Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Dec 1.
Published in final edited form as: J Psychiatr Res. 2017 Sep 8;95:299–307. doi: 10.1016/j.jpsychires.2017.09.008

Risperidone versus olanzapine among patients with schizophrenia participating in supported employment: Eighteen-month outcomes

Douglas L Noordsy 1, Shirley M Glynn 2,3, Catherine A Sugar 2,4, Christopher T O’Keefe 5, Stephen R Marder 2,3
PMCID: PMC5653420  NIHMSID: NIHMS907732  PMID: 28942217

Abstract

This study compares the efficacy and tolerability of olanzapine versus risperidone among patients with schizophrenia who are established in outpatient psychiatric care and entering supported employment. A multicenter, randomized, double-blind trial was conducted among 107 outpatients with schizophrenia, who were cross-titrated to flexible dose risperidone or olanzapine over 2 weeks. Clinical endpoints included time to hospitalization and persistence on assigned medication. Weight, laboratory tests, psychopathology, neurologic side effects, social adjustment and role functioning were assessed at 3–6 month intervals. Data were analyzed first by randomized treatment, and then reassessed controlling for prior medication treatment. The proportion of patients on assigned medication at 18 months was 30.9% for risperidone and 37.3% for olanzapine. Mean doses were 6.4+/−3.2 mg daily for risperidone, and 17.0+/−5.0mg daily for olanzapine. The groups did not differ significantly in time to medication discontinuation, first hospitalization or first employment. There were few differences in psychopathology, laboratory, or neurological assessments between groups at 18 months. Patients randomized to olanzapine gained modestly more weight. Controlling for pre-randomization medication suggested improvement in some aspects of psychopathology from switching medications; however, switching from olanzapine to risperidone was associated with more hospitalizations. Risperidone and olanzapine have similar efficacy and tolerability in patients with schizophrenia who are participating in supported employment. Randomization to olanzapine was associated with more weight gain, but randomization from olanzapine to risperidone appeared to be associated with a greater likelihood of hospitalization. Careful monitoring of metabolic effects and participation in supported employment may have contributed to minimal weight gain and metabolic effects.

Keywords: schizophrenia, antipsychotic medication, supported employment, BMI

Introduction

Guiding outpatients with schizophrenia in choosing optimal antipsychotic therapy for participating in psychosocial rehabilitation is complex and poorly understood (Salkever et al., 2006; Sungur et al., 2011). Antipsychotic medications may be associated with side effects such as sedation and EPS that can interfere with attention, learning and motor function (Kumar et al., 2013). There is some evidence that antipsychotics may have differential effects on cognitive function (Bilder et al., 2002; Keefe et al., 2007; Matsuda et al., 2014). Switching antipsychotic medications can also be disruptive and destabilizing (Takeuchi et al., 2017).

Large randomized trials such as the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study have demonstrated that the second-generation antipsychotics (SGAs) as a group have similar effects on psychopathology, with some advantages in treatment outcomes for clozapine, olanzapine, and risperidone (Lieberman et al., 2005; McEvoy et al., 2007, 2006; Stroup et al., 2006). However, the agents also differed in their side effects. A re-analysis of the CATIE phase 1 findings by Essock and colleagues (Essock et al., 2006) found that patients who were randomized to the same antipsychotic that they were taking before study entry tended to have better outcomes than those who were randomly assigned to take a different antipsychotic medication during the study. However, the authors did not evaluate the effects of continuing or changing antipsychotics on adverse effects. All of these medication effects were studied in isolation from any psychosocial rehabilitation that patients were receiving.

In the current trial, stable, unemployed outpatients with schizophrenia were randomly assigned to an 18-month double-blind comparison of olanzapine and risperidone at the point of initiating a supported employment program (Glynn et al., 2017). The purpose of this study was to compare two highly prescribed antipsychotic medications in the understudied clinical context of psychosocial rehabilitation. We examined the effectiveness, but also the side effects, of assigned medication treatment as a function of prior antipsychotic treatment.

Materials and methods

Study Setting and Design

We randomized 107 stable outpatients to double-blind risperidone vs. olanzapine. In addition, all participants received supported employment using the evidence-based Individual Placement and Support Program (IPS)(Becker and Drake, 1994), and half of these participants were randomly assigned to receive vocational maintenance skills training in the clinic (Glynn et al., 2017; Mueser et al., 2005). Patients were recruited from a Dartmouth-affiliated Community Mental Health Center (NH site) and an outpatient psychiatric clinic at UCLA and a Veterans Affairs Medical Center (LA site). The study was reviewed and approved by the institutional review boards at each site.

Participants

Eligible participants were 18 – 65 years old, living in the community, not currently competitively employed and interested in finding competitive work. Participants had a diagnosis of schizophrenia or schizoaffective disorder made by a research psychiatrist using DSM-IV criteria (American Psychiatric Association, 2000), and verified by the Structured Clinical Interview for DSM-IV (SCID) (First et al., 2002). Exclusion criteria included organic brain disease, mental retardation, active substance dependence within the prior 6 months, or a chronic medical illness which would prevent employment. Concomitant antidepressants and mood stabilizers were permitted. Participants were required to be competent to provide informed consent, and to demonstrate understanding of the informed consent document on a true-false test (Wirshing et al., 1998). After complete description of the study to the subjects, written informed consent was obtained.

The IPS model of supported employment used in this study encourages inclusion of interested patients regardless of psychiatric symptom severity (Becker and Drake, 1994). Therefore, while all participants were clinically stable outpatients at the point of entry into the study, there was no exclusion for severity of psychotic symptoms.

Interventions

Cross-titration from the prior antipsychotic medication to randomized, double-blind therapy with risperidone or olanzapine was to be completed within 2 weeks. Medications were blinded using a double-dummy strategy, with 10 dose levels available. Medication was distributed to patients weekly in a medication planner, and adherence to daily doses monitored at the end of each week by recording any doses remaining in the planner. The risperidone dose range was 1–10 mg daily, with an initial target of 4 mg once daily. The olanzapine dose range was 2.5–30 mg daily, with an initial target of 12.5 mg once daily.

