Change in Self-Rated Health and Mortality among the U.S. Elderly

Eric M. Vogelsang, University of Wisconsin-Madison

Existing research that identifies associations between self-rated health (SRH) and mortality often operationalize SRH as a simple two- or five-category indicator of general health status. This may be problematic since there are strong theoretical reasons to believe that both observed and perceived SRH changes may explain or modify associations between SRH and death. Using the oldest-old portion of the Health and Retirement Study over 13 years, I evaluate two different measures of dynamic SRH—reported SRH change and computed SRH change. My findings suggest that omitting trajectory measures may result in spurious associations between SRH and mortality. I also found that a retrospectively reported health decline was independently associated with a 10% greater hazard of mortality. Lastly, my results indicate that oldest seniors with more volatile health may be at a greater risk of death than those whose health remains relatively stable.