Schizophrenia Research 78 (2005) 161 – 169 www.elsevier.com/locate/schres Quality of life during treatment with haloperidol or olanzapine in the year following a first psychotic episode Stephen M. Strakowskia,*, Jacqueline L. Johnsonb, Melissa P. DelBelloa, Robert M. Hamerb, Alan I. Greenc, Mauricio Tohend,h, Jeffrey A. Liebermanb,e, Ira Glickf, Jayendra K. Patelg HGDH Research Group1 a Center for Imaging Research and Division of Bipolar Disorders Research, University of Cincinnati College of Medicine, 231 Albert Sabin Way (ML0583), Cincinnati, OH 45267-0583, USA b Department of Psychiatry, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA c Dartmouth Medical School, Lebanon, NH, USA d Eli Lilly and Co., Lilly Research Laboratories, Indianapolis, IN, USA e Columbia University, School of Medicine, New York, NY, USA f Stanford University School of Medicine, CA, USA g University of Massachusetts Medical Center, MA, USA h McLean Hospital, Harvard Medical School, Belmont, MA, USA Received 12 January 2005; received in revised form 13 April 2005; accepted 15 April 2005 Available online 13 June 2005 Abstract Objectives: Schizophrenia causes significant impairments of quality of life. As treatment approaches have advanced, more attention has been given to re-integrating patients into their psychosocial environments, rather than simply monitoring psychotic * Corresponding author. Tel.: +1 513 558 2958; fax: +1 513 558 7164. E-mail address: Stephen.Strakowski@uc.edu (S.M. Strakowski). 1 This paper was based in part on results from the study of the Comparative Efficacy and Safety of Atypical and Conventional Antipsychotic Drugs in First-Episode Psychosis by the HGDH Study Group sponsored by Eli Lilly and Company. The HGDH research group consists of Drs. Jeffrey Lieberman and Diana Perkins, Dept. of Psychiatry, Univ. of North Carolina School of Medicine, NC, USA; Drs. Joseph P. McEvoy, Cecil Charles and Richard Keefe, John Ulmstead Hospital, Duke University Health System, NC, USA; Drs. Robert B. Zipursky and Zafiris J. Daskalakis, Dept. of Psychiatry, University of Toronto School of Medicine, Ontario, Canada; Dr. Alan I. Green, Massachusetts Mental Health Center, Harvard Medical School, MA, USA; Dr. Charles B. Nemeroff, Dept. of Psychiatry, Emory University School of Medicine, GA, USA; Prof. Robin Murray and Dr. Tonmoy Sharma, Institute of Psychiatry, UK; Dr. Raquel E. Gur, Dept of Psychiatry, Univ. of Pennsylvania Medical Center, PA, USA; Drs. Bruce Cohen and Franca Centorrino, McLean Hospital, Harvard Medical School, Belmont, MA, USA; Prof. Dr. R.S. Kahn, University Hospital, Utrecht, The Netherlands; Drs. Wayne Goodman and John Kuldau, Dept. of Psychiatry, Univ. of Florida, FL, USA; Drs. Anthony J. Rothschild and Jayendra K. Patel, Dept. of Psychiatry, Univ. of Massachusetts Medical Center, MA, USA; Dr. Stephen M. Strakowski, Dept. of Psychiatry, Univ. of Cincinnati, OH, USA; Dr. Ira Glick, Dept. of Psychiatry, Stanford University School of Medicine, CA, USA; Dr. John De Quardo, Dept. of Psychiatry, Univ. of Michigan Medical Center, MI, USA; Drs. Gary Tollefson, Todd Sanger, Mauricio Tohen, Lilly Research Laboratories, IN, USA. 0920-9964/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2005.04.017 162 S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 symptoms. The development of the second-generation antipsychotics raised hope that these medications would provide better quality of life improvement than conventional antipsychotics. This improvement is particularly relevant early in the course of schizophrenia. Methods: To address these considerations, improvements in measures of general health and social function (determined using the SF-36) were assessed in 195 patients with first-episode schizophrenia for up to one year following randomization to either olanzapine or haloperidol in a double blind clinical trial. We hypothesized that olanzapine would demonstrate better improvement on these measures than haloperidol. In order to test this hypothesis, we used a repeated measure model with SF-36 scores as the outcome, and treatment group, time, time2, time-by-treatment group interaction, and time2-by-treatment group interaction as fixed effects. Results: Both treatments demonstrated similar changes on the SF-36. Independent of treatment, patients demonstrated significant improvements in most of the SF-36 subscales, which approached normative scores by the end of one year of treatment. Forty-six of 100 olanzapine-treated patients and 37 of 95 haloperidol-treated patients completed the one year of this study ( p b .4). Conclusions: