“I have to constantly prove to myself, to people, that I fit the bill”: Perspectives on weight and shape control behaviors among low-income, ethnically diverse young transgender women

Abstract

The impact of societal femininity ideals on disordered eating behaviors in non-transgender women has been well described, but scant research has explored these processes among transgender women. The present study explored weight and shape control behaviors among low-income, ethnically diverse young transgender women at high risk for HIV or living with HIV in a Northeastern metropolitan area. Semi-structured in-depth interviews were conducted with 21 participants (ages 18–31 years; mean annual income <$10,000; ethnic identity: Multiracial [n = 8], Black [n = 4], Latina [n = 4], White [n = 4], Asian [n = 1]). Interviews were transcribed and double-coded using a template organizing method, guided by ecosocial theory and a gender affirmation framework. Of 21 participants, 16 reported engaging in past-year disordered eating or weight and shape control behaviors, including binge eating, fasting, vomiting, and laxative use. Study participants described using a variety of strategies to address body image concerns in the context of gender-related and other discriminatory experiences, which shaped participants’ access to social and material resources as well as stress and coping behaviors. Disordered weight and shape control behaviors were discussed in relation to four emergent themes: (1) gender socialization and the development of femininity ideals, (2) experiences of stigma and discrimination, (3) biological processes, and (4) multi-level sources of strength and resilience. This formative study provides insight into disordered eating and weight and shape control behaviors among at-risk transgender women, illuminating avenues for future research, treatment, and public health intervention.

1. Introduction

The impact of societal femininity ideals on disordered eating behaviors in non-transgender women has been well described (Bordo, 1993; Thompson, 1999), but scant research has explored these processes among transgender women. In spite of limited empirical work, social science scholarship suggests that disordered eating and unhealthy weight and shape control behaviors constitute an area of significant public health concern in this underserved population. As feminist scholars have long noted, dominant Western constructions of the “ideal female body” are inextricably bound to historically contingent weight and shape ideals, as well as to long-standing hierarchies of gender, race, class, and physical ability (Bordo, 1993; Gilman, 1999). The hegemonic 20th/21st century Western definition of “feminine beauty” requires a thin and young body, light skin, and straight hair (Bordo, 1993; Patton, 2006). From this vantage point, culture is “not simply contributory but productive of eating disorders” (Bordo, 1993, p. 50). Epidemiological and psychological research with young cisgender (i.e., non-transgender) women has supported the claim that gendered societal thinness ideals contribute to body dissatisfaction (Paquette and Raine, 2004; Stice and Whitenton, 2002), which in turn contributes to unhealthy weight and shape control behaviors (Grigg et al., 1996; Sonneville et al., 2012).

Young transgender women (i.e., those whose gender identity—such as woman, transwoman, MtF, or genderqueer—differs from their assigned male sex at birth) are likely to be affected by these gendered beauty ideals promulgated in popular media and public discourse. At the same time, transgender women may face additional gender-related social stressors (Hendricks and Testa, 2012), including: discrimination in school, employment, or housing; sexual objectification; harassment and violence victimization; and lack of access to gender affirming health care (Bradford et al., 2012; Grant et al., 2011; Institute of Medicine (2011); Lombardi et al., 2001; White Hughto et al., 2015). In the context of such interpersonal and structural stigma, societal femininity norms may take on a heightened importance; indeed, for some transgender women, conforming to such norms may be part of “passing” as non-transgender, which may be essential to managing stigma, personal safety and even survival (Sevelius, 2013).

Exposure to psychosocial and material stressors, such as bullying and financial hardship, have been linked to disordered eating and weight control in cisgender populations (Copeland et al., 2015; Stephen et al., 2014). Further suggesting the role of social stress-related pathways, sexual minority youth have elevated risk of disordered weight and shape control compared to their heterosexual counterparts (Austin et al., 2013). Recent literature has also begun to illuminate the intersection of racism and gender in shaping body image and body dissatisfaction, particularly with regard to the experiences of Black and African American women as they negotiate both Black cultural beauty ideals and racist media messages that devalue or exoticize Black women's bodies (Burk, 2015; Capodilupo and Kim, 2014; Patton, 2006). There is evidence that adolescent boys of color have elevated risk of disordered weight control behaviors compared to white boys and that Latina and Native American/American Indian girls have similar or higher risk compared to white girls (Austin et al., 2011; Johnson et al., 2002; Neumark-Sztainer et al., 2002).

