A Review of Obesity Issues in Sexual Minority Women
Deborah J. Bowen1,3, Kimberly F. Balsam2 and Samantha R. Ender3

1Fred Hutchinson Cancer Research Center, Seattle, Washington, USA
2Department of Psychology, University of Washington, Seattle, Washington, USA
3Department of Social and Behavioral Sciences, Boston University, Boston, Massachusetts, USA
Correspondence: Deborah J. Bowen, ( Este endereço de e-mail está protegido de spam bots, pelo que necessita do Javascript activado para o visualizar )

Obesity is emerging as an important risk factor for multiple disease outcomes among women. In women, obesity is predictive of a diagnosis of multiple cancers, including breast, ovarian, colon, endometrial, and pancreatic, as well as cardiovascular disease and diabetes (1,2). Determining the demographic and cultural predictors of obesity is an important step in identifying the causes of obesity and potential interventions for preventing and reducing obesity. For example, obesity is more common among women from lower socioeconomic levels (3), which makes them a reasonable target for obesity prevention programs.

Sexual orientation is an important demographic factor that may be associated with higher risk for several major health outcomes, including cancer and cardiovascular disease (4). A recent review by the Institute of Medicine (5) called for research into differences between women of differing sexual orientations to determine health disparities between the groups. Several studies in the past several years have compared the obesity levels and associated behaviors among women of differing sexual orientations; however, patterns of differences between these demographic groups are not clear. Identifying any patterns of differences from the published literature can help to determine risk factor patterns and directions for needed interventions with sexual minority women (SMW).

Methods
In this paper, we focus on obesity and related issues among SMW and contrast these issues with available data on heterosexual women. These considerations will be useful in guiding the directions and content of interventions to reduce obesity in SMW. For a definition of SMW, we used any identified article that reported data on lesbians, bisexual women, or SMW, in the authors' chosen descriptive label. Sexual orientation was assessed in these studies using either self-reported sexual identity or the sex of sexual partners of participants. We searched Medline for relevant articles using the following steps. We first searched for articles published from 1993 through July 2006 using the following keywords: lesbian, bisexual, sexual minority, and sexual orientation. We then searched for articles over the same years using these keywords: obesity, body mass index, dietary intake, caloric intake, physical activity, and exercise. We then identified the intersection of those two searches, yielding 94 articles. We excluded articles that did not present empirical data (i.e., theoretical or review papers). We supplemented the Medline review with a search on PsycInfo and a search on Cumulative Index to Nursing and Allied Health Literature. We limited the articles to adults 18 years of age or older, and to North American samples. Nineteen articles met these criteria and were included in the final review group. We reviewed each article, noting key points in the accompanying tables. We report the ethnic distribution of the sample, where possible, as well as the method of determining sexual orientation. Only articles on women were included.

Results
Table 1 presents information from articles, listed first by design categories (large or population-based samples, medical record samples, and convenience samples), and then alphabetically by the last name of the first author. Relevant characteristics of the article are listed in the table, including the sampling method and characteristics, the type of sexual minority targeted, the type of measure included in the article, the control or comparison group used, and the relevant findings. We describe each of the articles below.

Populatio-based samples
Aaron et al. (6) conducted a large cross-sectional study (n = 1,010) of self-identified lesbians 18 years or older in the greater Pittsburgh, Pennsylvania area using a community sample (self-identified bisexual women were not included in the study sample). The study aimed to determine cardiovascular risk factor patterns in lesbians, using national data sets of general-population women as the comparison group (n = 88,191). The study sample included 88.7% white, 7.7% African American, and 3.6% other minority SMW, and utilized self-reported height and weight measures. A higher proportion of lesbians were overweight. Lesbians were also more likely than the general-population women to report vigorous activity in the last 3 days, but the proportion of sedentary participants was no different compared with population survey data.

Cochran et al. (7) combined data from seven large community convenience-based samples of lesbian and bisexual women collected over a 15-year period in order to examine health behaviors and risk factors. Data from all surveys were collected via anonymous, self-administered questionnaires. Combined, the surveys included 12,000 participants ranging in age from 18 to 50, with 85.9% of the participants identifying themselves as non-Hispanic whites and 14.1% identifying themselves as from another racial/ethnic background. Health and demographic variables were recoded so as to be comparable across studies. Results were then compared to data from population-based health surveys of U.S. women (n = 19,000). A total of 28% of the lesbian and bisexual women in the combined sample were obese, which is comparable to national general population-based data for women. However, when these values are standardized to take into account the demographic differences in samples, a greater proportion of SMW were obese than would be expected (P < 0.05). By contrast, after controlling for demographic differences, SMW were less likely than women in the general population to report that they consider themselves to be overweight (P < 0.05).

