Join ISPOR  | Sign up for mailing List  | Follow Us: LinkedIn Twitter Facebook YouTube
The Official News & Technical Journal Of The International Society For Pharmacoeconomics And Outcomes Research
Outcomes Assessment

Long-Acting Antipsychotic Medications and Adherence in Patients with Schizophrenia

Dawn N. Kim-Romo, PharmD, Haesuk Park, MS, Karen L. Rascati, PhD, Health Outcomes & Pharmacy Practice, The University of Texas at Austin, Austin, TX, USA

Medication Use and Adherence of Antipsychotics in Patients with Schizophrenia
Schizophrenia is a chronic and complex disorder typified by symptoms of hallucinations, delusions, and disorganized speech and behavior, and it has been shown to cause dysfunction with interpersonal relationships, self-care, work, and education. Characteristic symptoms of schizophrenia, which are differentiated into positive and negative symptoms, represent important diagnostic criteria for this disorder. While the onset of schizophrenia is typically around the late teens to the mid-30s, late-onset schizophrenia (after 45 years of age) also occurs [1].

Pharmacological treatment is recommended as a mainstay therapy for schizophrenia, especially in patients with first-episode schizophrenia, where longer periods of active psychosis before pharmacologic treatment are linked with worse clinical outcomes for as long as five years post treatment initiation [2,3]. Achievement of the long-term goals of treatment, which include minimization of relapse and re-hospitalization, is attributed to maintaining medication adherence. Second-generation antipsychotic medications (e.g., risperidone) are generally preferred over first-generation medications (e.g., haloperidol), due to a lower probability of some side effects, such as extrapyramidal side effects [2]. Poor rates of adherence to antispychotics have been cited, and one review study found that non-adherence to antipsychotic medications ranged from 4% to 72% with an average non-adherence rate of 40.5% (SD 18.5%) [4]. The need for increased treatment adherence in mental health patients has also been addressed by an executive summary from the Substance Abuse and Mental Health Services Administration and the National Institutes of Health [5].

Comparing Adherence of Patients Taking Short-Acting Versus Long-acting Antipsychotics
For patients who suffer from recurrent relapses due to partial or complete non-adherence, long-acting injectable antipsychotics are recommended. During the acute phase, patients may transition from oral to long-acting injectable antipsychotic formulations [2]. Risperidone was the first long-acting injectable second-generation antipsychotic available for the treatment of schizophrenia [2,6].

Studies have reported increased adherence with long-acting injectable risperidone over oral risperidone in first-episode schizophrenic patients [7,8]. Our study differs from earlier studies in that a retrospective database analysis was conducted to compare oral risperidone versus long-acting injectable risperidone medication adherence rates using three different measures: medication possession ratio (MPR), proportion of days covered (PDC), and persistence before a 30-day gap. Specifically, prescription claims of Texas Medicaid patients with schizophrenia were assessed, and statistical analyses involved the use of logistic regression for MPR and PDC and multiple regression for persistence. In general, MPR is defined as the sum of the days’ supply of the drug divided by the total number of days in the post-index period [9]. PDC is operationalized as the number of days with the drug available divided by the number of days in the post-index period. PDC has been found to be a more conservative calculation of adherence in patients with schizophrenia [10].

Data from 1470 Texas Medicaid patients were analyzed, and two cohorts were compared: 1) those with only oral risperidone prescriptions [oral cohort; N = 1338], and 2) those with any injectable risperidone prescriptions, where patients could also have oral risperidone prescriptions [injectable cohort: N = 132]. Females represented 46% and 34% of patients in the oral cohort and injectable cohort, respectively, and the average age for both cohorts was 37 years (SD 14 years). Based on MPR, 30% (405/1338) were adherent in the oral cohort and 43% (57/132) were adherent in the injectable cohort. For PDC, 26% (354/1338) were adherent in the oral cohort and 39% (52/132) in the injectable cohort were adherent.

A logistic regression analysis comparing MPR results, controlling for age, gender, and race, showed that patients in the injectable cohort were 77% significantly more likely to be adherent – defined as an MPR ≥ 80 % – (OR=1.77, 95% CI=1.22-2.56, p<0.05) compared to the oral cohort. Also, a logistic regression analysis comparing PDC results, controlling for the same demographic covariates, showed that patients in the injectable cohort were about 81% significantly more adherent – defined as PDC ≥ 80 % – than the oral cohort (OR=1.81, 95% CI=1.24-2.63, p<0.05). A multiple regression analysis was run in order to compare medication persistence. After controlling for age, gender, and race, mean persistence before a 30-day medication gap was 169.2 days in the oral cohort and 159.1 days in the injectable cohort; however, this was not a significant difference between the 2 cohorts (p>0.05).

Limitations Associated with Retrospective Claims Data
Some limitations should be considered when interpreting these findings. Firstly, the only population assessed was Texas Medicaid patients with schizophrenia; therefore, the generalizeability of the findings may be restricted. Secondly, a small sample size of patients in the injectable cohort was analyzed. Thirdly, as with most retrospective claims studies, it is not known whether study patients actually utilized their filled prescriptions. Finally, “selection bias” is an important limitation in retrospective studies. Selection bias occurs when there is an unequal distribution of clinical factors among cohorts that lead to certain study outcomes [11]. Although known differences (e.g., demographics) were statistically controlled for, other clinical factors are considered when prescribers determine the form of medication for their patients.

This study assessed medication adherence with oral versus long-acting injectable risperidone and found that Texas Medicaid patients using long-acting injectable risperidone were significantly more likely to be adherent based on MPR and PDC. On the other hand, it is important to note that the statistical analyses of persistence showed no significant differences between the oral and injectable cohorts, despite results to the contrary found in the literature. A South Korean study defined non-adherence as a medication gap of ≥ one week and found that patients taking long-acting injectable risperidone had significantly higher adherence rates at year-one and year-two compared to patients taking oral risperidone [7]. Also, a US study, which assessed non-adherence at a medication gap of ≥ 14 days, reported that the long-acting injectable risperidone group was significantly more adherent than the oral atypical antipsychotic group (89% vs. 59%) in the set of patients “as actually treated” by week 12 [8]. Therefore, data from larger databases with more complete clinical data may better address the medication adherence of patients who utilize oral risperidone solely versus those who utilize long-acting injectable risperidone alone or in conjunction with oral risperidone.

  Issues Index | 2012 Issues Index  

Contact ISPOR @  |  View Legal Disclaimer
©2014 International Society for Pharmacoeconomics and Outcomes Research. All rights reserved under International and Pan-American Copyright Conventions. 
Website design by Eagle Systems USA, Inc.