• schizophrenia;
  • polypharmacy;
  • antipsychotics;
  • treatment


  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References

Längle G, Steinert T, Weiser P, Schepp W, Jaeger S, Pfiffner C, Frasch K, Eschweiler GW, Messer T, Croissant D, Becker T, Kilian R. Effects of polypharmacy on outcome in patients with schizophrenia in routine psychiatric treatment.

Objective:  Evaluating the effects of different types of psychotropic polypharmacy on clinical outcomes and quality of life (QOL) in 374 patients with schizophrenia and schizoaffective disorder in routine care.

Method:  Psychotropic regimen, clinical outcomes, and QOL were assessed before discharge and after 6, 12, 18, and 24 months. Data were analyzed by mixed-effects regression models for longitudinal data controlling for selection bias by means of propensity scores.

Results:  At baseline 22% of participants received antipsychotic monotherapy (APM) (quetiapine, olanzapine, or risperidone), 20% more than one antipsychotic drug, 16% received antipsychotics combined with antidepressants, 16% antipsychotics plus benzodiazepines, 11.5% had antipsychotics and mood stabilizers, and 16% psychotropic drugs from three or more subclasses. Patients receiving APM had better clinical characteristics and QOL at baseline. Patients receiving i) antipsychotics plus benzodiazepines or ii) antipsychotics plus drugs from at least two additional psychotropic drug categories improved less than patients with APM.

Conclusion:  Combinations of antipsychotics with other psychotropic drugs seem to be effective in special indications. Nevertheless, combinations with benzodiazepines and with compounds from multiple drug classes should be critically reviewed. It is unclear whether poorer outcomes in patients with such treatment are its result or its cause.

Significant outcomes

  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References
  •  In clinical practice, antipsychotic combination therapy is more frequent than monotherapy and seems to be effective and well tolerated. Polypharmacy and combinations with benzodiazepines are associated with poorer outcomes.
  •  Longer course of illness was associated with higher total antipsychotic dose in chlorpromazine equivalents.
  •  Study designs should consider that combination of psychotropic drugs is a frequent issue and patients in routine care act differently to those in RCTs.


  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References
  •  The study has a naturalistic design with a non-random assignment. Changes of the medication over time were possible.
  •  For methodological reasons, at the beginning of the study, only patients with one of the three most frequently prescribed SGAs were included.
  •  The data reflect the therapeutic reality of psychiatric treatment in South Germany. Therefore, generalizability may be limited, and causal relationships cannot be drawn from our results.


  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References

Treatment guidelines from different countries differ in their recommendations regarding the use of first- and second-generation antipsychotics (1–3). However, there seems to be wide agreement that monotherapy has to be preferred because evidence for the efficacy of combinations of antipsychotics and other forms of polypharmacy is poor (4–10). Nevertheless, several recent studies indicate that different forms of antipsychotic combination therapy (i.e., combination of different antipsychotics) and polypharmacy (i.e., the use of adjunctive non-antipsychotic treatment) are common practice in routine psychiatric care (11) and that, in addition to the general increase in antipsychotic prescriptions (12), the trend to use combination therapy and polypharmacy is increasing (13–19).

Empirical research in open and randomized controlled prospective studies (6–8, 20, 21) has essentially been restricted to combinations with clozapine. The investigation of other types of polypharmacy has been rather limited, including a small number of observational trials and case reports (4, 9, 22). The conclusions of recent reviews regarding the effectiveness and safety of polypharmacy in schizophrenia treatment are inconsistent. On the basis of a meta-analysis, Barbui et al. (20) conclude that there is only weak evidence for the effectiveness of adding a second antipsychotic to clozapine. Cipriani et al. (7) in their Cochrane review on clozapine combined with other antipsychotic drugs identified three studies and found no evidence of superior effectiveness of any type of antipsychotic combination therapy over monotherapy. By contrast, Correll et al. (6) identified 19 RCTs comparing antipsychotic monotherapy (APM) with different types of polypharmacy and concluded that given specific circumstances antipsychotic co-treatment may be superior to monotherapy. Pandurangi & Dalkilic (8) reviewed eight RCTs and 66 case reports and found no empirical basis for general recommendations regarding the use of particular types of polypharmacy. Essock et al. (23) conducted a randomized trial on switching from a combination of two antipsychotics to monotherapy. They showed that switching was successful in two-thirds of the participants.

Few studies were conducted in terms of the augmentation of antipsychotics with psychoactive drugs other than antipsychotics. Zink et al. (5) in a narrative review of the literature conclude that there is limited evidence for beneficial effects of the augmentation of antipsychotic drugs with antidepressants while no evidence exists for combinations of antipsychotics with mood stabilizers (APMOOD) or for combinations of different classes of antipsychotics. In an extensive literature review aiming to revise the PORT Psychopharmacological Treatment Recommendations of Schizophrenia, Buchanan et al. (24) found that available studies did not yield sufficient evidence for the augmentation of antipsychotics with lithium and anticonvulsants for treatment-resistant positive symptoms. The conclusion was similar to the use of benzodiazepines in the treatment of anxiety and hostility. In terms of treatment of depression, no controlled trial was detected in which SGAs were augmented by new generation antidepressants.

