There have been mixed results from studies comparing mammography screening behaviors among sexual minority and heterosexual women. Although two studies have documented higher mammography rates for lesbian compared to heterosexual women [5, 20], the majority of studies have reported that sexual minority women were less likely than heterosexual women to have had a recent mammogram [8, 21, 22] or there were no differences in mammography rates by sexual orientation [23–26]. Cross-sectional data from at least three large population- [21, 22] and community-based [8] samples suggest that sexual minority women may be less likely to be screened for breast cancer than heterosexual women. For example, in multi-lingual population-based surveys of more than 18,000 adults in New York City, Kerker et al. [21] found that women who had sex with women (WSW) were less likely to have had a mammogram in the past 2 years (53 vs. 73 %) than other women. Next, using a combined sample of almost 47,000 women from the 2000–2007 Behavioral Risk Factor Surveillance Surveys (BRFSS), Buchmueller and Carpenter [22] found that women in same-sex relationships were significantly less likely to have had recommended mammograms (adjusted odds ratio [AOR] = 0.75; 95 % CI = 0.61, 0.92) than women in different-sex relationships. Finally, Cochran et al. [8] found that 73 % of women aged 40–49 years and 83 % of women aged 50–75 years who described themselves as lesbian or bisexual in a combined sample of nearly 12,000 women from seven U.S. surveys of sexual minority women reported ever receiving a mammogram, compared to 87–90 % of women in U.S. general population surveys [8].

Similar to mammography screening, there have been mixed results from studies comparing gynecological screening behaviors among sexual minority and heterosexual women. Several cross-sectional studies using both population-[21, 22, 29, 30] and community-based [5, 8, 25, 31] samples as well as a cohort study [32], have documented lower rates of Pap testing among sexual minority compared to heterosexual women. For instance, Agenor et al. found lower odds of Pap test use in the past 12 months (OR = 0.40, 95 % CI = 0.23, 0.68) among lesbians compared to heterosexual women in the 2006–2010 National Survey of Family Growth [30]. Combining 4 years of population-based data (2001, 2003, 2005, and 2007) from the California Health Interview Survey, Boehmer et al. found that lesbians (65 %) were less likely than heterosexual (74 %) and bisexual (77 %) women to report having had a Pap test in the prior year [29]. Kerkeret al. [21] reported that women who had sex with women (WSW) were less likely to have had a Pap test in the past 3 years (66 vs. 88 %) than other women in the New York City multi-lingual population-based surveys. In the combined BRFSS sample, Buchmueller and Carpenter [22] found that women in same-sex relationships were significantly less likely to have had a Pap test in the past 3 years (adjusted odds ratio [AOR]= 0.74; 95 % CI = 0.57, 0.97) than women in different-sex relationships. Data were similar for some community-based samples. For example, using data from a multisite survey of women 20–86 years, Matthews et al. [31] found that lesbians were less likely to report annual Pap test screening (49 vs. 66 %) and receipt of a Pap test every 3 years (81 vs. 90 %) compared to heterosexual women. Mays et al. [25] also reported that 2-year Pap test rates were lower for Hispanic (70 %) and African American (76 %) lesbians compared to Hispanic (81 %) and African American (81 %) heterosexual women in their Los Angeles sample of racial and ethnic minority women. Finally in the cohort of women who were part of the Growing Up Today Study, Charlton et al. found lower odds of receipt of a Pap test in the past year among lesbian (AOR = 0.25, 95 % CI = 0.12, 0.52) and mostly heterosexual/bisexual women (AOR = 0.70, 95 % CI = 0.54, 0.92) compared to heterosexual women [32].

To date, there have been few studies that have reported screening rates for cancer of the colon-rectum by sexual orientation, and these studies have generally found few differences. For example, in the WHI sample reported by Valanis et al. [6], receipt of a hemoccult screening within the prior 5 years among those 50–79 years was comparable for lifetime lesbians (59 %), adult lesbians (60 %), bisexual women (61 %), and heterosexual women (57 %). Similarly, among women 50 years and older in the Nurses’ Health Study, Austin et al. [27] reported no differences in rates of ever having had a colonoscopy or sigmoidoscopy by sexual orientation (lesbians = 42 %; bisexual women = 39 %; heterosexual women = 39 %). In the Massachusetts BRFSS sample  ≥ 50 years, Conron et al. [20] found no significant differences in ever having had a sigmoidoscopy or colonoscopy when they compared lesbian (58 %, AOR = 1.00, 95 % CI = 0.66, 1.51) and bisexual women (74 %, AOR = 2.16, 97 % CI = 0.96, 4.86) to heterosexual women (56 %). Blosnich et al. also found no differences in ever having had a colorectal cancer exam for lesbian (59 %), bisexual (64 %), and heterosexual (66 %) women in the combined BRFSS sample. Similarly, Boehmer et al. found no differences in ever having had a colorectal cancer exam for lesbian (74 %), bisexual (65 %), and heterosexual (68 %) women 50 years of age and older in the California population-based sample. Finally, in a community sample of 600 women, Clark et al. [26] reported that among women 50–75, 65 % of women who partnered with women or with both men and women and 67 % of women who partnered with men were on-schedule for colorectal cancer screening according to the American Cancer Society screening guidelines.

Despite the number of large-scale cohort, population- and community-based studies, it is not possible to unequivocally conclude whether or not there are differences in the prevalence of cancer screenings by sexual orientation. The strongest evidence presented is for cervical screening where the majority of relevant studies have documented lower rates among sexual minority compared to heterosexual women. There are a number of possible explanations for the mixed results for the other types of cancer screenings. First, studies differed in measures of sexual orientation. Some used identity to classify women but differed as to whether they combined lesbian and bisexual women together or analyzed them separately. Other studies used behavior (e.g., women who have sex with women versus women who have sex with men) to identify sexual minority women. Second, data were collected in different geographic regions. These geographic regions may differ in their screening rates due to density of screening facilities [33] as well as the availability and accessibility of cancer screening programs, particularly for uninsured and underinsured women [34]. Third, studies differed with regard to the age eligibility of women included in the analyses of the respective screening exams. For example, some studies of mammographic screening included all women regardless of age despite the fact that it is not recommended by any professional organization before age 40 years. Fourth, studies differed in the proportion of racial and ethnic minority women included in the sample. For example, some studies were comprised of almost all white women whereas other studies included higher proportions of racial and ethnic minority women consistent with the populations from which the samples were drawn (e.g., New York City, Massachusetts, California). Finally, studies differed in the definition of the screening outcome. For example, intervals for on-schedule Pap testing included the past 12 months, past 2 years and past 3 years. Similarly, colon-cancer screening differed in interval as well as type of test (e.g., lifetime colonoscopy versus past 5-year fecal occult blood test).