Demographic and Minority Stress Risk Factors for Obesity Among Sexual Minority Youth Assigned Female at Birth

Abstract

Purpose:

Sexual minority youth (adolescents and young adults) assigned female at birth (SM-AFAB) are at disproportionate risk of developing obesity compared with heterosexual cisgender youth AFAB. Grounded in minority stress theory, this study aimed to identify potential risk factors for obesity among SM-AFAB youth to inform the development of prevention and intervention efforts for this high-risk population.

Methods:

Data were collected in 2017 from 367 SM-AFAB youth (ages 16–20 years). Multinominal logistic regression was used to assess cross-sectional associations of race/ethnicity, sexuality, gender identity, household income, and sexual minority (SM) stressors (internalized stigma, microaggressions, and victimization) with weight status (normal, overweight, and obese).

Results:

Roughly half (53.1%) of participants' body mass index were in the normal weight range, with 24.8% in the overweight range and 22.1% in the obese range. Rates of obesity in Black and Latinx participants were 3–4.5 times those of White participants. Bisexual, pansexual, and queer individuals were at greater risk for obesity than gay/lesbian participants; only bisexual participants were at higher risk for overweight. Participants with a household income <$20,000 and between $20,000 and $39,000 were at greater risk for obesity than participants with household income >$80,000. Microaggressions were positively associated with obesity.

Conclusion:

Findings highlight risk for obesity among SM-AFAB youth, particularly for those who identify as racial minority, as low income, as being attracted to more than one gender, and for those who experience high levels of anti-SM microaggressions. Targeted obesity prevention and treatment programs should consider the unique needs, challenges, and strengths of SM-AFAB youth.

Introduction

Obesity is a major public health problem that is associated with increased risk for many health problems, including heart disease, diabetes, and some cancers.¹ Individuals with a higher weight also face weight stigma (e.g., societal stigmatized attitudes toward people in higher weight bodies), which itself is related to chronic disease and mortality.² Consistent evidence demonstrates that sexual minority youth (adolescents and young adults) assigned female at birth (SM-AFAB) at disproportionate risk of obesity compared with heterosexual cisgender youth AFAB.^3,4 Sexual minority (SM) individuals assigned male at birth do not share this risk.⁵ In this study, we aim to identify potential risk factors for obesity among SM-AFAB individuals to inform the development of targeted prevention and intervention efforts.

We focus on adolescence, when rates of obesity begin to rise across the population,^6,7 likely due to change in body composition (e.g., pubertal development encourages AFAB individuals to accumulate body fat) coupled with marked decreases in healthy eating and physical exercise.^8,9 Furthermore, adolescence is when SM disparities in obesity emerge.⁴ Prevention programs are most likely to be effective before the development of obesity; therefore, identification of risk and protective factors for SM-AFAB obesity at this age range may be particularly important.

Our exploration of risk factors for obesity among SM-AFAB youth is grounded in minority stress theory,¹⁰ which some have applied to SM women's increased risk of obesity.¹¹ Minority stress theory emphasizes how individuals with minoritized social identities (e.g., based on race, gender, or sexual identity) are at increased risk for health problems due to societal stigma (e.g., victimization and discrimination based on their identity). Furthermore, due to another type of societal stigma not discussed in minority stress theory, weight stigma, obese SM individuals may then be further marginalized due to their weight, leading to more minority stress and subsequent weight gain.²

Based on this theory, we explored whether obesity among SM-AFAB youth is associated with (1) holding an additional minoritized identity (related to income, race, gender, or specific sexual identity) and (2) experiences of SM stress (i.e., anti-SM victimization, microaggressions, and internalized stigma).

Many SM-AFAB youth hold additional minoritized identities that may be relevant to risk for obesity. Larger proportions of SM than heterosexual individuals live in poverty¹² and identify with minoritized racial groups.¹³ Though both of these demographic factors have been identified as risk factors for obesity in the general population,^14,15 little research has explored their effects among SM youth. A handful of studies suggest that Black and Latinx SM women have higher body mass index (BMI) than White SM women,^16,17 and lower income is associated with obesity among adults;¹⁸ however, these findings require replication in adolescence. Increasing numbers of SM-AFAB youth also identify as gender minorities (i.e., transgender or gender nonbinary [GNB]).¹⁹

Two studies suggest that transgender adults are more likely to be obese than cisgender participants,²⁰ particularly among AFAB individuals.²¹ However, we could find no research on obesity in transgender or GNB adolescents. Finally, mounting evidence suggests that bi+ women (i.e., bisexual, pansexual, queer, etc.) face more stigma than lesbian women,²² which might contribute to obesity risk. The little existing research, however, has not supported this possibility: Some research has found that lesbian adults were more likely to be obese than bisexual women,²³ and another found no difference.²⁴ It is important to explore these issues in today's youth, who are more likely than older cohorts to adopt bi+ sexual identities.²⁵

