Prediction of survival: a comparison between two subjective health measures in an elderly population.

A large amount of evidence shows that the subjective evaluation of health is a predictor of survival in many different populations. Subjective health (SH) is measured using different types of measures such as a general evaluation of health or a comparative evaluation of health. The aim of this study was to compare the prediction of survival by two measures of SH (a general measure and an age-related measure) and evaluate the association with other variables in an elderly population. The baseline survey was conducted during 1994, covering 1138 men and women aged over 70. The survival status was ascertained 7 years later. After adjustment for age, sex, education, marital status, perceived socioeconomic status and presence of diseases the two SH measures were found to be predictors of mortality, but only in men. In men, there was no significant difference between the two types of SH measures in their prediction of mortality. Also in men, when there was only one or no disease, being married had a protective effect compared with not being married when both types of SH measures were used. In elderly women, the association between the two types of SH and survival diminished after adjusting for the various variables. However, the general SH measure may be the preferable measure to use when needed. Education in women was associated with mortality only via the age-related SH measure.