Taken together, these factors suggest that there may be multiple, interacting pathways to the embodiment of health inequities related to harmful weight and shape control behaviors among ethnically diverse young transgender women. Indeed, preliminary evidence for eating disorders risk in transgender populations is troubling: in a study of 280,000 U.S. college students, transgender students had two times greater odds of a past-year diagnosis of anorexia nervosa or bulimia nervosa compared to cisgender heterosexual women (Diemer et al., 2015). Very little research has focused on transgender women, who may have distinct risk profiles (Ålgars et al., 2012; Vocks et al., 2009). To our knowledge, no studies have specifically explored harmful weight and shape control behaviors (including binge eating, diet pill use, vomiting, and laxative abuse) from the perspective of young transgender women or examined the contextual factors that might drive these behaviors. Our study sought to fill this gap.

1.1. Theory and study aims

This study draws on two theoretical frameworks relevant to the study of transgender health inequities: ecosocial theory and the gender affirmation framework. Ecosocial theory draws attention to the importance of embodiment, or the ways that people incorporate, biologically, their lived social and material conditions into their bodies (Krieger, 2012). Ecosocial theory articulates multiple “pathways of embodiment” that influence distribution of health and illness in a population at multiple levels over the life course and across historical generations. For example, ecosocial theory encourages scholars to explicitly focus on the drivers of health inequities (e.g., the disproportionate burden of eating disorders among women compared to men). Multilevel pathways to the embodiment of high risk weight and shape control behaviors could include adverse exposure at the individual, institutional, and structural levels, in such domains as: social trauma (e.g., gender and racial discrimination), social and economic deprivation, and targeted marketing of harmful commodities (e.g., harmful products to change weight and shape). This theory was used as an overarching framework to structure inquiry, including informing interview guide development and creation of the preliminary analytic codes (described below).

The gender affirmation framework was developed to contextualize high-risk health behaviors among transgender women of color (Sevelius, 2013). Gender affirmation is “an interpersonal, interactive process whereby a person receives social recognition and support for their gender identity and expression” (Sevelius, 2013, p. 676). The framework posits that insufficient access to gender affirmation together with intersecting forms of social oppression (racism, sexism, transphobia, and poverty) create high-risk contexts, which lead to high-risk health behaviors. Applying the gender affirmation framework to the broader structure of ecosocial theory suggested pathways of embodiment of high-risk weight and shape control behaviors that might be particularly relevant to transgender women of color (see Fig. 1). Informed by these theories, we designed a qualitative, descriptive research study to explore:

1. How do young transgender women engage with weight and shape control behaviors?

2. What are the contexts that influence body image and structure the use of high-risk weight and shape control behaviors among young transgender women?

Fig. 1.

Conceptual framework for potential pathways to embodiment of high-risk weight and shape control behaviors in low-income transgender women.

2. Methods

2.1. Setting & participants

All participants in this study were recruited from an ongoing HIV behavioral intervention trial (hereafter, “the parent study”), developed by and for young transgender women (Garofalo et al., 2012). Eligibility criteria for the trial were: age 16–29 years at enrollment; living in the Boston metropolitan area; able to understand English; assigned a male sex at birth and currently identify as a transgender woman, female, or on a transfeminine spectrum; and at risk for HIV or living with HIV (e.g., engaged in unprotected anal or vaginal sex in the four months prior to enrollment). To be eligible for our study, participants had to be ages 18 years or older and had completed their 12-month follow-up visit for the parent study. Staff from the parent study provided study flyers to eligible participants at the end of a parent study visit or via email or text message; participants then met with the lead investigator to ask questions and enroll. At the time we started recruitment for this study, 100 participants were enrolled in the parent study, of whom 51 had completed or would soon complete their 12-month follow-up and were thus immediately eligible to receive recruitment materials (see Supplemental Table S1). This study was approved by the Fenway Health Institutional Review Board.

As shown in Table 1, the data reported here include 21 participants, ages 18–32 years (mean age = 24 years, SD = 4.3 years). The majority of participants were women of color (n = 17) and born in the U.S. or a U.S. territory (n = 19). Participants were very low income (16 reported an annual income in 2013 less than $6000), about half (n = 10) reported unstable housing in the prior six months (including living in a hotel, group home, on the street, or having no fixed address), and only three reported current full-time employment. For comparison, a probability sample of Massachusetts adults (18–64 years), found that nearly one-third of trans-gender adults were living at or below the poverty level and one-third were unemployed (compared to 9% and 12% of non-transgender adults in Massachusetts, respectively) (Conron et al., 2012).

