Variation In Use of $4 Generic Program And Potential Savings Among Medicare Beneficiaries---Biostatistics Student's Internship Exit Report

As an option to fulfill the MS thesis requirement at the Department of Biostatistics, I worked as an intern under the supervision of Dr. Yuting Zhang at the Department of Health Policy & Management, Graduate School of Public Health, University of Pittsburgh, from January to June 2010. During the internship, I have been fully involved in some of Dr. Zhang's projects and have made the following contributions. First, I consolidated different pharmacy event data and medical claims data obtained from multiple sources into several analytic databases for those projects. The end products in this step included the analytic datasets, data dictionary for each corresponding dataset, and the SAS programming codes. After completion of the dataset construction, I had opportunities to fully apply the statistical skills I have learned during my coursework on a specific project, entitled "Variation in the use of $4 generic prescription and potential savings among Medicare beneficiaries." Under the supervision of Dr. Zhang as well as collaborating with other colleagues, I played the leading role in data analysis, the interpretation of results and writing of a manuscript for publication.Public Health Relevance: Our research on these projects focused on evaluating the strengths and weaknesses of the Medicare prescription drug program, especially its effects on vulnerable American populations such as under-served minorities, patients with severe mental health and multiple medical conditions. Through our research, public policy might be improved to eliminate health disparities in populations. Our findings from the project have important policy implications for optimizing cost-effective use of prescription plans to the public.Through this half-year long internship, I have had great opportunities to learn study design, data management, statistical analysis and hypothesis testing in a real world setting, to apply statistics/econometrics knowledge to large existing data, to evaluate the effects of health care policy and interventions on medical spending and health outcomes. In addition, I have practiced advanced SAS programming skills in manipulating the large datasets