Dementia Prevalence over Time: Proximate Causes and Consequences

点击翻译按钮获取中文摘要

The goals of the R01 grant “Dementia Prevalence over Time: Proximate Causes and Consequences,” include quantifying several consequences of Alzheimer’s Disease and Alzheimer’s Disease Related Dementias (AD/ADRD) in the older population. Specific Aim 5 of the R01 is to “Forecast the use and availability of informal care and how it will impact nursing home use and costs.” To conduct this research, we had planned to rely on the RAND HRS Family Data. This data set, which cleans and streamlines HRS information on respondents’ children, parents, siblings, and caregivers, enables analyses of caring for persons with AD/ADRD from two angles: when an HRS respondent receives care for AD/ADRD; and when an HRS respondent provides care to parents with AD/ADRD. Unfortunately, and unknown to us at the time of our original R01 grant application, funding support for the RAND HRS Family Data files ceased after incorporation of the 2014 HRS data. Hence, we will have to clean and process HRS 2016 and HRS 2018 family data to accomplish fully Aim 5. These data are critical for forecasting future AD/ADRD effects as proposed in aim 5 of the R01 for two reasons. First, using more recent data will make forecasts of use and availability of informal care more accurate. Second, the more recent waves of the HRS family data include the first waves of the Late Baby Boomer (1960-1965) cohort, as well as a substantial amount of additional information on older cohorts. Processing these data will be a major effort given the more than 300 raw variables per wave that go into the RAND HRS Family Data, as well as the complexities of linking children’s records longitudinally. Prior to 2002, IDs for respondents’ children were not unique, leading to difficulties linking the information from earlier waves as well as difficulties in linking later records with unique IDs to difficult-to-link data from earlier waves. While it is a complex and substantial operation to process the 2016 and 2018 data, we will be able to build on the existing programs from prior releases of the RAND HRS Family Data. Dr. Rohwedder, investigator on the R01 and director of the RAND HRS Data development efforts, will work with a team of experienced programmers to accomplish this effort. We will examine and improve cross-wave linkages of children’s records. Working with available data, we identified situations where we can augment the longitudinal matching algorithms. This is important for obtaining reliable estimates of who is likely to become a caregiver, particularly given that such duties typically fall to just one or very few family members. To enhance the transparency and reproducibility of our research, we will document our work in the RAND HRS Family Data codebook, including any cross-wave differences we find in the 2016 and 2018 data resulting from changes to the HRS survey instrument or in how data are recorded. We will make the augmented RAND HRS Family Data available to the research community. The RAND HRS Family Data have received 6,800 downloads since first being made available in 2011. The addition of these latest waves will help leverage additional research on the provision of care for those living with AD/ADRD.