 allele frequency in European Caucasians, Asians, Africans and their descendants. Pharmacogenomics. 2002;3:229-243 Crossref
  • Buchanan et al., 2009 R.W. Buchanan, J. Kreyenbuhl, D.L. Kelly, J.M. Noel, D.L. Boggs, B.A. Fischer, S. Himelhoch, B. Fang, E. Peterson, P.R. Aquino, W. Keller. The 2009 schizophrenia PORT psychopharmacological treatment recommendations and summary statements. Schizophr Bull. 2009;36:71-93
  • Chen et al., 2000 R.S. Chen, P.M. Nadkarni, F.L. Levin, P.L. Miller, J. Erdos, R.A. Rosenheck. Using a computer database to monitor compliance with pharmacotherapeutic guidelines for schizophrenia. Psychiatr Serv. 2000;51:791-794 Crossref
  • Correll et al., 2009 C.U. Correll, C. Rummel-Kluge, C. Corves, J.M. Kane, S. Leucht. Antipsychotic combinations vs monotherapy in schizophrenia: a meta-analysis of randomized controlled trials. Schizophr Bull. 2009;35:443-457 Crossref
  • Ito et al., 2005 H. Ito, A. Koyama, T. Higuchi. Polypharmacy and excessive dosing: psychiatrists' perceptions of antipsychotic drug prescription. Br J Psychiatry. 2005;187:243-247 Crossref
  • Japanese Society of Psychiatry and Neurology, 2008 Japanese Society of Psychiatry and Neurology. the second edition in: Treatment Guidelines for Schizophrenia. (Igaku Shoin, Tokyo, 2008) [in Japanese]
  • Jeste et al., 1995 D.V. Jeste, M.P. Caligiuri, J.S. Paulsen, R.K. Heaton, J.P. Lacro, M.J. Harris, A. Bailey, R.L. Fell, L.A. McAdams. Risk of tardive dyskinesia in older patients. A prospective longitudinal study of 266 outpatients. Arch Gen Psychiatry. 1995;52:756-765 Crossref
  • Kane et al., 2009 J.M. Kane, C.U. Correll, D.C. Goff, B. Kirkpatrick, S.R. Marder, E. Vester-Blokland, W. Sun, W.H. Carson, A. Pikalov, S. Assuncao-Talbott. A multicenter, randomized, double-blind, placebo-controlled, 16-week study of adjunctive aripiprazole for schizophrenia or schizoaffective disorder inadequately treated with quetiapine or risperidone monotherapy. J Clin Psychiatry. 2009;70:1348-1357 Crossref
  • Kapur and Mamo, 2003 S. Kapur, D. Mamo. Half a century of antipsychotics and still a central role for dopamine D2 receptors. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27:1081-1090 Crossref
  • Kinon et al., 1993 B.J. Kinon, J.M. Kane, C. Johns, R. Perovich, M. Ismi, A. Koreen, P. Weiden. Treatment of neuroleptic-resistant schizophrenic relapse. Psychopharmacol Bull. 1993;29:309-314
  • Koen et al., 2008 L. Koen, P. Magni, D.J. Niehaus, A. le Roux. Antipsychotic prescription patterns in Xhosa patients with schizophrenia or schizoaffective disorder. Afr J Psychiatry. 2008;11:287-290
  • Lee and Walker, 2008 B. Lee, V. Walker. Polypharmacy as the initial second-generation antipsychotic treatment. Psychiatr Serv. 2008;59:717 Crossref
  • Lehman et al., 2004 A.F. Lehman, J.A. Lieberman, L.B. Dixon, T.H. McGlashan, A.L. Miller, D.O. Perkins, J. Kreyenbuhl. Practice guideline for the treatment of patients with schizophrenia, second edition. Am J Psychiatry. 2004;161:1-56
  • Lemmens et al., 1999 P. Lemmens, M. Brecher, B. Van Baelen. A combined analysis of double-blind studies with risperidone vs. placebo and other antipsychotic agents: factors associated with extrapyramidal symptoms. Acta Psychiatr Scand. 1999;99:160-170 Crossref
  • Leslie and Rosenheck, 2004 D.L. Leslie, R.A. Rosenheck. Adherence of schizophrenia pharmacotherapy to published treatment recommendations: patient, facility, and provider predictors. Schizophr Bull. 2004;30:649-658 Crossref
  • Pandurangi and Dalkilic, 2008 A.K. Pandurangi, A. Dalkilic. Polypharmacy with second-generation antipsychotics: a review of evidence. J Psychiatr Pract. 2008;14:345-367 Crossref
  • Procyshyn et al., 2010 R.M. Procyshyn, W.G. Honer, T.K. Wu, R.W. Ko, S.A. McIsaac, A.H. Young, J.L. Johnson, A.M. Barr. Persistent antipsychotic polypharmacy and excessive dosing in the community psychiatric treatment setting: a review of medication profiles in 435 Canadian outpatients. J Clin Psychiatry. 2010;71:566-573 Crossref
  • Rajji et al., 2010 T.K. Rajji, H. Uchida, Z. Ismail, W. Ng, D.C. Mamo, G. Remington, B.G. Pollock, B.H. Mulsant. Clozapine and global cognition in schizophrenia. J Clin Psychopharmacol. 2010;30:431-436 Crossref
