SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

Objective:  The aim of this study was to describe factors associated with achieving a minimally symptomatic status outcome in outpatients with schizophrenia.

Methods:  Data were analysed from a 3-year, prospective observational study that examined outcomes in 7658 patients with schizophrenia. Minimally symptomatic status was defined as a postbaseline score of 1 or 2 on the Clinical Global Impressions Severity Scale-Schizophrenia version (CGI-SCH).

Results:  Baseline CGI-SCH score had the strongest association with minimally symptomatic status followed by age, geographical region and hospitalisation status. The probability of becoming minimally symptomatic was consistently higher in the olanzapine and clozapine monotherapy groups compared with the quetiapine, risperidone or haloperidol groups. The olanzapine group achieved the minimally symptomatic status in a shorter period of time than the other treatment groups (p ≤ 0.016).

Conclusion:  The likelihood of patients achieving a minimally symptomatic status was greater in younger patients with lower baseline clinical severity and in patients whose treatment included olanzapine.


What’s known

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

Achievement and maintenance of minimally symptomatic status is a vital issue for patients with schizophrenia. Despite more rigorous diagnostic criteria and better treatment, the long-term course and eventual outcome of schizophrenia is still variable and is of ongoing concern (2–5). Predictors of response to treatment are desirable for future management strategies.

What’s new

This subanalysis of the huge Schizophrenia Outpatients Health Outcomes (SOHO) study dataset identified factors associated with achieving a minimally symptomatic status outcome in outpatients with schizophrenia, which are important considerations for individual treatment planning. Although the probability of becoming minimally symptomatic varied between the atypical antipsychotic treatment groups, the likelihood of patients achieving a minimally symptomatic status was greater in younger patients with lower baseline clinical severity and in patients whose treatment included olanzapine.

Introduction

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

Currently there is no aetiological cure for schizophrenia and the clinical aim of treatment is directed towards recovery and sustained remission (1). Despite more rigorous diagnostic criteria and better treatment, the long-term course and eventual outcome of schizophrenia is still variable and is of ongoing concern (2–5). Predictors of response to treatment are desirable for future management strategies and have been reported from various studies. In a historical, prospective study of culturally diverse-treated patients with schizophrenia (follow-up of 15 and 25 years), the strongest predictor of outcome was the time spent experiencing psychotic symptoms (6). Baseline variables of younger age, family involvement, history of drug use and a lack of social interaction had some relationship to poor outcome but further studies are necessary to clarify these relationships. Local environment also played a significant role in symptoms and social disability (7–9). Better cognitive performance and a shorter duration of psychosis prior to study entry have been associated with improved symptom remission, adequate social/vocational functioning and full recovery in some studies (10).

There has been a shift in therapy for schizophrenia from the use of typical antipsychotics, such as haloperidol, to atypical agents (11,12) such as olanzapine, quetiapine and risperidone, which appear to have a broader spectrum of clinical efficacy with less risk of extrapyramidal symptoms (EPS) and movement disorders (13–17). The course and/or outcome measurements in studies of schizophrenia have included global assessment scales, such as Clinical Global Impression-Severity in Schizophrenia (CGI-SCH), and symptom domain-specific scales, such as the Scale for the Assessment of Negative Symptoms, and the assessment of functional status such as occupation and social relationships. Remission is generally assessed using an absolute threshold of severity of the symptoms of schizophrenia (18). In a recent expert working group (Remission in Schizophrenia Working Group), the authors suggested that criteria for symptomatic remission should be based on distinct thresholds for reaching and maintaining improvement. Such a definition of improvement, through threshold criteria rather than change from baseline, enables direct cross-trial comparisons (19). The descriptive validity and accordance with clinical and functional outcome measures of these criteria were investigated. The results showed high descriptive validity and strong association with physician reported outcomes (20). However, in this analysis, the Minimally Symptomatic criteria was applied because it was predefined in the protocol of the Intercontinental Schizophrenia Outpatients Health Outcomes (IC-SOHO) study based on the previous research suggestions (21) and the proposed criteria of remission by Remission in Schizophrenia Working Group was published only years later in 2005. In addition to it, the IC-SOHO study was a non-interventional naturalistic study, where CGI-Severity scales were used and the Brief Psychiatric Rating Scale or the Positive and Negative Syndrome Scale (PANSS) were not used: these scales are used for definitions of response and remission mostly in clinical trial settings (22). Observational studies with their naturalistic, non-interventional features have some advantages over randomised, controlled trials, including longer follow-up and a wider range of patients. Observational studies are also useful for evaluating the general applicability of the results of randomised, controlled trials in a wider population (23). The main objective of the IC-SOHO study was to compare the effectiveness of olanzapine with other antipsychotics and explore related patient outcomes, and the results were reported elsewhere (24). In this analysis, we report data on the probability of patients acquiring and maintaining minimally symptomatic status from the 3-year observational IC-SOHO study (F1D-SN-HGJR) of 7658 patients, examining the treatment of patients with schizophrenia in a naturalistic setting.