Valanis et al. (8) analyzed baseline data from 93,311 participants in the Women's Health Initiative, a multi-site investigation of disease outcomes in women aged 50–79 in the United States. Participants for this study were recruited through a number of sources, including media advertisements, community and health organizations, and direct unsolicited mailing. Sexual orientation was assessed at baseline by a single question that examined the gender(s) of a participant's lifetime sexual partners, yielding five categories: heterosexual, bisexual, lifetime lesbian (only ever had sex with women), adult lesbian (had sex only with women after the age of 45), and never had adult sex. Adult lesbian, lifetime lesbian, bisexual, and heterosexual women were relatively similarly distributed across ethnic groups, with 85% non-Hispanic white recorded among heterosexual women and 90% non-Hispanic white recorded among adult lesbian women. Participants identifying themselves as black ranged from 6.2% in the adult lesbian group to 8.5% in the heterosexual group. Participants reporting no adult sex were more likely to identify as "other" (7.0%) or Hispanic (4.2%) than any of the other groups. Clinic staff measured height and weight at the screening visit, and BMI was calculated. Participants' dietary intake was assessed using the Food Frequency Questionnaire. Physical activity was assessed with questions about the frequency and duration of walking and other exercise activities. SMW status (lesbian and bisexual combined) was associated with a greater likelihood of obesity (odds ratio = 1.25, P < 0.001) and with less likelihood of consuming four to six servings of vegetables per day (odds ratio = 0.77, P = 0.006), compared to heterosexual women.

Utilizing data from the Nurse's Health Study II on risk factors for chronic disease among a cohort of 90,823 middle-aged nurses, aged 32–51, Case and colleagues (9) compared self-reported lesbian and bisexual participants with self-reported heterosexual participants. The majority of all three groups identified as white non-Hispanic, with 92.4% for bisexual and heterosexual women to 95.7% for lesbian women. Both lesbians and bisexual women reported a higher current BMI, using self-reported height and weight, compared with heterosexual women, as well as a higher BMI at age 18. Lesbians were also slightly more likely to report strenuous exercise as compared to heterosexual women.

Medical record based samples
Dibble et al. (10) conducted a retrospective medical chart review of 433 self-reported lesbians and 586 self-reported heterosexual women to determine differential levels of obesity and other variables, while assessing the risk for ovarian cancer. The overall sample was 70% white, with significantly fewer black women and more of white women identifying themselves as lesbian. Bisexual women were not included in this study sample. Participants ranged in age from 35 to 75 years, and height and weight were both measured by clinic providers. No ethnicity data were provided in the article. The clinical setting was a community clinic with a relatively large proportion of lesbian clientele. These authors found that BMI was higher among lesbians when sexual orientation was noted by means of a medical chart (gathered via self-report).

Roberts and colleagues (11) compared the medical records of lesbian (n = 433) and heterosexual (n = 586) women at a lesbian-focused community clinic to determine differences in breast cancer risk factors. Women who identified as bisexual were excluded from the sample. The women in the overall sample (n = 1,019) ranged from 35 to 75 years of age, and the majority of them were white (70%). The self-identified lesbian group included a higher percentage of white women (74.6%) as compared to the heterosexual group (62.5%), as well as a significantly lower percentage of black women (8.3%, as compared to 17.9% of the heterosexual group). Asian American, Latina, other, and missing ethnicities were similarly represented among the groups. According to the review of provider-measured height and weight data, lesbians had higher BMIs compared to heterosexual patients.

