Self- and physician-rated general health in relation to symptoms and diseases among women

Abstract

Self-rated general health has been demonstrated as an important predictor of future morbidity and mortality, but the determinants of perceived good or poor general health are less clear. The aim of the present study was to investigate self- and physician-rated general health status in relation to self-reported symptoms and diseases (n = 488) as well as to physician-reported health problems based on medical examinations (n = 369) among 43-year-old women representing the general population of a Swedish community. Although most women had good general health as reflected by their own as well as by the physicians' ratings, the physicians rated about twice as many women (12%) to have poor health compared to the women themselves (6%). The agreement between the womens' self-reported general health and the physicians' ratings was significant but low (22%). Bivariate correlations showed that many of the symptoms and diseases in the present sample were significantly associated with both self- and physician-rated general health, but the explained variance was low, varying from 1 to 12%. The exception was for a self-reported sense of powerlessness that explained 20% of the variance in self-rated general health. When taken together and analysed by multivariate logistic regression, it was found that only self-reported symptoms predicted the self-rated general health of the women, contributing to 24% of the variance (p < 0.001). In contrast, both physician-reports of health problems (11%) and self-reports of symptoms (13%), respectively, significantly predicted the physician-rated general health of the women, together contributing to 24% of the variance (p < 0.001). Self-reported diseases did not predict either the self-rated or the physician-rated general health of the women. The findings from the present study support the notion that general health status, when rated by a physician or by the woman herself, is a complex and multidimensional concept that is only partly predicted by symptoms and conditions. Thus, future studies should focus on additional tentative factors in determining self-rated health, as well as on how evaluations of health are communicated to, and interpreted by physicians.