Methods

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

The IC-SOHO study was a 36-month, prospective, observational study on health outcomes associated with antipsychotic medication treatment of outpatients suffering from schizophrenia. The study was conducted in 27 countries across five regions: Africa (Algeria); Asia (Korea, Malaysia, Taiwan); Central and Eastern Europe (Czech Republic, Hungary, Lithuania, Poland, Romania, Russia, Slovakia, Slovenia and Turkey); Latin America (Argentina, Chile, Colombia, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Peru, Puerto Rico and Venezuela); and the Middle East (Egypt, Israel and Saudi Arabia). By definition, the study was non-interventional and the treatment groups were non-randomised. The main results of this study, reported elsewhere (25), confirmed that these naturalistic data from less-studied outpatient communities highlight the variability in clinical and functional outcomes associated with long-term antipsychotic treatment. The study was conducted in accordance with the local ethical and regulatory requirements of each country. All participants provided informed consent as required by local regulations.

Patient selection and treatment groups

At the normal care outpatient visit, 7658 patients, who met criteria for schizophrenia (ICD-10 or DSM-IV) (26), were at least 18 years of age, had provided consent, and were enrolled in this study. To ensure that the study reflected a naturalistic setting, participating psychiatrists were instructed to first make treatment decisions, independent of the study, using standard clinical practice guidelines. The choice of antipsychotic and the dose prescribed were at the psychiatrists’ discretion. Patients were permitted to use concomitant medications such as anticholinergics, antidepressants, anxiolytics and mood stabilisers, as clinically indicated. Patients attended eight outpatient visits: at baseline, and at 3, 6, 12, 18, 24, 30 and 36 months. Patients were included in this treatment group analysis for as long as they remained on the originally prescribed monotherapy during the 36 months of treatment, i.e. the analysis only included the patients’ data while they were still on their original monotherapy. Details of the study’s methodology have been previously published elsewhere (27). Briefly, the objective of the IC-SOHO study was to compare the effectiveness of olanzapine with other antipsychotics and explore related patient outcomes. Therefore, the study was designed to provide two patient cohorts of approximately equal size. As such, patients were enrolled using an alternating entry structure (until a block of 10 patients was reached, i.e., five in each group) into one of the following two patient cohorts: (i) patients who were initiated on or switched to olanzapine therapy (monotherapy or in combination with other agents), or (ii) patients who initiated or switched to any other antipsychotic therapy other than olanzapine. Hence, as a result of the study design, approximately half of enrolled patients were initially treated with olanzapine. This resulted in a stratified sample with the olanzapine group as the ‘over-sampled’ stratum. While this imbalance does not create bias or invalidates the comparison of the groups, it needs to be recognised that estimates and comparisons involving these groups are less precise than those made with the larger olanzapine and risperidone groups, as indicated by the accompanying confidence intervals (24). As low precision reduces statistical power, we may have failed to detect meaningful differences in comparisons involving the smaller subgroups. Furthermore, the smaller treatment subgroups may be less representative of the patient population as a whole, thus reducing the external validity of the comparisons. In the analyses presented here, patients in the non-olanzapine cohort are separated into treatment groups based upon the antipsychotic prescribed at study entry.

Outcome measures

Disease severity (CGI-SCH) was assessed at every visit during the follow-up period. Minimally symptomatic status was defined as an overall CGI-SCH score of 1 (‘Normal, Not ill’) or 2 (‘Minimally ill’) at any post baseline visit. Patients with a missing baseline CGI-SCH score or a baseline score of 1 or 2 (rated as not ill or minimally ill at baseline) were excluded from this analysis. This study reports the time period from baseline until patients became minimally symptomatic, the time period they remained minimally symptomatic and the factors associated with a minimally symptomatic status. Because of study data collection schedule, the first change to minimally symptomatic status could be seen at approximately 3 months, corresponding to the target time of the first postbaseline visit. At the following visit, patients could lose this status if the CGI-SCH increased to 3 (‘Mildly ill’) or higher. The definition does not require maintaining the minimally symptomatic status for a few consecutive visits because of long intervals (3 and 6 months) between visits. Patients who discontinued the study or changed baseline medication prior to becoming minimally symptomatic were included in the analysis up to the time of their last study visit, at which point their data were censored.