Table 1.

Demographic characteristics of study participants (n = 21).

	Total
Gender identity
Female	11
Transgender	4
Male-to-Female (MTF)	3
Transsexual	1
Genderqueer	1
Not listed: Demi-girl	1
Race/Ethnicity
Black/African American	4
Latina	4
White	4
Asian/Asian American	1
Multiracial	8
Nativity
U.S. (incl. Puerto Rico)	19
Outside U.S.	2
Sexual orientation identity
Heterosexual or Straight	11
Lesbian, Gay or Homosexual	2
Bisexual	2
Queer	2
Not listed: Pansexual	2
Not listed: Transgender woman	1
Don't know	1
Education (highest level completed)
Middle school	1
Some high school	3
HS grad or GED	9
Some college	5
College degree	3
Income, past year
<$6000	16
$6000–$11,999	2
$18,000–$23,999	1
$24,000–$29,999	1
Not reported	1
Employment status
Full-time (>30 h/wk)	3
Part-time	5
Unemployed	13
Disabled	3
Student	3
Body Mass Index (BMI)
Underweight (<18.5)	1
Recommended (18.5–24.9)	7
Overweight (25.0–29.9)	8
Obese (30.0 and above)	4
Not reported	1
Age in years (mean [SD])	24.4 [4.3]

Notes. Employment status counts do not add to 21 because categories were not exclusive (check all that apply). BMI (kg/m²) was calculated using self-reported height and weight and categories were constructed using CDC guidelines for each category.

2.2. Interview guide

An interview guide was developed based on the previously described theoretical frameworks and review of the literature on (1) disordered weight control behaviors in cisgender women (Stice, 2002; Stice and Whitenton, 2002; Thompson, 1999), and (2) contextual influences on health risk behaviors in transgender women (Clements-Nolle et al., 2001; Nemoto et al., 2004; Reisner et al., 2014). The interview guide was revised following pilot-testing and review by both transgender- and cisgender-identified study staff who had extensive experience working within trans-gender communities. Interview topic areas covered: body dis/satisfaction; needs for gender affirming body change; stress and coping behaviors; social influences on feelings about appearance; and past or current experiences with weight and shape control behaviors (including dieting, exercise, laxative use, vomiting, fasting, diet pills or other over-the-counter weight loss products, and prescription medications). Reports of weight and shape control behaviors were followed up with probes about the behavior and context in which it occurred. A post-interview questionnaire was developed to capture relevant demographic characteristics, body image, and weight and shape control information using standardized measures from population-based and topical health surveys (Eaton et al., 2010; Field et al., 2004; Grant et al., 2011; Neumark-Sztainer et al., 2014). Interview guide and post-interview questionnaire are provided as electronic Supplemental Materials.

2.3. Data collection

Following the informed consent process, 21 semi-structured in-depth interviews were conducted in a private location at a community health center in a U.S. city in June–September 2014. In-depth interviews were selected as the optimal method for gathering formative information on individuals' perceptions of body image and weight and shape control behaviors (Ulin et al., 2005). In addition, because peer relationships can mediate or exacerbate body image concerns (Thompson, 1999), we decided against using social methods such as focus groups at this time. Study visits lasted 55–90 min, including approximately 45–75 min for the in-depth interview and 10 min for the questionnaire. Participants received a $50 gift card for their participation. All interviews were conducted by the lead author, who identifies as a white cisgender queer woman. Interviews were audio recorded and transcribed verbatim for analysis; permission to record and transcribe was included in the consent process.

2.4. Analysis

Analysis was conducted in iterative phases using a template organizing style, which pairs a deductive approach structured by an underlying theoretical framework with the flexibility to follow emergent themes (Crabtree and Miller, 1992). The lead author reviewed all transcripts and proposed preliminary codes guided by ecosocial theory and the gender affirmation framework, which were revised by the study team. All transcripts were read and double-coded by the principal investigator (AG) and the parent study project manager (JWH), using an iterative process that involved periodic meetings to identify gaps, resolve discrepancies, and finalize codes (resulting in 10 key codes; see Table S2). Then, following an immersion and crystallization approach (Borkan, 1992), the lead author iteratively reviewed narratives within each of the ten key codes to identify themes and sub-themes related to weight and shape control. To improve validity, interpretations were discussed by the study team and all transcripts were reviewed for counter-narratives within each theme. NVivo 10 software (QSR International) was used for data management.