Obesity among SM-AFAB youth may also be associated with experiences of SM stress,¹⁰ defined as unique social stress based on stigma against SM individuals, including acts of anti-SM victimization and microaggressions (i.e., everyday verbal and nonverbal communications that convey negativity toward SM individuals).¹⁰ SM youth experience minority stressors frequently: More than two-thirds reported verbal harassment based on their sexual orientation and 70% reported hearing antigay remarks during the past year.²⁶ Consequently, many SM youth internalize stigma about their own sexuality.²⁷ Consistent with minority stress theory, these experiences raise risk for many adverse health outcomes, including chronic diseases, poor general health, and mental health problems.²⁸

Minority stress may be similarly associated with obesity: Stress increases cortisol, which is associated with weight gain²⁹ and influences obesity-linked behaviors (e.g., food consumption and physical activity).³⁰ Minority stress may raise risk through reduced community cohesion, social support, and access to quality health care.^10,31 Although prior study has theorized that minority stress may contribute to obesity among SM-AFAB individuals, only one study has tested this empirically.

Controlling for demographic risk factors, health behaviors, and depression, lesbian women with obesity had 2.49 higher odds of experiencing heterosexist discrimination than those of normal weight.³² Further research is needed to explore this issue in youth and with a broader range of minority stressors, including victimization, microaggressions, and internalized stigma.

The current study

This study aimed to explore whether minority identities and minority stress experiences are associated with risk for obesity among SM-AFAB youth (ages 16–20 years). Grounded in minority stress theory, we hypothesized that participants who identified as bisexual, transgender, or Black, and had a lower household income would have a higher likelihood of overweight and obesity than those who did not. We also hypothesized that higher levels of SM victimization, microaggressions, and internalized stigma would be associated with a higher likelihood of overweight and obesity.

Methods

Participants and procedures

Participants were drawn from the adolescent cohort of FAB400 (n = 400), a longitudinal cohort study of sexual and gender minority individuals assigned female at birth (SGM-AFAB) youth, which includes cisgender SM women, transgender men, and nonbinary AFAB youth. Inclusion criteria required participants to be AFAB, speak English, be aged 16–20 years, and identify as a sexual or gender minority. The sample is diverse in race/ethnicity, sexual orientation, and gender identity (Table 1). Participants were recruited at various SM venues and through online social media advertisements. Then, enrolled participants could refer up to five peers to the study and were paid US $10 for each peer they successfully recruited.

Table 1.

Characteristics of the Analytic Sample (N = 367)

Variable	n (%)
Race/ethnicity
Black	104 (28.3)
Latinx	96 (26.2)
Other	60 (16.3)
Asian	22 (6.0)
Multiracial	37 (10.1)
Other	1 (0.3)
White	107 (29.2)
Sexual identity
Gay/Lesbian	70 (19.1)
Bisexual	140 (38.1)
Pansexual	76 (20.7)
Queer	47 (12.8)
Other	34 (9.3)
Unsure/questioning	19 (5.2)
Asexual	11 (3.0)
Not listed	4 (1.1)
Gender identity
Cisgender female	264 (71.9)
Transgender identity	35 (9.5)
Male	19 (5.2)
Transgender	16 (4.4)
Nonbinary/GNC	68 (18.5)
Gender nonconforming	23 (6.3)
Genderqueer	25 (6.8)
Nonbinary	15 (4.1)
Not listed	5 (1.4)
Household income
<$20,000	67 (18.3)
$20,000–$39,000	68 (18.5)
$40,000–$49,000	42 (11.4)
$50,000–$59,000	40 (10.9)
$60,000–$69,000	33 (9.0)
$70,000–$79,000	20 (5.4)
>$80,000	95 (25.9)

	Mean (SD)
Age	18.53 (1.34)
Body mass index	26.46 (6.87)
SM microaggressions	1.77 (0.62)
SM internalized stigma	1.76 (0.60)
LGBT victimization	0.23 (0.32)

GNC, gender nonconforming; SD, standard deviation; SM, sexual minority.

We used data from the baseline assessment, completed in 2017, for which participants were paid $50. The study was approved by the institutional review board at Northwestern University with a waiver of parental permission for participants under 18 years of age under the code of federal regulation's requirements for permission by parents or guardians and for assent by children. Participants provided written informed consent, and a federal certificate of confidentiality was obtained. For these analyses, we removed 29 participants categorized as underweight (BMI ≤18.49 kg/m²), who may have unique risk factors for unhealthy weight. Because the minority stress measures were specific to sexual identity, four gender minority participants who were not sexual minorities were excluded, yielding an analytic sample of 367 SM-AFAB youth.

Measures

Weight status

Self-reported weight was converted to kilograms (kg) and divided by height in meters squared to calculate BMI. For participants aged 20 years and under, BMI was adjusted for age and sex using the Centers for Disease Control and Prevention's (CDC's) BMI growth charts (BMIz). Per the CDC's guidelines, participants were assigned to three weight categories: normal weight (BMIz ≥5th and <85th; BMI ≥18.5 and <25), overweight (BMIz ≥85th and <95th; BMI ≥25 and <30), and obese (BMIz ≥95th; BMI ≥30).³³ We acknowledge that BMI has various deficiencies as a measure of obesity;³⁴ however, it remains a widely used approach to assessing obesity. For the remainder of the article, we will refer to both BMIz and BMI as BMI.