Outcomes

At baseline, all participants received medical, metabolic, symptom and psychosocial assessments. Participants were weighed at every clinic visit, and blood tested at baseline, 3, 6, 12, 18 and 24 months. Body Mass Index (BMI) was calculated using height measured at baseline. Psychiatric symptoms were measured every 3 months using the Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1962), the Schedule for the Assessment of Negative Symptoms (SANS) (Andreasen, 1990), the Symptom CheckList-90 (SCL-90) (Derogatis and Cleary, 1977), and the Hamilton Depression Scale (HAM-D) (Hamilton, 1960).

Neurologic side effects were rated every 3 months using the Abnormal Involuntary Movement Scale (AIMS) (Lane et al., 1985) the Barnes Akathisia Scale (BAS) (Barnes, 1989), The Subjective Extrapyramidal Rating Scale (SERS) (Van Putten et al., 1981), and the Modified Simpson-Angus Scale (S-A) (Simpson and Angus, 1970). Social and role functioning were rated every 6 months using the Social Adjustment Scale (SAS) (Weissman et al., 2001) and the Quality of Life scale (QOL) (Heinrichs et al., 1984). Time on study medication, time to first hospitalization and time to competitive employment were also recorded for each subject.

Inter-rater reliability was established by training all raters to the standards of the UCLA Diagnostic Assessment Laboratory (kappa coefficient > 0.8), and maintaining regular conference calls among study raters. Regular reliability checks of study raters were conducted throughout the study via examination of video taped study assessments.

In this group of stable outpatients identifying interest in employment, we expected that risperidone and olanzapine would have similar effects on participants’ continuance on assigned medication, psychiatric symptoms, likelihood of relapse and ability to obtain competitive employment. We also hypothesized that differences in side effects between the two medications would influence functional outcomes and quality of life. We evaluated continuation of assigned medication as an indicator of outcome that incorporates both efficacy and tolerability, consistent with other studies (Kahn et al., 2008; Lieberman et al., 2005). Hospitalization was used as a proxy for relapse.

Data Analysis

Demographic variables were examined for differences by both site and study medication using t-tests and chi-squared tests. Comparative analyses were performed using two sets of patient groupings: (a) assigned study medication (risperidone or olanzapine) and (b) the combination of pre-randomization and assigned study medication (RSP-RSP, RSP-OLZ, OLZ-RSP, OLZ-OLZ, N-RSP, N-OLZ), where RSP-RSP indicates that the participant was taking risperidone naturalistically prior to randomization and then was randomized to blinded risperidone treatment during the trial, RSP-OLZ indicates prior treatment with risperidone followed by randomization to olanzapine, and so forth. N indicates the subject was on a medication other than risperidone or olanzapine prior to randomization, which primarily was a first generation antipsychotic. Subjects who were on both risperidone and olanzapine pre-randomization were grouped with the olanzapine subjects. We made this decision based on an assumption that a patient’s metabolic status was more likely to be related to olanzapine rather than risperidone. The purpose of the second set of analyses was to determine whether subjects who switched medications during the study had different outcomes than those who were randomized to continue on their prior medication.

Continuation on assigned study medication, time to relapse (hospitalization) and time to obtain employment were analyzed by estimating survival curves stratified by medication group using all available data for each subject (SAS PROC LIFETEST). Subjects who remained on their randomized medication or who failed to relapse until they left the study were considered censored. Median survival times are reported to avoid bias incurred in the mean due to censoring in the longest continuing subjects. Overall comparisons were made using the log-rank test for equality over strata and 18-month survival rates were compared using Z-tests. For subjects who obtained a job, the median number of weeks worked is also reported. The primary analytic tool for other outcomes was mixed effects regression with site, medication group, time, and a medication group by time interaction as fixed effects and random intercept and slope effects within subjects to account for the repeated measures. Data were used only for those time points during which patients remained on their assigned study medication. The mixed model approach automatically handles missing observations, producing unbiased parameter estimates as long as values are missing at random. All models were fit with SAS PROC MIXED using REML estimation and an unstructured covariance matrix. The primary tests of interest were the time by medication interactions and between medication group contrasts.

Results

Randomized Medication Group Analysis (Two Groups)

The demographic composition of the sample is presented in Table 1. There were no differences in any of the demographic variables between the two randomized treatment groups. There was a significant difference in pre-randomization medication by site, with patients recruited in LA more often taking risperidone, and patients recruited in NH more often taking olanzapine. There were also site differences in race, gender, education and prior psychiatric hospitalizations (Glynn et al., 2017). Therefore, we included an indicator for site in the mixed effects models.

Table 1.