3. Results

Participant responses to standardized measures of disordered eating and weight control and their narratives yielded complementary empirical evidence regarding pathways to embodiment of high-risk weight and shape control behaviors.

3.1. Embodiment of high-risk weight and shape control behaviors

Based on the post-interview questionnaire, the majority of participants (n = 16) reported having engaged in some level of disordered eating or high-risk weight or shape control behavior in the previous year. Table 2 presents these frequencies. The most commonly reported form of disordered eating behavior was overeating, or the consumption of a large amount of food in a discrete period of time (n = 13); of these participants, two reported overeating with a sense of loss of control, a criteria for binge eating disorder (Striegel-Moore, 2011). Table S3 provides narratives illustrating high-risk forms of weight and shape control behaviors in this sample.

Table 2.

Disordered eating and disordered weight and shape control behaviors among a sample of young, low-income transgender women ages 18–32 years as reported on post-interview questionnaire (n = 21).

	N
Currently trying to …
Lose weight	8
Gain weight	3
Stay the same weight	3
Change shape without changing weight	4
Not trying to do anything about weight	2
Refused	1
Disordered eating behaviors in past year
Overeating with loss of control (binge eating)	2
Any overeating	13
None	5
< Once/month	7
1–3 times/month	3
Once/week	1
> Once/week	2
Behaviors to lose or maintain weight in past year
Fasting (at least a day)	8
< Once/month	3
1–3 times/month	1
Once/week	1
> Once/week	3
Vomiting	3
Laxative use	2
Diet pill use	4
Any of above disordered eating or weight control	16

3.2. Pathways of embodiment: four emergent themes

Four main themes emerged from analysis of the contexts surrounding weight & shape control behaviors in this sample: (1) Gender socialization processes and the development of femininity ideals; (2) Experiences of stigma and discrimination; (3) Biological processes; and (4) Resilience processes. These themes and their sub-themes are described below with illustrative examples. A common context for these themes was an awareness of violence directed against transgender women who do not conform to gender expectations (Lombardi et al., 2001; Miller and Grollman, 2015), whereby many participants described their constant vigilance when flirting or dating and a few participants alluded to experiences of physical violence in the context of sexual encounters or intimate relationships.

3.3. Theme 1: gender socialization, femininity, and weight and shape control

Participants represented a diverse array of gender transition trajectories, including those who lived openly as a girl since middle school, those who transitioned socially (e.g., changed name, pronoun, gender expression) or medically (e.g., accessed gender affirming hormones or surgery) in their 20s, and those who currently identified on a transfeminine spectrum while presenting as gender-fluid or male (i.e., “in boy mode”). Across this diversity, all participants (n = 21) described engaging with femininity from a young age—both their own and cultural notions of “ideal” femininity—with implications for body image and weight or shape control behaviors.

3.3.1. Sub-theme 1.1: feminine thinness ideals and weight and shape control

For some, notions of ideal womanhood and related gender transition goals were intimately linked with weight and shape. As a 27-year old participant (ID 9) explained, “When I started my transition, I was obviously—I wanna be skinny, and this is how I visualize my body. Have been working to where I wanna get my body.” In her narrative, the ideal body is so normative that it is “obvious”; thus, as she transitioned she visualized a skinny body as her goal. Other participants described internalizing the femininity norms of their peers and subsequently adopting weight control behaviors present in those peer groups. As one participant, who also reported a great deal of familial shaming around her eating and weight, commented about high school:

A lot of girls would come up with these inventive ways of how to stay thin like the crash diet. They'd be like, “I just eat cottage cheese. I just have a Tic Tac today. I just had an apple. I just had a piece of bread.” … I wanted to be these girls that these guys saw. What they thought was pretty. I would be like, if I lose weight, or I can fit in this size … (Age 31 years, ID 11)

This narrative integrates the participant's identification with her (cisgender) female peers, who are engaging in weight control behaviors to embody feminine thinness ideals, and her desire to be attractive to her male peers, who she viewed as subscribing to these ideals. She later linked these observations to her own attempts at crash dieting and restricted eating, resulting in episodes of passing out in school. Another striking case of peer-based gender socialization increasing vulnerability to disordered eating came from a participant who self-described as having formerly engaged in (undiagnosed) binge eating and purging. She described learning about negative body talk and disordered weight control from friends at school (before socially transitioning):

