| | Risk of discontinuation of atypical antipsychotic agents in the treatment of schizophreniaReceived 15 November 2006; received in revised form 25 April 2007; accepted 26 April 2007. published online 28 June 2007. Abstract ObjectivesTo compare discontinuation rates of atypical antipsychotic agents in patients with schizophrenia. ResultsAt one-year follow-up, most patients discontinued their antipsychotic medication (90.4% adjusted mean discontinuation). The hazard ratio (HR) for discontinuing therapy in patients starting treatment on aripiprazole, risperidone, or ziprasidone was not significantly different from olanzapine [HR 1.047, 0.973 and 0.990, respectively]. Quetiapine was associated with significantly higher hazard of discontinuation [HR 1.130] than olanzapine. Covariates associated with significantly lower discontinuation were being male [HR 0.899], older age [HR 0.997] and being on concurrent medication when initiating therapy [HR 0.225]; having a previous hospitalization was associated with significantly higher discontinuation hazard [HR 1.276]. Results were robust in the sensitivity analysis. ConclusionsDiscontinuation rates were high at one-year follow-up and did not differ significantly for patients on aripiprazole, olanzapine, risperidone, or ziprasidone. The higher hazard of discontinuation associated with quetiapine when compared to olanzapine is consistent with that observed in Phase I of the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE). 1. Introduction  Continuous treatment at therapeutic doses is an important goal for the effective management of schizophrenia, while discontinuation of medication is considered a negative outcome (Jaffe and Levine, 2003, Keith and Kane, 2003, Naber and Vita, 2004), resulting from factors such as medication intolerability or inadequate efficacy, and poor patient insight or attitude regarding the need for medication (Lacro et al., 2002, Keith and Kane, 2003). Various studies have associated nonadherence to or discontinuation of antipsychotic treatment with relapse, rehospitalization, and longer durations of inpatient hospital stays (Sullivan et al., 1995, Svarstad et al., 2001, Weiden et al., 2004, Eaddy et al., 2005). As a result, nonadherence also contributes to a substantial increase in the cost of schizophrenia management (Svarstad et al., 2001, Thieda et al., 2003, Knapp et al., 2004, Eaddy et al., 2005). Weiden et al., in 1995, projected that the increase in hospital costs due to noncompliance in schizophrenia over a two-year time period would reach $705 million (Weiden and Olfson, 1995). Until publication of the landmark NIMH-funded Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study, evidence comparing atypical antipsychotic discontinuation rates was inconclusive, in part, due to the non-equivalence of doses used in previous head-to-head trials and the indirect comparisons involved in available meta-analyses of clinical trial drop-out rates (Gilbody et al., 2000, Santarlasci and Messori, 2003, Pelagotti et al., 2004, Kemmler et al., 2005). The CATIE study was an important step forward in evaluating the relative effectiveness of antipsychotic treatments because it directly compared olanzapine, quetiapine, risperidone, ziprasidone (and perphenazine in Phase 1) in a randomized double-blind study that was powered to detect differences in discontinuation rates (Lieberman et al., 2005, Stroup et al., 2006). In the initial randomization phase, after statistical adjustment for multiple comparisons, time to discontinuation was longer for olanzapine relative to quetiapine and risperidone but not perphenazine or ziprasidone (Lieberman et al., 2005). In the subsequent randomization of patients entering Phase 2 (patients who discontinued their medication in the first phase of CATIE), time to discontinuation was longer for risperidone and olanzapine treated patients than for quetiapine or ziprasidone treated patients (Stroup et al., 2006). In both phases of the CATIE study, discontinuation rates were high for all medications (64–82% after 18 months in phase 1 and 64–84% in phase 2 (Lieberman et al., 2005, Stroup et al., 2006). As with previous discontinuation studies, many clinicians consider the relative doses of medications used in the CATIE study to have been unbalanced thereby raising questions about whether observed differences in discontinuation rates reflect differences in effectiveness in actual clinical practice (Tamminga, 2006). Further observational studies using larger samples in routine clinical practice may be useful in understanding differences in time-to-discontinuation across medications. The objective of this study was to compare time to all-cause-discontinuation across five atypical antipsychotics in a large and representative sample of Medicaid patients diagnosed with schizophrenia. 2. Methods 2.1. Data set The dataset contained all pharmacy and medical service encounter and fee-for-service (FFS) claims from the Maryland Medicaid FFS and HealthChoice programs for recipients prescribed at least one pharmacy claim for aripiprazole, olanzapine, quetiapine, risperidone, or ziprasidone between January 1, 2001 and December 31, 2003. Demographic data included age, sex, race, and duration of Medicaid eligibility across the three-year study period. 