Demographic variables at baseline by site and medication assignment

Variable LA (N=45) NH (N=62) p RSP1 (N=55) OLZ2 (N=52) p Total (N=107)
Education (yrs) 13.4 11.6 <.0001 12.4 12.4 n.s. 12.4
Age of onset (N=102, yrs) 26.8 23.5 .0526 24.0 25.9 n.s. 24.9
Prior Hospitalizations (N=100) 5.5 8.6 .0448 7.6 6.9 n.s. 7.3
Age (yrs) 42.2 41.9 n.s. 41.8 42.2 n.s. 42.0
Illness duration (N=102, yrs) 15.4 18.5 n.s. 17.9 16.5 n.s. 17.3
Ethnicity N((%) <.0001 n.s.
 White 23 (51.1%) 60 (96.8%) 40 (72.7%) 43 (82.7%) 83 (77.6%)
 Latino 2 (4.4) 0 (0) 1 (1.8) 1 (1.9) 2 (1.9)
 Asian 3 (6.7) 0 (0) 1 (1.8) 2 (3.9) 3 (2.8)
 Black 15 (33.3) 1 (1.6) 10 (18.2) 6 (11.5) 16 (15.0)
 Other 2 (4.4) 1 (1.6) 3 (5.5) 0 (0) 3 (2.8)
Gender N(%) .0105 n.s.
 Male 42 (93.3%) 46 (74.2%) 45 (81.8%) 43 (82.7%) 88 (82.2%)
 Female 3 (6.7) 16 (25.8) 10(18.2) 9 (17.3) 19 (17.8)
Marital status N(%) n.s. n.s.
 Never 28 (62.2%) 30 (49.2%) 30 (54.6%) 28 (54.9%) 58 (52.4%)
 Divorced 6 (13.3) 8 (13.1) 7 (12.7) 7 (13.7) 14 (13.3)
 Married 11 (24.4) 23 (37.7) 18 (32.7) 16 (31.4) 34 (32.4%)
Pre-randomization medication <.0001 n.s.
 RSP1 18(40%) 8 (13%) 12(22%) 14(27%) 26 (24%)
 OLZ2 6 (13) 35 (56) 20 (36) 21 (40) 41 (38)
 Other 21 (47) 19 (31) 23 (42) 17 (33) 40 (37)
Open in a new tab
1

RSP = risperidone

2

OLZ = olanzapine

Persistence on study medication

The mean daily dose of study medication at 18 months was 6.4 ± 3.2 mg for risperidone and 17.0 ± 5.0 mg for olanzapine. Overall, 42 (39.6%) of the 107 patients remained on assigned medication and in the study at one year and 36 (34.0%) at 18 months (Fig 1). In the risperidone group 20 (36.4%) of 55 patients remained on study medication at 1 year and 17 (30.9%) remained on study medication at 18 months. In the olanzapine group, 22 (43.1%) of 51 patients remained on study medication at 1 year, and 19 (37.3%) at 18 months. Kaplan-Meir survival rates were 0.49 for risperidone and 0.57 for olanzapine at 18 months. Reasons for medication discontinuation in participants who continued the vocational portion of the study included lack of efficacy (RSP 6/OLZ 8), intolerable side effects (RSP 7/OLZ 7), participant decision (RSP 6/OLZ 2) and poor adherence (RSP 0/OLZ 1). Side effects that led to termination of risperidone included weight gain, lipid elevation, cognitive slowing, priapism, seizure and neuroleptic malignant syndrome. Side effects that led to termination of olanzapine included weight gain, anxiety, restlessness and insomnia. There was no evidence of a significant difference in persistence on study medication by group (Overall log rank test: X2 = 0.92, NS; difference in 18-month survival rate: Z = 0.77, NS; Fig 2a), nor in reasons for discontinuation.

Figure 1.

Figure 1

Open in a new tab

Flow Diagram of Participants in the Randomized Control Trial

Figure 2.

Figure 2

Figure 2

Open in a new tab

Figure 2a. 18-month survival on study medication (2 group analysis)

X2 = .9271, df = 1, p-value .3356

R = risperidone, O = olanzapine

Figure 2b. 18-month survival on study medication by pre-randomization medication (6 group analysis)

X2 = 6.0042, df = 5, p-value .3058

R = risperidone, O = olanzapine, N = other antipsychotic medication

Effectiveness

Psychopathology

There were no significant differences between the risperidone and olanzapine groups in the rate of change (i.e. medication by time interaction) on the BPRS or any of its subscales, nor on the SANS or any of its subscales (Table 2). The risperidone group demonstrated a greater rate of improvement on the SCL-90 somatization subscale up to 18 months (medication by time interaction p <0.05). There were no other between groups differences on the remaining SCL-90 subscales, nor in depression as measured by the HAM-D. There was significant improvement over time but no medication effect (time p’s < .05, negative slope; medication and interaction terms NS) on the BPRS psychosis and suspiciousness subscales and on the SANS avolition subscale. There were site differences indicating that participants in NH had lower mean BPRS total, BPRS psychosis and SANS alogia subscale scores, but higher SANS affective flattening subscales scores relative to their counterparts in LA after adjusting for other factors in the model.

Table 2.

Eighteen-month psychopathology, side effect, laboratory and social outcomes, adjusted for site and medication adjustment period (2 group analysis)

Variable Time1 Group Group x Time
Pyschopathology Measures
Brief Pyschiatric Rating Scale (BPRS)2 n.s. n.s. n.s.
BPRS-D (Depression) n.s. n.s. n.s.
BPRS-S (Thought Disturbance) .0442 n.s. n.s.
BPRS-R (Retardation) n.s. n.s. n.s.
BPRS-P (Paranoid Suspiciousness) .0001 n.s. n.s.
Negative Symptoms (SANS) .0802 n.s. n.s.
Affect Flattening n.s. n.s. n.s.
Alogia .0982 n.s. n.s.
Avolition .0089 n.s. n.s.
Anhedonia n.s. n.s. n.s.
Symptom Checklist (SCL)-90 Somatization .0073 n.s. .03963
Hamilton Depression Scale (HAM-D) n.s. n.s. n.s.
Side Effect Measures
Body Mass Index (BMI) n.s. n.s. 0.01764
Abnormal Involuntary Movement (AIMS) n.s. n.s. n.s.
Barnes Akathisia Scale (BAS) n.s. n.s. n.s.
Simpson-Angus Scale (SA) n.s. n.s. n.s.
Subjective Extrapyramidal Rating Scale (SERS) n.s. n.s. n.s.
Laboratory Measures
Glucose n.s. n.s. n.s.
HbA1c n.s. n.s. n.s.
Cholesterol n.s. 0.03395 n.s.
Triglycerides n.s. n.s. n.s.
LDL Cholesterol n.s. n.s. n.s.
HDL Cholesterol 0.0812 n.s. n.s.
Social Adjustment and Quality of Life Measures
SAS: General Adjustment n.s. n.s. n.s.
QOL: Total Score .0314 n.s. n.s.
Open in a new tab
1

All changes over time are in the direction of improvement.

