The effect of adjunctive mood stabilizers on antipsychotic utilization pattern and health resource utilization for Medicaid enrollees with schizophrenia

Background: Prescribing adjunctive mood stabilizers to manage schizophrenia is prevalent, despite the lack of substantial evidence to support the long-term use of this treatment regimen. Objective: The objective of this study was to assess the impact of using adjunctive mood stabilizers on antipsychotic utilization, total health expenditures, inpatient hospitalizations, long-term care stays, and emergency room (ER) visits for patients with schizophrenia. Methods: Georgia Medicaid claims from 1999 through 2001 were analyzed to identify recipients diagnosed with schizophrenia (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM]: 295.XX). The treatment groups consisted of subjects who received combination therapy of mood stabilizers and antipsychotics (including both atypical and typical medications), while the comparison group consisted of subjects who were on antipsychotic medications without exposure to the mood stabilizers under investigation. Four treatment groups (valproate, lithium, carbamazepine, and combination mood stabilizer therapy) were formed based on the mood stabilizers patient received. Differences in annual health care use and expenditures were estimated between propensity score matched treatment and comparison groups controlling for comorbidity, prior utilization, demographic, and health provider specialty. Results: During the 1-year observation period, subjects in treatment groups filled an average of 200-days supply of adjunctive mood stabilizers. These adjunctive mood stabilizer recipients had significantly longer antipsychotic treatment durations than the subjects who did not have exposure to mood stabilizers (valproate + antipsychotic vs. antipsychotic only, net difference: 56.47 days, p < 0.0001; lithium + antipsychotic vs. antipsychotic only, net difference: 90.25 days, p < 0.0001; carbarnazepine + antipsychotic vs. antipsychotic only, net difference: 41.27 days, p = 0.0439; multiple mood stabilizers + antipsychotic vs. antipsychotic only, net difference: 83.14 days, p < 0.0001). The intensive pharmacotherapy associated with treatment groups resulted in $900-$1300 higher pharmacy costs than the comparison groups (valproate + antipsychotic vs. antipsychotic only, net difference: $1218.43, p < 0.0001; lithium + antipsychotic vs. antipsychotic only, net difference: $985.79, p = 0.0015; carbarnazepine + antipsychotic vs. antipsychotic only, net difference: $911.63, p = 0.0497; multiple mood stabilizers + antipsychotic vs. antipsychotic only, net difference: $1281.91, p < 0.0047). However, there were no statistically significant differences for total health expenditures, hospitalizations, emergency room visits, and nursing home admissions between propensity-matched treatment and control groups. Conclusions: There were no differences in health care costs or utilization of ER, long-term care, and inpatient services between schizophrenia patients who did and did not receive adjunctive mood stabilizer; however, longer antipsychotic treatment durations were observed in patients receiving adjunctive mood stabilizers. Interpretation of these results is limited by the unknown selection bias between the treatment and the comparison groups and the relatively small number of patients in some treatment groups. The development of a better-controlled study to further evaluate this treatment regimen is warranted.