2.2. Study population Maryland Medicaid recipients aged 18 to 64 having a claim for schizophrenia (295.xx) any time during the three-year study period were included in the sample if their first (index) prescription for any of the five atypicals (aripiprazole, olanzapine, quetiapine, risperidone, or ziprasidone) occurred between July 1, 2001, and September 30, 2003. Recipients were required to have continuous Medicaid eligibility for a minimum of 6 months prior to their index prescription. Patients with a prescription for more than one atypical antipsychotic were categorized as users of the atypical agent with the earliest index date. Patients taking clozapine were not included in this analysis because use of that agent is limited to treatment-refractory patients, and the blood monitoring required would likely affect persistence on the medication. Age was calculated as of the index date. African American or Caucasian race was identified in order to explore its association with discontinuation. The numbers of recipients of other races were too small to provide sufficient sample sizes for analysis and were excluded. 2.3. Data analysis 2.3.1. Measures 2.3.1.1. Discontinuation Time to discontinuation was calculated as the number of days between the index date and the date of the first gap of more than 14 days between prescriptions. Prescribing gaps were calculated by first adding days' supply to the dispensing date of each prescription to identify the date the prescription should have been exhausted. The first gap in prescribing of 15 days or more was identified as a discontinuation event, even if the patient was switched to another antipsychotic. Other events could have occurred during patient follow-up: disenrollment from Medicaid, psychiatric-related hospitalization1, nonpsychiatric-related hospitalization, or persistence until the end of the follow-up period. Patients who fell into these other event categories were considered censored. The first event occurring after the drug index date determined the recipient's status. 2.3.1.2. Prior health care utilization A variable was created to examine health care utilization prior to the index date as a proxy for the severity of schizophrenia. We counted the number of unique inpatient hospitalization episodes that were associated with a mental health diagnosis (ICD-9-CM =290.xx to 319.xx) during a 180-day pre-index period. 2.3.1.3. Concomitant drug use To account for the concomitant use of psychotropic drugs, we compared individuals who used other psychotropic medications while initiating on the index atypical antipsychotic agent with those who received monotherapy. Concomitant medication use examined ten psychotropic drug classes.2 Psychotropic drug use was considered concurrent with the studied atypical antipsychotic's index date if the prescription for the psychotropic medication began within 30 days of the index date of the atypical, and at least 30 days prior to the atypical agent's end date, to ensure that the two drugs were used concurrently. Use of at least one concurrent psychotropic medication was sufficient to classify the patient as a concurrent psychotropic medication user. 2.3.2. Statistical analysis Descriptive statistics were calculated to compare the distribution of age, gender, race, the occurrence of at least one hospitalization within 180 days prior to the index date, and the use of concurrent medications across the five atypical antipsychotics. Chi-square tests were used to test the differences across medications for categorical variables. Patients were followed for up to 365 days after the initiation of therapy. The survival time for each recipient was calculated as days from the treatment initiation date to the date of discontinuation. Recipients who disenrolled, had psychiatric- or nonpsychiatric-related hospitalizations, or reached the end of the study period before they discontinued antipsychotic treatment were considered as “censored” in the survival analysis because the subsequent prescription data were unobservable. Cox proportional hazard models were run to assess the relative effects of the five atypical antipsychotic agents selected on the hazard ratios of discontinuation, while controlling for other confounding factors. Olanzapine was used as the reference group, and the hazard of discontinuation of therapy on each other atypical antipsychotic was compared to the hazard of discontinuation on olanzapine. Covariates in the models included age, gender, black race, and the occurrence of a mental health hospitalization within 180 days prior to the index date. Adjusted survival curves also were generated using the means of covariates technique (Nieto and Coresh, 1996). 