2

BPRS and SANS totals are sums; Subscale scores are averages of their respective items.

3

The model estimates a rate of decrease of .077 points/year on the SCL-90 SOM scale in the OLZ group and a rate of decrease of .292 points/year in the RSP group.

4

Patients randomized to OLZ had an estimated rate of increase of .650 BMI points/year while those randomized to RSP had an estimated rate of decrease of 1.095 points/year.

5

Average cholesterol scores of OLZ subjects are 18.673 points higher than RSP subjects.

Relapse

Twenty-eight of the 107 participants were hospitalized at some point during their study participation, 4 at the Los Angeles site and 24 at the New Hampshire site for a total of 50 hospital admissions. Sixteen of these subjects were in the group randomized to risperidone versus 12 to olanzapine. Twenty-three of these subjects experienced their first hospitalization within the first 18 months. Forty-five of these admissions occurred while patients were receiving double-blind study medications. There was no significant difference in time to first hospitalization between the risperidone and olanzapine groups (Overall log rank test: X2 = 1.46, NS; difference in 18 month survival rate Z = 1.20, NS). However the difference in hospitalization rates between the two sites was significant (X2 = 4.91, p < .05)

Side effects

Adverse effects

Patients randomized to olanzapine had an estimated rate of increase of .650 BMI points/year while those randomized to risperidone had an estimated rate of decrease of 1.095 BMI points/year, a statistically significant difference (Table 2). There were no significant interactions between medication group and time for any measures of neurological side effects. Applying Schooler-Kane criteria to the AIMS data, we identified 6 cases of treatment-emergent tardive dyskinesia (all of mild severity) during the 18-month assessment period, 3 of whom were in the risperidone group and 3 in the olanzapine group (NS). This represents a treatment-emergent tardive dyskinesia rate of 3.6%/year for risperidone, and 3.8%/year for olanzapine. There were site differences indicating that participants in NH had higher mean BMI scores and lower mean AIMS scores than their counterparts in LA after adjusting for other factors in the model.

Serious adverse events

There was one death in the risperidone group due to respiratory arrest and 1 death in the olanzapine group due to brain hemorrhage, both while on assigned study medication. There were no suicide attempts or completed suicides during study participation.

Lab measures

There were no significant differences in the rate of change in any laboratory studies by medication group over the course of the study (Table 2). The olanzapine group had higher mean total cholesterol levels overall. Eliminating observations when subjects were on concurrent lipid-lowering medications revealed no systematic differences in the findings. There was a trend towards increasing HDL in both groups over time. Patients at the NH site had significantly higher mean triglycerides and HDL levels than those at the LA site after adjusting for other factors in the model.

Functional outcomes

Social and role functioning

There were no significant differences in either mean or rate of change between the medication groups in terms of social adjustment or overall quality of life (Table 2). However, there was significant improvement over the 18-month study period on the QOL total score (time p-value < .05). There was also a significant difference between the LA and NH sites, with subjects in LA having better average scores on the SAS general adjustment scale.

Employment

Sixty-seven of the 107 subjects obtained competitive employment during the study period (Glynn et al., 2017). Thirty-two of these subjects were in the group randomized to olanzapine and 35 were in the group randomized to risperidone. All but one of the subjects who obtained jobs had done so by 18 months, and the corresponding the Kaplan-Meier estimates of employment rates at 18 months were .73 for the olanzapine group and .80 for the risperidone group. There were no significant differences in time to first employment by medication group (Overall log rank test: X2 = 0.31, NS; difference in 18-month employment rate: Z = 0.44, NS). The estimated median time to obtain a job was 182 days (mean 216.4 +/− 21.9 days) in the olanzapine group and 225 days (mean 274.2 +/− 24.3 days) in the risperidone group (Z = 0.74, NS). Among subjects who did obtain competitive employment the median number of weeks during the two-year follow-up with at least some work hours was 43 in both medication groups.

Pre-randomization By Randomized Medication Group Analysis (Six Groups)

To account for potential differences between subjects who switched medications and those who were randomized to remain on their prior medication, we reran the above analyses using a 6-class pre-post randomization medication grouping. There were no statistically significant differences in demographic variables among these groups, except for ethnic composition. This was attributable to the fact that the New Hampshire site had both a higher proportion of white subjects and a higher prevalence of olanzapine use pre-randomization. Overall 12 subjects switched from risperidone to olanzapine (RSP-OLZ), 20 switched from olanzapine to risperidone (OLZ-RSP), 33 subjects remained on their pre-randomization medication (12 RSP-RSP and 21 OLZ-OLZ) and 40 subjects were placed on risperidone or olanzapine for the first time at randomization (23 N-RSP, 17 N-OLZ). While sample sizes for the 6 group analyses are small, limiting our power to detect group by time interactions in the mixed models with high confidence, the repeated measures give us enough information to identify some suggested areas of interest. Specific details are given below.

Persistence on study medication

Median survival time on assigned medication was similar across the 6 medication groups to 18 months, and there was no statistically significant difference in survival rate across treatment groups (Log rank test: X2 = 6.00, NS; Fig 2b). Visually, the RSP-OLZ, N-OLZ and N-RSP groups clustered together, followed by the OLZ-OLZ, RSP-RSP, and OLZ-RSP groups.

Effectiveness

Psychopathology

There were suggestions of significant differences in response among the groups on several psychopathology variables using the 6-group model (Table 3). However, none of them were strong enough to survive correction for multiple testing. Measures with possible medication by time interactions were the BPRS psychosis subscale (favoring the OLZ-RSP and RSP-OLZ groups) and SCL-90 total score (favoring the OLZ-OLZ, OLZ-RSP and N-RSP groups). See Table 3 for details.

Table 3.

