Self-report and primary care medical record documentation of mammography and Pap smear utilization among low-income women

J Natl Med Assoc. 2004 Dec;96(12):1632-9.

Abstract

Purpose: To determine whether self-report of mammography and Pap smear utilization was accurate and to determine whether racial/ethnic differences existed.

Methods: Face-to-face surveys were administered to 314 consecutively selected women over 40 attending two low-income inner-city family practice sites. Medical records were reviewed for documentation of mammography and Pap smear utilization. Level of agreement between self-report and chart review was reported. Sensitivity, specificity, positive predictive value and negative predictive value were calculated.

Main findings: Puerto Rican women had lower income levels and were less educated than African-American and non-Latina white women. Self-report of mammograms and Pap smears were higher than medical record documentation. Level of agreement was higher for more recent tests. Negative predictive values for mammography were high (75-95.5%). Lower sensitivity, specificity, positive predictive value and negative predictive value for mammography were seen among Puerto Rican women compared to African-American and non-Latina white women.

Conclusions: High negative predictive values suggest that asking women about mammography use may be an inexpensive, easy intervention in the primary care setting to increase screening among women currently not being screened by increasing conversations between patients and providers to address personal barriers to screening.

Publication types

  • Multicenter Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • African Americans / statistics & numerical data
  • Documentation
  • European Continental Ancestry Group / statistics & numerical data
  • Female
  • Hispanic Americans / statistics & numerical data
  • Humans
  • Mammography / statistics & numerical data*
  • Middle Aged
  • New York
  • Papanicolaou Test*
  • Poverty / statistics & numerical data
  • Primary Health Care
  • Sensitivity and Specificity
  • Urban Population / statistics & numerical data
  • Vaginal Smears / statistics & numerical data*