A latent variable approach to studying risk factors for Alzheimer's disease

The objective of this study is to investigate and to improve measurements for the study of suspected animal and occupational exposure risk factors for Alzheimer's disease (AD). A subsample (n = 200) of a longitudinal study sample from research on risk factors for Alzheimer's disease was selected by systematic probability sampling. Participants were men born between 1917 and 1927, selected from the National Academy of Sciences-National Research Council (NAS-NRC) Registry of white male twins who served in the United States Armed Forces during the World War II Korean Conflict era. A questionnaire was administered via telephone interview. A total of 25 animal exposure items and 28 occupational exposure items were selected from the risk factor questionnaire administered to the men. Measurement for animal or occupational exposure was ascertained based upon the participants' responses to the survey questions. Latent variable analyses extracted 3 animal exposure factor scales: pet, farm/livestock and exposure to wild animals (e.g, while hunting). The occupational exposure data led to extraction of 2 factor scales: medical exposures and manual work involving metal. Latent variable analysis revealed that many items (potential risk factors) are highly correlated with each other and can be conceptualized as correlated observable manifestations of more unitary underlying constructs. In future research, it may be advantageous to assess correlated items such as these in relation to an underlying dimension of exposure, rather than as discrete exposures. Copyright (C) 2001 S. Karger AG, Basel.