The CGI-SCH scale was adapted from the CGI scale (24,28). The previous research has shown that the CGI-SCH scale is a valid and sensitive tool for assessing disease severity and treatment response in schizophrenia. Given its simplicity, brevity and clinical face validity, the scale is appropriate for use in observational studies and routine clinical practice.

Statistical analyses

Patients were grouped according to the monotherapy they received/were prescribed at the baseline visit, and patients were included in treatment group analyses for as long as they remained on the originally prescribed monotherapy during the 36 month follow-up period.

Categorical variables were summarised in terms of the number and percentage of patients in each category for each treatment group. Continuous variables were summarised using means (unadjusted), SD and median. These summaries were based on all of the patients with available visit data. Unadjusted analyses were performed for baseline comparisons between groups.

As the treatment groups were not balanced in terms of baseline characteristics, an adjusted analysis was an attempt to correct for some of the most important differences when postbaseline comparisons were conducted. Baseline patient and disease characteristics associated with the time to becoming minimally symptomatic were explored with the following procedures. The variables identified as being potentially clinically significant were: the treatment prescribed at baseline (five groups), time since diagnosis, previous use of depot typical treatment (within 6 months of baseline), previous use of clozapine (within 6 months of baseline), previous use of atypical antipsychotics (olanzapine, risperidone, clozapine, or other atypicals within 6 months), neuroleptic-naïve, hospital admission in the 6 months before the baseline visit, sex, age, CGI overall score, CGI positive score, CGI negative score, CGI depressive score, CGI cognitive score, presence of EPS at baseline, presence of tardive dyskinesia (TD) at baseline, mood stabilisers at the baseline visit, patient’s compliance, social activity, spouse/partner relationship, housing status, work status, suicide attempts, substance/alcohol abuse, hostility, sexual dysfunction, body mass index (BMI), geographical region and country effect. As the first step, each variable was tested separately using the log-rank test to select a smaller subset of the variables to include in the simultaneous model. Variables significant at the 10% level were retained for testing for simultaneous association with reaching minimally symptomatic status. A stepwise regression procedure was used to identify jointly significant variables. Investigator (> 100 investigators) and country effects (27 countries) were not tested in the joint model to avoid over-fitting. The Cox proportional hazards survival regression model (25) was used to estimate hazard ratios (HR) of becoming minimally symptomatic and associated confidence intervals. The proportionality of the hazards assumption was assessed graphically for all of the variables in the final model.

Kaplan–Meier survival curves were used to show the probability of becoming minimally symptomatic. For ease of interpretation, the graphs were inverted so that there was a zero probability up to the first study visit at 3 months and the cumulative probability increased from one visit to the next. The lines were also smoothed for better visual effect. Patients who discontinued the study prior to becoming minimally symptomatic were included in the analysis up to the time of their last study visit; at which point their data were censored. Patients who changed to a different drug prior to becoming minimally symptomatic were included in the analysis up to the study visit when the change was recorded and then their data were censored. Patients who were minimally symptomatic at baseline and patients with missing data for any of the covariates retained in the final model were not included in the estimation of the HR. No imputations were used for missing data. The calculated p-values are presented without adjustment for multiple comparisons. Statistical analyses were performed using SAS® Version 8.02 for Windows (SAS Institute, Cary, NC, USA).

Results

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

The results are based on those patients who maintained their originally prescribed baseline monotherapy, irrespective of dose change, up to the time of the assessment unless otherwise stated. The numbers of patients who were enrolled, prescribed one of the five antipsychotic agents at baseline, and remained on monotherapy up to 36 months are presented in Figure 1. At baseline, there were 13 patients with a missing CGI-SCH overall score, 22 had a score of 1, and 210 had a score of 2. These patients (approximately 5% of the total sample) were omitted from the modelling.

image

Figure 1.  Flowchart of numbers of patients and treatment at 36 months. Twenty-one patients had no prescription information at baseline.

Download figure to PowerPoint

Baseline characteristics

Table 1 shows the baseline characteristics of patients treated with the five different antipsychotic agents prescribed as monotherapy at baseline. Of the 7658 patients enrolled in the IC-SOHO study, 5836 were prescribed monotherapy at baseline. As a result of the study design, the majority of patients (3222) were prescribed olanzapine at baseline. The number of patients who were treated with risperidone (1117) was greater than those treated with clozapine (237), quetiapine (189) and haloperidol (257). After 36 months, 61.8% of the evaluable patients treated with olanzapine remained on monotherapy and this was a greater proportion than in any other treatment group (clozapine: 53.8%, risperidone: 47.6%, quetiapine: 41.1% and haloperidol: 26.2%). Comparison of the odds of remaining on baseline monotherapy shows that the olanzapine group have a greater odds of remaining on monotherapy and it was significantly different from clozapine (p = 0.006) and three other groups (all p < 0.001).