Convenience samples
Cogan (12) recruited a non-random sample of 181 self-identified lesbian and bisexual women at a gay pride festival in California. The participants ranged from 17 to 58 years of age. Nearly 75% of the participants identified as white, 8% as Latina, 7% as Asian American, 6% as African American, 4% as Native American, and 2% as mixed race. Participants completed a survey examining BMI, body image, satisfaction with weight, eating disorder symptoms, types of fitness activities, reasons for exercise, and feminist self-labeling and attitudes. No heterosexual comparison group was included. Participants had engaged in four fitness activities in the previous month, with walking, dancing, and hiking/camping being the most frequent activities. The women's primary reasons for exercising were functional and non-aesthetic. On average, participants reported a mean BMI of 24.5. The participants' mean weight of 167.9 differed significantly from their ideal weight of 136.6. Older and heavier women were more dissatisfied with their weight than younger and thinner women. As far as eating disorders were concerned, 10% of participants currently reported bulimic symptoms. On average, participants sometimes engaged in dieting behavior and often felt dissatisfied with their bodies. Feminist self-labeling and greater acceptance of the women's movement were each associated with less body dissatisfaction, less discrepancy between real and ideal weight, less exercising for aesthetic reasons, and fewer dieting behaviors.

Heffernan (13) recruited a non-random sample of 203 self-identified lesbians from community organizations and festivals. Participants completed a survey examining weight, attitudes toward social norms, disordered eating, body esteem, importance of attractiveness, and reasons for exercise. No heterosexual comparison group was included. The participants ranged in age from 17 to 65 years, with a mean age of 34 years. White participants accounted for the majority of the sample (84.2%), while 4.9% identified themselves as African American, 3.0% as Latina or Asian American, 2.9% as Native American, and the remainder as Middle Eastern or multiracial/of unspecified ethnicity. One percent of participants met the criteria for bulimia nervosa, and 0.49% met criteria for anorexia nervosa. The average BMI of participants was 27, and the mean discrepancy between actual and desired weight was 27 pounds. A total of 48% of participants reported consciously restricting food intake on half of the days in the previous month, and 27% reported at least one episode of binge eating in the previous month. Overall body esteem was significantly related to self-esteem but not to sexual attractiveness, physical condition, or concern with weight. Close to two-thirds (63%) of participants said that it was important to them to be with a physically attractive partner. Participants who were highly involved in lesbian/gay activities reported less weight concern than those who were not involved in such activities. Participants rated health and fitness as the primary reasons for their exercise activities.

Herzog and colleagues (14) examined the relationship between sexual orientation and weight, body image, and eating attitudes in a small convenience sample of homosexual and heterosexual women (n = 109). Of this sample, 64 identified themselves as heterosexual, while 45 identified themselves as homosexual. The mean age of the heterosexual women was 24.3 years, while the mean age for the homosexual women was 27.4 years. The large majority in both samples was found to be white (91% of the heterosexual group and 96% of the homosexual group). Weight and height were self-reported. Researchers found that homosexual women were heavier, reported less desire to lose weight, and reported a heavier ideal weight, compared to heterosexual controls.

Koh (15) recruited 524 lesbian, 143 bisexual, and 637 heterosexual women from physician's offices and medical clinics. Participants completed anonymous, self-report questionnaires, which included questions about demographics, health status, and health-related behaviors. Sexual orientation was assessed using a single question that addressed the participants' self-identification. The mean age of participants was 40; 83% of participants were white. In an analysis controlling for demographic differences, no sexual orientation differences in BMI were found (P = 0.09).

A survey study of self-identified lesbians and bisexual women (n = 188) (16), ranging in age from 18 to 55 years, found that women with a self-rated "feminine" appearance and with more heterosexual friends reported less satisfaction with their bodies compared to those with a more "masculine" appearance and to participants with more lesbian friends.BMI values were similar among the cultural groups. No ethnicity data were collected.

Moore and Keel (17) recruited 45 self-identified lesbians and 47 heterosexual women from the community for a survey on body image and eating problems. Bisexual women were not included in the study sample. The mean age of participants was 32.6 years, and 92% of the participants were white. Participants were compared according to sexual orientation and to age on the Body Esteem Scale, the Eating Disorders Inventory-2, the Reasons for Exercise Inventory, and BMI. No sexual orientation differences were found based upon BMI. Lesbians reported less Drive for Thinness and less Weight Control on the Reasons for Exercising scale. No interaction was established between age and sexual orientation.

