Person variables, psychosocial state variables, and reported health behaviors : relationship to preterm delivery

Abstract

The purpose of this study was to use a conceptual model to examine selected possible relationships among person variables, health behavior variables, and psychosocial state variables in relationship to pre term delivery. The study utilized data from an existing data set of pregnant women at risk for low binhweight infants. Factors which were· conceptualized as person variables were age and race. Psychosocial state variables included anxiety, depression, life events, mastery, self-esteem, stress, and social suppon. Reponed health behavior variables included smoking, drinking, drug use, prepregnant _! weight for height, and weight gain during pregnancy. The dependent variable was preterm delivery, which was defined as delivery before 37 weeks gestation. The analyses used two subgroups of women. The first subgroup consisted of 1163 women who delivered moderately preterm infants (32-37 weeks gestation). The second subgroup consisted of 1258 women who delivered both moderately preterm and very preterm infants (27-37 weeks gestation). Data were analyzed utilizing both univariate and multivariate statistics, with logistic regression as the principle multivariate technique. As a group, person variables and psychosocial state variables had direct relationships, as posited in the hypotheses and supponed in the literature, to preterm delivery. Health behavior variables were directly related to preterm delivery in the variable set that contained both moderately preterm and very preterm infants. Indirect relationships were supponed for the hypothesis that added psychosocial state to health behaviors. Individual variables that were associated with preterm delivery were self-esteem and mastery. Results of this study were significantly different than results of a parallel study using this same data set, but examining the association between psychosocial variables and intrauterine growth retardation.