Centorrino et al. (4) reported that patients prescribed antipsychotic polypharmacy received higher medication doses and reported more adverse effects than patients who received monotherapy while the clinical improvement rates were similar in both groups. The authors concluded that antipsychotic polypharmacy has no additional clinical benefits when compared to monotherapy.

Aims of the study

The study was planned as a prospective observational study on antipsychotic drug treatment of patients with schizophrenia in routine psychiatric care in Germany. In this article, effects of different types of psychotropic polypharmacy vs. APM on clinical outcome indicators and subjective quality of life (QOL) in the routine treatment of patients with schizophrenia or schizoaffective disorder are analyzed.

Material and methods

  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References

Study design

The effects of long-term atypical neuroleptic treatment under routine conditions (ELAN) study is a multi-site prospective observational trial following patients with schizophrenia and schizoaffective disorder under routine psychiatric treatment conditions over 24 months after discharge from psychiatric inpatient treatment. Recruitment occurred from 2005 to 2006, and data collection continued until December 2008. Baseline assessments were conducted in the last week before hospital discharge and follow-up assessments were performed at 6, 12, 18, and 24 months.


Inclusion criteria were a first line diagnosis of schizophrenia or schizoaffective disorder based on the International Diagnosis Checklists for ICD-10 and DSM-IV, an age of 18 or above, and a recommendation for antipsychotic maintenance treatment with quetiapine, olanzapine, or risperidone for a minimum of 12 months. These drugs could be combined with any other psychoactive drug or antipsychotic, but not with each other at the time of inclusion. In the follow-up period, any drug prescription and combination was allowed according to physicians’ and patients’ preferences. Exclusion criteria comprised a first line diagnosis of substance dependence, organic psychiatric disorder, or developmental/learning disability. Study participants who were eligible according to these criteria were selected on the basis of inpatient admission files in nine psychiatric hospitals located in South Germany. In the recruitment process, compatibility with inclusion criteria was established by researchers according to a standardized screening manual. After complete description of the study to the subjects, written informed consent was obtained. The study protocol was approved by the Ethical Committees of the University of Tübingen and the University of Ulm and the State Medical Chamber of Baden-Wuerttemberg.


Psychopathological symptoms were assessed by means of the Positive and Negative Syndrome Scale (PANSS; 25). Functional capacity was measured by the Global Assessment of Functioning (GAF) scale from DSM-IV (26). Medication side effects, including those resulting from combination therapy, were rated by means of the Simpson-Angus Scale for extrapyramidal motor side effects (SAS) (27) and the Abnormal Involuntary Movement Scale (AIMS) for tardive dyskinesia (28).

Body weight and body mass index were recorded by the researchers at discharge and at all follow-ups using a balance and the measure of body height noted in the charts. Blood samples, ECGs, and serum levels of the prescribed antipsychotic with the highest dose were investigated at discharge, at 12-month follow-up and at 24-month follow-up (results not reported here).

Quality of life was measured by the Lancashire Quality of Life Profile LQoLP (29). Adherence to medical treatment was assessed by the Medication Adherence Rating Scale for Psychosis (MARS) (30). Information about psychiatric service use including consumption of psychotropic medication was obtained by means of the Client Sociodemographic and Services Receipt Interview (CSSRI) (31). The total chlorpromazine equivalent (CPZ) dosages of antipsychotic drugs were computed from CSSRI data using suggestions from recent publications (32–35).

Definition of medication groups

The comparison of patient groups receiving one of the three antipsychotics recommended at discharge (quetiapine or olanzapine or risperidone) did not show any differences. Detailed results will be presented elsewhere. For further analyses, six patient subgroups were defined on the basis of medication regimes in accordance with recent suggestions (14): (1) APM, including all patients who received a single antipsychotic drug; (2) antipsychotic combination therapy (APCOMB), including all patients who received at least one antipsychotic drug in combination with at least one additional first- or second-generation antipsychotic concomitantly; (3) antipsychotic and antidepressant (APAD) including all patients who received at least one antipsychotic drug augmented by at least one antidepressant but without any other type of psychotropic drug; (4) antipsychotics and benzodiazepines (APBENZ) including all patients who received at least one antipsychotic drug in combination with at least one benzodiazepine or other type of tranquilizer (lorazepam 93%, diazepam 5%, others 2%). (5) APMOOD including all participants who received at least one antipsychotic drug augmented by sodium valproate or valproic acid (43%), lithium (25%), carbamazepine (20%), lamotrigine (8%), or others (4%), and (6) psychotropic polypharmacy (POLY) including all study participants who were prescribed compounds belonging to three or more sub-classes of psychoactive drugs.