  • Ray et al., 2009 W.A. Ray, C.P. Chung, K.T. Murray, K. Hall, C.M. Stein. Atypical antipsychotic drugs and the risk of sudden cardiac death. N Engl J Med. 2009;360:225-235 Crossref
  • Rettenbacher et al., 2004 M.A. Rettenbacher, A. Hofer, U. Eder, M. Hummer, G. Kemmler, E.M. Weiss, W.W. Fleischhacker. Compliance in schizophrenia: psychopathology, side effects, and patients' attitudes toward the illness and medication. J Clin Psychiatry. 2004;65:1211-1218 Crossref
  • Santone et al., 2011 G. Santone, C. Bellantuono, P. Rucci, A. Picardi, A. Preti, G. de Girolamo. Patient characteristics and process factors associated with antipsychotic polypharmacy in a nationwide sample of psychiatric inpatients in Italy. Pharmacoepidemiol Drug Saf. 2011;20:441-449 Crossref
  • Sernyak and Rosenheck, 2004 M.J. Sernyak, R. Rosenheck. Clinicians' reasons for antipsychotic coprescribing. J Clin Psychiatry. 2004;65:1597-1600 Crossref
  • Sernyak et al., 2003 M.J. Sernyak, D. Dausey, R. Desai, R. Rosenheck. Prescribers' nonadherence to treatment guidelines for schizophrenia when prescribing neuroleptics. Psychiatr Serv. 2003;54:246-248
  • Stahl and Grady, 2006 S.M. Stahl, M.M. Grady. High-cost use of second-generation antipsychotics under California's Medicaid program. Psychiatr Serv. 2006;57:127-129 Crossref
  • Suzuki et al., 2008 T. Suzuki, H. Uchida, K. Watanabe, S. Nakajima, K. Nomura, H. Takeuchi, A. Tanabe, G. Yagi, H. Kashima. Effectiveness of antipsychotic polypharmacy for patients with treatment refractory schizophrenia: an open-label trial of olanzapine plus risperidone for those who failed to respond to a sequential treatment with olanzapine, quetiapine and risperidone. Hum Psychopharmacol. 2008;23:455-463
  • Takeuchi et al., 2010 H. Takeuchi, H. Uchida, T. Suzuki, K. Watanabe, H. Kashima. Changes in metabolic parameters following a switch to aripiprazole in Japanese patients with schizophrenia: one-year follow-up study. Psychiatry Clin Neurosci. 2010;64:104-106 Crossref
  • Uchida and Mamo, 2009 H. Uchida, D.C. Mamo. Predicting age-specific dosing of antipsychotics. Clin Pharmacol Ther. 2009;86:360-362 Crossref
  • Uchida et al., 2003 H. Uchida, T. Suzuki, K.F. Tanaka, K. Watanabe, G. Yagi, H. Kashima. Recurrent episodes of perceptual alteration in patients treated with antipsychotic agents. J Clin Psychopharmacol. 2003;23:496-499 Crossref
  • Uchida et al., 2008 H. Uchida, T. Suzuki, D.C. Mamo, B.H. Mulsant, A. Tanabe, A. Inagaki, K. Watanabe, G. Yagi, M. Tomita. Effects of age and age of onset on prescribed antipsychotic dose in schizophrenia spectrum disorders: a survey of 1,418 patients in Japan. Am J Geriatr Psychiatry. 2008;16:584-593 Crossref
  • Uchida et al., 2009a H. Uchida, S. Kapur, B.H. Mulsant, A. Graff-Guerrero, B.G. Pollock, D.C. Mamo. Sensitivity of older patients to antipsychotic motor side effects: a PET study examining potential mechanisms. Am J Geriatr Psychiatry. 2009;17:255-263 Crossref
  • Uchida et al., 2009b H. Uchida, D.C. Mamo, B.H. Mulsant, B.G. Pollock, S. Kapur. Increased antipsychotic sensitivity in elderly patients: evidence and mechanisms. J Clin Psychiatry. 2009;70:397-405 Crossref
  • Uchida et al., 2009c H. Uchida, T. Suzuki, D.C. Mamo, B.H. Mulsant, T. Kikuchi, H. Takeuchi, M. Tomita, K. Watanabe, G. Yagi, H. Kashima. Benzodiazepine and antidepressant use in elderly patients with anxiety disorders: a survey of 796 outpatients in Japan. J Anxiety Disord. 2009;23:477-481 Crossref
  • Uchida et al., 2009d H. Uchida, T. Suzuki, H. Takeuchi, T. Arenovich, D.C. Mamo. Low dose vs standard dose of antipsychotics for relapse prevention in schizophrenia: meta-analysis. (Schizophr Bull, 2009)
  • World Health Organization, 1992 World Health Organization. The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. (WHO, Geneva, 1992)
  • Xiang et al., 2007 Y.T. Xiang, Y.Z. Weng, C.M. Leung, W.K. Tang, G.S. Ungvari. Clinical and social determinants of antipsychotic polypharmacy for Chinese patients with schizophrenia. Pharmacopsychiatry. 2007;40:47-52 Crossref


