Abstract
Background: Self-rated health (SRH), an important indicator of cognitive appraisal of health, consistently predicts mortality, morbidity, and health services utilization. However, few explanations account for how these cognitive appraisals of health might differ within a population of midlife adults with chronic illnesses who may be at risk for further illnesses over time.
Objectives: The purpose of this study was two-fold: (a) to uncover classes of chronically ill midlife adults who shared unique profiles of characteristics that predicted SRH over time and (b) to reveal the predictive factors of SRH for each class over time.
Methods: Using 5 waves of data (1992-2000) from the Health and Retirement Study, the sample included 6,335 respondents (ages 51 to 61 at baseline) who reported at least one chronic illness. Selected components of the Interaction Model of Client Health Behavior guided the inclusion of relevant predictors of SRH from the literature. Latent class regression was employed to simultaneously classify respondents and identify factors that predicted SRH for each class over time.
Results: The final model reflected 3 distinct profiles of SRH over time: positive health, average health, and negative health. Four time-varying predictors differed significantly across the 3 classes: overweight, work limitation, depressed mood, and living with a partner. Three time-varying predictors-comorbidity, vigorous activity less than 3 times per week, and current smoking-had the same influence on all 3 classes.
Discussion: The differential effects of these predictors on SRH over time distinguish these results from prior research. In future studies, profiles of SRH that are unique to each class could be used to develop class-specific targeted interventions to improve cognitive appraisal of health, whereas generic interventions would be based on the class-independent predictors of SRH.