Process and outcome: lack of correlation in a primary care model.

Records of 87 hypertensive patients in a group family practice were reviewed to assess the validity of the process audit in determining medical care quality. Although a significant reduction in mean diastolic blood pressures was found at two follow-up intervals, physician performance scores showed no significant correlation with this outcome. Nor could an association between medical process and outcome be demonstrated when control of blood pressure to less than 95 mmHg was used as an outcome measure. The reliability between two chart auditors was poor, with complete agreement being achieved in only 29 percent of cases. Accurate quality assessment will require more practical methods of review which relate directly to patient outcomes.