Effects of antipsychotic medication on psychiatric service utilization and cost.

BACKGROUND: Based on randomized clinical trials, consensus has been emerging that the first line of treatment for individuals with psychotic disorders should be the newer atypical or second generation antipsychotic medications rather than the older neuroleptics. Given that acquisition costs of atypical antipsychotics are generally higher than typical antipsychotics, uncertainty exists whether the newer atypicals are cost effective alternatives when used in ordinary practice settings.

AIMS OF THE STUDY: The introduction of newer atypical antipsychotic agents has prompted evaluation of their overall effectiveness in reducing health care costs given their higher acquisition costs. This paper focuses on the effects of differing classes of atypical versus typical antipsychotic medications on psychiatric service utilization and cost for persons with serious mental illness treated in usual practice settings.

METHODS: Descriptive statistics are used to compare patient characteristics, service rates and costs across psychotropic medication groups. Prediction equations employing ordinary least squares regression models are used to explain variation in cost due to pharmacy group membership controlling for demographics, clinical diagnoses and symptoms. Subjects were 338 Medicaid clients with serious mental illness from Florida, Pennsylvania and Oregon treated in ordinary clinical settings. Resource utilization and costs were operationalized using administrative databases to measure consumption of treatment services and pharmaceuticals for a six month period.

RESULTS: Inpatient service use was significantly higher for individuals on atypical only and combination atypical/typical medications compared to those on typical medications only, whereas outpatient use was highest for those on typicals. Furthermore, six-month costs for both pharmacy and psychiatric services were significantly greater for persons in the atypical only (USD 6528) and combination typical/atypical groups (USD 6589) compared to those on typicals only (USD 3463). There were still significantly higher costs associated with atypical only and the combination typical/atypical users after multivariate controls were used.

DISCUSSION: This study showed that Medicaid clients in community settings using atypical only and typical/atypical combination medications had the highest costs both in pharmacy and service use when compared to those on typical only medications. However, this study design does not allow us to ascribe a causal relationship between medication group and service costs. Given that olanzapine was the most recent medication in the compendium of available drugs at the time of this study, it is possible that those in the olanzapine only group were failing on other drugs. Caution must be used in drawing policy implications regarding cost effectiveness of newer medications since individuals who are getting the newer atypical or combination medications in community mental health center settings may be unstable on the older medications.

IMPLICATIONS FOR FUTURE RESEARCH: A longer follow-up period is needed to determine if the cohort remaining on current atypical medications stabilize over time while those taking the newest drug on the market become the most costly population.