Table 1.   Patients’ baseline characteristics
 OlanzapineClozapineRisperidoneQuetiapineHaloperidol
  1. *Patients who were characterised as not at all ill or borderline ill (CGI-SCH overall scores of 1 or 2). p < 0.001: c vs. clozapine, h vs. haloperidol, r vs. risperidol. p-Value from: ANOVA – age, duration of diagnosis, CGI; logistic regression – gender, first use of antipsychotic, functional status. CGI, clinical global impressions; SCH, severity in schizophrenia; SD, standard deviation. (%) is calculated based on the number of patients with information about specified characteristic available.

Age, mean ± SD (years)34.9 ± 12.134.3 ± 11.935.9 ± 12.234.4 ± 12.035.0 ± 11.5
Gender (female), n (%)1438 (44.9)c72 (30.8)541 (48.6)c93 (50.0)c121 (47.1)c
Duration of diagnosis, mean ± SD (years)8.5 ± 9.79.2 ± 9.29.0 ± 10.09.0 ± 10.19.5 ± 9.8
First use of antipsychotic, n (%)565 (18.0)22 (9.4)194 (17.7)19 (10.2)44 (17.6)
Married/permanent partner, n (%)1109 (35.8)c55 (24.0)415 (38.7)c,h74 (41.8)c66 (26.7)
Independent residence, n (%)1056 (32.8)c51 (21.5)382 (34.4)c69 (36.7)c71 (27.8)
Employed, n (%)724 (22.7)35 (14.8)249 (22.4)40 (21.3)45 (17.6)
Involved in social activities, n (%)1831 (58.0)125 (53.4)657 (59.6)114 (61.0)132 (52.0)
Overall CGI-SCH, mean ± SD4.4 ± 1.1c,r4.6 ± 1.04.2 ± 1.0c4.4 ± 1.14.4 ± 1.1
Minimally symptomatic*, n (%)153 (4.7)8 (3.4)48 (4.3)12 (6.4)11 (4.3)

Symptom severity at baseline

There were some differences between the treatment groups in the severity of symptoms at baseline, as measured by the CGI-SCH scores. The mean baseline CGI-SCH overall score for the group of clozapine-treated (4.6 ± 1.0) patients was higher than the corresponding scores for all other treatment groups and this was statistically significant (p < 0.001) compared with olanzapine (4.4 ± 1.1) and risperidone (4.2 ± 1.0). The risperidone-treated group was characterised by a significantly lower overall CGI-SCH score compared with the olanzapine group (p < 0.001). The clozapine group had the smallest percentage of patients who were minimally symptomatic at baseline (3%) and the quetiapine group had the highest (6%).

Proportion of minimally symptomatic patients after baseline

The percentage of patients who were minimally symptomatic was noticeably greater in the olanzapine group after only 3 months of treatment and this trend persisted up to 36 months (Table 2). The clozapine and risperidone treatment groups also had significantly higher proportions of patients who were minimally symptomatic at 12 months than those receiving haloperidol. Because there were differences in the baseline characteristics and numbers of patients in each of the five treatment groups, the following analyses for the time to becoming and remaining minimally symptomatic were performed.

Table 2.   Numbers of patients who were minimally symptomatic at 3, 12, 24 and 36 months
TimeOlanzapineClozapineRisperidoneQuetiapineHaloperidol
  1. Comparisons are from logistic regression analyses. p < 0.001: c vs. clozapine, r vs. risperidone, q vs. quetiapine, h vs. haloperidol. N, number of patients remaining on monotherapy.

Overall symtoms
3 months (N)2497183820133179
 n (%)653 (26.2)r,q,h29 (15.8)161 (19.6)13 (9.8)19 (10.6)
12 months (N)18881335367497
 n (%)1002 (53.1)q,h49 (36.8)h247 (46.1)h23 (31.1)18 (18.6)
24 months (N)1451973755154
 n (%)875 (60.3)c,r,h39 (40.2)175 (46.7)20 (39.2)17 (31.5)
36 months (N)1101812744933
 n (%)626 (56.9)35 (43.2)133 (48.5)17 (34.7)12 (36.4)

