Sex differences in factors associated with use of medical care and alternative illness behaviors

In many previous investigations of illness behavior the dependent variable has been restricted to use or nonuse of physicians' services. The purpose of this study was to determine how males and females differ in the tendency to respond to their symptoms when self-treatment and lay consultation as well as medical care are considered. The sample consisted of all white, married individuals (n = 532) participating in a household survey who reported at least one symptom episode during a 4-week recall period. An important feature of this survey data set is that there were no proxy respondents. Similar distributions of behavioral responses to symptom episodes were found for men and women. Multinomial logit analysis was used to identify the factors associated with specific illness behaviors. The independent variables were: (1) characteristics of the individual and family such as number of children. social class and employment status; (2) characteristics and perceptions of the symptom episode including discomfort, disability and belief that a physician could do something to relieve it; and (3) access to, and attitudes toward medical care. Belief that a physician could do something to relieve the symptoms, number of days of disability and number of component symptoms in the complex were positively related to use of medical care and duration of the symptom episode was negatively related to use of medical care for both men and women. Several of the independent variables affected only one sex group or had differential effects on males and females. Sex differences were most pronounced with respect to use of lay consultants. Among men, higher social class was associated with use of medical care, whereas, among women, those in higher social classes were more likely to use lay consultants. Other things being equal, women who rated their symptom complexes high on discomfort and seriousness visited a physician, those worried about their symptoms sought lay consultation. An unexpected finding was that the presence of children in the home had a significant effect on men but not women. Employment status, number of children and age of the youngest child did not have significant effects on illness behaviors of men or women These findings should not be generalized beyond similar groups of white, primarily working class adults living in nuclear families in the early seventies. However, this data set provided an opportunity to explore gender differences in the factors associated with alternative illness behaviors including self-treatment and lay consultation in addition to use of medical care. In this sample sex differences in illness behavior were most pronounced with respect to the factors associated with use of the lay network to cope with illness. Further study is warranted.