2.3.3. Sensitivity analysis In the primary analysis, where patients may have used multiple atypical antipsychotics, the drug with the first index date was selected as the atypical antipsychotic the patient received. Through sensitivity analyses, we sought to examine whether selecting the atypical agent with the last index date, or selecting all atypicals for that patient, would influence our results. For example, consider a patient who initiates olanzapine therapy and then switches to aripiprazole therapy after 4 months. In the original analysis, the patient is categorized as an olanzapine user and time to olanzapine discontinuation is studied. In the first sensitivity analysis, the patient is categorized as an aripiprazole user and time to discontinuation of aripiprazole is evaluated. In the second sensitivity analysis, both treatment episodes for the patient are included in the analysis, with the patient contributing to both evaluation of time to discontinuation of olanzapine, and time to discontinuation of aripiprazole. 3. Results A total of 5898 patients were included in this study; 1705 were prescribed olanzapine, 1580 risperidone, 1467 quetiapine, 700 ziprasidone, and 446 aripiprazole, which reflects the differences in their market shares at the time of the study. Of the 5898 patients in the study, 50.2% used one or more psychotropic drugs concurrently with the atypical antipsychotic they were prescribed (Table 1). With respect to the nature of concomitant medications, 15.0% of the study population used another atypical antipsychotic, 11.0% used a conventional antipsychotic, 29.2% used an antidepressant, while 31.5% used other psychotropic agents (results not shown). Differences in prior health care utilization, a proxy for severity of illness, and racial differences, existed across the patient groups (Table 1). Only 16.6% of patients prescribed aripiprazole had a previous psychiatric hospitalization, compared to a range of 29.4 to 35.4% of the patients prescribed other atypicals. Consistent with the literature on racial differences in prescribing patterns (Kuno and Rothbard, 2002, Opolka et al., 2004, Mallinger et al., 2006), black patients were less likely to be prescribed the newer agents aripiprazole and ziprasidone (46.4 and 51.3% respectively), compared to olanzapine and risperidone (60.9% and 61.5% respectively). Patients taking aripiprazole were more likely to be taking another psychiatric medication (62.8%) when compared to patients initiating on other agents (44% to 55.2%). Although the groups appeared similar with respect to age, gender, and type of schizophrenia, the differences between groups were statistically significant, likely due to the large sample size (Table 1). These statistical differences in age and gender distributions are unlikely to be clinically important. | a Based on Chi-squared statistic comparing the five drugs. |
After adjusting for differences in baseline characteristics using a Cox proportional hazards regression, patients taking ziprasidone, risperidone, aripiprazole, and olanzapine had an almost identical one-year hazard of discontinuation (HRs compared to olanzapine were: 0.99, p=0.85 for ziprasidone; 0.97, p=0.50 for risperidone; and 1.05, p=0.49 for aripiprazole, Table 2). Patients taking quetiapine had a significantly greater risk of discontinuation than patients taking olanzapine (HR=1.13, p<.01). Other covariates significantly influenced discontinuation. Those who used concurrent medications had a 77.5% lower risk of discontinuation than those who were initiated on monotherapy, and male patients had a 10.1% lower risk of discontinuation than female patients. Older patients had a marginally lower risk of discontinuation than younger patients (HR=0.997). Finally, having a history of psychiatric hospitalization increased the risk of discontinuation by 27.6% compared to having no previous hospitalization. | ⁎ Statistically significant with p<0.05 based on the Wald's Chi-square statistic. |
Survival curves for each atypical antipsychotic after adjustment for covariate differences are shown in Fig. 1, and reflect the relatively higher discontinuation rate for quetiapine, in comparison to the other atypicals examined. A drop in survival is observed in the curves at day 30 as a result of a substantial loss of patients that discontinue therapy at their first refill date. Adjusted continuation rates at 180 and 365 days are shown in Table 3. Low rates of medication continuation at 180 days (ranging from 19.1% to 21.5%) were generally observed for the atypicals included in the analysis, with rates dropping even further towards the end of the one-year period after initiation of drug therapy (declining to the range of 7.4% to 10.7% continuation). The results of the sensitivity analyses were similar to that of the primary analysis. Using olanzapine as the reference category, quetiapine was associated with significantly greater rates of discontinuation when either a patient's last index date or when all index dates were used (HR=1.127, p<.01; HR=1.130, p<0.01, respectively). No other drugs were associated with significantly different rates of discontinuation compared to olanzapine at the α=0.05 significance level. 