Correlates of underutilization of gynecological cancer screening among lesbian and heterosexual women

Alicia K. Matthews, Dana L. Brandenburg, Timothy P. Johnson, Tonda L. Hughes

Research output: Contribution to journalArticlepeer-review

73 Scopus citations


Background. Study aims were to examine cervical cancer risk factors, screening patterns, and predictors of screening adherence in demographically similar samples of lesbian (N = 550) and heterosexual women (N = 279). Methods. Data are from a multisite survey study of women's health conducted from 1994 to 1996. Results. Differences in sexual behavior risk factors for cervical cancer were observed with lesbians reporting earlier onset of sexual activity (P < 0.05), more sexual partners (P < 0.001), and lower use of safer sex activities (P < 0.01). Lesbian and heterosexual women were equally likely to have ever had a Pap test; however, lesbians were less likely to report annual (P < 0.001) or routine (P < 0.001) testing. Multivariate analyses were used to determine the associations between demographics, health care factors, health behaviors, and worry about health and screening behaviors. Individual predictors of never screening included younger age, lower income, and lack of annual medical visits. Independent predictors of both recent and annual screenings included history of an abnormal Pap test, being heterosexual, and annual medical visits. Conclusion. Data indicate that lesbians are at risk for cervical cancer, yet underutilize recommended screening tests. Findings have implications for research, education, and cancer control among lesbians.

Original languageEnglish (US)
Pages (from-to)105-113
Number of pages9
JournalPreventive medicine
Issue number1
StatePublished - Jan 2004
Externally publishedYes


  • Cervical cancer risk
  • Cervical cancer screening
  • Sexual orientation

Fingerprint Dive into the research topics of 'Correlates of underutilization of gynecological cancer screening among lesbian and heterosexual women'. Together they form a unique fingerprint.

Cite this