These results suggest an important initial treatment goal for patients with new onset schizophrenic disorders, namely that they can expect to recover significant quality of life and social function at least initially in treatment. D 2005 Elsevier B.V. All rights reserved. Keywords: Functional outcome; SF-26; First-episode psychosis; Schizophrenia 1. Introduction Although it has been recognized for many years that antipsychotic medications improve psychotic symptoms in patients with schizophrenia, improvements in other aspects of this illness have gained increasing attention with the development of newer medications. For example, the negative symptoms and cognitive impairments of schizophrenia have been identified as better predictors of long-term outcome than positive (psychotic) symptoms, and the newer medications may provide increased benefit in these domains relative to classic neuroleptics (Sharma and Antonova, 2003; Moller, 2003). As advances have continued in the assessment of clinical treatment response, a patient’s ability to re-integrate into his or her psychosocial environment after acute psychosis, and despite chronic symptoms, has increasingly been recognized as representing the most sensitive indicator of clinical improvement. This re-integration is assessed through measures of quality of life and functional status, and such assessments have become commonplace in the study of medical conditions, generally, and increasingly in psychiatric disorders (Nasrallah et al., 2004). There has been hope that the second-generation antipsychotics will demonstrate better clinical efficacy in these areas of recovery relative to traditional neuroleptics, although studies to date are relatively few (Awad and Voruganti, 2004; Nasrallah et al., 2004). Olanzapine demonstrated modest improvements in quality of life measurements, which were significantly greater than haloperidol in two one-year studies (Revicki et al., 1999; Tunis et al., 1999), but not in a third (Rosenheck et al., 2003). Olanzapine also demonstrated similar effects on quality of life outcome compared with clozapine (Naber et al., 2005) and risperidone (Ritsner et al., 2004). Similar to the latter study, a naturalistic comparison of first- and secondgeneration antipsychotics in quality of life improvements in patients with schizophrenia found no differences between these drug classes (Kilian et al., 2004). These studies have predominantly included patients with chronic schizophrenia, in whom treatment response may be limited. Symptomatic treatment response appears to be more robust in new onset patients, particularly immediately following a first psychotic episode (Lieberman et al., 1998). Similarly, improvements in functional outcome and quality of life may be more robust in response to treatment early, rather than later, in the course of schizophrenia. Although there is no absolute dgold standardT of measurement of functional and health status the Medical Outcomes Study 36-item Short-Form health survey (SF-36) (Ware et al., 1993, 1994) has become a widely used, well-validated measure of health-related quality of life and functional outcome. The SF-36 is a generic health status measure designed to evaluate S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 functioning and health-related quality of life in a wide variety of chronic medical conditions (Ware et al., 1993, 1994). It has been shown to be reliable and valid in schizophrenia (Meijer et al., 2002; Russo et al., 1998; Pukrop et al., 2003) and has successfully used to measure functional and health status in several studies in schizophrenic populations (Nasrallah et al., 2004; Revicki et al., 1999; Gureje et al., 2003; Tunis et al., 1999). Consequently, we utilized this instrument in the current study. With these considerations in mind, we compared changes in one-year quality of life and functional outcome as assessed with the SF-36 in 195 patients with first-episode schizophrenic spectrum psychoses treated with olanzapine or haloperidol. Quality of life in this study refers primarily to general health outcomes, as assessed by the SF-36. Our primary hypothesis was that the second-generation antipsychotic (olanzapine) would provide greater improvement in these outcome measurements than the first-generation medication (haloperidol) consistent with findings in more chronic populations (Revicki et al., 1999; Tunis et al., 1999). 