Characteristics of Nursing Home Residents with Item Nonresponse in their Prescribed Medicine Data: Analyzing the 1996 Medical Expenditure Panel Survey Nursing Home Component

• Activity Number: 74
• Type: Contributed
• Date/Time: Monday, August 12, 2002 : 8:30 AM to 10:20 AM
• Sponsor: Section on Survey Research Methods*
• Author(s): Denys Lau*+ and D.E.B. Potter and Francesca Dominici
• Affiliation(s): Agency for Healthcare Research and Quality and Agency for Healthcare Research and Quality and Bloomberg School of Public Health, Johns Hopkins University
• Keywords: MEPS ; item non-response ; nursing home ; prescribed medicine

Abstract

The 1996 Medical Expenditure Panel Survey Nursing Home Component (MEPS NHC) was designed to provide unbiased national and regional estimates of the NH use and expenditures of all NH users in the U.S. at any time in 1996. Data were collected on facility and resident characteristics, and health services use while residents were in NHs. Prescribed medicine (Rx) use data were obtained from administration records and included Rx name, form, strength, dosage, and frequency of administration according to each calendar month when residents were in an NH. In this paper, the characteristics of residents who had item nonresponse in their Rx data are compared to those who had non-missing Rx data. Detailed information on the socio-demographic and health factors of the residents, their length of stay, and characteristics of NHs where they stayed in 1996 is used to inform the multivariate analyses. When appropriate, therapeutic classes of Rx taken are also considered. The paper discusses implications of study findings on strategies for estimating Rx use among NH residents adjusting for item nonresponse.