Individual and organizational predictors of patient care: A multilevel analysis of the English National Health Service Staff Survey.

Abstract

Growing evidence attests to the importance of doctors’ psychosocial working conditions in relation to the provision of good quality patient care. However, few studies have explored the mechanisms of this relationship, including the role of work-related wellbeing as a mediator. Even fewer have done so from a multilevel perspective with organization-level predictors or informed by a theoretical framework. Drawing on the job demands-resources (JD-R) model, we ran multilevel structural equation models to test: (i) whether individual and organization-level psychosocial predictors influence doctors’ self-reported perception of care provided; and (ii) whether psychological strain and work engagement mediate these relationships. Data was drawn from a national survey of doctors (n=14,066) and metrics from all 157 English hospital organizations. At the individual level, work overload and job control predicted quality of care, while workplace aggression and manager support did not. Indirect effects were observed for all examined psychosocial predictors and quality of care as mediated by work engagement. At the organization level, only work engagement mediated the relationships between the number of emergency admissions and the quality of organization and individual care. These findings emphasize the importance of positive manifestations of work-related wellbeing and the improvement of hospital working conditions. The JD-R model provides a useful framework at the individual level, although study design limitations may explain its incompatibility at the organization level.