Time to and factors associated with becoming minimally symptomatic

The cumulative probability of patients becoming minimally symptomatic (CGI-SCH score of 1 or 2), for each treatment group, up to 36 months from baseline is shown in Figure 2. On the figure, the estimated probability curves were higher at all time points in the olanzapine and risperidone monotherapy groups than in the other groups (overall comparison between treatment groups p < 0.001). Based on the log-rank test from univariable models, the characteristics of the following baseline variables had a strong association with the time required to reach minimally symptomatic status: treatment prescribed (olanzapine, clozapine, risperidone, quetiapine or haloperidol); duration of prior antipsychotic treatment; neuroleptic-naïvety; age; CGI-SCH scores (overall, positive, negative, depressive and cognitive); presence of EPS or TD; social activity; spouse/partner; housing status; work; region; country; and investigator. The following four additional baseline characteristics, that had a borderline association with time to reaching minimally symptomatic status, were included in the joint model: use of clozapine; hospital admission in the 6 months prior to baseline; patient compliance with medication; and sexual dysfunction. The following baseline factors were not associated with the time required to reach minimally symptomatic status: previous treatment with depot typical antipsychotics; gender; the use of mood stabilisers; suicide attempts; substance/alcohol abuse; hostility and BMI.

image

Figure 2.  Cumulative probability of becoming minimally symptomatic by treatment prescribed at baseline over 36 months

Download figure to PowerPoint

Table 3 shows the association between the baseline variables and time until patients became minimally symptomatic based on the final joint model. The baseline CGI-SCH scores had a strong association with the time until patients became minimally symptomatic. Furthermore, younger patients, patients with lower baseline CGI-SCH scores, and patients without admission to hospital in the 6 months prior to the baseline visit reached minimally symptomatic status earlier than other patients. In addition, differences between geographical regions in time to reaching minimally symptomatic status were seen in the model. The olanzapine-treated patients tended to reach minimally symptomatic status earlier than those prescribed clozapine, risperidone, quetiapine or haloperidol. The HR relative to olanzapine treatment was closest for risperidone treatment, but already reached statistical significance (p < 0.001). The HR of becoming minimally symptomatic for olanzapine was almost twice that for haloperidol (p < 0.001). Hazard ratios for quetiapine and clozapine are between those for risperidone and haloperidol, but these are only statistically significant at the 5% level, possibly because of the smaller sample sizes of these groups.

Table 3.   Hazard ratios of becoming minimally symptomatic and associated confidence intervals estimated by the Cox proportional hazards survival regression model
FactorEstimated hazard ratio95% CI lower bound95% CI upper boundp-Value
  1. AMEA, Africa and Middle East; CEE, Central and Eastern Europe; CGI, clinical global impressions; CI, confidence interval; SCH, severity in schizophrenia.

Age at baseline0.950.920.980.004
Hospital admission1.121.021.230.014
Overall treatment effect
 Olanzapine1Treatment estimates are relative to olanzapine
 Risperidone0.790.710.88< 0.001
 Quetiapine0.690.520.910.009
 Clozapine0.760.610.950.016
 Haloperidol0.570.430.74< 0.001
CGI-SCH sores
 Overall0.720.690.76< 0.001
 Positive0.750.710.80< 0.001
 Negative0.720.670.77< 0.001
 Depressive0.930.881.000.038
 Cognitive0.670.630.72< 0.001
Region effect
 Latin America1Region estimates are relative to Latin America
 CEE0.840.750.930.001
 AMEA0.890.791.000.057
 Asia0.840.720.970.018

Probability of remaining minimally symptomatic

The cumulative probability of remaining minimally symptomatic, for patients who remained on the same treatment by which this status was achieved, is shown in Figure 3. This figure shows that patients in the clozapine group were most likely to remain minimally symptomatic (overall comparison between treatment groups p < 0.001). In our study, the lack of/incomplete effectiveness was the most common reason for treatment modification, followed by patient’s request, intolerability and non-compliance.

image

Figure 3.  Cumulative probability of remaining minimally symptomatic by continuing treatment on which minimally symptomatic status was achieved over 36 months

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

Achievement and maintenance of minimally symptomatic status is a vital issue for patients with schizophrenia. One of the aims of this sub-analysis of the long-term IC-SOHO observational study database was to report data on the probability of patients acquiring and maintaining minimally symptomatic status during the study’s 3-year follow-up period. The IC-SOHO study encompassed a heterogeneous, large population of patients with a broad geographical coverage.