Statistical analyses

Bias control.  The propensity-score method (36–38) was applied to control for selection bias. Propensity scores for the conditional probability of receiving one of the treatment regimes defined earlier were computed by means of a multinomial logistic regression model (37, 39) with the type of drug-therapy as dependent variable and the following independent variables: antipsychotic drug-therapy before hospital admission; outpatient treatment setting before hospital admission; primary diagnosis of schizoaffective disorder; a comorbid diagnosis of substance abuse or personality disorder; a history of bipolar affective disorder, postschizophrenic depression, or anxiety disorder from clinical records; involuntary admission at index hospitalization; number of previous inpatient episodes and number of suicide attempts; self-rated compliance with medical treatment before admission, GAF score at baseline; body weight at baseline; age and gender; employment status before admission; education; living in own apartment; having a partner; nationality; occupational training; job status; foreign mother language.

The multinomial logistic regression analysis was conducted by means of the mlogit module of stata 10 (40).

Statistical analyses of medication effects.  Mixed-effects regression models were fitted for each outcome measure including a random linear time effect, and fixed effects for treatment group, treatment*time interaction, additional anticholinergic medication, interaction between anti-Parkinson medication and time, total CPZ dosage of all antipsychotic drugs, interaction between the CPZ dosage and time, interaction between the baseline measure of the dependent variable and time, and propensity scores for three of the four baseline medication groups. Missing values were assumed as random, and therefore, all cases were included in mixed model analyses.

Mixed model regression analyses were conducted by using the xtmixed module of stata 10 (40).

Antipsychotic monotherapy was chosen to be the reference category in the mixed-effects model. Owing to this parameterization, regression coefficients for the time effect indicate the linear change of the outcome variable among patients who received APM. The regression coefficients for the main effects of medication group variables indicate the baseline deviation of the value of the outcome variable in each medication group in comparison with APM. The regression coefficients for the treatment*time interaction for each medication group indicate deviations of the linear time effect (time slope) from the linear time effect in the APM group.


  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References

Characteristics of the study sample and study flow

Five hundred and thirty patients were found to be eligible for inclusion and asked for study participation. Three hundred and seventy-four patients (71%) agreed to participate and gave informed consent. One hundred and fifty-six patients (29%) refused to participate. Patients who refused to participate, in comparison with those who participated, were significantly younger and had fewer inpatient episodes and a lower rate of schizoaffective disorder diagnoses. The socio-demographic and clinical characteristics of study participants are presented in Table 1.

Table 1.   Description of the study sample characteristics at baseline
 TotalMedication groupSignificance
  1. APM, antipsychotic monotherapy; APCOMB, antipsychotic polypharmacy; APAD, antipsychotic(s) + antidepressant(s); APBENZ, antipsychotic(s) + benzodiazepine(s); APMOOD, antipsychotic(s) + mood stabilizer(s); POLY, antipsychotic(s) + 2 other psychopharmacologic subclasses; None, no psychotropic drug; GAF, global assessment of functioning.

 n1783525303325302 < 4,5,6
Having a job
Own apartment
 n3124243434334351 < 3, 4, 5 2 < 5
Having a partner
 n1111418182515211 < 4,5,6
Foreign nationality
Lower education
Schizoaffective disorder
 n106131016728321,2,3,4 < 5,6
 %28.316.913.526.211.765.154.24 < 3
Involuntary admitted
Body mass index
Number of inpatient admissions
 Mean6. < 5,6 2 < 6 3 < 6
GAF score
PANSS total score
Total CPZ dosage
 Mean706.56507.38900.46558.86847.69666.95761.411 < 2,4,6 3 < 2,4

A total of 257 participants (69%) were assessed at all measurement points. Twenty-nine patients (8%) left the study after the baseline assessment. Three hundred and twenty-three (86%) patients participated in the first follow-up at 6 months, 314 (84%) participated in the second follow-up at 12 months, 302 patients (81%) participated in the third follow-up at 18 months, and 300 patients participated in the fourth follow-up at 24 months. Comparisons of baseline sample characteristics with each follow-up sample failed to reveal any systematic differences and confirmed the missing at random assumption.

Medical treatment groups

At baseline, 77 participants (21%) received APM, 74 participants (20%) received combinations of at least two antipsychotic drugs (APCOMB), 61 participants (16%) received antipsychotics combined with antidepressant medication (APAD), 60 (16%) patients were treated with antipsychotics and tranquilizer (APBENZ), 43 patients (11%) received antipsychotics augmented by mood-stabilizing drugs (APMOOD), and 59 (16%) of the participants received at least one antipsychotic combined with compounds from two or more other subclasses of psychotropic drugs (POLY) (see Table 2).