2. Methods 2.1. Subjects Patients were recruited as part of a double blind, randomized, multi-site, international 2-year comparison study of olanzapine versus haloperidol in 263 patients with first-episode psychosis (the HGDH study; 132 were randomized to haloperidol, 131 to olanzapine). Detailed descriptions of the design and methods of the HGDH study have been published elsewhere (Lieberman et al., 2003; Green et al., 2004; Perkins et al., 2004), so a relatively brief discussion of aspects of the design relevant to the current analysis will be presented here. Patients, 16–40 years old, were recruited from 14 academic medical centers in North America and Western Europe at the time they presented for treatment of psychosis. Patients were included if they met DSM-IV criteria (assessed using the SCID-I/P; First et al., 1995) for schizophrenia, schizoaffective disorder or schizophreniform disorder. Patients were defined as dfirst-episodeT if they had less then 5 years of psychotic symptoms and had not 163 recovered from this first episode for a period more than 6 months, and they had received V 16 cumulative weeks of antipsychotic treatment (Lieberman et al., 2003). Entry into the study required active psychotic symptoms characterized by at least two Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987) psychosis items z 4 or a score on one psychosis item z 5, and a CGI score z 4 (moderately ill). For this analysis, patients had to have completed both a baseline and at least one follow-up SF-36, as well as have measures of the covariates included in the analysis; 195 (74%) of the 263 patients met this criterion. These 195 patients were similar to the total sample in average age (24 years in both), percent men (82% v. 80%, total v. this sample, respectively), percent white (53% v. 55%), and diagnosis (59% v. 61% with schizophrenia). 2.2. Study design and procedures Each participating institution’s human subjects review board approved this study and all subjects provided written informed consent after procedures had been explained in full, and after they had been determined to be competent to provide consent. After initial screening and a 2–14 day washout (in patients who were on antipsychotics at the time of screening), patients were randomized to haloperidol 2–6 mg/day or olanzapine 5–20 mg/day, adjusted during the first 12 weeks of the trial (Lieberman et al., 2003). Additional medications for agitation or insomnia (lorazepam, diazepam, chloral hydrate) or antipsychotic side effects (benztropine, biperiden, propranolol, procyclidine) were allowed on an as needed basis (for details, see Lieberman et al., 2003). Subjects were then seen at regular intervals for symptom and outcome ratings for up to 92 additional weeks (104 weeks total). However, because only a small number of subjects completed the second year of this study (N = 30 in the olanzapine group and N = 14 in the haloperidol group), this analysis is restricted to one year of outcome only. At one year, 88 patients (46 of 100 patients on olanzapine and 37 of 95 on haloperidol) remained in the study (m2 = 0.99, df = 1, p b .4). Reasons for study discontinuation have been reviewed elsewhere (Lieberman et al., 2003), but most commonly was related to the patient’s or physician’s decision not to continue for a wide variety of reasons. 164 S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 Medication adherence was monitored by pill counts at each visit and non-adherence was a priori defined as z 5 days of not taking medications. Overall, about 50% of subjects met this definition at some point during the study; although the haloperidol group was 1.4 times more likely to meet this definition of non-adherence, this difference was not significant ( p = .11; Perkins et al., submitted for publication). Additionally, after acute treatment, certain thymoleptics were allowed and were similarly prescribed between groups. In the haloperidol group, 20 patients received fluoxetine, and one each received lithium and divalproex; in the olanzapine group 19 patients received fluoxetine, 2 received lithium and one received divalproex. 