Personality and the physician-patient relationship as predictors of quality of life of cardiac patients after rehabilitation

Background: Numerous studies document the influence of psychosocial variables on the course of coronary heart disease. This study examines the influence of personality traits (trait anger, cynicism) and aspects of the physician-patient relationship (promoting patient participation by the physician, active communication behavior of the patient, trust in the physician) on the health related quality of life (HRQOL) of cardiac patients after rehabilitation. Methods: N = 331 patients with chronic ischemic heart disease were surveyed using questionnaires at two time points (beginning and end of 3-weeks inpatient rehabilitation). In addition, characteristics of the disease and cardiac risk factors were provided by the physician. HRQOL was measured using a total of six scales and three instruments: SF-12, MacNew questionnaire, and SAQ. Hierarchical regression analyses were carried out to predict HRQOL after rehabilitation, in which the baseline values of HRQOL, sociodemographic variables, characteristics of the disease and risk factors, personality traits, and finally the aspects of the physician-patient relationship were included stepwise. As a number of variables were used for the regression models, multiple imputation was conducted. Results: The baseline values explain most of the variance (42%-60%). After controlling the baseline values, the sociodemographic variables explain up to 5% incremental variance of HRQOL, with income being the most important predictor. The characteristics of the disease and cardiac risk factors explain between 0.4% and 3.8% incremental variance, however, variance increase is often not significant. The personality traits added in the fourth step explain up to 2% additional variance; trait anger is a significant predictor of HRQOL in three of the six scales. The features of the physician-patient relationship included in the last step lead to a significant increase in explained variance (between 1.3% and 3.9%) for all six scales. In particular, the physician's promotion of patient participation has a significant influence. The overall explanation of variance for HRQOL is between 50% and 64%. Conclusions: Low income, a high level of trait anger, and low patient participation are significant risk factors, even if a number of potential confounders are adjusted. Research is needed that shows which causal pathway low income functions on and what therapies in rehabilitation can mitigate the disadvantage of persons with a high level of trait anger. The providers should implement measures to actively integrate rehabilitation patients in treatment (e.g. encourage them to ask questions).