This post hoc analysis identified several baseline variables with strong influence on time to reach minimally symptomatic status, such as atypical antipsychotic treatment prescribed at baseiline (vs. haloperidol), lengths of antipsychotic treatment, neuroleptic treatment naive status, age, lower severity of positive, negative, depression, cognitive and overall scores of CGI-SCH, absence of EPS, and certain social factors, like social activity, presence of partners, homing status, as well as regional effects. Younger pateints, with no hospitalisation in the past 6 months relative to basline, with less psychopathological burden (less depressive symptoms, as well lower severity scores on positive, negative, overall and cognitive CGI-SCH scores, and being treated in South America had the best chance to reach faster the minimally symptomatic status. The interpretation of the findings that younger patients, patients with lower baseline CGI-SCH scores and patients without admission to hospital in the 6 months prior to the baseline visit reached minimally symptomatic status earlier than other patients and differences between geographical regions seen in the model should be interpreted with caution. However, these findings may imply a potentially more successful intervention strategy in these populations where the revealed predictors for a better functional treatment outcome emphasise the need for early psychosocial and vocational rehabilitation in schizophrenic patients. These factors also suggest that minimally symptomatic is a realistic goal in the treatment of schizophrenia. Attaining and sustaining this status may warrant better clinical and functional outcomes for patients.

The proportion of patients reaching minimally symptomatic status was greater in the olanzapine group and the clozapine group than in haloperidol group, but the differences between all groups were not statistically significant at 36 months. Similarly, in the CATIE (Clinical Antipsychotic Trials in Intervention Effectiveness) trial, the advantage of olanzapine treatment compared with the other drugs on PANSS and CGI-SCH scores, observed initially, diminished over 18 months (17).

Once minimally symptomatic status had been achieved, the likelihood of maintaining it consistently declined with time. In the first 12 months of the IC-SOHO study, patients treated with either olanzapine or risperidone were significantly less likely to experience relapse than those receiving haloperidol (29). Figure 3 displays the length of time for which they remained minimally symptomatic. The probability to remain minimally symptomatic was the highest for clozapine followed by olanzapine. Until now, clozapine has mainly been restricted to patients with schizophrenia who are resistant to other antipsychotic treatment. The current analysis on the longer-term outcomes shows that the increasing probability of patients becoming minimally symptomatic over the first 12 months of treatment continued, but was less pronounced, over the next 24 months. After 36 months of treatment, olanzapine and risperidone was associated with greater weight gain than haloperidol, but the differences were not statistically significant (30), partly because of the low power at 3-year comparison because of discontinuations. A greater proportion of patients remained on therapy with the atypical antipsychotics compared with the typical antipsychotic, haloperidol. The highest proportion remained on olanzapine treatment; this may reflect the faster rate at which these patients reached minimally symptomatic status compared with those on the other treatments. It is likely that patients doing well on a particular treatment will remain on it. Compliance with medication was not collected in this observational study, because of potential biases in this setting where pill count or other objective measure is not available. Sexual dysfunction results were reported elsewhere for the IC-SOHO study (30). The present results are consistent with the concept that this particular group of treatment-resistant patients differs from other schizophrenic patients in the long-term outcomes of the disease.

Differences in the efficacy of the various atypical antipsychotics have not been well established over the longer-term. At 6-month olanzapine resulted in significantly greater improvements in all outcomes compared with quetiapine, or risperidone (25), and after 12 months of treatment, olanzapine- and risperidone-treated patients had approximately a 3- to 4-times higher odds ratio of response than those receiving haloperidol (29). The results of a meta analysis of 12 controlled studies indicated that clozapine exhibited superiority over typical antipsychotics (haloperidol and chlorpromazine) in terms of efficacy and safety (31).

In other studies, overall remission was much lower than that associated with probability of remaining minimally symptomatic in this study. In a sample comprised of patients with a first episode schizophrenia (70%) and schizoaffective disorder (30%), symptom remission for more than 2 years was noted in approximately 39% of patients (9). In another study of middle-aged and elderly patients, who had previously met the DSM-III-R or DSM-IV criteria for schizophrenia, only 8% of patients living independently had sustained remission (27). However, comparison of results from different studies is complicated because outcome measures as well as design and measurement vary considerably. The apparent differences between outcomes in the IC-SOHO study and those of others could be attributable to the rigour of definitions of remission, and different treatments and baseline characteristics.

In terms of analysis, the time-to-event approach adopted in this study is a powerful method as it allows the inclusion of patients who change their medication or do not complete the trial, by making comparisons at a number of time-points (32). An adjusted analysis was used in an attempt to correct for the most important differences in baseline characteristics when postbaseline comparisons were conducted. The study sample analysed were those patients who remained on the same antipsychotic medication throughout the 36-month treatment period and, as patients changed drug or dropped out during the study, the numbers of patients analysed at each visit decreased. In the quetiapine and haloperidol groups only 51 and 38 patients respectively, remained making comparisons less reliable. Very few patients remained on the initially assigned drug and estimates at the end of the study for those treatment groups with few patients should be interpreted with caution. The impact of discontinuation on the conclusions from analyses will be small, because using survival analyses methodology censors patients if they change treatment or discontinue the study prior to becoming minimally symptomatic. In addition, time to reaching minimally symptomatic status is shorter in the groups with lower level of discontinuation’ (30).