a Department of Neuropsychiatry, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo, 160-8582, Japan

b Department of Psychiatry, Asakadai Mental Clinic, 1-2-10 Hamasaki, Asaka-shi, Saitama, 351-0033, Japan

c Geriatric Mental Health Program, Centre for Addiction and Mental Health, 1001 Queen Street West, Toronto, ON, Canada M6J 1H4

d Department of Psychiatry, Ohizumi Hospital, 6-9-1, Ohizumigakuen-cho, Nerima-ku, Tokyo, 178-0061, Japan

e Multimodal Imaging Group, PET Centre, Centre for Addiction and Mental Health, 250 College Street, Toronto, ON, Canada M5T 1R8

f Department of Psychiatry, University of Toronto, 250 College Street, Toronto, ON, Canada M5T 1R8

g Department of Psychiatry, Ongata Hospital, 105, Nishiterakata-cho, Hachiohji-shi, Tokyo, 192-0153, Japan

h Department of Psychiatry, Ohizumi Mental Clinic, 1-30-7, Higashi-ohizumi, Nerima-ku, Tokyo, 178-0063, Japan

i Institute of Psychiatry, King's College, London, De Crespigny Park, London, SE5 8AF, United Kingdom

lowast Corresponding author at: Department of Neuropsychiatry, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo, 160-8582, Japan. Tel.: +81 3 5363 3829; fax: +81 3 5379 0187.