4. Discussion Results of this study suggest uniformly high rates of discontinuation and relatively short treatment duration for all antipsychotics studied, with no medication being continued for as long as 1 year in the majority of patients. Rates of discontinuation were similar for aripiprazole, olanzapine, risperidone, and ziprasidone in a one-year survival model although quetiapine-treated patients had a somewhat higher rate of discontinuation (13% higher) than olanzapine, the reference drug in the analysis. These results may be helpful in understanding the effectiveness of atypical antipsychotics in clinical practice because they are derived from a large, Medicaid pharmacy claims database with a broad representation of clinicians and patients. The sample permitted meaningful tests of the influence of patient characteristics on medication discontinuation. Several variables significantly predicted medication adherence. In particular, the effect size of concurrent use of psychotropic medications (HR of 0.225) was large in magnitude. To further explore this, we used a more restrictive definition of concurrency in a sensitivity analysis.3 The effect of concurrent medication was equivalent in magnitude and direction across the drug classes we examined. However, whether concurrent medications have a causal relationship on lowering discontinuation rates cannot be determined from this data. Prescribers may add medications to regimens in patients whom they deem more likely to comply with complex therapy, or longer treatment durations may give prescribers more opportunity to increase the complexity of treatment regimens. Prospective randomized studies are needed to determine the effect of concurrent medication on treatment duration and likelihood of discontinuation. In addition to concurrent psychotropic medication use, older age, male gender, and having no prior psychiatric hospitalizations, were associated with significantly lower risk of medication discontinuation. However, due to the small magnitude of the HR for age (0.997), the effect of age on discontinuation is unlikely to be clinically significant. In light of the observed covariate effects, coupled with the lack of difference in discontinuation across medication groups (with the exception of quetiapine), our results suggest that core aspects of schizophrenia patients, rather than the specific antipsychotic agent, may be the overriding determinants of treatment duration. Results of the current analysis are not fully consistent with the recent CATIE schizophrenia study. Although the CATIE study also found high rates of medication discontinuation at 18 months among all medications studied, our adjusted discontinuation rates were higher at one-year follow-up (Table 3). The discontinuation differences observed between our study and CATIE could be attributed to several factors. Our study was observational, reflecting treatment in real world practice settings where clinical resources and levels of patient monitoring may be less than in a randomized control trial such as CATIE. In addition, racial and gender differences between our study sample and the CATIE sample may have contributed to differences in discontinuation rates. There was also a wide variation in median treatment duration across antipsychotic medications in CATIE, ranging from 3.5 to 9.2 months in Phase 1 and 2.8 to 7.0 months in Phase 2. Conversely, the adjusted median treatment durations in the present study did not vary greatly, with a range of 54 to 61 days observed for aripiprazole, olanzapine, risperidone, and ziprasidone. The equivalent treatment duration for the antipsychotic medications included in this study may be due to the fact that treatment doses in a pharmacy prescription claims database reflect doses used by clinicians in routine treatment, in contrast to the mean modal doses used in Phase 1 and Phase 2 of the CATIE study, which differed somewhat from clinical practice (Tamminga, 2006). The current findings are consistent with an analysis of antipsychotic medication persistence using a prescription claims database obtained from Verispan (Lung et al., 2006). Verispan data demonstrated high discontinuation rates for all first-line atypicals used in current practice with no apparent differences in treatment duration. Although olanzapine was shown to be associated with a somewhat greater treatment duration than select agents in other studies, including CATIE, this effect has not been conclusive (Gilbody et al., 2000, Santarlasci and Messori, 2003, Pelagotti et al., 2004, Lieberman et al., 2005). Taken together, these results suggest that no single first-line atypical antipsychotic is distinctly superior with respect to duration of treatment in routine clinical practice. Given the lack of consistent evidence for differential treatment persistence across antipsychotic medications, it would appear that safety and tolerability differences should be a primary consideration in treatment selection, since it is well recognized that antipsychotic medications differ in their safety and tolerability profiles (ADA, APA, AACE, NAASO. Consensus development conference on antipsychotic drugs and obesity and diabetes, 2004). This study adds to the totality of evidence and clinical consensus that medication discontinuation rates are a major problem in schizophrenia therapeutics. The factors leading to medication discontinuation in the schizophrenia population need to be better understood in order to develop more effective ways of managing and maintaining continuous treatment. There are some limitations to this study. Selection bias is always a concern with