2.3. Assessment of quality of life and psychosocial function Quality of life (QOL) and psychosocial function were assessed using the Medical Outcomes Study Short-Form 36 Healthy Survey (SF-36; Ware et al., 1993). The SF-36 is a widely used, well-established measure of general health and functional status that has been shown to be valid and reliable in patients with schizophrenia (Revicki et al., 1999; Pukrop et al., 2003; Nasrallah et al., 2004; Russo et al., 1998; Gureje et al., 2003; Meijer et al., 2002; Tunis et al., 1999). In schizophrenia, the SF-36 has been shown to have very good internal consistency (Chronbach’s a = 0.71–0.89) and is responsive to symptom improvements (Pukrop et al., 2003). The SF-36 consists of 36 questions that produce 8 subscales: physical function, bodily pain, role limitations due to physical problems (Role–Physical), vitality, general health perceptions, role limitations due to emotional problems (Role–Emotional), mental health, and social function. Each of these subscales is linearly transformed into a 0–100 scale in which higher scores correspond with better health status and function (i.e. QOL; Ware et al., 1993). Additionally, two summary scores, based on these eight subscales, can be calculated: the physical component summary and the mental component summary (Ware et al., 1994). These summary scores were constructed such that the general population’s mean score is 50 with a standard deviation of 10. As noted, the patients included in this analysis (N = 195) completed a baseline SF-36 and at least one additional SF-36 during the first year of followup, as well as having information for all covariates (14 patients were missing one or more covariate). Similar percentages of patients in each treatment group met this criterion [100 / 131 (76%) for olanzapine and 95 / 132 (72%) for haloperidol]. Subjects were scheduled to complete the SF-36 at baseline, 3 months, 6 months, and one year. Trained clinicians or interviewers administered all patient interviews. 2.4. Statistical analysis Two-sided Wilcoxon tests or Fisher’s exact tests were used to compare baseline characteristics between treatment groups. In order to test our primary hypothesis and to evaluate therapy group differences in SF-36 assessments across the major time points (i.e., from baseline to 3, 6, and 12 months), for each SF-36 scale, we fit a repeated measure model with SF-36 scores as the outcome and treatment group, time, time2, time-by-treatment group interaction, and time2-by-treatment group interaction as fixed effects. We assumed unstructured covariance for the time repeated measure. The model accommodates missing time points, provides repeated assessments at different time points, and provides a therapy contrast for both linear and quadratic change in SF-36 assessments over one year. In this model, the primary contrast of interest is the interaction term of time-by-treatment group, to identify whether one treatment leads to better health status than the other. Additionally, to accommodate potential non-linear interactions, we also fit a quadratic term to these data (i.e., time2) in which the primary variable of interest would therefore be time2-by-treatment group. To control for the potential confounds of other patient characteristics, we covaried by study site, age, sex, ethnicity, duration of illness (time from disease onset to study enrollment), baseline symptom ratings (PANSS Total score, MADRS total score), baseline substance use (yes or no), and baseline premorbid adjustment (the Premorbid Adjustment Scale, PAS; Cannon-Spoor et al., 1982). We also adjusted for side effects that differed among treatments — weight gain, EPS (Simpson Angus scale, Barnes Akithisia Scale, and AIMS). To control for multiple contrasts of the eight SF-36 subscale scores, a Bonferroni correction was used to define significance as p b .00625 (i.e., 165 S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 variables, although the patients randomized to haloperidol had a longer mean duration of illness. Additionally, the two treatment groups showed similar baseline SF-36 subscale and summary item scores (Table 2), although the haloperidol group showed a significantly better rating on general health, which is also considered a clinically meaningful difference (i.e., N5 points; Ware et al., 1993). p = .05 / 8). Significance for the two summary scores was defined as p b .05. SF-36 subscale scores were compared to normative data using t-tests and a Bonferroni correction for significance ( p b .00625). Normative data were taken from published tables (Ware et al., 1994), but were weighted to the age and sex distribution of this current study sample. Finally, other statistical contrasts were performed as necessary for completeness. 