Eighteen-month psychopathology, side effect, laboratory and social outcomes, adjusted for site and medication adjustment period (6 group analysis)

Variable Time1 Group Group x Time
Pyschopathology Measures
Brief Pyschiatric Rating Scale (BPRS)2 n.s. .0833 n.s.
BPRS-D (Depression) n.s. .0484 n.s.
BPRS-S (Thought Disturbance) .0850 n.s. .0488
BPRS-R (Retardation) n.s. n.s. n.s.
BPRS-P (Paranoid Suspiciousness) .0004 n.s. n.s.
Negative Symptoms (SANS) .0758 n.s. n.s.
Affect Flattening n.s. n.s. n.s.
Alogia .0678 n.s. n.s.
Avolition .0113 n.s. n.s.
Anhedonia n.s. n.s. n.s.
Symptom Checklist (SCL)-90 Positive Symptoms n.s. n.s. .0142
Hamilton Depression Scale (HAM-D) n.s. n.s. n.s.
Side Effect Measures
Body Mass Index (BMI) n.s. n.s. 0.0737
Abnormal Involuntary Movement (AIMS) n.s. n.s. 0.0664
Barnes Akathisia Scale (BAS) n.s. 0.0435 n.s.
Simpson-Angus Scale (SA) n.s. n.s. n.s.
Subjective Extrapyramidal Rating Scale (SERS) n.s. n.s. n.s.
Laboratory Measures
Glucose n.s. n.s. n.s.
HbA1c n.s. n.s. n.s.
Cholesterol n.s. n.s. 0.0699
Triglycerides n.s. 0.0195 n.s.
LDL Cholesterol n.s. n.s. n.s.
HDL Cholesterol n.s. n.s. 0.0300
Social Adjustment and Quality of Life Measures
SAS: General Adjustment n.s. n.s. n.s.
QOL: Total Score .0538 n.s. n.s.
Open in a new tab
1

All changes over time are in the direction of improvement.

2

BPRS and SANS totals are sums; Subscale scores are averages of their respective items.

Relapse

The OLZ-RSP group had a significantly higher rate of hospitalization than the remaining 5 groups (59.5% by 18 months as compared to 12.5–28.9% for the other groups, Log rank test: X2 = 12.23 p=0.03). In contrast to the other groups, no participant in the OLZ-RSP group remained in the study and on their assigned study medication without a hospitalization for 600 days or longer, and the majority were hospitalized or dropped out in under a year. As subjects at the NH site were more likely to be on olanzapine prior to randomization and there was also a lower threshold for hospitalization at that site, we reran the analyses comparing only participants who entered the study on olanzapine (OLZ-OLZ vs. OLZ-RSP groups, total n=41). The OLZ-RSP group appeared to again have a shorter time to first hospitalization, although the difference did not quite achieve statistical significance (Log rank test: X2 = 2.99, p = .0838; Rate difference at 18 months: Z = 1.62, p = .1052).

Side Effects

There was a trend towards pre-randomization medication influencing change in BMI in response to randomized medication in the expected direction, with those who were randomized to the same medication that they were taking pre-randomization showing the least change in BMI (p = .0737). There were no significant differences in the rate of change in the side effect scales by medication group in the 6-group analysis. The only laboratory measures with suggested effects were triglyceride levels (group effect, no interaction) and HDL cholesterol (group-by-time interaction, with decreasing HDL only in the N-OLZ group and the largest increase in HDL in the N-RSP group.) Eliminating observations when subjects were on concurrent lipid-lowering medications again revealed no systematic differences. See Table 3 for details.

Functional outcomes

Social adjustment and role functioning

Again, there were no significant differences in rate of change in the social adjustment and role functioning subscales by medication group in the 6 group analyses. There was again borderline evidence of improvement over time on the QOL total score, and of better general social adjustment at the LA site. See Table 3 for details.

Employment

Median time to employment was similar across the 6 medication groups to 18 months, and there was no statistically significant difference in employment rates (Overall log rank test: X2 = 0.93, NS).

Discussion

The primary analysis comparing the 2 randomized treatment groups confirmed our hypothesis that risperidone and olanzapine have similar effectiveness in the context of supported employment. The low 18-month assigned medication continuation rates, consistent with those found in CATIE phase-1 (Lieberman et al., 2005), are somewhat surprising in this study where participants were actively involved in psychosocial rehabilitation. Study physicians were actively working with rehabilitation staff to reinforce obtaining and maintaining competitive employment. We might have expected stable outpatients who are focused on achieving vocational goals to be less reliant on medication manipulation to achieve satisfactory illness management. This finding may reflect the continual churn in medication many prescribers and patients undergo to find the elusive optimal trade-off between functioning, symptoms, and side effects. It may also simply reflect the difficulty study participants have in sustaining blinded, randomized treatment.

Our finding of differences in BMI change is consistent with the finding of greater weight gain during olanzapine treatment relative to other first-line SGAs from several other studies (Pagsberg et al., 2017; Zhao et al., 2016). However, we found substantially less weight gain than CATIE (where patients on olanzapine gained approximately 2 pounds per month). Moreover, we did not find the changes in lipids that appeared in other studies (Grootens et al., 2011; Patel et al., 2009). We also found no support for our hypothesis that differences in the effects of the two medications would influence functional outcomes and quality of life. This may be related to the careful monitoring of metabolic effects that took place during this study (Marder et al., 2004). It may have also been related to increased physical activity among this unique population that was participating in supported employment. We have previously reported that reversal of antipsychotic-associated weight gain was associated with vocational activity (O’Keefe et al., 2003). Finally, olanzapine and risperidone may be too similar in their effects on functional outcomes and quality of life to result in differences in employment parameters that are apparent beyond the robust effect of supported employment.