nonrandomized study designs, and it is possible that, despite our attempt to control for observable factors in our statistical analysis, discontinuation rates were influenced by unobserved differences between drug groups that were not accounted for in the analysis. For example, we were unable to control for the presence of a substance abuse disorder, which can influence adherence to medications. Also, although use of concomitant medications and a history of hospitalization were used as proxies for illness severity, direct assessment of severity of illness was not possible. Daily dosing information was not available in our study, making it difficult to decipher if discontinuation differences were the result of inappropriate dosing. Reasons for discontinuation were also not observed. It would be interesting to know the relative contribution in this population of lack of efficacy, poor tolerability or other factors, to medication discontinuation, and whether these reasons differed among the atypical antipsychotics studied. Furthermore, clinical validation of a schizophrenia diagnosis was not possible using Medicaid claims. Finally, limitations arise as a result of using Medicaid pharmacy claims' refill patterns to determine discontinuation of therapy. Refill patterns do not directly measure compliance or even persistence with certainty. There are alternative operational definitions that have been used to measure compliance with medication, although no single method has proven to be a standard method of choice (Steiner and Prochazka, 1997, Sikka et al., 2005). Further research is needed to replicate these findings in other populations such as the elderly, commercially insured patients, and those with bipolar disorder. 5. Conclusion In a Medicaid population of schizophrenia patients, those initiating treatment on aripiprazole, risperidone, olanzapine, and ziprasidone had similar rates of discontinuation at 1 year; quetiapine-users had higher rates when compared to olanzapine. Further research is needed to examine the specific medication-related and patient-related factors that influence treatment discontinuation in schizophrenia patients. Role of funding source Funding for this study was provided by Pfizer Inc, New York NY, USA. Contributors C. D. Mullins designed the study. Author N. Obeidat managed the literature searches, and authors N. Obeidat and J. Naradzay performed the statistical analyses. Authors C. D. Mullins, N. Obeidat, and J. Naradzay wrote the first draft of the manuscript. All authors listed have contributed to all subsequent drafts, and have approved the final manuscript. Conflict of interest Dr. Mullins is a consultant and Advisory Board member for Pfizer. Drs. Cuffel and Loebel are employees of Pfizer. Acknowledgments The data in this study were provided through cooperation with the Maryland Department of Health and Mental Hygiene. Portions of this work were presented at the International Society for Pharmacoeconomics and Outcomes 11th Annual International Meeting, May 20–24, 2006. References American Diabetes Association, 2004. 1.American Diabetes Association, American Psychiatric Association, American Association of Clinical Endocrinologists, North American Association for the Study of Obesity . Consensus development conference on antipsychotic drugs and obesity and diabetes. Diabetes Care. 2004;27(2):596–601. MEDLINE |
CrossRef
Eaddy et al., 2005. 2.Eaddy M, Grogg A, Locklear J. Assessment of compliance with antipsychotic treatment and resource utilization in a Medicaid population. Clin. Ther. 2005;27(2):263–272. Abstract |
Full-Text PDF (1070 KB)
|
CrossRef
Gilbody et al., 2000. 3.Gilbody SM, Bagnall AM, Duggan L, et al. Risperidone versus other atypical antipsychotic medication for schizophrenia. Cochrane Database Syst. Rev. 2000;3:. Jaffe and Levine, 2003. 4.Jaffe AB, Levine J. Efficacy and effectiveness of first- and second-generation antipsychotics in schizophrenia. J. Clin. Psychiatry. 2003;64(Suppl 17):3–6. Keith and Kane, 2003. 5.Keith SJ, Kane JM. Partial compliance and patient consequences in schizophrenia: our patients can do better. J. Clin. Psychiatry. 2003;64(11):1308–1315. MEDLINE |
CrossRef
Kemmler et al., 2005. 6.Kemmler G, Hummer M, Widschwendter C, et al. Dropout rates in placebo-controlled and active-control clinical trials of antipsychotic drugs: a meta-analysis. Arch. Gen. Psychiatry. 2005;62(12):1305–1312.
CrossRef
Knapp et al., 2004. 7.Knapp M, King D, Pugner K, et al. Non-adherence to antipsychotic medication regimens: associations with resource use and costs. Br. J. Psychiatry. 2004;184:509–516. MEDLINE |
CrossRef
Kuno and Rothbard, 2002. 8.Kuno E, Rothbard AB. Racial disparities in antipsychotic prescription patterns for patients with schizophrenia. Am. J. Psychiatry. 2002;159(4):567–572.
CrossRef
Lacro et al., 2002. 9.Lacro JP, Dunn LB, Dolder CR, et al. Prevalence of and risk factors for medication nonadherence in patients with schizophrenia: a comprehensive review of recent literature. J. Clin. Psychiatry. 2002;63(10):892–909. MEDLINE Lieberman et al., 2005. 10.Lieberman JA, Stroup TS, McEvoy JP, et al. Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N. Engl. J. Med. 2005;353(12):1209–1223.
