OBJECTIVES:  Elderly people with multiple chronic conditions usually take a substantial number of medication every year. Thus in order to decrease the out-of-pocket spending on prescription drug, they may choose generic alternatives over the branded drug since generics cost less. The objective of this project aims to characterize the association of number of chronic conditions and generic utilization among elderly people in the United States. METHODS:  We used 20% Medicare data of 2012, which included the Prescription Drug Event file, and other files containing information about the beneficiary and their health status. The outcome is defined as the generic drug share of the total drug use at the person-year level. We used IMS drug classification to categorize claims of generic and branded drug. The main variable of interest is the number of chronic conditions of the beneficiary, which was defined as the total number of different chronic diseases according to Chronic Condition Warehouse definition. Other covariates include age, gender, out-of-pocket spending for drug during the year of the beneficiary, plan type that the beneficiary was enrolled in, and whether the beneficiary received any subsidy for health care and drug use. We ran a multivariate linear regression model to examine the association. RESULTS:

• The mean age of the sample population is 69.7 years old and on average, each beneficiary had more than 6 chronic conditions and about 72% of the total claims in the year was generic drug.
• The coefficient on comorbidity is 0.008 and it’s statistically significant, indicating that with one more chronic conditions, the use of generic drug may increase by 0.8%.

CONCLUSIONS: Elderly people with more comorbidity may tend to use more generic drug in order to save on the out-of-pocket spending.