Can you use a sequential sample of patients as a substitute for a full practice audit?: Study of mammography screening rates in 20 family practices in Ontario.

OBJECTIVE To compare rates of mammography screening among women in family practices, based on a sequential sample of eligible women presenting to the practices during an 8-week period, with rates found in a full audit of all eligible patients. DESIGN Chart review. SETTING Twenty community-based family practices in south-central Ontario. PARTICIPANTS Family physicians and their female patients 52 to 71 years old who had had at least 1 visit to the office during the past 3 years. INTERVENTION Eligible patients were sampled by 2 approaches: sequential sampling of patients coming for appointments during an 8-week period and a full practice audit of all eligible women. MAIN OUTCOME MEASURE Mammography rates found using the 2 approaches. RESULTS The mean time-appropriate rate of mammography screening based on the sequential sample was 66.4%. The mean time-appropriate rate of mammography screening for the full practice audit was 58.8%. The sequential sample rate was higher than that of the full audit by 7.6%; differences ranged from -6.5% to 24.9% among practices. Regression analysis indicated a positive and significant correlation between rates based on the data generated by the 2 different approaches (r2 = 0.50). CONCLUSION A rate of mammography screening based on a sequential sample can reasonably approximate the actual rate of mammography screening that would be found based on a full practice audit.