3.2. Outcome measures Using the repeated measures model described previously, none of the SF-36 subscale or summary scores exhibited significant time-by-treatment group effects, thereby failing to support our primary hypothesis. In contrast, significant improvement over time (i.e., a significant time effect) was observed for all patients for most of the SF-36 variables. Specifically, significant improvement over the one-year period, after controlling for covariates, was observed for 5 of the 8 subscales: bodily pain [ F(1, 416) = 22.0, p b .0001], general health [ F(1, 411) = 6.2, p b .02], 3. Results 3.1. Baseline and clinical characteristics The baseline clinical and demographic characteristics of the 195 patients analyzed in this report are listed in Table 1. These patients were predominantly young men with and average of 14 months of symptoms. Approximately 72% had prior antipsychotic exposure of an average of 6 weeks duration. The two treatment groups were similar on most of these Table 1 Baseline demographic and clinical information for 195 patients with first-episode schizophrenia–spectrum psychosis participating in a two-year treatment study comparing olanzapine with haloperidol Variable All patients (N = 195) Olanzapine- treated patients (N = 100) Haloperidol-treated patients (N = 95) Age, years Sex, N(%) male Ethnicity, N(%) White African-American Other Diagnosis, N(%) Schizophrenia Schizophreniform Schizoaffective PANSS total PAS total Duration of illness, weeks Duration of previous antipsychotic use, weeksa Substance use disorder, N(%) Hospitalized at index, N(%) Active subjects, N(%) at: 3 months 6 months 12 months 24 (5) 156 (80) 24 (5) 76 (76) 24 (5) 80 (84) 108 (55) 69 (35) 18 (10) 54 (54) 35 (35) 11 (11) 54 (57) 34 (36) 7 (7) Treament differences, p-value N.8 N.2 N.6 N.14 119 58 18 81 0.33 65 6 (61) (30) (9) (15) (0.16) (62) (10) 54 35 11 81 0.33 54 6 (54) (35) (11) (15) (0.16) (55) (7) 65 23 7 81 0.22 77 6 (69) (24) (7) (15) (0.15) (67) (13) N.8 N.7 b.007 N.4 15 (8) 112 (57) 8 (8) 60 (60) 7 (7) 52 (55) 1.0 N.4 154 (79) 123 (63) 83 (46) 85 (85) 70 (70) 46 (46) 69 (73) 53 (56) 37 (39) b.04 b.06 b.4 Variables listed are mean (standard deviation), unless otherwise noted. a Calculated only for patients with prior antipsychotic use: olanzapine (N = 76), haloperidol (N = 64). 166 S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 100 mental health [ F(1, 414) = 44.4, p b .0001], role–emotional [ F(1, 415) = 60.2, p b .0001], and social functioning [ F(1, 416) = 87.0, p b .0001]. These improvements in ratings over time are illustrated in Fig. 1 (in this figure only points for baseline and one year are provided, to improve presentation; changes from baseline to 1 year were roughly linear). Significant improvement was also observed over time in all subjects in the Mental Summary scale score [ F(1, 411) = 72.0, p b .0001], but not the Physical Summary scale score [ F(1, 411) = 0.6, p N .4]. These are illustrated for each treatment group separately in Fig. 2. The quadratic model (i.e., time2, time2*treatment group) provided essentially the same findings. 90 80 70 60 50 Baseline one year normative 40 30 20 10 y lit ta th ta M en er en G Vi lh he al ea l th in al ily io B eol R od ot em fu al ci So pa na l n tio nc ys e- ol Ph ys ic 3.3. Comparisons with normative data R al fu nc Ph tio ni ic ng al 0 Baseline scores were significantly below normative scores for all subscales except physical functioning [t(194) N 4.7, p b .001] (Fig. 1). At one year, only vitality [t(86) = 3.8, p b .001], role–emotional [t(86) = 3.4, p b .005], and role–physical [t(86) = 3.6, p b .001] remained significantly less than norms. Fig. 1. Adjusted (least-square) mean scores for SF-36 subscales from 195 patients with first-episode schizophrenia–spectrum disorders at baseline and after one year of treatment with olanzapine or haloperidol, compared with normative data weighted for age and sex to the study sample. All subscales exhibited significant improvement ( p b .00625) between baseline and one year except physical functioning, role–physical and vitality. Baseline scores were significantly ( p b .00625) below normative scores for all subscales except physical functioning. At one year, only role–emotional, role–physical, and vitality remained significantly ( p b .00625) below normative values. 4. Discussion In this one-year study of patients with new onset schizophrenic disorders, treatment with both olanzapine and haloperidol was associated with significant improvements in health status and functional outcome as measured with the SF-36, with no significant differences between therapies. Therefore, our hypothesis that greater improvement would be observed with Table 2 Baseline SF-36 transformed subscale and summary scores for 195 patients with first-episode schizophrenia–spectrum psychosis participating in a two-year treatment study comparing olanzapine with haloperidol Variable Subscale scores: Bodily pain General health Mental health Physical functioning Role–Emotional Role–Physical Social functioning Vitality Summary