PREDICTORS AND OUTCOMES OF PSYCHOTROPIC TREATMENT AMONG MEDICAID CHILDREN
Open Access
- Author:
- Taylor, Jessica Pohl
- Graduate Program:
- Health Policy and Administration
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- July 27, 2005
- Committee Members:
- Dr E Michael Foster, Committee Chair/Co-Chair
Karen Linn Bierman, Committee Member
Peter Kemper, Committee Member
Dennis Patrick Scanlon, Committee Member - Keywords:
- Psychotropics
adherence
medication management
predictors
outcomes
Medicaid - Abstract:
- ABSTRACT This dissertation is composed of three related studies of children’s psychotropic treatment among Medicaid recipients. The first study identifies predictors of children’s psychotropic treatment by examining adherence to medication and attendance to medication management visits. The second study examines whether treatment improves children’s behavioral outcomes. The third study examines the effect of treatment on children’s utilization and expenditures of mental health services. The objective of the first study was to examine predictors of psychotropic medication adherence and quarterly medication management among Medicaid children. TennCare claims data were used in combination with survey data and data from the Area Resource File. Claims-based measures of adherence and the receipt of medication management services were used. Descriptive analysis included examining characteristics of the sample and correlation between the dependent variables. Multivariate logistic regression was used to determine which factors influenced the likelihood of a child being adherent or receiving medication management quarterly. The results of this study suggest that the quality of children’s psychotropic treatment in the TennCare program was suboptimal. While approximately one-quarter of children did not receive quarterly medication management, adherence to medication was particularly poor; only one-third of children were adherent. Logistic regression results indicated a child’s prior medication management service use, race, diagnosis, and parental education were significant predictors of adherence. The number of classes of medication a child was prescribed during the study period was the only significant predictor of the likelihood of quarterly medication management receipt. The objective of the second study was to examine the effect of adherence and receiving quarterly medication management on children’s behavioral outcomes. TennCare claims and charge data were used in combination with survey data and data from the Area Resource File. The outcome was measured as the change in a child’s total Child Behavior Check List (CBCL) score from baseline to the end of the study (duration of 18 months). Propensity score analysis was used to control for confounders in estimating the treatment effect of receiving appropriate psychotropic treatment on children’s behavior. The results from study 2 indicated that children who received medication management within three months of a medication claim had better outcomes than children who did not, though not at conventional levels of significance (P=0.07). No relationship was found between adherence to medication and behavioral outcomes. These results indicate the benefit of quarterly monitoring and raise the issue of whether lower doses of medication may be equally effective in the psychotropic treatment of children. The objective of the third study was to examine the treatment effect of receiving quarterly medication management and being adherent on children’s mental health service utilization and expenditures. A retrospective pre-post analysis of children’s interview and administrative Medicaid claims data was conducted. The outcomes were measured as the logged sum of a child’s total mental health-related Medicaid service use and charges. Propensity score analysis was used to control for confounders in estimating the treatment effect of receiving appropriate psychotropic treatment. The results from study 3 indicate that quarterly receipt of medication management resulted in higher utilization of mental health services and charges in the post-study period. While no relationship was found between adherence and utilization of mental health services, adherence was positively related to charges but not at conventional levels of significance (P=0.06). The provision of quarterly medication management visits resulted in moderately higher mental health services utilization and charges. As a result, the potential benefits of improved outcomes may outweigh the additional cost of increasing the frequency of follow-up; previous analyses suggested that quarterly medication management visits resulted in better behavioral outcomes. Optimizing dosing through frequent medication management visits may enable providers to achieve better patient outcomes at lower dosing, and ultimately produce cost savings. In conclusion, the contribution of this research to the literature was to address questions regarding children’s psychotropic treatment that were previously unanswered. The financial burden of covering costly psychotropic drugs to treat children’s mental illness is a threat to the long-term viability of the Medicaid program. As a result, Medicaid policy makers and providers will continue to be challenged to provide quality services in a cost-effective manner. These results emphasize that better outcomes can be achieved at moderately higher costs.