Our findings of significant improvement in thought disorder, suspiciousness, avolitional symptoms, and quality of life across both medication groups over time are intriguing, leading to several potential explanations. This could result from both risperidone and olanzapine being more effective than previous antipsychotic treatment, however, less than 40% of the sample entered the study on medications other than risperidone or olanzapine, and the 6-group analyses demonstrated little advantage for patients switched from other antipsychotics to these agents. It could also result from more frequent or effective psychiatric care during the study relative to previous care, although recruitment took place among patients in case management services in well-organized clinics. Finally, it is possible that involvement in supported employment was helpful in reducing symptom levels, potentially by engaging patients attention outside of their internal thought processes for significant periods of time (Jäckel et al., 2017). The latter explanation, together with the finding of attenuated changes in weight and positive changes in lipids, would be consistent with synergistic interactions between medication and rehabilitation (Kane et al., 2016; Noordsy et al., 2000).

We found rates of treatment-emergent tardive dyskinesia are higher than previously reported with risperidone and olanzapine (Mari et al., 2004; Schooler et al., 2005), but consistent with the findings of recent analyses among patients in maintenance treatment (Carbon et al., 2017; Yoshida et al., 2014). While our study was not powered for detection of treatment-emergent tardive dyskinesia, this finding should remind clinicians to remain vigilant for signs of tardive dyskinesia even among stable patients in maintenance treatment.

The secondary analysis comparing the 6 pre-randomization-by-randomized treatment groups suggested that patients who were changed from olanzapine to risperidone had a significantly shorter time to first hospitalization, consistent with the Essock reanalysis of the CATIE data (Essock et al., 2006). There was also a weak trend towards those who were randomized to the same medication that they were taking pre-randomization showing the least change in BMI. These findings suggest that clinical practice may result in medication selection that is better matched to patient specific responses than random assignment. However, they also suggest a dilemma for clinicians who are managing patients who are clinically stable on olanzapine, but who are also overweight, insulin resistant, or have elevated lipids. That is, changing medication is associated with a risk of psychotic exacerbation.