I'd be with the other girls at the table … and they'd talk amongst themselves, and they'd … critique each other's physical appearance. They'd say, “Oh, my gosh, like you lost so much weight.” “Look at her; she's so pretty now … Look at how skinny she is.” Then once that specific person would leave the table, they'd be like, “Well, I heard she was resorting to bulimia. She was hospitalized for a while. That's why she's been so long out of school. That's why she got skinny so fast.” That's how … I knew what bulimia was, and what I knew what, allegedly, other girls were doing to lose weight. That's how [I] kind of got motivated to do what I was doing, which was a really unhealthy, dangerous habit of bingeing and purging everything that I'd eat. (Age 21 years, ID 2)

This participant perceived these conversations as prompting her to engage in what she described as unhealthy and dangerous behaviors. These narratives also illustrate the ways that adoption of dominant femininity ideals can provide gender affirmation through affiliating oneself with femininity-linked disordered weight control behaviors.

3.3.2. Sub-theme 1.2: race/ethnicity and alternative femininity ideals

A minority of the Latina- and Black/African American-identified participants talked explicitly about their racial/ethnic identities in relation to alternative femininity ideals. For example, the 27-year old participant above (ID 9) also noted: “That's how I envision my body. Not big breasts, but breasts that would fit my body. Then my stomach to be skinny. Then big hips and a big butt. In the Latin community, that's pretty much what you see.” Some women described a “thicker” body as more, rather than less, feminine or as a personal goal:

My ideal body is just what I have right now, just to lose my stomach and just be nice and curvaceous. I don't wanna be skinny, skinny. I wanna be nice and thick, really proportioned … just looking good and feeling good. I like to be thick. It's not too big. All in the right places … That's my ideal body. (Age 31 years, ID 6)

While this thicker feminine beauty ideal stands in contrast to the pressures of the white/Western thinness ideal described by several of the participants, it did not come without its own pressures. Some participants reported teasing in school for being too thin, while others reported being criticized by other transgender women or gay men for lacking curves.

3.4. Theme 2: stigma & discrimination, gender affirmation, and weight and shape control

All (n = 21) participants described growing up in social environments that were homophobic and transphobic—and for many, weight-stigmatizing (n = 12). For a few participants, the stress of sexual orientation and gender identity development in this context was inextricable from weight control. One participant recalled the stresses of weight stigma and sexual orientation stigma compounding each other during elementary school, a time when she lived as a boy and was grappling with feelings of same-sex attraction:

I was battling my mind back then with what I was thinking, how I was feeling with other guys, and stuff like that … It was just stressful. My elementary school and middle school years was very, very, very, very stressful. I was trying to find myself, and trying to fix my weight. (Age 22 years, ID 20)

Several participants explicitly connected their fears of family rejection to coping-related eating and weight control. One participant described the onset of her binge eating habits when she was around age 16 years:

Cuz it was right after I had lost my virginity. And then I even … I was afraid then, like of, “What if my mother or father ever found out that I just slept with a man?” Like, that was—it was my biggest concern. I was so scared … and that guilt, you know, it just—I started eating more and more. (Age 26 years, ID 7)

A participant whose father had destroyed all of her “non-male-conforming” clothing when she was in high school described a home environment in which weight and gender stigma were mutually reinforcing:

Well, cuz when I'm over there [father's house], I'm more in male mode, in a sense, just cuz I haven't told him yet, but I kind of told him—the hormone meds cuz he always asks, “What are they for?” I'm, like, “Oh, they're for my [health condition],” cuz I don't really wanna tell him yet. He was saying how I'm really gaining weight and I'm gonna have to work out or something so I can lose those bitch tits, in a sense. That was, like, male-male. [Chuckles] … It's the way he thinks. It's how he thinks, so I'm cool with it, but, yeah, he doesn't really know that I'm trans. … Just, like, “Yeah, I'm working on it, trying to lose weight.” (Age 22 years, ID 21)

The above narrative suggests some parents or peers might manage their anxieties or prejudices about gender identity by channeling them into more socially acceptable narratives about weight gain and fitness. This participant, in turn, may have been managing her father's transphobia and weight stigma and protecting herself by affirming the narrative of weight loss rather than disclosing her gender status.