Table 2.   Distribution of psychotropic medication groups during the course of the study
  1. APM, antipsychotic monotherapy; APCOMB, antipsychotic polypharmacy; APAD, antipsychotic(s) + antidepressant(s); APBENZ, antipsychotic(s) + benzodiazepine(s); APMOOD, antipsychotic(s) + mood stabilizer(s); POLY, antipsychotic(s) + 2 other psychopharmacologic subclasses; None, no psychotropic drug.


As shown in Table 2, the proportion of patients receiving APM or APAD increased slightly in the course of the study while the proportion of participants who received APCOMB, APBENZ, or POLY decreased. The proportion of study participants receiving antipsychotics augmented by mood stabilizers remained stable throughout the course of the study. From the first follow-up assessment onwards a small group of 13–18 patients did not receive any kind of psychotropic medication. In addition to psychotropic drugs, 5–8% of the participants received anti-Parkinson medication to reduce extrapyramidal motor side effects. Patient sample characteristics (see Table 1) indicate significant differences between subgroups with regard to gender, living situation, partnership, diagnostic group, and number of previous psychiatric inpatient admissions.

Results of the multinomial logistic regression model indicate that three variables influenced the likelihood of receiving any drug combination rather than APM: i) medication before hospital admission, ii) the number of previous inpatient episodes, and iii) patient adherence with drug treatment. The type of medication or medication combination before hospital admission determined the subsequent medication considerably. With the exception of APMOOD, the likelihood of receiving polypharmacy rather than APM increased with the number of previous inpatient episodes. Patients who rated themselves as compliant with medical treatment before index hospitalization had a higher probability of receiving more than one substance.

Dosages of antipsychotic drugs

As shown in Table 1, the average total CPZ dosage of antipsychotic drugs was 706.56 (SD = 461.38). CPZ dosage was lowest in participants who received APM while the highest dosages were found in the APCOMB, APBENZ, and POLY subgroups. Table 3 shows change in total dosages for each medication group in the course of the study. CPZ dosages were lowest among patients receiving APM or APAD and highest in patients receiving APCOMB or APBENZ. After 24 months, CPZ dosages had decreased in patients receiving all kinds of medication except for APAD.

Table 3.   Means and standard deviations of the chlorpromazine equivalent doses of antipsychotic drugs
  1. APM, antipsychotic monotherapy; APCOMB, antipsychotic polypharmacy; APAD, antipsychotic(s) + antidepressant(s); APBENZ, antipsychotic(s) + benzodiazepine(s); APMOOD, antipsychotic(s) + mood stabilizer(s); POLY, antipsychotic(s) + 2 other psychopharmacologic subclasses; None, no psychotropic drug.


Clinical outcomes

Results of the mixed-effects regression analyses are presented in Table 4. For the illustration of the longitudinal results adjusted means and 95% CI for the PANSS total score predicted from the mixed-effects model are presented in Table 5.

Table 4.   Results of the mixed-effects regression analysis
  1. Time, linear time effect; APM, antipsychotic monotherapy; APCOMB, antipsychotic polypharmacy; APAD, antipsychotic(s) and antidepressant(s); APBENZ, antipsychotic(s) and benzodiazepine(s); APMOOD, antipsychotic(s) and mood stabilizer(s); POLY, antipsychotic(s) and 2 other psychopharmacologic subclasses.

  2. Controlled for: anti-Parkinson medication (yes/no), anti-Parkinson medication* time, chlorpromazine equivalent (CPZ) dosage, CPZ* time, propensity scores for the probability of receiving APCOMB, APAD, APBENZ, APMOOD or POLY instead of APM at baseline (coefficients not shown).

  3. AIMS, Abnormal Involuntary Movement Scale; GAF, global assessment of functioning.

APMReference category
APM*timeReference category
Table 5.   Means and standard deviations of the Positive and Negative Syndrome Scale total score
  1. APM, antipsychotic monotherapy; APCOMB, antipsychotic polypharmacy; APAD, antipsychotic(s) + antidepressant(s); APBENZ, antipsychotic(s) + benzodiazepine(s); APMOOD, antipsychotic(s) + mood stabilizer(s); POLY, antipsychotic(s) + 2 other psychopharmacologic subclasses; None, no psychotropic drug.

None  57.9426.5344.7915.4847.3816.1342.1719.69

Baseline differences of outcome variables

Regression coefficients for fixed treatment effects reveal that participants who received combinations of different antipsychotics (APCOMB) had lower baseline values of the GAF (b = −2.84; P = 0.039) and the QOL total score (b = −2.68; P = 0.007). Participants who received antipsychotics combined with benzodiazepines (APBENZ) had higher baseline values for the PANSS total score (b = 3.96; P = 0.022) and lower baseline scores for the GAF (b = −4.04; P = 0.008) in comparison with those who received APM.