A convenience sample of 525 lesbians and 258 heterosexual controls (18), categorized by self-reporting of sex of sexual partners, provided survey data on weight, height, frequency of exercise behavior (answers ranged from "never" to "daily"), and attitudes toward eating and weight. Bisexual women, as they were determined by researchers' dependence upon survey responses, were not included in the study sample. Ages of participants ranged from 20 to 86 years, with a mean age of 42.6 years. Seventy-eight percent of the lesbians and seventy-four percent of the heterosexual women identified themselves as European American, while eleven percent of the lesbians and fourteen percent of the heterosexual women were African American. The remaining 11% of lesbians and 12% of heterosexual women were equally distributed across ethnic groups. Lesbians had slightly higher BMIs than did heterosexual women (29.6 vs. 28.2), and a slightly higher percentage of lesbians were above a BMI of 27.3 (36.8% for lesbians vs. 30.1% for heterosexual women). Lesbians had slightly more positive body images than did heterosexual women.

Patton et al. (19) recruited 71 lesbians and 77 heterosexual women between the ages of 30 and 50 years from community sources. Sexual orientation was determined by self-identification, and bisexual women were not included in the study. All participants identified themselves as white. Health professionals took the height and weight measurements. BMI was similar in the two groups. Lesbians reported taking part in a higher proportion of both regular and of weightbearing exercise. Among the exercisers, the number of years they had been exercising and the number of minutes per week they currently exercised were found to be similar for lesbians and heterosexual women.

A community sample of 570 self-identified lesbians was recruited using convenience sampling to identify the rates of breast cancer risk factors in SMW (20,21). The sample consisted of 72% white, 19.8% black, 3.9% Native American, 2.8% Latina, and 1.6% Asian/Pacific Islander or other groups. Sexual orientation, ethnicity, and height and weight measures were all self-reported. The prevalence of overweight, defined as a BMI of 27.3 in this sample, was 26.8%.

Roberts et al. (22) used convenience sampling methods to recruit self-identified lesbians over age 40 and their heterosexual sisters as a control group. A total of 324 sister pairs completed a 90-item self-report questionnaire. Lesbians reported a significantly higher BMI (26.5 vs. 25.4, P = 0.016), higher waist circumference (34.2 vs. 32.4, P < 0.001), and higher waist-to-hip ratio (0.82 vs. 0.80, P < 0.001). These differences remained even after controlling for age and education. Lesbians were also significantly less likely to have eaten red meat in the previous one year (P < 0.001), but they did not significantly differ from their heterosexual sisters in their intake of fat or vegetarian food. Lesbians were more likely than their sisters to report a history of gaining and losing 10 pounds more than once (43% vs. 34%, P = 0.013). Lesbians and their heterosexual sisters did not differ in the number of times they exercised per week, the length of the exercise session, or vigor of exercise; however, lesbians were more likely to exercise at least weekly (80.8% vs. 72.2%, P < 0.01).

Schneider et al. (23) surveyed a randomly selected sample of 805 employees at a large health care corporation, and selected demographically comparable groups of self-identified lesbians, heterosexual women and men, and gay men to compare weight, ideal weight, several aspects of satisfaction with weight, and disordered eating (e.g., binging, purging, overeating, etc.). The stratified sample included 25 lesbians, 50 gay men, 75 heterosexual women, and 75 heterosexual men. The mean ages of the groups ranged from 34.6 to 36.3 years; ethnicity data were not collected. Height and weight measurements were self-reported. The BMIs measured varied across groups. Heterosexual men had the highest BMI (24.5), while heterosexual women and gay men had similar BMIs (23.1 and 23.0, respectively). Lesbians reported a mean BMI of 22.4. In general, lesbians were more satisfied with their current weight than either heterosexual women or gay men and reported less disordered eating than either of these two groups. In fact, lesbians resembled heterosexual men in several aspects related to attitudes toward eating and weight.

Strong et al. (24) recruited 89 self-identified lesbians from the student population and community and recruited 116 heterosexual women for comparison from undergraduate classes at a university. Participants completed an anonymous survey. The heterosexual females surveyed had a median age of 21, and 94 of the 112 were white. Of the 89 lesbians surveyed, 70 were white and the median age was 23. Compared to heterosexual women, lesbians reported higher BMIs (23.08 vs. 21.48, P < 0.05) but had less concern with their appearance and less perceived influence from the media on attitudes towards thinness. Lesbians and heterosexual women did not differ as regards eating disorder symptoms.

