Background: Those who must make health care decisions on behalf of persons who lack decision-making capacity often have too little information. The purpose of this study was to determine whether and to what degree sociodemographic factors, social support, religious involvement, and functional status served as predictors of the health-related values and preferences of geriatric patients.
Methods: A retrospective chart review involving 178 cognitively intact patients enrolled at a geriatric clinic at a university medical center was conducted. Patient variables included age, race, sex, marital status, years of education, participation in formal religious activities, self-reported functional status, and adequacy of social and family support. These were compared with responses to a modified version of the Values History questionnaire, an established method of evaluating patient values that includes advance directives.
Results: Patients were likely to value quality (82%) over quantity of life (18%) and the ability to think clearly (64%) over 13 other specific health-related values. They were likely (93%) to want to be taken to a hospital emergency department on losing consciousness or becoming confused. Seventy-eight percent expressed a desire to be resuscitated using cardiopulmonary resuscitation (CPR), and 76% expressed a preference for use of a respirator, if necessary. Most (85%) considered a permanent vegetative state to be worse than death. High functional status predicted a preference for quality of life over length of life and the perception of a persistent vegetative state as worse than death (positive predictive value = 89%). Persons with better social and family support were more likely to accept treatment with CPR or a respirator and less likely to consider a permanent vegetative state to be worse than death. No single patient factor was strong enough to increase the probability of a particular value or preference by more than 17% above baseline.
Conclusions: Sociodemographic and functional status variables are relatively weak predictors of personal values and directives. This reinforces the importance of routinely eliciting patient values and preferences and of updating the information, particularly following changes in functional status or family support. Baseline information regarding the health-related values and preferences of this primary care geriatric clinic population may provide valuable information about the values and preferences of decisionally impaired older patients.