Estimating Continuous Long-Term Daily Health Behavior Using Missing by Design Methods with Application for Medication Adherence of People Living with HIV

• Activity Number: 379 - Understanding HIV/AIDS Epidemics with Newly Available Data Sources
• Type: Invited
• Date/Time: Tuesday, August 1, 2017 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Ofer Harel*
• Companies: University of Connecticut
• Keywords: Incomplete data ; missing by design ; HIV ; adherence ; multiple imputation

Abstract

Daily diary methods have been used to study a variety of emotions, cognitions, health symptoms, and health behaviors. Daily diary methods allow for relatively easy repeated measures of experiences and behaviors and can reduce recall bias in the reporting of these. Time-stamped electronic data collection methods, such as the use of daily data collection via smart phones or text messaging have become an optimal choice for several research settings. However, it is complicated to design long instruments and repeat the data collection for long periods of times. These complications arise mostly due to cost and patients burden. In this project we use missing by design approach to reduce the number of days each participant has to respond, while estimating the daily relationship between food insecurity and medication adherence among people living with HIV.