| he Well-Elderly Study was a randomized clinical trial to evaluate the effectiveness of occupational therapy on healthy elderly people. There were three experimental conditions in this study: receiving occupational therapy treatment, participating in social activity group, and remaining untreated. Early results from this study indicated statistically significant benefits in many outcome measures attributable to the occupational therapy intervention when compared to the combined control group. Based on those findings, we have continued to investigate the cost-effectiveness of occupational therapy.;Theoretically, the ratio of marginal effects of treatment indicator variable on health status and health care costs is the incremental cost-effectiveness ratio. To estimate the marginal effects, we constructed a statistical model. The variables of health care costs and health status interact with each other. Ignoring such endogeneity would lead to a specification bias. In addition, there are difficulties in modeling panel data in our study. The simultaneous error-component model in this paper is developed to correct the endogeneity bias and to effectively model panel data. We employed the error-component three-stage least squares (EC3SLS) estimation procedure to obtain consistent and most efficient estimates. The two endogenous variables for the model are monthly global rating scale for health status and monthly health care costs.;Due to empirical problems of our data set, such as a small sample size, missing data, short intervention and follow-up periods, and drawbacks of the global rating scale, insensitivity and being unrepresentative, the results of the EC3SLS are inclusive. Using the SF-36 scores, we further conducted a cost-effectiveness analysis with simple covariates adjustment. Results based on the SF-36 scores clearly indicated that occupational therapy treatment is very cost-effective when compared to social intervention (ICER is... |
