Purpose

Methods

We conducted a group-randomized effectiveness trial in which firms within Veterans Administration general internal medicine clinics served as units of randomization, intervention, and analysis. Respondents to a baseline health inventory were regularly mailed the 36-Item Short Form (SF-36) and, as relevant, questionnaires about six chronic conditions (ischemic heart disease, diabetes, chronic obstructive pulmonary disease, depression, alcohol use, and hypertension) and satisfaction with care. Data were reported to primary providers at individual patient visits and in aggregate during a 2-year period.

Results

Baseline forms were mailed to 34,050 patients; of the 22,413 respondents, 15,346 completed and returned follow-up surveys. Over the 2-year study, the difference between intervention and control groups (as measured by difference in average slope) was –0.26 (95% confidence interval [CI]: –0.79 to 0.27; P = 0.28) for the SF-36 Physical Component Summary score and –0.53 (95% CI: –1.09 to 0.03; P = 0.06) for the SF-36 Mental Component Summary score. No significant differences emerged after adjusting for deaths. There were no significant differences in condition-specific measures or satisfaction between groups after adjustment for provider type, panel size, and number of intervention visits, or after analysis of patients who completed all forms.

Conclusion

An elaborate, sustained audit/feedback program of general and condition-specific measures of health/satisfaction did not improve outcomes. To be effective, such data probably should be incorporated into a comprehensive chronic disease management program.