Progress in studying the relationship between religion and health has been hampered by the absence of an adequate measure of religiousness and spirituality. This article reports on the conceptual and empirical development of an instrument to measure religiousness and spirituality, intended explicitly for studies of health. It is multidimensional to allow investigation of multiple possible mechanisms of effect, brief enough to be included in clinical or epidemiological surveys, inclusive of both traditional religiousness and noninstitutionally based spirituality, and appropriate for diverse Judeo-Christian populations. The measure may be particularly useful for studies of health in elderly populations in which religious involvement is higher. The measure was tested in the nationally representative 1998 General Social Survey ( N = 1,445). Nine dimensions have indices with moderate-to-good internal consistency, and there are three single-item domains. Analysis by age and sex shows that elderly respondents report higher levels of religiousness in virtually every domain of the measure.