scores: Physical summary Mental summary Adjusted normative scores All patients (N = 195) Olanzapine- treated patients (N = 100) Haloperidol- treated patients (N = 95) Treatment differences, p-value 82 78 75 94 83 91 85 64 79 66 54 91 33 72 47 51 81 63 52 90 34 75 46 50 77 69 57 91 33 70 48 52 N.1 b.05 N.08 N.9 N.7 N.4 N.6 N.4 (20) (17) (17) (14) (31) (23) (21) (19) 50 (10) 50 (10) (27) (21) (20) (18) (40) (39) (31) (21) 55 (10) 32 (12) (27) (21) (19) (20) (39) (38) (30) (19) 55 (10) 31 (11) Variables listed are mean (standard deviation), unless otherwise noted. (26) (20) (21) (17) (40) (41) (32) (22) 55 (10) 33 (12) N.7 N.3 S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 60 50 40 30 20 10 PS-HAL PS-OLZ MN-OLZ Normative MN-HAL 0 Baseline 12 weeks 6 months 1 year Fig. 2. Changes in SF-36 Physical Summary (PS) and Mental Summary (MS) mean scores in 195 patients with first-episode schizophrenia–spectrum disorders after one year of treatment with olanzapine or haloperidol. The MS, but not the PS, demonstrates significant improvement over time ( p b .0001). olanzapine was not supported. However, both groups showed marked improvement during this first year of treatment, consistent with the recognized antipsychotic responsiveness of first-episode schizophrenic spectrum disorders. This treatment responsiveness of new onset patients in general may have diminished the possibility of finding treatment differences between groups. The greater improvement in functional outcomes observed with atypical agents compared to conventional antipsychotics may be limited to chronic patients who show limited functional improvement with conventional agents. Consequently, results from the current sample of young patients early in their course of illness may differ from results from chronic samples. In fact, the lack of differences between haloperidol and olanzapine in this study is consistent with some previous studies in chronic schizophrenic patients (Rosenheck et al., 2003; Kilian et al., 2004), although not others (Revicki et al., 1999; Tunis et al., 1999). Notably, in these latter reports in which olanzapine was found to be superior, treatment differences were generally rather modest (i.e., absolute changes b 10 points with effect sizes b0.4). Importantly, the SF-36 was not specifically designed to study quality of life or functional outcome in schizophrenia; this 167 instrument is primarily focused on physical health issues that may not be relevant to a young sample of patients with schizophrenia. A more specific instrument might be able to identify more subtle differential treatment effects. The treatment groups did not differ in functional outcome despite previously reported differences in symptom improvement at 12 weeks (Lieberman et al., 2003). In that previous report, the effect sizes for differences between olanzapine and haloperidol were modest (in the range of 0.3–0.4 typically), suggesting that a greater magnitude of symptom differences may be needed to produce corresponding differences in functional improvement. Additionally, this observation raises the point that medications alone may be inadequate to significantly impact quality of life without other systematic interventions (e.g., psychosocial and rehabilitative therapies). Although there were minimal differences between haloperidol and olanzapine treatment in SF-36 measures in this study, overall the patients demonstrated marked and significant improvement in most of these measures over time. Most of the subscale scores that significantly differed from normative values at baseline showed significant improvement by the end of one year, and all of the subscales showed numeric improvement. The largest improvements observed were in subscale scores of role limitations due to emotional problems, social functioning, and mental health, likely reflecting improvement in psychiatric symptomatology, which was also observed in this study (see Lieberman et al., 2003). Moreover, the Mental summary score similarly showed marked improvement. In contrast, there were relatively minimal changes in specific physical scales, with the physical functioning subscale and Physical summary score being normative at baseline and changing minimally over time. These findings were possibly expected, given the known benefit of effective antipsychotic therapy on psychotic symptoms of schizophrenia, and the relative physical health of these primarily young patients; nonetheless, normalization of these measures supports the ability of effective treatments to assist patients with re-integration into their psychosocial