Yancey et al. (25) conducted a community-based anonymous survey of ethnically diverse bisexual women and lesbians living in Los Angeles County (n = 1,152). Self-reported height and weight, and sexual orientation measures were used. Almost 17% of the sample women were under 30 years of age, while those from the ages of 30–39 and 40–49 each made up 30% of the sample while the 50 or older age group made up 23% of the sample. Seven hundred and ninety-one of the women in the sample identified themselves as white, 162 as Hispanic, 127 as African American, and 72 as Asian/Pacific Islander. The mean BMI of participants was 27.0, with 48.1% of participants reporting a BMI > 25. Significant ethnic/racial differences were found, with African American and Hispanic women reporting higher BMIs than white and Asian/Pacific Islander women. The majority of participants reported that their current weight was greater than their desired weight. A total of 45.3% of participants reported exercising at least three times weekly. Significant ethnic/racial differences in exercising were found, with Hispanic and African American women exercising significantly less than Asian/Pacific Islander and white women.

Discussion
This review has documented the available and published literature on body weight, obesity, and obesity-related issues in SMW. The studies reviewed here begin to reveal patterns in obesity differences between sexual minority and heterosexual women.

One pattern that can be seen in Table 1 is that, according to the reviewed studies, more lesbians than heterosexual women are overweight or obese. The studies cited here suggest, but do not confirm, this difference. Of 19 studies, nine found higher weight or obesity rates among lesbians than control heterosexual samples, five found no differences in obesity or overweight levels, and four studies did not report comparisons. The differences found range from 1 to 5 pounds, so the exact difference is difficult to pinpoint. Also, some studies reported weight levels and compared them among sub-samples, while others focused on obesity (weight or BMI over a certain cutoff point) as the outcome. Despite these variations in reporting, the emerging picture suggests that obesity is an important problem among SMW.

A large flaw in all of the studies cited is the lack of high-quality methodological choices in the study of SMW. None of the samples recruited were population based and, as such, cannot claim to represent the population from which they were drawn. This is a considerable problem, as has been noted previously (26). Other methodological flaws include an almost exclusive reliance on cross-sectional data, the lack of consistent and thoughtfully recruited heterosexual control groups in five of the studies, and lack of consistent measures of sexual orientation. We cannot say which of these issues contributes to the findings as presented, because there are no studies that are actually conducted well enough in all cases that can be used in making such statements.

This association between obesity, which is a clear health hazard, and the demographic category of minority sexual orientation should lead to strong policy recommendations and increased funding for research in this community. Despite a clear need for further research, the Strategic Plan for National Institutes of Health Obesity Research, which was issued by the National Institutes of Health in 2004, makes no mention of sexual orientation disparities in its discussion of priority funding areas for obesity research (27). Often in minority and underserved groups, such as ethnic minority groups, recent immigrants, and others outside the dominant culture, specific targeted interventions that attend to culturally appropriate messages and motivators are called for. This could be an additional area of future research. Access to culturally sensitive care within health care systems has long been a barrier to full and appropriate care for SMW (4,28,29). Here, too, in considering aid for weight control in the SMW community so as to reduce this potential health burden, we must consider specific interventions sensitive to the community's needs, issues, and culture. Future research should examine the relative efficacy of general vs. culturally specific interventions among SMW.

The data for differences between SMW and heterosexual women on fitness-related variables were mixed. Of the four studies that measured and reported exercise levels in SMW compared to control groups, two found higher levels of overall exercise among SMW, one found no differences overall but higher levels of vigorous exercise, and one found that SMW reported lower levels of exercise. None of these studies recruited a representative sample and all were cross-sectional, highlighting again the need for better and more consistent methodology in future research. These confusing patterns indicated that we do not yet have enough research to make conclusions about differences in exercise level according to sexual orientation. It may be that the differences reported above are simply due to sampling methods and that there are no true differences in activity between these two groups of women. Or, alternatively, it may be that while subgroups of SMW participate in high levels of activity (i.e., "jocks" or "athletes"), the majority of SMW engage in activity similar to or less than their heterosexual peers.

