Abstract
Recent research has reported that even though the incidence of Alzheimer’s disease (AD) in the U.S. has been decreasing, prevalence of the disease has been increasing over the past 25 years. At the same time, studies based on nationally representative survey data have reported that the rates of cognitive impairment (CI), the natural outcome of AD, have been on the decline. In this study, we identify the reasons behind this ostensible contradiction through analysis of trends in AD-related epidemiologic characteristics. Using data from the Multiple Cause of Death database and the Health and Retirement Study we found that between 1999–2011 AD-specific mortality rate patterns show an annual percent change (APC) of 1.84% for males and 2.62% for females. The prevalence of AD has been increasing at an APC of 2.46% (3.04% for males and 2.32% for females) up to 2010 after which it begins to level off. Over the same time period, CI as measured by the Telephone Interview for Cognitive Status (TICS) shows that the time-trends in CI are highly sensitive to the threshold used to define CI. Under the usually accepted cut-off score of 8 a decline in the time-trend was detected; however this trend reversed when the cut-off point was varied in either direction. This general pattern held for gender-, race-, and education level-specific analysis. We conclude that there is a high level of heterogeneity in the time-trends of AD and the measures representing CI; this heterogeneity is a strong contributor to the observed contradiction in time-trends.