environments. These results suggest an important initial treatment goal for patients with new onset schizophrenic disorders, namely that they can expect to recover significant quality of life and social function at least initially in treatment. Future work will 168 S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 be important to determine how these measures change over longer periods of time and which patients are at risk for psychosocial functional and health status deterioration longitudinally. As with any clinical study, limitations must be considered when interpreting these results. As already noted, the SF-36 was not designed for the study of schizophrenia, so therefore may be insensitive to some aspects of the psychosocial and health-related impairments of this disorder. This insensitivity might have contributed to failure to identify differences between treatment groups or to detect more subtle differences that are important to patients, but not measured with the SF-36. Additionally, over 50% of the sample discontinued study participation by the end of one year; it is possible that these patients were no longer representative of the entire sample and, therefore, treatment differences were obscured. In fact, the original study goal was to evaluate 2-year outcomes, but the limited number of 2-year completers precluded this analysis. Patients on haloperidol were more likely to discontinue than those on olanzapine (Lieberman et al., 2003), so that if SF-36 measures were strongly related to discontinuation, differences between treatment groups would have been expected to emerge. The absence of these differences suggests that patients that discontinued were scoring similarly on the SF-36 between groups. Nonetheless, the patients doing the least well may have been particularly prone to discontinue from both treatment groups, thereby elevating the overall responsiveness of these patients on SF-36 measures. Additionally, the most severely ill patients typically do not participate in research studies, due to issues related to obtaining informed consent as well as meeting strict entry criteria. Therefore, these results may represent a dbest outcomeT assessment. In this study, patients were not formally re-diagnosed after the baseline evaluation. Although the diagnoses of schizophrenia and schizoaffective disorder tend to be stable, schizophreniform disorder is an unstable diagnosis that tends to change over time (Strakowski, 1994). Therefore, it is conceivable that differences in evolving diagnoses influenced this study. However, most cases of schizophreniform disorder progress to schizophrenia or schizoaffective disorder, so that these patients would continue to fit the study’s overall goals. Moreover, the distribution of schizophreniform disorder was similar between the treatment groups. Finally, the definition of dfirst-episodeT in this study was relatively liberal, in that patients with illness as long as five years, and who had received 16 weeks of treatment, could be included. However, the average subject had only slightly more than one year of illness and 6 weeks for treatment, and nearly one third of subjects still met criteria for schizophreniform disorder, suggesting that many of these patients would meet more conservative definitions of dfirst-episode psychosisT as well. In summary, olanzapine and haloperidol did not substantially differ in their effects on psychosocial functional and health status improvement in the first year of treatment following the onset of a schizophrenic psychosis. Both treatments were effective for normalizing measures of general health and functional status, thereby providing an important treatment goal and expectation in this population. Acknowledgements This work was support by an investigator-initiated (JAL) grant from the Eli Lilly Research Foundation. References Awad, A.G., Voruganti, L.N., 2004. New antipsychotics, compliance, quality of life, and subjective tolerability — are patients better off? Canadian Journal of Psychiatry 49, 283 – 284. Cannon-Spoor, H.E., Potkin, S.G., Wyatt, R.J., 1982. Measurement of premorbid adjustment in chronic schizophrenia. Schizophrenia Bulletin 8, 470 – 484. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1995. Structured Clinical Interview for DSM-IV Axis I Disorders— Patient Editions (SCID-I/P). Biometrics Research Department, New York State Psychiatric Institute, 722 West 168th Street, New York, NY 10032. Green, A.I., Tohen, M.F., Hamer, R.M., et al., 2004. First episode schizophrenia-related psychosis and substance use disorders: acute response to olanzapine and haloperidol. Schizophrenia Research 66, 125 – 135. Gureje, O., Miles, W., Keks, N., et al., 2003. Olanzapine vs. risperidone I the management of schizophrenia: randomized double-blind trial in Australia and New Zealand. Schizophrenia Research 61, 303 – 314. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia Bulletin 13, 261 – 276. Kilian, R., Dietrich, S., Toumi, M., Angermeyer, M.C., 2004. Quality of life in persons with schizophrenia in out-patient S.M. Strakowski et al. / Schizophrenia Research 78 (2005) 161–169 treatment with first- or second-generation antipsychotics. Acta Psychiatrica Scandinavica 110, 108 – 118. Lieberman, J.A., Sheitman, B., Chakos, M., et al., 1998. The development of treatment resistance in patients with schizophrenia: a clnical and pathophysiological perspective. Journal of Clinical Psychopharmacology 18, 205 – 245. Lieberman, J., Tohen, M., Green, A.I., et al., 2003. Comparative efficacy and safety of atypical and conventional antipsychotic drugs in first-episode psychosis: a randomized double-blind trial of olanzapine vs. haloperidol. American Journal of Psychiatry 160, 1396 – 1404. Meijer, C.J., Schene, A.H., Koeter, M.W., 2002. Quality of life in schizophrenia measured by the MOS SF-36 and lancashire quality of life profile: a comparison. Acta Psychiatrica Scandinavica 105, 293 – 300. Moller, H.J., 2003. Management of the negative symptoms of schizophrenia: new treatment options. CNS Drugs 17, 793 – 823. Naber, D., Riedel, M., Klimke, A., et al., 2005. Randomized double blind comparison of olanzapine vs. clozapine on subjective well-being and clinical outcome in patients with schizophrenia. Acta Psychiatrica Scandinavica 111, 106 – 115. Nasrallah, H.A., Duchesne, I., Mehnert, A., Janagap, C., Eerdekens, M., 2004. Health-related quality of life in patients with schizophrenia during treatment with long-acting, injectable risperidone. Journal of Clinical Psychiatry 65, 531 – 536. Perkins, D.O., Johnson, J.L., Hamer, R.M., et al., submitted for publication. Predictors of antipsychotic medication adherence in patients recovering from a first psychotic episode. Perkins, D., Lieberman, J., Gu, H., et al., 2004. Predictors of antipsychotic treatment response in patients with first-episode schizophrenia, schizoaffective and schizophreniform disorders. British Journal of Psychiatry 185, 18 – 24. Pukrop, R., Schlaak, V., Möller-Leimkühler, A.M., et al., 2003. Reliability and validity of quality of life assessed by the shortform 36 and the modular system for quality of life in patients with schizophrenia and patients with depression. Psychiatry Research 119, 63 – 79. 169 Revicki, D.A., Genduso, L.A., Hamilton, S.H., et al., 1999. Olanzapine versus haloperidol in the treatment of schizophrenia and other psychotic disorders: quality of life and clinical outcomes of a randomized clinical trial. Quality of Life Research 8, 417 – 426. Ritsner, M., Gibel, A., Perelroyzen, G., Kurs, R., Jabarin, M., Ratner, Y., 2004. Quality of life outcomes of risperidone, olanzapine, and typical antipsychotics among schizophrenia patients treated in routine clinical practice: a naturalistic comparative study. Journal of Clinical Psychopharmacology 24, 582 – 591. Rosenheck, R., Perlick, D., Bingham, D., et al., 2003. Effectiveness and cost of olanzapine and haloperidol in the treatment of schizophrenia: a randomized controlled trial. Journal of the American Medical Association 290, 2693 – 2702. Russo, J., Trujillo, C.A., Wingerson, D., et al., 1998. The MOS 36item short form healthy survey: reliability, validity and preliminary findings in schizophrenic outpatients. Medical Care 36, 752 – 756. Sharma, T., Antonova, L., 2003. Cognitive function in schizophrenia. Deficits, functional consequences, and future treatment. Psychiatric Clinics of North America 26, 25 – 40. Strakowski, S.M., 1994. Diagnostic validity of schizophreniform disorder. American Journal of Psychiatry 151, 815 – 824. Tunis, S.L., Croghan, T.W., Heliman, D.K., Johnstone, B.M., Obenchain, R.L., 1999. Reliability, validity, and application of the medical outcomes study 36-item short-form health survey (SF-36) in schizophrenic patients treated with olanzapine versus haloperidol. Medical Care 37, 678 – 691. Ware, J.E., Snow, K.K., Kosinski, M., Gandek, B., 1993. SF-36 Health Survey Manual and Interpretation Guide. The Health Institute, New England Medical Center, Boston, MA. Ware, J.E., Koskinsk, M., Keller, S.D., 1994. SF-36 Physical and Mental Health Summary Scales: A User’s Manual. The Health Institute, New England Medical Center, Boston, MA.