3.4.1. Sub-theme 2.1: unmet needs for gender affirmation and weight and shape control

Participants described the pressures of being held to stricter standards than cisgender women, pressures which might lead to disordered eating. One participant explained that even among her family,

I have to be like this ultra-feminine woman … My family's always nitpicking everything I do, like, “Don't sit that way. Women don't sit that way” … Meanwhile, when my other cis cousins go about doing that stuff, nobody tells anything to them. “Cause at the end of the day, they don't stop being women because they were born women. Meanwhile, I have to constantly prove to myself, to people, that I fit the bill.” (Age 21 years, ID 2)

For this participant, who was quoted earlier regarding her binge eating and purging behaviors in high school, lack of gender affirmation led to heightened vigilance, a feeling that she had to “constantly prove” herself and faced rejection if she did not “fit the bill” of her family's femininity ideals.

3.5. Theme 3: biological processes in interaction with weight and shape

Ecosocial theory describes embodiment as a process of dynamic engagement between the biological and social worlds. This held for the embodiment of weight and shape control behaviors among study participants. The primary sub-theme that emerged concerned the effect of gender affirming cross-sex hormones on body satisfaction and dissatisfaction. The majority of participants reported having currently or previously taken hormones in order to change their appearance (n = 15). Of these, 10 participants described hormones as affecting their weight—typically, through increased appetite or weight, although one participant reported significant weight loss.

Most participants taking hormones reported increased body satisfaction and an increased sense of wellbeing. Participants largely attributed this satisfaction to hormones' gender affirming effects (e.g., feeling “so much better” after hormones contributed to a more “feminine physique”). However, for about half, hormone use also accompanied increased weight dissatisfaction. For some, this centered on frustration with increased appetite; as one participant put it, after taking a hormone pill “you're ready to eat the entire world” (Age 20 years, ID 18). Belly fat was particularly salient for some of the participants, sometimes with a sense of inevitability and solidarity (“we call it 'mone belly sometimes”) and sometimes with great frustration and even pain. Several described increases in weight or belly fat as a kind of foil to ideal womanhood:

Being on hormones, your weight is such a struggle as a woman. Because as a woman, you put on weight a lot quicker. It's a lot harder to take weight off. You maintain more fat. Through the hormones, it makes you eat more … I put on 65 pounds. As a woman, it makes me feel gross and disgusting. (Age 31 years, ID 11)

While facing the biological reality of a changing metabolism and fat distribution due to hormone use, some participants were also navigating broader Western cultural scripts of “overweight” female bodies as deviant and disgusting. Thus, the interaction of cultural femininity ideals and hormone-related weight change presented a challenge to the benefits of gender affirming hormone use.

Several participants talked about how hormone-related weight or appetite increases motivated them to want to be more “healthy,” which for some included healthful strategies like physical activity, but for others prompted the dangerous misuse of products, such as laxatives, as for this participant: “My estrogen says to eat. My hormones say to eat, honey, and I just maybe just try to live healthy now. I just maybe go to—I try to fight it, and I just do a herbal lifer. A detoxifying flush. When I have a few extra bucks, it's like it cleans ya” (Age 24 years, ID 1).

One participant highlighted another dimension of biology: weight control as a response to psychological distress due to undergoing puberty in the wrong gender. Specifically, this participant, who was the only participant who had received treatment for an eating disorder, reported:

[The binge eating and purging started] not long before I was there [at the residential treatment facility] for the self-injury and stuff. Just because I'd see other people and girls, the way their bodies were forming and mine and I'm becoming too masculine. Like, this is something—I need to do something. (Age 20 years, ID 3)

This participant articulated a link between their own increasing gender dysphoria as male puberty progressed, body comparison with other girls, and conscious decision to remedy this dissonance through binge eating and purging.

3.6. Theme 4: stress, resilience, and positive body image development

While many participants reported previously or currently engaging in some degree of unhealthy weight or shape control behavior, only a minority of participants described severe disordered eating (n = 5). In fact, agency and resilience threaded throughout the weight and shape control narratives. At the individual level, some grappled with ways to deflect negative comments from others and reinforce body satisfaction through positive body talk: “Worry about your own weight, because at the end of the day I wanna be thicker, but I'm so happy with my body. Just because I want little things changed doesn't mean I'm not satisfied with who I am” (Age 19 years, ID 5). Some participants employed critical analysis and media literacy to protect themselves from pervasive marketing of unrealistic femininity ideals. One participant described growing up watching her mother “go through every single popular diet plan” of the 1980s–2000s. Her conclusion: “That every diet plan that I've seen, with the exception of just ‘eat what you need' was just to get money” (Age 28 years, ID 13). Another participant explained that she finds popular media to be “very objectifying of women” and noted “lots of fatphobia. Lots of photoshopping people so they look more suitable for their magazine,” to the extent that “over the past few years I've just completely cut myself off from any news sources … I just don't like the messages that they're trying to send to their audience” (Age 20 years, ID 16).

