The association between race, ethnicity, and socioeconomic status and cervical cancer screening rates and cervical intraepithelial neoplasia in Monroe county .

Background/Significance: More research is needed to determine the association between contextual socioeconomic status (SES), racial/ethnic clustering and cervical cancer screenings and/or cervical precancer rates. If an association exists, appropriate public health tools can be developed to address those disparities. 

Objectives/Purpose: To determine the association between race/ethnicity, SES and cervical cancer screenings rates and cervical precancer rates. 

Methods: The 2010 cervical cancer screening rate was determined and two additional methods of estimating Pap rates were developed. Two ecological and a matched case control studies were conducted. Poisson regression models were constructed to assess the association between race/ethnicity and cervical cancer screening rates and cervical precancer rates. A conditional logistic regression model was constructed to determine the association between SES and cervical precancer. 

Results: The 2010 cervical cancer screening rate was 29.4%. Census block groups whose largest racial/ethnic group was non-Hispanic black was associated with a lower cervical cancer screening rate as compared to non-Hispanic white census block groups (IRR = 0.6750, p < 0.0001). Higher SES was independently associated with increased cervical cancer screening rates (IRR = 1.0697, p < 0.0001). There was no association between census block groups whose largest racial/ethnic group was “other” as compared to non-Hispanic white census block groups (IRR = 1.2511 p = 0.1422). Increased SES was significantly associated with decreased cervical precancer rates within census block groups (IRR =0.8216, p < 0.0141); however, no association existed between SES and the individual risk of cervical precancer (OR = 1.005 p = 0.9757). Women who were ever smokers had an increased risk of cervical precancer development as compared to women who never smoked (OR = 1.778 p = 0.0322). 

Discussion/Conclusion: Over-reporting is an issue with self-reported data used to calculate screening rates and this type of data is often used to estimate screening rates in the population. Increased precancer rates on an ecologic level, but not on the individual level, were associated with decreased SES. The direction of the association between privately insured women and the risk of cervical precancer, however, indicates privately insured women had elevated risk of cervical precancer. This may suggest that after an abnormal pap, privately insured women are more likely to continue follow-up care.