Limitations

We lacked a more distinct control medication such as clozapine or a first-generation antipsychotic medication, which are challenging to implement in controlled trials. Sample size may have also limited our power to detect differences between the medication treatment groups. All of the above findings must be considered exploratory given the fact that we did not correct for multiple comparisons. They should be considered most generalizable to patients who are established in psychiatric care and interested in competitive employment. Dropout from assigned medication condition was substantial over time. While there was no evidence of a differential rate of dropout among treatment groups, only a third of patients remained on their assigned medication at 18 months, creating potential for biases due to dropouts who were poor responders or had difficulty tolerating their assigned medication.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. 4 2000. [Google Scholar]
  2. American Psychiatric Association, Washington, DC. Andreasen, N.C. Methods for Assessing Positive and Negative Symptoms1. 1990;24:73–88. doi: 10.1159/000418013. [DOI] [PubMed] [Google Scholar]
  3. Barnes TR. A rating scale for drug-induced akathisia. Br J Psychiatry. 1989;154:672–676. doi: 10.1192/bjp.154.5.672. [DOI] [PubMed] [Google Scholar]
  4. Becker DR, Drake RE. Individual placement and support: A community mental health center approach to vocational rehabilitation. Community Ment Health J. 1994;30:193–206. doi: 10.1007/BF02188630. [DOI] [PubMed] [Google Scholar]
  5. Bilder RM, Goldman RS, Volavka J, Czobor P, Hoptman M, Sheitman B, Lindenmayer JP, Citrome L, McEvoy J, Kunz M, Chakos M, Cooper TB, Horowitz TL, Lieberman JA. Neurocognitive effects of clozapine, olanzapine, risperidone, and haloperidol in patients with chronic schizophrenia or schizoaffective disorder. Am J Psychiatry. 2002;159:1018–1028. doi: 10.1176/appi.ajp.159.6.1018. [DOI] [PubMed] [Google Scholar]
  6. Carbon M, Hsieh CH, Kane JM, Correll CU. Tardive dyskinesia prevalence in the period of second-generation antipsychotic use: A meta-analysis. J Clin Psychiatry. 2017;78:e264–e278. doi: 10.4088/JCP.16r108832. [DOI] [PubMed] [Google Scholar]
  7. Derogatis LR, Cleary PA. Confirmation of the dimensional structure of the scl-90: A study in construct validation. J Clin Psychol. 1977;33:981–989. doi: 10.1002/1097-4679(197710)33:4&#x0003c;981::AID-JCLP2270330412&#x0003e;3.0.CO;2-0. [DOI] [Google Scholar]
  8. Essock SM, Covell NH, Davis SM, Stroup TS, Rosenheck RA, Lieberman JA. Effectiveness of Switching Antipsychotic Medications. Am J Psychiatry. 2006;163:2090–2095. doi: 10.1176/ajp.2006.163.12.2090. [DOI] [PubMed] [Google Scholar]
  9. First M, Spitzer R, Gibbon M, Williams J. Biometrics Research. New York State Psychiatric Institute; New York: 2002. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Research Version, Patient Edition With Psychotic Screen (SCID-I/P W/PSY SCREEN) [Google Scholar]
  10. Glynn SM, Marder SR, Noordsy DL, O’Keefe C, Becker DR, Drake RE, Sugar CA. An RCT Evaluating the Effects of Skills Training and Medication Type on Work Outcomes Among Patients With Schizophrenia. Psychiatr Serv. 2017;68:271–277. doi: 10.1176/appi.ps.201500171. [DOI] [PubMed] [Google Scholar]
  11. Grootens KP, van Veelen NMJ, Peuskens J, Sabbe BGC, Thys E, Buitelaar JK, Verkes RJ, Kahn RS. Ziprasidone Vs Olanzapine in Recent-Onset Schizophrenia and Schizoaffective Disorder: Results of an 8-Week Double-Blind Randomized Controlled Trial. Schizophr Bull. 2011;37:352–361. doi: 10.1093/schbul/sbp037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hamilton M. A Rating Scale for Depression. J Neurol Neurosurg Psychiatry. 1960;23:56–62. doi: 10.1136/jnnp.23.1.56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Heinrichs DW, Hanlon TE, Carpenter WT. The Quality of Life Scale: An Instrument for Rating the Schizophrenic Deficit Syndrome. Schizophr Bull. 1984;10:388–398. doi: 10.1093/schbul/10.3.388. [DOI] [PubMed] [Google Scholar]
  14. Jäckel D, Kupper Z, Glauser S, Mueser KT, Hoffmann H. Effects of Sustained Competitive Employment on Psychiatric Hospitalizations and Quality of Life. Psychiatr Serv Wash DC appips201600083. 2017 doi: 10.1176/appi.ps.201600083. [DOI] [PubMed] [Google Scholar]
  15. Kahn RS, Fleischhacker WW, Boter H, Davidson M, Vergouwe Y, Keet IP, Gheorghe MD, Rybakowski JK, Galderisi S, Libiger J, Hummer M, Dollfus S, López-Ibor JJ, Hranov LG, Gaebel W, Peuskens J, Lindefors N, Riecher-Rössler A, Grobbee DE. Effectiveness of antipsychotic drugs in first-episode schizophrenia and schizophreniform disorder: an open randomised clinical trial. The Lancet. 2008;371:1085–1097. doi: 10.1016/S0140-6736(08)60486-9. [DOI] [PubMed] [Google Scholar]
  16. Kane JM, Robinson DG, Schooler NR, Mueser KT, Penn DL, Rosenheck RA, Addington J, Brunette MF, Correll CU, Estroff SE, Marcy P, Robinson J, Meyer-Kalos PS, Gottlieb JD, Glynn SM, Lynde DW, Pipes R, Kurian BT, Miller AL, Azrin ST, Goldstein AB, Severe JB, Lin H, Sint KJ, John M, Heinssen RK. Comprehensive Versus Usual Community Care for First-Episode Psychosis: 2-Year Outcomes From the NIMH RAISE Early Treatment Program. Am J Psychiatry. 2016;173:362–372. doi: 10.1176/appi.ajp.2015.15050632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Keefe RSE, Bilder RM, Davis SM, Harvey PD, Palmer BW, Gold JM, Meltzer HY, Green MF, Capuano G, Stroup TS, McEvoy JP, Swartz MS, Rosenheck RA, Perkins DO, Davis CE, Hsiao JK, Lieberman JA CATIE Investigators, Neurocognitive Working Group. Neurocognitive effects of antipsychotic medications in patients with chronic schizophrenia in the CATIE Trial. Arch Gen Psychiatry. 2007;64:633–647. doi: 10.1001/archpsyc.64.6.633. [DOI] [PubMed] [Google Scholar]
  18. Kumar A, Datta SS, Wright SD, Furtado VA, Russell PS. Atypical antipsychotics for psychosis in adolescents. Cochrane Database Syst Rev. 2013:CD009582. doi: 10.1002/14651858.CD009582.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. 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;173:353–357. doi: 10.1097/00005053-198506000-00005. [DOI] [PubMed] [Google Scholar]
  20. Lieberman JA, Stroup TS, McEvoy JP, Swartz MS, Rosenheck RA, Perkins DO, Keefe RSE, Davis SM, Davis CE, Lebowitz BD, Severe J, Hsiao JK. Effectiveness of Antipsychotic Drugs in Patients with Chronic Schizophrenia. N Engl J Med. 2005;353:1209–1223. doi: 10.1056/NEJMoa051688. [DOI] [PubMed] [Google Scholar]