In addition to internal, individual-level resources, many participants reported their body confidence being bolstered by the gender affirming support of intimate partners, friends and community members. About one-third of participants described feeling supported and affirmed in the context of their intimate relationships in ways that benefited their body image and weight perceptions. For some this came in the form of an external counter-narrative to negative self-talk, as for the participant who reported disliking her weight but noted, “my man person, boo thing doesn't really want me to lose weight … I don't think he's saying it just to say it. I think he says it cuz he believes it. That helps me believe it” (Age 20 years, ID 18). For others, a partner might directly address disordered eating behaviors, such as the participant whose partner would encourage her to eat when she was avoiding food: “He'll be like, ‘At least eat half of the TV dinner, because I don't want you getting sick and anything happening to you’” (Age 27 years, ID 9). Friends constituted another important source of community-level resilience, as this participant illustrated:

My friends are like congratulating me and cheering me on, like, “Ooh, your butt's getting bigger. Your boobs getting bigger. You're filling out. You're looking good. You're gaining weight.” Not even from just boys, my friends, too. [Laughter] That part feels good. (Age 21 years, ID 19)

In this instance, her friends' positive body talk helped the participant to feel good about herself. Other participants described the importance of receiving feedback and gender transition-related information through transgender community networks. Some, particularly those with limited access to in-person transgender community due to geographic isolation, mental health, or living in foster care, described the importance of finding online support.

It must be noted that community networks represented important sources of stress as well as strength and that many participants described judgment and policing of what was considered “appropriately feminine” within their own communities:

People have said, “Oh, you're hard.” Or “You have a big back.” Or “Your hands are very big.” Or “… Look at your jaw line.” … It's pretty cruel for another peer of yours to or another transgender peer to look at you and say, “Oh, you're not that feminine.” Or … “You don't fit into some category of being real or realness,” or just all these different catty categories. (Age 24 years, ID 1)

A few participants acknowledged their own biases against transgender women who they perceived as insufficiently womanly or “passable.” Such experiences point to the internalization of transphobia and the normalization of this internalized transphobia at a community level. These within-community stressors could have important implications if weight and shape control behaviors are largely inseparable from gender affirmation.

4. Discussion

This study sheds light on an underserved population at risk for disordered weight and shape control behaviors: socially and economically marginalized adolescent and young adult transgender women and particularly women of color. Past research on disordered weight control has focused almost exclusively on cisgender women (Smink et al., 2012; Thompson, 1999); only a few recent studies have focused on transgender individuals (Ålgars et al., 2012; Diemer et al., 2015; Witcomb et al., 2015). This study found that 16 of 21 participants reported some level of disordered weight and shape control. The ubiquity of disordered weight and shape control behaviors is consistent with the sample's high levels of economic and housing instability, social marginalization, and high risk of psychological distress, each of which has been linked to disordered eating in the general population (Austin et al., 2013; Fournier et al., 2009; Johnson et al., 2002; Neumark-Sztainer et al., 2002). For example, in a representative sample of Massachusetts high school students, those who were unstably housed had 2–3 times greater odds of reporting disordered weight control behaviors relative to non-homeless students; among unstably housed students, 11% reported purging, 25% fasting for weight loss, and 13% diet pill use (compared to 5%, 10% and 4% of non-homeless students, respectively) (Fournier et al., 2009).

The central finding that emerged from our analysis of the contexts of weight and shape control behaviors in this sample was that weight and shape control desires and behaviors were situated at the intersection of (1) gender socialization and femininity ideals, (2) stigma and discrimination, particularly transphobia, (3) biological processes in interaction with social processes, and (4) individual- and community-level resilience. Three key implications stem from this analysis. First, consistent with previous work (Nuttbrock et al., 2009; Sevelius, 2013), gender affirmation is vital to understanding the health and wellbeing of young transgender women, including disordered weight and shape control behaviors. Sexual objectification coupled with stigmatization was a commonly reported experience among women in this study and such experiences have been shown to increase self-objectification and body consciousness, which in turn increase psychological strain, body dissatisfaction, and disordered weight control (Fredrickson et al., 1998; Hebl et al., 2004). This analysis suggested that gender affirmation has a critical and multi-faceted role to play in any interventions seeking to prevent high-risk weight and shape control behaviors among transgender women.