Baseline values for the SAS score measuring extrapyramidal motor side effects were higher for participants who received an antipsychotic combined with antidepressant drugs (APAD) (b = 0.13; P = 0.000), and for those prescribed an antipsychotic combined with mood-stabilizing drugs (APMOOD) (b = 0.09; P = 0.031) or POLY (b = 0.14; P = 0.000). Baseline values at the AIMS side effects scale were higher for participants who received APCOMB (b = 0.61; P = 0.006), APAD (b = 0.61; P = 0.010), APBENZ (b = 0.50; P = 0.042), and POLY (b = 0.72; P = 0.006). Baseline values of adherence measured by the MARS scale were lower in participants who received APBENZ (b = −0.67; P = 0.004) and in those who quit antipsychotic drug intake after baseline (b = −2.32; P = 0.010). Baseline values for BMI did not differ significantly across medication groups.

Changes in outcome variables

Regression coefficients for linear time effects indicated that, over the course of the trial, both psychopathological symptoms (PANSS total score; b = −1.51; P = 0.001) and extrapyramidal motor side effects (SAS score; b = −0.02; P = 0.022) decreased significantly in patients who received APM. Functioning (GAF score; b = 1.62; P = 0.000) and adherence with drug treatment (MARS score; b = 0.14; P = 0.018) improved significantly in the APM subgroup. In contrast, there were no significant changes over time for QOL, tardive dyskinesia (AIMS), or Body Mass Index (BMI) among patients receiving APM.

Treatment*time interaction effects

In contrast to patients who received APM, those who received antipsychotics plus benzodiazepines (APBENZ) deteriorated with regard to PANSS (b = 2.92; P = 0.000), GAF (b = −2.33; P = 0.001), and QOL (b = −1.44; P = 0.004). Also, participants who received POLY deteriorated while those who received APM improved with regard to PANSS (b = 2.03; P = 0.005) and GAF (b = −1.60; P = 0.011). Compared with APM patients, extrapyramidal motor side effects decreased significantly more among those treated with antipsychotics plus antidepressants (b = −0.03; P = 0.042). MARS adherence ratings improved significantly less in patients with multiple psychotropic drug treatment (POLY vs. APM; b = −0.18: P = 0.049).

Effects of CPZ dosage

Regression coefficients for CPZ dosage (not shown) indicate that higher dosages of antipsychotic drugs at baseline were related to higher baseline values for PANSS (b = 0.003; P = 0.010), lower baseline values for GAF (b = −0.004; P = 0.000), and QOL (b = −0.002; P = 0.011), and lower baseline AIMS scores (b = −0.0003; P = 0.049). However, the interaction effects between CPZ dosage and time were not significant indicating that changes in these outcome variables were not related to the dosage of antipsychotic drugs.


  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References

To our knowledge, this is the first study examining the effects of polypharmacy under routine psychiatric treatment conditions using a prospective study design. In an observational trial, results must be interpreted with a view to limitations resulting from non-randomization and non-blinding. On the other hand, the 24 month observation period and repeated assessments with a comprehensive set of clinical outcome measures are strengths of this study with respect to identifying long-term treatment effects.

Data analysis, in this trial, is controlled for selection bias by means of a propensity-score adjustment. By use of this method, a wide range of clinical variables relevant to receiving psychotropic polypharmacy, sociodemographic variables, and data on illness history can be controlled without reducing the power of the main statistical model. Nevertheless, the limitations of this approach result from the potential effects of unknown not included confounding variables.

This study indicates that during a 2-year period, an average proportion of just 21–33% of patients with psychotic disorders in routine posthospital care were treated with APM while the majority of patients were treated with several combinations of psychotropic drugs. These results are in accordance with those of recent studies that found 25.7–45% of patients with schizophrenia to be prescribed at least two antipsychotic drugs (41, 42). For patients with first-episode schizophrenia, Nielsen et al. report that 56.2% of the patients received antipsychotic polypharmacy (15). Studies have also found almost 50% of patients on APM to be treated concurrently with antidepressants, mood stabilizers (35.3%), or benzodiazepines (37.8%) (41). Similar prevalence rates for antipsychotic polypharmacy were identified by Constantine et al. (43) in an analysis of Medicaid claims in Florida.

In this study, it was only patients treated with antipsychotics plus benzodiazepines (APBENZ) or with drug combinations from three or more psychotropic sub-classes (POLY) that improved significantly less or deteriorated in the course of the study when compared with patients treated with APM. Combinations of antipsychotics at baseline were associated with a longer duration of illness, and they implied about a two-fold antipsychotic dose in comparison with the monotherapy group. Clinicians claim that patients with long course of illness and severe clinical features require higher dosage of antipsychotics to achieve clinical response. Our results yield some evidence that clinicians might obtain the effects of higher doses by combining antipsychotic drugs, thus avoiding excessive dosing of single antipsychotic compounds.