Similarly, very few studies reported differences in food, energy, or fruit/vegetable intake between SMW and heterosexual women. Of the two studies that reported actual eating patterns of SMW compared to heterosexual controls, one found that SMW reported poorer eating patterns (fewer fruits and vegetables per day) and one found reports of more healthy eating patterns (less red meat) in SMW. Again, these studies do not point to major and consistent deficits in eating patterns among SMW, but the small number of studies hampers our ability to draw strong conclusions at this time.

There are both limitations and strengths to this review that should be considered when using these findings. One key limitation occurred both in the published literature and in our review process. Most of the studies combined the bisexual women together with lesbians in the study analyses. In our review of the studies, there were not usually enough data on bisexual women to report them separately. It is likely that bisexual women may be very different on some variables, compared to women who partner exclusively with women. For example, Taub (30), in a qualitative investigation of beauty norms among bisexual women, found that the majority of participants reported having more concern about their appearance when they were dating men than when they were dating women. This issue certainly deserves research attention, as the few studies that do compare self-labeled lesbians with bisexual women (e.g., Valanis et al. (8) do find differences among the multiple sexual orientations).

Another complication in conducting the review was the sampling quality and diversity seen in these studies. Sampling quality has been identified as a major problem in the study of the health of SMW (4,31). Among the 19 studies reviewed here, there were no real population-based sampling methods used. This makes it difficult, if not impossible, to estimate rates of obesity and overweight, for example, in the general population. Until we are able to confirm the findings of this review in a representative sample, we must use care when quantifying the findings. This has been a difficulty in much of the published research on SMW (31). The strengths of the review are the assembly of a body of studies in this hard-to-study target group, the number of consistent findings in some of the areas, and the identification of opportunities for research into new areas of health promotion in an underserved community. With these studies we can move forward into research on interventions that could focus on weight management through eating and activity. Intervention research focused specifically on the SMW population is relatively rare in the literature, especially interventions that have been rigorously evaluated with strong randomized designs. These are needed to determine the efficacy of methods to prevent or reduce overweight and obesity in SMW. Also, given the relative lack of efficacy of traditional weight loss programs (32), exploring new and untested directions in preventing overweight or obesity in SMW could prove quite valuable.

How might this review be useful in designing prevention interventions for SMW? This review points to a disparity in a condition (obesity) that might have social and behavioral causes, such as overeating and lack of exercise, potential body image differences, and others. The next step is to determine the mechanism of higher levels in obesity in this target group. Typically, a strong motivation for women to participate in interventions to change eating behaviors or physical activity behaviors is that they might lose weight. If SMW are more overweight, then addressing SMW to change their behaviors might be reasonable. Identification of the motivations and issues surrounding obesity in SMW would be the next step. Indeed, a focus on health and health maintenance might even promote more mental health and well-being in all women, as the focus on ideal body weight has been hypothesized to relate to disordered eating in general samples of women (23).

Disclosure
The authors declared no conflict of interest.

