Frequency and Determinants of Triennial Pap Smear Screening Rates

Abstract

Purpose: Before 1993, no data on either 3 year rates of Papanicolaou (Pap) screening or factors determining screening rates were available for large populations except for results of self-reported patient surveys containing known inaccuracies. The purpose of this study was to investigate the factors determining 3 year rates of Pap screening among health plan members throughout Kaiser Permanente in Southern California. Methods: We analyzed computer files of enrollment, clinical encounters, and Pap reports. Results: Overall, 74.5% of the study cohort received ≥1 Pap smear during the 3 year study interval. Screening rates varied by patient’s age, median income, copayment status, and number of visits to primary care departments. Women who had either their own assigned personal physician, a female personal physician, or both were more likely to have received Pap screening. Screening rates were inversely related to age of personal physician. Women who had personal physicians practicing in family practice (FP) departments were more likely to have received Pap screening than women who had personal physicians practicing in internal medicine (IM) departments. Multivariate analysis of women who had a personal physician assigned to them showed that women were more likely to receive Pap screening if they had a female FP personal physician (odds ratio (OR) = 1.13) or a female IM personal physician (OR = 1.23) than if they had a male IM or FP personal physician. We detected no statistically significant effect of provider age or Hispanic ethnicity when controlling for other variables. Conclusions: High rates of triennial Pap screening rates were documented. Major factors determining receipt of Pap screening were patient age and number of visits to primary care departments. Differences in screening rates associated with median income and personal physician’s gender were smaller than those reported in other settings. Unlike patients in other settings, patients in this population who had older providers, greater out-of-pocket medical expenses, or Hispanic ethnicity did not have lower 3 year screening rates after controlling for other factors.