The study design has several limitations. Some of these are intrinsically associated with observational studies, such as the non-randomised design and lack of blinding. Observational studies, such as this one, are useful in accessing a wider, more realistic community-based population than that used in randomised control studies (23). The strengths and limitations of the design, the rationale and overall objectives for this study have been discussed in a previous publication of the 12-month results (29). These limitations have the potential to introduce bias, but were reduced in this study by adjusting comparisons using baseline covariates that were recognised to have clinical significance. Nevertheless, it is acknowledged that these baseline corrections may not completely address the potential biases resulting from the design of this study. The relatively large size of the olanzapine group enhanced the precision of the estimates for the outcomes associated with olanzapine treatment. As a result of the large number of tests undertaken and the absence of formal adjustments for multiple comparisons, results should be interpreted with caution. The clinical assessments were made by clinicians of different professional and cultural backgrounds and, therefore, the diagnosis of schizophrenia and the application of the CGI-CH scale may have differed between countries or regions (33). This could have led to specific limitations in relation to the factors affecting the minimally symptomatic status. In addition, it was not possible to determine the extent to which this variability may have influenced the treatment decisions made by the clinicians.

In conclusion, this multinational, observational study provides detailed long-term information on the effectiveness of monotherapy treatment with typical and atypical antipsychotics in achieving and maintaining minimally symptomatic status. During 36 months, more patients remained on olanzapine monotherapy and had a higher probability of becoming minimally symptomatic than patients on any other atypical (clozapine, risperidone or quetiapine) or typical (haloperidol) antipsychotic; CGI-SCH scores at baseline were the strongest predictor of outcome followed by age, geographical region and hospitalisation status. Because of the exploratory nature of this study and numerous comparisons of outcomes, the results of the analysis should be interpreted with caution.

Acknowledgements

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

This study was supported by a research grant from Eli Lilly & Company, Indianapolis, IN, USA. The authors thank all the members of the IC-SOHO study team for their valuable contributions and ongoing commitment to the study.

Author contributions

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References

All authors participated in the interpretation of the results of this sub-analysis, and in the drafting, critical revision and approval of the final version of the manuscript. In addition, Y. Dyachkova was the statistician, who performed the statistical analysis.