The clinicians’ view seems correct in light of our results, as a close link seems to exist between a longer course of illness and use of higher dosages (in CPZ equivalents). Of course, the results of observational studies do not allow conclusions in terms of causality. That is, a worse clinical course might not be the outcome but a cause of polypharmacy. Future RCTs should take into account the issue of dosage in more detail than has been done so far. Thus, it should be examined whether the use of equivalent CPZ dose long-term monotherapy leads to more favorable effects than combination therapy (44, 45).

Nevertheless, a combination of different antipsychotics in an individual patient should be preceded by several trials of monotherapy with different agents. Thereby, dosage should be optimized, and the duration should be long enough. When using combinations, therapy should be closely monitored and success should be routinely controlled. Combination therapy should only be continued if there are clear therapeutic benefits. A good example for a practicable clinical approach is shown by Essock et al. (23). The randomized trial shows that monotherapy is sufficient and even brings positive effects concerning BMI for many patients. Nevertheless, Essock et al. claim that ‘patients should be free to return to polypharmacy if an adequate trial on APM proves unsatisfactory’.

Combinations of antipsychotic drugs with antidepressants or with mood-stabilizing drugs, in this study, were associated with similar outcomes (as APM). This is probably due to the fact that such treatment regimens were frequently prescribed to patients with marked affective features, including schizoaffective psychosis, who tend to have better outcomes (46, 47).

All in all, the selection of medication in the patients followed in this study seemed to match the clinical necessities, because the outcomes of combination therapy did not differ significantly from the outcomes of monotherapy. Only for the combination of antipsychotic drugs with benzodiazepines or with more than one other type of psychoactive compound, outcomes were worse, not supporting the use of this type of polypharmacy.

Side effects assessed by the SAS score declined more in patients treated with a combination of antipsychotics with antidepressants than in those on APM, but no other effects of polypharmacy on EPS side effects were found. This result is plausible because EPS side effects are expected to be mainly related to CPZ dosage that is controlled in our mixed-effects model.

A limitation of the study is based on the not anticipated high frequency of giving benzodiazepines at discharge and in the course of the study as well as the not anticipated entire extent of polypharmacy. That is why, in hindsight, the target criteria were not completely attuned to this result. Here is room for improvement in a follow-up study. On the other hand, it is to bear in mind that our study is a real world study, in which statements about the ideas of the practitioner could only be made with an incomparable overhead. Also must be remembered that already many items were collected and that the number of items can’t be extended unlimitedly.

This study does not answer the question of whether combined antipsychotic medication or polypharmacy generally lead to net benefit or harm in the treatment of psychotic disorders. However, in view of the widespread use of combinations and polypharmacy, there is an urgent need for randomized controlled trials of different types of psychotropic polypharmacy that should take into account the total antipsychotic dosage. Such trials could add substantial evidence for future guideline recommendations on psychotropic drug combinations in clinical practice.


  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References

The ELAN study was funded as an investigator-initiated research project by a grant from AstraZeneca Deutschland to the University of Tübingen (Project Nr. 229/2004V – Version 2, 27.09.04). AstraZeneca had no role in the development of the research questions, the design of the study, the collection, analyses and interpretation of data, and the writing of the manuscript. AstraZeneca had the right to comment on the final draft of the article before the submission to the journal.

We wish to thank all participants who participated in the study. In addition, we thank Max Schmauß, Jochen Bügler, and Frank Schwärzler for supporting the patient recruitment. We also thank Heike Wiesner, Filiz Özfirat, Tanja Gieselmann, and Simone Triem for data collection and data entry and Ildiko Baumgartner for her work in processing data. We also thank Juan Valdes for his advice in computing CPZEQ equivalent dosages.

Declaration of interest

  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References

Reinhold Kilian, Gerhard Längle, Daniela Croissant, Carmen Pfiffner, Susanne Jaeger, and Prisca Weiser have no interest conflicts. Wiltrud Schepp has received a fund for attending a conference from AstraZeneca. Tilman Steinert has once received funding for attending a congress from Astra Zeneca and has received funding for lectures in his hospital from drug companies Astra Zeneca, Eli Lilly, Bristol Myers Squibb, and Janssen-Cilag. Thomas Messer has received funding for attending a congress from Eli Lilly and received honoraria for lectures from Bristol Myers Squibb, Janssen-Cilag, Pfizer, and Bayer Vital Health Care. Karel Frasch has received funding for conference travels and honoraria for lectures from Astra Zeneca, Janssen-Cilag, Eli Lilly, and Pfizer. He furthermore holds shares of the pharmaceutical company Stada. Gerhard W. Eschweiler conducts clinical studies on Alzheimer’s dementia for the ACI Immune and Janssen-Alzheimer-Immunotherapy companies. Thomas Becker reports research funding to the department (unrestricted research grants) from AstraZeneca, GlaxoSmithKline, Pfizer and Affectis Pharmaceuticals AG for clinical trials and investigator-initiated trials. The statistical analyses were performed by Reinhold Kilian.


