Abstract
The National Dementia Workforce Study was designed to improve our understanding of the individuals and systems who care for people with dementia, but designing and implementing such a study is challenging due to the large number of patient care organizations, clinical and direct care roles, and locations in which care is provided. Specifically, developing a probability sample of organizations and staff caring for people with dementia is a complex and difficult process. While there are national sampling frames available for federally certified nursing homes (i.e., via data from the Center for Medicare and Medicaid Services), there are no national sampling frames for assisted living communities or home care agencies. The latter frames must be developed through querying state-level regulatory agencies and through other, supplemental strategies such as working with professional organizations, large employers, and organizations that provide services (e.g., payroll services) to this sector. Further, since there are no national sampling frames that allow for direct sampling of staff working in any of these types of organizations, we opted for a two-stage design. In the first stage, organizations are identified, sampled, recruited to participate in an organizational-level survey, and asked to provide a roster of eligible staff. In the second stage, individual staff members are recruited for a staff-level survey. We describe the plan for sampling and recruitment procedures to be used in each stage and discuss limitations, including implications for coverage of the target population. Data collected through these surveys will be available to the research community.