CrossRef
Lung et al., 2006. 11.Lung C, Loebel AD, Lombardo ID, et al. Treatment Persistence and Antipsychotic Therapy: Results of a Longitudinal Pharmacy Claims Database Analysis. In: Poster presented at the annual meeting of the American Psychiatric Association, Toronto, Canada. 2006;. Mallinger et al., 2006. 12.Mallinger JB, Fisher SG, Brown T, et al. Racial disparities in the use of second-generation antipsychotics for the treatment of schizophrenia. Psychiatr. Serv. 2006;57(1):133–136. MEDLINE |
CrossRef
Naber and Vita, 2004. 13.Naber D, Vita A. Tools for measuring clinical effectiveness. Eur. Neuropsychopharmacol. 2004;14(Suppl 4):S435–S444.
CrossRef
Nieto and Coresh, 1996. 14.Nieto FJ, Coresh J. Adjusting survival curves for confounders: a review and a new method. Am. J. Epidemiol. 1996;143(10):1059–1068. MEDLINE Opolka et al., 2004. 15.Opolka JL, Rascati KL, Brown CM, et al. Ethnicity and prescription patterns for haloperidol, risperidone, and olanzapine. Psychiatr. Serv. 2004;55(2):151–156. MEDLINE |
CrossRef
Pelagotti et al., 2004. 16.Pelagotti F, Santarlasci B, Vacca F, et al. Dropout rates with olanzapine or risperidone: a multi-centre observational study. Eur. J. Clin. Pharmacol. 2004;59(12):905–909. MEDLINE |
CrossRef
Santarlasci and Messori, 2003. 17.Santarlasci B, Messori A. Clinical trial response and dropout rates with olanzapine versus risperidone. Ann. Pharmacother. 2003;37(4):556–563. MEDLINE Sikka et al., 2005. 18.Sikka R, Xia F, Aubert RE. Estimating medication persistency using administrative claims data. Am. J. Manag. Care. 2005;11(7):449–457. MEDLINE Steiner and Prochazka, 1997. 19.Steiner JF, Prochazka AV. The assessment of refill compliance using pharmacy records: methods, validity, and applications. J. Clin. Epidemiol. 1997;50(1):105–116. Abstract |
Full-Text PDF (1273 KB)
|
CrossRef
Stroup et al., 2006. 20.Stroup TS, Lieberman JA, McEvoy JP, et al. Effectiveness of olanzapine, quetiapine, risperidone, and ziprasidone in patients with chronic schizophrenia following discontinuation of a previous atypical antipsychotic. Am. J. Psychiatry. 2006;163(4):611–622.
CrossRef
Sullivan et al., 1995. 21.Sullivan G, Wells KB, Morgenstern H, et al. Identifying modifiable risk factors for rehospitalization: a case-control study of seriously mentally ill persons in Mississippi. Am. J. Psychiatry. 1995;152(12):1749–1756. Svarstad et al., 2001. 22.Svarstad BL, Shireman TI, Sweeney JK. Using drug claims data to assess the relationship of medication adherence with hospitalization and costs. Psychiatr. Serv. 2001;52(6):805–811. MEDLINE |
CrossRef
Tamminga, 2006. 23.Tamminga CA. Practical treatment information for schizophrenia. Am. J. Psychiatry. 2006;163(4):563–565.
CrossRef
Thieda et al., 2003. 24.Thieda P, Beard S, Richter A, et al. An economic review of compliance with medication therapy in the treatment of schizophrenia. Psychiatr. Serv. 2003;54(4):508–516. MEDLINE |
CrossRef
Weiden and Olfson, 1995. 25.Weiden PJ, Olfson M. Cost of relapse in schizophrenia. Schizophr. Bull. 1995;21(3):419–429. MEDLINE Weiden et al., 2004. 26.Weiden PJ, Kozma C, Grogg A, et al. Partial compliance and risk of rehospitalization among California Medicaid patients with schizophrenia. Psychiatr. Serv. 2004;55(8):886–891. MEDLINE |
CrossRef
a University of Maryland School of Pharmacy, 220 Arch Street, Baltimore, MD 21201, United States b Pfizer Inc, New York NY, United States  Corresponding author. Tel.: +1 410 706 0879; fax: +1 410 706 5394.
PII: S0920-9964(07)00191-0 doi:10.1016/j.schres.2007.04.035 © 2007 Elsevier B.V. All rights reserved. | |
|
FULL TEXT