REFERENCES
Dubbert PM. Physical activity and exercise: recent advances and current challenges. J Consult Clin Psychol 2002;70:526–536.
United States Public Health Service Office of the Surgeon General, National Center for Chronic Disease Prevention and Health Promotion (U.S.), President's Council on Physical Fitness and Sports (U.S.). Physical activity and health: a report of the Surgeon General. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention President's Council on Physical Fitness and Sports, 1996.
Ball K, Crawford D. Socioeconomic status and weight change in adults: a review. Soc Sci Med 2005;60:1987–2010.  ISI
Committee on Lesbian Health Research Priorities. Lesbian Health: Current Assessment and Directions for the Future. Institute of Medicine, National Academy Press: Washington DC, 1999.
Department of Health and Human Services, National Institutes of Health, Gay and Lesbian Medical Association. Scientific Workshop on Lesbian Health 2000: Steps for Implementing the IOM Report. Department of Health and Human Services: Washington, DC, 2000.
Aaron DJ, Markovic N, Danielson ME et al. Behavioral risk factors for disease and preventive health practices among lesbians. Am J Public Health 2001;91:972–975. 
Cochran SD, Mays VM, Bowen DJ et al. Cancer-related risk indicators and preventive screening behaviors among lesbians and bisexual women. Am J Public Health 2001;91:591–597. 
Valanis BG, Bowen DJ, Bassford T et al. Sexual orientation and health: comparisons in the Women's Health Initiative sample. Arch Fam Med 2000;9:843–853. 
Case P, Austin SB, Hunter DJ et al. Sexual orientation, health risk factors, and physical functioning in the Nurses' Health Study II. J Womens Health (Larchmt) 2004;13:1033–1047.
Dibble SL, Roberts SA, Robertson PA, Paul SM. Risk factors for ovarian cancer: lesbian and heterosexual women. Oncol Nurs Forum 2002;29:E1–E7.
Roberts S, Dibble S, Scanlon J et al. Difference in risk factors for breast cancer: lesbian and heterosexual women. J Gay Lesbian Med Assoc 1998;2:93–101.
Cogan JC. Lesbians walk the tightrope of beauty: think is in but femme is out. J Lesbian Studies 1999;3:77–89.
Heffernan K. Eating disorders and weight concern among lesbians. Intl J Eat Disord 1996;19:127–138.
Herzog DB, Newman KL, Yeh CJ, Warshaw M. Body image satisfaction in homosexual and heterosexual women. Intl J Eat Disord 1992;11:391–396.
Koh AS. Use of preventive health behaviors by lesbian, bisexual, and heterosexual women: questionnaire survey. West J Med 2000;172:379–384. 
Ludwig MR, Brownell KD. Lesbians, bisexual women, and body image: an investigation of gender roles and social group affiliation. Intl J Eat Disord 1999;25:89–97.
Moore F, Keel PK. Influence of sexual orientation and age on disordered eating attitudes and behaviors in women. Intl J Eat Disord 2003;34:370–374.
Owens LK, Hughes TL, Owens-Nicholson D. The effects of sexual orientation on body image and attitudes about eating and weight. J Lesbian Studies 2003;7:15–33.
Patton CL, Millard PS, Kessenich CR et al. Screening calcaneal ultrasound and risk factors for osteoporosis among lesbians and heterosexual women. J Womens Health 1998;7:909–915. 
Rankow EJ. Lesbian health issues and cultural sensitivity training for providers in the primary care setting: results of a pilot intervention. J Gay Lesbian Med Assoc 1997;1:227–234.
Rankow EJ, Tessaro I. Mammography and risk factors for breast cancer in lesbian and bisexual women. Am J Health Behav 1998;22:403–410.
Roberts SA, Dibble SL, Nussey B, Casey K. Cardiovascular disease risk in lesbian women. Womens Health Issues 2003;13:167–174.
Schneider JA, O'Leary A, Jenkins SR. Gender, sexual orientation, and disordered eating. Psychol Health 1995;10:113–128.
Strong SM, Williamson DA, Netemeyer RG, Geer JH. Eating disorder symptoms and concerns about body differ as a function of gender and sexual orientation. J Soc Clin Psychol 2000;19:240–255.
Yancey AK, Cochran SD, Corliss HL, Mays VM. Correlates of overweight and obesity among lesbian and bisexual women. Prev Med 2003;36:676–683.
Bowen DJ, Boehmer U. The lack of cancer surveillance data on sexual minorities and strategies for change. Cancer Causes Control 2007;18:343–349.
National Institutes of Health. Strategic Plan for NIH Obestiy Research: A Report of the NIH Obesity Research Task Force. In. Edited by Health NIo: US Department of Health and Human Services; 2004.
Cochran SD, Mays VM. Disclosure of sexual preference to physicians by black lesbian and bisexual women. West J Med 1988;149:616–619. 
Kaiser Permanente National Diversity Council, Kaiser Permanente National Diversity Department. A Provider's Handbook on Culturally Competent Care: Lesbian, Gay, Bisexual and Transgendered Population. Kaiser Permanente: Oakland, CA, 2000.
Taub J. Bisexual women and beauty norms: a qualitative examination of the effects of coming out and gender of partner. In: Cogan JC, Erickson JM (eds). Lesbians, Levis, and Lipstick: The Meaning of Beauty in our Lives. Haworth Press: New York, 1999, pp 27–36.
Bowen D, Bradford J, Powers D. Comparing sexual minority status across sampling methods and populations. Women Health 2006;44:121–134.
National Institutes of Health: National Heart Lung and Blood Institute. Clinical Guidelines on the Indentification, Evaluation, and Treatment of Overweight and Obesity in Adults: The Evidence Report. Rockville, MD: National Institutes of Health: National Heart Lung and Blood Institute; 1998.

Ler artigo no contexto original