References

  1. Top of page
  2. Summary
  3. What’s known
  4. Introduction
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. Author contributions
  10. References
  • 1
    Auslander LA, Jeste DV. Sustained remission of schizophrenia among community-dwelling older outpatients. Am J Psychiatry 2004; 161: 14903.
  • 2
    James RL, May PR. Diagnosing schizophrenia: professor Kraepelin and the research diagnostic criteria. Am J Psychiatry 1981; 138: 510504.
  • 3
    Hegarty JD, Baldessarini RJ, Tohen M, Waternaux C, Oepen G. One hundred years of schizophrenia: a meta-analysis of the outcome literature. Am J Psychiatry 1994; 151: 140916.
  • 4
    Bleuler M. The Schizophrenic Disorders: Long-Term Patient and Family Studies (1972). London: Yale University Press, 1978.
  • 5
    Modestin J, Huber A, Satirli E, Malti T, Hell D. Long-term course of schizophrenic illness: Bleuler’s study reconsidered. Am J Psychiatry 2003; 160: 22028.
  • 6
    Harrison G, Hopper K, Craig T et al. Recovery from psychotic illness: a 15- and 25-year international follow-up study. Br J Psychiatry 2001; 178: 50617.
  • 7
    Craig TJ, Siegel C, Hopper K, Lin S, Sartorius N. Outcome in schizophrenia and related disorders compared between developing and developed countries. A recursive partitioning re-analysis of the WHO DOSMD data. Br J Psychiatry 1997; 170: 22933.
  • 8
    Harding CM. Course types in schizophrenia: an analysis of European and American studies. Schizophr Bull 1988; 14: 63343.
  • 9
    Robinson DG, Woerner MG, McMeniman M, Mendelowitz A, Bilder RM. Symptomatic and functional recovery from a first episode of schizophrenia or schizoaffective disorder. Am J Psychiatry 2004; 161: 4739.
  • 10
    Bassett AS, Addington D. Canadian clinical practice guidelines for the treatment of schizophrenia. (The Canadian Psychiatric Association). Can J Psychiatry 1998; 43 (Suppl. 2): 25S40S.
  • 11
    Lehman AF, Steinwachs DM. Translating research into practice: the Schizophrenia Patient Outcomes Research Team (PORT) treatment recommendations. Schizophr Bull 1998; 24: 110.
  • 12
    Kennedy A, Jain S, Vinogradov S. Atypical antipsychotics for schizophrenia: their collective role and comparative profiles. Formulary 2001; 36: 50017.
  • 13
    Tarsy D, Baldessarini RJ, Tarazi FI. Effects of newer antipsychotics on extrapyramidal function. CNS Drugs 2002; 16: 2345.
  • 14
    Tandon R, Jibson MD. Efficacy of newer generation antipsychotics in the treatment of schizophrenia. Psychoneuroendocrinology 2003; 28 (Suppl. 1): 926.
  • 15
    Schillevoort I, De Boer A, Herings RM, Roos RA, Jansen PA, Leufkens HG. Risk of extrapyramidal syndromes with haloperidol, risperidone, or olanzapine. Ann Pharmacother 2001; 35: 151722.
  • 16
    Leucht S, Pitschel-Walz G, Abraham D, Kissling W. Efficacy and extrapyramidal side-effects of the new antipsychotics olanzapine, quetiapine, risperidone, and sertindole compared to conventional antipsychotics and placebo. A meta-analysis of randomized controlled trials. Schizophr Res 1999; 35: 5168.
  • 17
    Lieberman JA, Stroup TS, McEvoy JP et al. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med 2005; 353: 120923.
  • 18
    Frank E, Prien RF, Jarrett RB et al. Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence. Arch Gen Psychiatry 1991; 48: 8515.
  • 19
    Andreasen NC, Carpenter WT Jr, Kane JM, Lasser RA, Marder SR, Weinberger DR. Remission in schizophrenia: proposed criteria and rationale for consensus. Am J Psychiatry 2005; 162: 4419.
  • 20
    Van OS J, Drukker M, à Campo J, Meijer J, Bak M, Delespaul P. Validation of remission criteria for schizophrenia. Am J Psychiatry 2006; 163: 200002.
  • 21
    Beasley CM Jr, Sutton VK, Hamilton SH et al. A double-blind, randomized, placebo-controlled trial of olanzapine in the prevention of psychotic relapse. J Clin Psychopharmacol 2003; 23: 58294.
  • 22
    Leucht S, Davis JM, Engel RR, Kissling W, Kane JM. Definitions of response and remission in schizophrenia: recommendations for their use and their presentation. Acta Psychiatr Scand Suppl 2009; 438: 714.
  • 23
    Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. N Engl J Med 2000; 342: 187886.
  • 24
    Dossenbach M, Pecenak J, Szulc A et al. Long-term antipsychotic monotherapy for schizophrenia: disease burden and comparative outcomes for patients treated with olanzapine, quetiapine, risperidone, or haloperidol monotherapy in a pan-continental observational study. J Clin Psychiatry 2008; 69: 190115.
  • 25
    Dossenbach M, Erol A, El Mafoud Kessaci M et al. Effectiveness of antipsychotic treatments for schizophrenia: Interim 6-month analysis from a prospective observational study (IC_SOHO) comparing olanzapine, quetiapine, risperidone and haloperidol. J Clin Psychiatry 2004; 65: 31221.
  • 26
    World Health Organization. The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. Geneva, Switzerland: World Health Organisation, 1992.
  • 27
    Jager M, Messer T, Laux G et al. Standardized remission criteria in schizophrenia: descriptive validity and comparability with previously used outcome measures. Pharmacopsychiatry 2008; 41: 1905.
  • 28
    American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn. Washington, DC: American Psychiatric Association, 1994, pp. 273315.
  • 29
    Dossenbach M, Arango-Davila C, Silva Ibarra H et al. Response and relapse in patients with schizophrenia treated with olanzapine, risperidone, quetiapine, or haloperidol: 12-month follow-up of the Intercontinental Schizophrenia Outpatient Health Outcomes (IC-SOHO) study. J Clin Psychiatry 2005; 66: 102130.
  • 30
    Lima MS, Soares BG, Paoliello G et al. The Portuguese version of the Clinical Global Impression-Schizophrenia Scale: validation study. Rev Bras Psiquiatr 2007; 29: 2469.
  • 31
    Chakos M, Lieberman J, Hoffman E, Bradford D, Sheitman B. Effectiveness of second-generation antipsychotics in patients with treatment-resistant schizophrenia: a review and meta-analysis of randomized trials. Am J Psychiatry 2001; 158: 51826.
  • 32
    Cox DR, Oakes D. Analysis of Survival Data. London, England: Chapman & Hall, 2001.
  • 33
    Haro JM, Kamath SA, Ochoa S et al. The Clinical Global Impression-Schizophrenia scale: a simple instrument to measure the diversity of symptoms present in schizophrenia. Acta Psychiatr Scand Suppl 2003; 416: 1623.