Preventive screening. What factors influence testing?

OBJECTIVE To determine factors associated with having preventive screening tests in a population-based sample of Ontario women.

DESIGN Secondary analysis of data from Statistics Canada's National Population Health Survey linked to data from the Ontario Health Insurance Plan to ascertain whether women aged 20 or older had Pap smears, mammography, bone densitometry, or cholesterol testing. Factors associated with having testing were subjected to logistic regression analysis.

SETTING Ontario.

PARTICIPANTS Women aged 20 or older; from 19,600 Canadian households, 2232 Ontario women gave consent to linkage of administrative databases.

MAIN OUTCOME MEASURES Age-specific population screening rates. Odds ratios and probabilities of having screening in relation to socioeconomic, geographic, and physician-associated factors.

RESULTS Having screening was associated with age, income, education, and place of residence. Women with regular physicians were more likely to have Pap smears (odds ratio [OR] 4.4, range 1.7 to 12), densitometry (OR 22, range 3.6 to 140), and cholesterol testing (OR 8.0, range 2.3 to 29). Women who had periodic health examinations were more likely to have Pap smears (OR 6.7, range 4.6 to 9.8), mammograms (OR 3.7, range 2.3 to 5.9), densitometry (OR 3.7, range 1.3 to 10.5), and cholesterol testing (OR 3.0, range 2.0 to 4.5). The probability of having testing increased with number of visits a year to a doctor, but ceased to increase after three visits.

CONCLUSION Having screening tests was associated with socioeconomic factors including income, education, and place of residence. Patients who went to doctors for episodic care only were less likely to have preventive screening than patients who went for periodic health examinations.