  1. Top of page
  2. Abstract
  3. Significant outcomes
  4. Limitations
  5. Introduction
  6. Material and methods
  7. Results
  8. Discussion
  9. Acknowledgements
  10. Declaration of interest
  11. References
  • 1
    Gaebel W, Weinmann S, Sartorius N, Rutz W, McIntyre JS. Schizophrenia practice guidelines: international survey and comparison. BJPsych 2005;187:248255.
  • 2
    American Psychiatric Association. Practice guideline for the treatment of patients with schizophrenia, 2nd edn. Arlington, VA: American Psychiatric Association, 2004.
  • 3
    Kreyenbuhl J, Buchanan R, Dickerson FB, Dixon LB. The Schizophrenia Patient Outcomes Research Team (PORT): updated treatment recommendations 2009. Schizophr Bull 2010;36:94103.
  • 4
    Centorrino F, Goren JL, Hennen J, Salvatore P, Kelleher JP, Baldessarini RJ. Multiple versus single antipsychotic agents for hospitalized psychiatric patients: case-control study of risks versus benefits. Am J Psychiatry 2004;161:700706.
  • 5
    Zink M, Englisch S, Meyer-Lindenberg A. Polypharmacy in schizophrenia. Curr Opin Psychiatry 2010;23:103111.
  • 6
    Correll CU, Rummel-Kluge C, Corves C, Kane JM, Leucht S. Antipsychotic combinations vs monotherapy in schizophrenia: a meta-analysis of randomized controlled trials. Schizophr Bull 2009;35:443457.
  • 7
    Cipriani A, Boso M, Barbui C. Clozapine combined with different antipsychotic drugs for treatment resistant schizophrenia. Cochrane Database Syst Rev 2009;8(3):CD006324.
  • 8
    Pandurangi AK, Dalkilic A. Polypharmacy with second-generation antipsychotics: a review of evidence. J Psychiatr Pract 2008;14:345367.
  • 9
    Faries D, Ascher-Svanum H, Zhu B, Correll C, Kane J. Antipsychotic monotherapy and polypharmacy in the naturalistic treatment of schizophrenia with atypical antipsychotics. BMC Psychiatry 2005;5:26.
  • 10
    Stahl SM, Grady MM. A critical review of atypical antipsychotic utilization: comparing monotherapy with polypharmacy and augmentation. Curr Med Chem 2004;11:313326.
  • 11
    Baandrup L, Allerup P, Lublin H, Nordentoft M, Peacock L, Glenthoj B. Evaluation of a multifaceted intervention to limit excessive antipsychotic co-prescribing in schizophrenia out-patients. Acta Psychiatr Scand 2010;122:367374.
  • 12
    Verdoux H, Tour M. Antipsychotic prescribing trends: a review of pharmaco-epidemiological studies. Acta Psychiatr Scand 2010;121:410.
  • 13
    Clark RE, Bartels SJ, Mellman TA, Peacock WJ. Recent trends in antipsychotic combination therapy of schizophrenia and schizoaffective disorder: implications for state mental health policy. Schizophr Bull 2002;28:7584.
  • 14
    Messer T, Tiltscher C, Schmauss M. [Polypharmacy in the treatment of schizophrenia]. Fortschr Neurol Psychiatr 2006;74:377391.
  • 15
    Nielsen J, Le Quach P, Emborg C, Foldager L, Correll CU. 10-year trends in the treatment and outcomes of patients with first-episode schizophrenia. Acta Psychiatr Scand 2010;122:356366.
  • 16
    Zhu B, Ascher-Svanum H, Faries DE, Correll CU, Kane JM. Cost of antipsychotic polypharmacy in the treatment of schizophrenia. BMC Psychiatry 2008;8:19.
  • 17
    Mojtabai R, Olfson M. National trends in psychotropic medication polypharmacy in office-based psychiatry. Arch Gen Psychiatry 2010;67:2636.
  • 18
    Kreyenbuhl JA, Valenstein M, Mccarthy JF, Ganoczy D, Blow FC. Long-term antipsychotic polypharmacy in the VA health system: patient characteristics and treatment patterns. Psychiatr Serv 2006;58:489495.
  • 19
    Centorrino F, Cincotta SL, Talamo A et al. Hospital use of antipsychotic drugs: polytherapy. Compr Psychiatry 2008;49:6569.
  • 20
    Barbui C, Signoretti A, Mule S, Boso M, Cipriani A. Does the addition of a second antipsychotic drug improve clozapine treatment? Schizophr Bull 2009;35:458468.
  • 21
    Kroken RA, Johnsen E, Ruud T, Wentzel-Larsen T, Jorgensen HA. Treatment of schizophrenia with antipsychotics in Norwegian emergency wards, a cross-sectional national study. BMC Psychiatry 2009;9:24.
  • 22
    Freudenreich O, Goff DC. Antipsychotic combination therapy in schizophrenia: a review of efficacy and risks of current combinations. Acta Psychiatr Scand 2002;106:323330.
  • 23
    Essock SM, Schooler NR, Stroup TS et al. Effectiveness of switching from antipsychotic polypharmacy to monotherapy. Am J Psychiatry 2011;168:702708.
