This study aims to measure the validity and reliability of patient preferences for life-sustaining treatment and states near to or worse than death. This information is needed to assess the potential validity and reliability of advance directives for medical care, such as living wills. Specifically, we will identify the medical interventions, and health and social states which form the domain of content for advance directives, as well as the factors which influence the acceptability of advance directives. We will also evaluate the construct validity test- retest reliability, and internal consistency of patient treatment preferences, and the ability of patients to predict their preferences for states and interventions they have not previously experienced (predictive validity). This study has three phases. In Phase I, 50 patients, 10 well-adults, 15 physicians, and 15 nurses will receive a semi- structured interview to identify the relevant content for advance directives and to pretest methods for Phase II. In Phase II, 250 patients and 100 well adults will be administered a structured interview to scale their preferences for states and medical interventions, and evaluate the construct validity, predictive validity, and internal consistency of their preferences. In Phase III, patients, and well-adults will be reinterviewed two weeks and six months following their enrollment in Phase II to evaluate the test- retest reliability and predictive validity, respectively, of their preferences. Patients will include persons with acquired immunodeficiency syndrome, terminal illness, elderly with chronic disease, cognitively impaired, and nursing home residents. Physicians will be selected from a variety of primary care and subspecialty practices. Nurses will be selected from intensive care, inpatient, outpatient, home health, and nursing home practices. These results will be directly applicable to the development of valid, efficacious, and effective advance care directives. In addition, they will provide an improved basis for understanding the validity and reliability of preferences contained in advance directives.