  21. Marder SR, Essock SM, Miller AL, Buchanan RW, Casey DE, Davis JM, Kane JM, Lieberman JA, Schooler NR, Covell N, Stroup S, Weissman EM, Wirshing DA, Hall CS, Pogach L, Pi-Sunyer X, Bigger JT, Friedman A, Kleinberg D, Yevich SJ, Davis B, Shon S. Physical Health Monitoring of Patients With Schizophrenia. Am J Psychiatry. 2004;161:1334–1349. doi: 10.1176/appi.ajp.161.8.1334. [DOI] [PubMed] [Google Scholar]
  22. de Mari JJ, Lima MS, Costa AN, Alexandrino N, Rodrigues–Filho S, de Oliveira IR, Tollefson GD. The prevalence of tardive dyskinesia after a nine month naturalistic randomized trial comparing olanzapine with conventional treatment for schizophrenia and related disorders. Eur Arch Psychiatry Clin Neurosci. 2004;254:356–361. doi: 10.1007/s00406-004-0514-1. [DOI] [PubMed] [Google Scholar]
  23. Matsuda Y, Sato S, Iwata K, Furukawa S, Hatsuse N, Watanabe Y, Anzai N, Kishimoto T, Ikebuchi E. Effects of risperidone and aripiprazole on neurocognitive rehabilitation for schizophrenia. Psychiatry Clin Neurosci. 2014;68:425–431. doi: 10.1111/pcn.12147. [DOI] [PubMed] [Google Scholar]
  24. McEvoy JP, Lieberman JA, Perkins DO, Hamer RM, Gu H, Lazarus A, Sweitzer D, Olexy C, Weiden P, Strakowski SD. Efficacy and Tolerability of Olanzapine, Quetiapine, and Risperidone in the Treatment of Early Psychosis: A Randomized, Double-Blind 52-Week Comparison. Am J Psychiatry. 2007;164:1050–1060. doi: 10.1176/ajp.2007.164.7.1050. [DOI] [PubMed] [Google Scholar]
  25. McEvoy JP, Lieberman JA, Stroup TS, Davis SM, Meltzer HY, Rosenheck RA, Swartz MS, Perkins DO, Keefe RSE, Davis CE, Severe J, Hsiao JK. Effectiveness of Clozapine Versus Olanzapine, Quetiapine, and Risperidone in Patients With Chronic Schizophrenia Who Did Not Respond to Prior Atypical Antipsychotic Treatment. Am J Psychiatry. 2006;163:600–610. doi: 10.1176/ajp.2006.163.4.600. [DOI] [PubMed] [Google Scholar]
  26. Mueser KT, Aalto S, Becker DR, Ogden JS, Wolfe RS, Schiavo D, Wallace CJ, Xie H. The Effectiveness of Skills Training for Improving Outcomes in Supported Employment. Psychiatr Serv. 2005;56:1254–1260. doi: 10.1176/appi.ps.56.10.1254. [DOI] [PubMed] [Google Scholar]
  27. Noordsy DL, Torrey WC, Mead S, Brunette M, Potenza D, Copeland ME. Recovery-oriented psychopharmacology: Redefining the goals of antipsychotic treatment. J Clin Psychiatry. 2000;61:22–29. [PubMed] [Google Scholar]
  28. O’Keefe CD, Noordsy DL, Liss TB, Weiss H. Reversal of antipsychotic-associated weight gain. J Clin Psychiatry. 2003;64:907–912. doi: 10.4088/jcp.v64n0808. [DOI] [PubMed] [Google Scholar]
  29. Overall JE, Gorham DR. THE BRIEF PSYCHIATRIC RATING SCALE. Psychol Rep. 1962;10:799–812. doi: 10.2466/pr0.1962.10.3.799. [DOI] [Google Scholar]
  30. Pagsberg AK, Tarp S, Glintborg D, Stenstrøm AD, Fink-Jensen A, Correll CU, Christensen R. Acute Antipsychotic Treatment of Children and Adolescents With Schizophrenia-Spectrum Disorders: A Systematic Review and Network Meta-Analysis. J Am Acad Child Adolesc Psychiatry. 2017;56:191–202. doi: 10.1016/j.jaac.2016.12.013. [DOI] [PubMed] [Google Scholar]
  31. Patel JK, Buckley PF, Woolson S, Hamer RM, McEvoy JP, Perkins DO, Lieberman JA Investigators, for the C. Metabolic profiles of second-generation antipsychotics in early psychosis: Findings from the CAFE study. Schizophr Res. 2009;111:9–16. doi: 10.1016/j.schres.2009.03.025. [DOI] [PubMed] [Google Scholar]
  32. Salkever D, Slade E, Karakus M. Differential effects of atypical versus typical antipsychotic medication on earnings of schizophrenia patients: estimates from a prospective naturalistic study. Pharmaco Economics. 2006;24:123–139. doi: 10.2165/00019053-200624020-00003. [DOI] [PubMed] [Google Scholar]
  33. Schooler N, Rabinowitz J, Davidson M, Emsley R, Harvey PD, Kopala L, McGorry PD, Van Hove I, Eerdekens M, Swyzen W, et al. Risperidone and haloperidol in first-episode psychosis: a long-term randomized trial. Am J Psychiatry. 2005;162:947–953. doi: 10.1176/appi.ajp.162.5.947. [DOI] [PubMed] [Google Scholar]
  34. Simpson GM, Angus JW. A Rating Scale for Extrapyramidal Side Effects. Acta Psychiatr Scand. 1970;45:11–19. doi: 10.1111/j.1600-0447.1970.tb02066.x. [DOI] [PubMed] [Google Scholar]
  35. Stroup TS, Lieberman JA, McEvoy JP, Swartz MS, Davis SM, Rosenheck RA, Perkins DO, Keefe RSE, Davis CE, Severe J, Hsiao JK. Effectiveness of Olanzapine, Quetiapine, Risperidone, and Ziprasidone in Patients With Chronic Schizophrenia Following Discontinuation of a Previous Atypical Antipsychotic. Am J Psychiatry. 2006;163:611–622. doi: 10.1176/ajp.2006.163.4.611. [DOI] [PubMed] [Google Scholar]
  36. Sungur M, Soygür H, Güner P, Üstün B, Çetin I, Falloon IR. Identifying an optimal treatment for schizophrenia: a 2-year randomized controlled trial comparing integrated care to a high-quality routine treatment. Int J Psychiatry Clin Pract. 2011;15:118–127. doi: 10.3109/13651501.2011.554987. [DOI] [PubMed] [Google Scholar]
  37. Takeuchi H, Thiyanavadivel S, Agid O, Remington G. Gradual vs. wait-and-gradual discontinuation in antipsychotic switching: A meta-analysis. Schizophr Res. 2017 doi: 10.1016/j.schres.2017.02.011. [DOI] [PubMed] [Google Scholar]
  38. Van Putten T, May PRA, Marder SR, Wittmann LA. Subjective Response to Antipsychotic Drugs. Arch Gen Psychiatry. 1981;38:187–190. doi: 10.1001/archpsyc.1981.01780270073010. [DOI] [PubMed] [Google Scholar]
  39. Weissman MM, Olfson M, Gameroff MJ, Feder A, Fuentes M. A Comparison of Three Scales for Assessing Social Functioning in Primary Care. Am J Psychiatry. 2001;158:460–466. doi: 10.1176/appi.ajp.158.3.460. [DOI] [PubMed] [Google Scholar]
  40. Wirshing DA, Wirshing WC, Marder SR, Liberman RP, Mintz J. Informed Consent: Assessment of Comprehension. Am J Psychiatry. 1998;155:1508–1511. doi: 10.1176/ajp.155.11.1508. [DOI] [PubMed] [Google Scholar]
  41. Yoshida K, Bies RR, Suzuki T, Remington G, Pollock BG, Mizuno Y, Mimura M, Uchida H. Tardive dyskinesia in relation to estimated dopamine D2 receptor occupancy in patients with schizophrenia: Analysis of the CATIE data. Schizophr Res. 2014;153:184–188. doi: 10.1016/j.schres.2014.01.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zhao YJ, Lin L, Teng M, Khoo AL, Soh LB, Furukawa TA, Baldessarini RJ, Lim BP, Sim K. Long-term antipsychotic treatment in schizophrenia: systematic review and network meta-analysis of randomised controlled trials. Br J Psychiatry Open. 2016;2:59–66. doi: 10.1192/bjpo.bp.115.002576. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES