We elicit numerical expectations for late-onset dementia in the Health and Retirement Study. Our elicitation distinguishes between precise and imprecise probabilities, while accounting for rounding of reports. Respondents quantify imprecision using probability intervals. Nearly half of respondents hold imprecise dementia probabilities, while almost a third of precise-probability respondents round their reports. We provide the first empirical evidence on dementia-risk perceptions among dementia-free older Americans and novel evidence about imprecise probabilities in a nationally-representative sample. We show, in a specific framework, that failing to account for imprecise or rounded probabilities can yield incorrect predictions of long-term care insurance purchase decisions.currently incurable, that affects millions of Americans every year. In 2018, 5 to 6 million Americans were estimated to live with ADRD, the large majority over 65 years old (Alzheimer's Association, 2018). 6 ADRD is one of the top-10 causes of death in the United States. In 2017, it ranked 6 th overall, 5 th among the over 65, and 3 rd among the over 85 (Heron, 2019). Furthermore, ADRD has been estimated to impose a high economic burden on households and government programs due to the associated need for prolonged specialized care by the sick (Hurd et al., 2013).Important economic decisions of older individuals can potentially be affected by their perceptions of dementia risk, including precautionary savings choices, the timing of retirement, and the purchase of long-term care insurance.Producing reliable forecasts of the future prevalence of dementia has proved challenging, as forecasts are sensitive to the historical data and modelling assumptions used (Hudomiet, Hurd and Rohwedder, 2018). Nevertheless, most forecasts agree that the prevalence of the illness will grow with the increasing size and proportion of the U.S. population over 65, also due to improved rates of early detection, treatment, and prevention of major causes of death other than dementia. If this is correct, the already high demand for LTC services such as nursing homes will rise.
Current statistics indicate that over 40% of residents in long-term care facilities haveAlzheimer's Disease and over 70% have some cognitive condition (National Center for Health Statistics, 2013). Recent economic literature using the HRS has found a strong link between the decision to buy LTC insurance and people's subjective probabilities of moving to a nursing home (e.g., Finkelstein and McGarry (2006)). However, to the best of our knowledge, nothing is known about people's beliefs of developing dementia as they age, or about how these beliefs are related to LTC decisions and outcomes. This data gap motivated our choice of measuring respondents' subjective probabilities of developing dementia in the future.When thinking about elicitation of perceptions for dementia risk, it is important to recognize that, in general, available research on the prevalence of dementia does not provide evidence on th...