Comparing breast cancer risk between lesbians and their heterosexual sisters

Abstract

Purpose

The purpose of this study was to explore the similarities and differences between lesbians and their heterosexual sisters in the established risks for developing breast cancer.

Methods

The design for this study was a matched (lesbian with heterosexual sister) cross-sectional, mail-back, anonymous survey. We distributed the surveys throughout the state of California to English-speaking women who identified themselves as lesbians, age 40 and older, and their sisters. Using the modified Gail Breast Cancer Risk model as well as other well-established factors associated with the development of breast cancer, we compared the breast cancer risk potential for 324 sister pairs (N = 648). Data were analyzed using paired t-tests, analysis of covariance (ANCOVA), McNemar's χ², or the Bowker statistic, as appropriate for the level of data.

Main findings

The lesbians had significantly higher 5-year (p < .0001) and lifetime (p = .001) risk for developing breast cancer. The reasons for lesbians' predicted rate of breast cancer were most likely their higher scores on all pregnancy-related variables and the relatively high number of breast biopsies they reported. The lesbians had used birth control pills less (p <. 0001), had significantly fewer pregnancies (p < .0001), children (p < .0001), abortions (p < .0001), and miscarriages (p < .0001) as well as significantly more breast biopsies (p = .02) than did their heterosexual sisters.

Conclusions

A lesbian who comes out to her clinician is relying on the clinician to be informed and be open to discuss her life. When a lesbian has a lump or a suspicious mammogram, she needs her clinician to advocate for her within the health care system because she is at higher risk for having cancer than a heterosexual woman.

Introduction

In 2004, an estimated 215,990 women are expected to be diagnosed with breast cancer, and 40,110 women will die as a result of this disease in the United States (Jemal et al., 2004). A portion of these women will be lesbians. In the early 1990s, a noted epidemiologist reported that risk for developing breast cancer is two to three times greater among lesbians than among heterosexual women (Haynes, 1994). These results were interpreted by the media as an epidemic or plague of breast cancer among lesbians (see Plumb [1997] for a discussion of this history), and caused enormous concern and trepidation within the lesbian communities. The purpose of this study was to explore the similarities and differences between lesbian and heterosexual women in the established risks for developing breast cancer.

The number of lesbians diagnosed with breast cancer is unknown at this time. The National Cancer Institute (NCI) maintains a registry of all cancers diagnosed in 11 geographic areas in the United States, which covers about 14% of the US population (NCI, 2003). This registry is called the Surveillance, Epidemiology, and End Results Program, and information about the incidence and death rates for specific subpopulations are created from their data. Data about sexual orientation are not collected in these registries; therefore, the prevalence of lesbians diagnosed with breast cancer has not been established. Studies are needed to determine whether breast cancer risk is different for lesbians and heterosexual women just as studies have determined that the rate of breast cancer deaths for African American women (35.9/100,000) is different than that for Asian/Pacific Islander women (12.5/100,000) (Jemal et al., 2004).

Recent research suggests that lesbians may be at greater risk for developing breast cancer than heterosexual women (Cochran et al., 2001, Dibble, et al., 1999, McTiernan et al., 2001, Rankow and Tessaro, 1998, Roberts et al., 1998). The findings of these studies are in question, however, because many of them have significant limitations, including small sample sizes, clinic sampling bias, and inadequate use of appropriate control groups.

Clinical risk assessment holds great promise for identifying individuals and groups of individuals who might benefit from preventive interventions. In the case of breast cancer, statistical and genetic models, including the Gail and Claus models, have been developed to assess an individual woman's future risk of developing disease (see McTiernan et al. [2001] for an excellent discussion of these tools). The Gail Breast Cancer Risk Model is the most widely used breast cancer risk assessment tool. It was developed by statisticians based on data from the Breast Cancer Detection and Demonstration Project, which was a mammography screening project involving over 280,000 women conducted during the 1970s. The software program for the Gail Breast Cancer Risk Model as modified by the researchers at the National Surgical Adjuvant Breast and Bowel Project (NSABP), a clinical trials cooperative group supported by the NCI (1998), predicts the 5-year and lifetime rates of both invasive and noninvasive breast cancer, using the following factors: age, ethnicity, age at menarche, age at first live birth, number of first-degree relatives with breast cancer including daughters, number of breast biopsies, and presence of atypia in a biopsy specimen (Costantino et al., 1999, Gail et al., 1989). Although controversy exists about the omission of some risk factors and the overestimation of risk of developing breast cancer in young women (Sakorafas, Krespis, & Pavlakis, 2002), the Gail Breast Cancer Risk Model has been found to be a fairly accurate predictor developing breast cancer in four validation studies (Bondy et al., 1994, Costantino et al., 1999, Gail and Benichou, 1992, Spiegelman et al., 1994). The Gail Risk model has never been applied to a sample of lesbians who were not at high risk for developing breast cancer.

Although other items have been identified as behavioral risk factors for breast cancer, they are not included in the model because evidence indicates that their contribution to breast cancer risk is uncertain, and researchers have been unable to determine with precision how much these factors contribute to breast cancer risk. These risk factors include use of hormone replacement therapy (HRT); obesity and high-fat diets; increased use of alcohol; radiation exposure; smoking; and environmental pollutants (Brekelmans, 2003, Brody and Rudel, 2003, McTiernan, 2003). Protective factors against the development of breast cancer may include breast feeding and physical activity (McTiernan, 2003).