Second, gender affirmation must be considered in relation to both dominant and culturally specific femininity ideals when conceptualizing determinants of weight and shape control behaviors. Gender socialization and gender transition require that women, particularly women of color, negotiate both Western femininity ideals (thinness, whiteness, youthfulness) and competing femininity ideals rooted in racial/ethnic identity or community norms. For transgender women, the stakes are heightened because conforming to societal standards of femininity and “passing” as non-transgender can be necessary for daily safety and survival (Grant et al., 2011; Miller and Grollman, 2015). Across types of relationships (family, peers, partners), gender affirmation was often bundled with narrow prescriptions for socially acceptable expressions of femininity, including acceptable forms of female weight and shape. On the other hand, several women in the study found ways to reject these narrow prescriptions and assert their right to define femininity for themselves. Although beyond the scope of this study, an analysis of transgender women's body image and weight and shape control behaviors within and across racial/ethnic identities and in relation to both racist Western femininity ideals and the specific strengths and constraints of non-dominant cultural femininity ideals is greatly needed.

Third, these findings emphasized biology as an integral component of positive body image development and weight perceptions for young transgender women in the sample. This was particularly striking in participants' descriptions of the effects of hormone use, which was largely viewed as positive while also generating some dissatisfaction related to undesired weight change or fat redistribution. There are several biological pathways through which hormones might influence weight and appetite, although studies of exogenous hormone use in cisgender women have found limited or no evidence that either cause weight gain (Hirschberg, 2012). Clinical guidelines have identified weight gain as a risk of feminizing hormones (Coleman et al., 2012) but there has been scant research on the weight or appetite effects of hormones for transgender populations. One community sample of 16 trans-gender women in Los Angeles found no significant weight change at 6 months post-initiation of hormones (Deutsch et al., 2015). The long-term effects of hormone use on weight and/or appetite in transgender youth and adults is an understudied clinical issue that warrants future research with implications for understanding body image development and disordered weight and shape control behaviors.

4.1. Strengths & limitations

Three primary limitations are noted. First, as a cross-sectional study, this study could not characterize developmental processes or assess temporality, such as timing of potential influences in relation to engagement in weight and shape control behaviors. Instead, our approach was to situate experiences as they were reported to have occurred in broad segments of the life course (e.g., childhood, high school), which allowed for a recognition, if not an analysis, of life course variation. In a qualitative study, generaliz-ability is neither feasible nor sought. However, a second limitation is that this sample was restricted to a group of low-income and ethnically diverse young adult transgender women who were participating in an HIV risk reduction intervention study and who already may have had heightened exposure to discrimination and trauma. The degree of harmful weight control behaviors may not be substantially different if compared to cisgender women in similar social and economic circumstances—but it's nevertheless valuable to focus on this underserved population. Importantly, even within a segment of the broader population of transgender women, there was substantial heterogeneity. Last, this study used an “etic” approach—that is, the study was largely designed, conducted, and analyzed by researchers who are “outside” the study population (as compared with an “emic” approach in which the researchers can count themselves as members of the group being studied). There are strengths and limitations to both approaches and we recognize that, were this study to be replicated from an emic perspective, additional insights might emerge.

5. Conclusion

To our knowledge, this the first study to explicitly examine weight and shape control experiences, as well as multi-level stressors and strengths, among a group of young transgender women. Further, the insights and experiences of the transgender women in this study made visible some of the processes of gender socialization and affirmation that are frequently obscured in U.S. conversations about gender and inequity. Our findings suggest several directions for future inquiry and possible intervention. There is a need for research exploring themes outlined in this study among economically diverse populations of transgender women and among transgender men, who may face distinct gender-related pathways to weight and shape control behaviors. Epidemiologic research is needed to assess absolute and relative risk of disordered eating behaviors among gender minorities across gender identity, race/ethnicity, and socioeconomic position. Efforts are mounting to incorporate gender identity questions on population-based surveys in the U.S. that include topics relevant to weight control, such as the Youth Risk Behavior Surveys. More work is also needed to understand the perspectives of health care providers in the two fields bridged by this analysis: (a) those providing primary care to transgender populations, who may not be screening patients for disordered weight and shape control behaviors; and (b) those providing eating disorders treatment, who may not have expertise in providing care for transgender patients. Finally, study participants described multiple forms of individual- and community-level resilience. These represent important strengths that may be starting points for the development of intervention efforts seeking to facilitate access to safe gender affirming care and the promotion of positive body image development relevant to the needs of young transgender women.