The recognition of family caregivers and the consequence of caregiving for Alzheimer's disease (AD) patients have resulted in some measurement tools for caregiver burden and fulfillment. The breadth of the caregiving experience and burden demands reliable, comprehensive and validated measurement tools suited to the complicated and dynamic patient-caregiver relationships. Only then can they be reliably assessed and so that caregivers be appropriately assisted in their provision of care and protected from complications of excess stress. This type of tool can be useful to assess a caregiver's situation by clinicians and others in early detection efforts to screen for caregivers at risk and to direct proper and timely interventions. The proposed project will utilize modern advanced psychometric methods, i.e., item response theory (IRT), to evaluate the core battery of measures used in the Resources for Enhancing Alzheimer's Caregiver Health (REACH) initiative. A primary strength of this National Institute on Aging (NIA) and the National Institute on Nursing Research (NINR) sponsored initiative is the large (N = 1,222) racially and ethnically diverse sample of caregivers (non-Hispanic White, non-Hispanic Black, and Hispanic). Given the limited availability of caregiver health assessment tools that are sensitive to the needs of linguistically, culturally, and educationally diverse caregivers, this proposed project will focus on language and culture aspects. Our two specific aims are: 1) to test the psychometric properties of the REACH core battery of measures and to examine the effect of language and culture on item response; and 2) to compare the measurement equivalence of the socio-demographic and clinical groups of caregivers on the core battery of measures in terms of other psychometric characteristics, and to explore the sensitivity to clinical change over time. In summary, the purpose of this secondary data analysis project is twofold. The first purpose is to perform IRT-based item analyses of the REACH core battery of instruments in order to identify items that show measurement bias. Knowledge about such DIP impacts will provide a basis for more effective use of the instruments in culturally diverse AD caregiver populations. The second purpose is to examine the effects on scale-level statistical properties of maintaining or removing items with significant differential item functioning (DIP). The completion of this study may provide useful data, on the basis of a large population-based sample on various scale-level measures. ' ? ? ?