  • 24
    Buchanan RW, Kreyenbuhl J, Kelly DL et al. The 2009 schizophrenia PORT psychopharmacological treatment recommendation and summary statements. Schizophr Bull 2010;36:7193.
  • 25
    Kay SR, Opler LA, Lindemeyer JP. The Positive and Negative Syndrome Scale (PANSS): rationale and standardisation. Br J Psychiatry 1989;155:5965.
  • 26
    American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edn, Text Revision. Edited by Washington DC: American Psychiatric Association, 2000.
  • 27
    Simpson G, Angus J. A rating scale for extrapyramidal side effects. Acta Psychiatr Scand 1970;212:1119.
    Direct Link:
  • 28
    Lane RD, Glazer WM, Hansen TE, Berman WH, Kramer SI. Assessment of tardive dyskinesia using the Abnormal Involuntary Movement Scale. J Nerv Ment Dis 1985;173:353357.
  • 29
    Gaite L, Vázquez-Barquero JL, Arrizabalaga A et al. Quality of life in schizophrenia: development, reliability and internal consistency of the Lancashire Quality of Life Profile – European Version. EPSILON Study 8. European Psychiatric Services: inputs linked to outcome domains and needs. Br J Psychiatry Suppl 2000;39:s49s54.
  • 30
    Thompson K, Kulkarni J, Sergejew AA. Reliability and validity of a new Medication Adherence Rating Scale (MARS) for the psychoses. Schizophr Res 2000;42:241247.
  • 31
    Chisholm D, Knapp M, Knudsen HC et al. Client Sociodemographic and Service Receipt Inventory- European Version. Br J Psychiatry Suppl 2000;39:S28S33.
  • 32
    Woods SW. Chlorpromazine equivalent doses for the newer atypical antipsychotics. J Clin Psychiatry 2003;64:663667.
  • 33
    Andreasen NC, Pressler M, Nopoulos P, Miller D, Ho BC. Antipsychotic dose equivalents and dose-years: a standardized method for comparing exposure to different drugs. Biol Psychiatry 2010;67:255262.
  • 34
    Gardner DM, Murphy AL, O’donnell H, Centorrino F, Baldessarini RJ. International consensus study of antipsychotic dosing. Am J Psychiatry 2010;167:686693.
  • 35
    Atkins M, Burgess A, Bottomley C, Riccio M. Chlorpromazine equivalents: a consensus of opinion for both clinical and research. Psychiatr Bull 1997;21:224226.
  • 36
    Swanson JM, Hinshaw SP, Arnold LE et al. Secondary evaluations of MTA 36-month outcomes: propensity score and growth mixture model analyses. J Am Acad Child Adolesc Psychiatry 2007;46:10031014.
  • 37
    Kilian R, Angermeyer MC. The effects of antipsychotic treatment on quality of life of schizophrenic patients under naturalistic treatment conditions: an application of random effect regression models and propensity scores in an observational prospective trial. Qual Life Res 2005;14:12751289.
  • 38
    D’agostino RB. Tutorial in biostatistics. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17:22652281.
  • 39
    Greene WH. Econometric analysis, 5th edn. Upper Saddle River: Prentice Hall, 2003.
  • 40
    Stata Corporation. Stata statistical software: release 10. College Station, TX: StataCorp LP, 2007.
  • 41
    Procyshyn RM, Honer WG, Wu TK et al. 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;71:566573.
  • 42
    Baandrup L, Gasse C, Jensen VD et al. Antipsychotic polypharmacy and risk of death from natural causes in patients with schizophrenia: a population-based nested case-control study. J Clin Psychiatry 2010;71:103108.
  • 43
    Constantine RJ, Andel R, Tandon R. Trends in adult antipsychotic polypharmacy: progress and challenges in Florida’s Medicaid Program. Community Ment Health J 2010;46:523530.
  • 44
    Stahl SM, Freudenreich O, Goff DC. A successful antipsychotic combination trial. Quo Vadis? Acta Psychiatrica Scand 2004;110:241242.
  • 45
    Stahl SM. Antipsychotic polypharmacy, evidence based or eminence based? Acta Psychiatr Scand 2002;106:321323.
  • 46
    Möller HJ, Bottlender R, Wegner U, Wittmann J, Strauss A. Long-term course of schizophrenic, affective and schizoaffective psychosis: focus on negative symptoms and their impact on global indicators of outcome. Acta Psychiatr Scand 2000;407:5457.
  • 47
    Harrow M, Grossman LS, Herbener ES, Davies EW. Ten-year outcome: patients with schizoaffective disorders, schizophrenia, affective disorders and mood-incongruent psychotic symptoms. Br J Psychiatry 2000;177:421426.