Identifying pathways from quality of care to outcomes and satisfaction in Primary Care using structural equation modelling
Ricci Cabello I., Stevens S., Dalton ARH., Griffiths RI., Campbell J., Valderas JM.
Purpose To obtain empirical evidence on the relationships between the different dimensions of quality of primary healthcare in family practices in England in order to inform quality improvement evaluation and measurement of performance. Method Cross-sectional study at the practice level (7,607 practices). We examined 136 quality indicators from the Quality and Outcome Framework and the GP Patient Survey, and allocated them to domains of processes of care (“quality assurance”, “education and training”, “medicines management”, “access to the practice”, “clinical management” and “patient centred care”), health outcomes (“intermediate outcomes” and “patient-reported health status”) and patient satisfaction. The relationships between the domains were hypothesized in a conceptual model which was subsequently empirically tested using a hybrid structural equation model that combined factor and path analyses. Results The model supported two distinct paths. In the first path, “education and training” was associated with “clinical management” (β=0.32), which in turn was strongly associated with “intermediate outcomes” (β=0.69). In the second path, “access to the practice” was associated with “patient centred care” (β=0.63), which in turn was strongly associated with “patient satisfaction” (β=0.88). These two paths were substantially independent, with weak associations being observed between “clinical management” and either “patient centred care” (β=0.08), or “patient satisfaction” (β=0.05). “Patient-reported health status” was weakly associated with “patient centred care” (β=-0.05), and “patient satisfaction” (β=0.09), but not with “clinical management” or “intermediate outcomes”. Conclusions The evaluation of performance of primary care services needs to take into account these three domains of quality which seem to be independent from each other: clinical management, patient experiences and patient reported outcomes.