Demographic variables

Participants described their sexual identity as gay, lesbian, bisexual, queer, unsure/questioning, straight/heterosexual, pansexual, asexual, or not listed (please specify). For analyses, responses were coded into gay/lesbian, bisexual, queer, pansexual, and other. There were no straight/heterosexual participants. Participants described their gender identity as male, female, transgender, gender nonconforming (GNC), genderqueer, nonbinary, or not listed (please specify). We coded responses into three groups: (1) cisgender women (female), (2) transgender (male or binary transgender), or (3) nonbinary/GNC (any other gender identity).

Based on self-reported ethnicity and race, participants were classified as Black, White, Latinx, or Other (all other races combined due to low numbers). As recommended by the National Institutes of Health,³⁵ all who selected a Latinx ethnicity were classified as Latinx regardless of race.

Household income

Participants were asked, now, think about who pays for things in your life (food, rent or mortgage, clothing, transportation, school, entertainment, etc.). This may be just you, or may be your parents, or may be both you and a partner. What is the total annual income of these people together? Participants selected one of the following on a Likert-type scale: <$20,000, $20,000–$39,999, $40,000–$49,999, $50,000–$59,999, $60,000–$69,999, $70,000–$79,999, >$80,000.

Minority stress variables

Victimization

Participants rated how frequently they experienced each of 10 types of victimization over the past 6 months³⁶ (e.g., “Have you been verbally insulted because you are, or were thought to be gay, lesbian, bisexual, or trans?”; 0 = never to 5 = more than 10 times). Scores reflect average rating across items (alpha = 0.82).

Sexual orientation microaggressions

On the Sexual Orientation Microaggressions Scale,³⁷ participants rated how frequently they experienced 23 microaggressions over the past month (e.g., “You were told you just haven't found the right person of the opposite sex”; 1 = not at all to 5 = 21–30 times [almost everyday]). Scores reflect the mean rating across items (alpha = 0.94).

SM internalized stigma

Participants rated how strongly they agreed with eight statements³⁸ (e.g., “I think that if I were straight, I would probably be happier”; 1 = strongly disagree to 4 = strongly agree). Transgender and male identified participants were not presented with two items referring to attractions in ways that do not fit with their identities. Higher scores (mean ratings across items; alpha = 0.87) indicated more internalized stigma.

Statistical analysis

In SPSS version 27, we first used single-predictor multinomial logistic regression models to estimate bivariate associations of each of the demographic and minority stress variables with weight category (predicting overweight and obese groups, with normal weight as the reference group). Categorical demographic variables were dummy coded; reference groups were White, lesbian/gay, cisgender, and income >$80,000. Then, to test for unique effects of each proposed risk factor, we ran a multinominal logistic regression with the four demographic variables and three minority stress variables entered simultaneously as predictors of weight category.

Results

Roughly half (53.1%) of the participants' BMIs were in the normal weight range, with 24.8% in the overweight range and 22.1% in the obese range.

Results of single-predictor multinomial regression models are presented in Table 2. Compared with White participants, Black and Latinx participants were approximately twice as likely to be overweight; Black participants were 4.7 and Latinx participants were 3.8 times as likely to be obese. Regarding sexual identity, bisexual individuals were at greater risk for overweight and pansexual individuals at greater risk for obesity than gay/lesbian participants (odds ratio [OR] = 2.31 and 2.42, respectively). Participants in the lowest two income groups were at greater risk for obesity than participants with household incomes >$80,000. Gender identity was not associated with obesity or overweight. No minority stressors were associated with being overweight; only microaggressions were associated with obesity.

Table 2.

Zero-Order Associations of Demographic Characteristics and Minority Stressors with Weight Status

Odds ratios	Obesity (N = 81)	95% CI	SE	Overweight (N = 91)	95% CI	SE
Race
Black	4.67***	2.18–9.99	0.39	2.16*	1.10–4.24	0.34
Latinx	3.77***	1.72–8.23	0.40	2.21*	1.13–4.35	0.34
Other	1.64	0.65–4.16	0.47	1.17	0.53–2.57	0.40
Sexual identity
Bisexual	1.96	0.90–4.30	0.40	2.31*	1.12–4.76	0.37
Queer	1.55	0.58–4.12	0.50	1.31	0.51–3.39	0.49
Pansexual	2.42*	1.04–5.62	0.43	1.49	0.64–3.48	0.43
Other	1.63	0.54–4.85	0.56	1.77	0.65–4.86	0.52
Gender identity
Transgender identity	0.67	0.27–1.64	0.46	0.42	0.15–1.16	0.52
Nonbinary/GNC	0.75	0.37–1.51	0.36	0.87	0.46–1.65	0.33
Household income
<$20,000	3.00*	1.28–7.03	0.44	1.65	0.78–3.50	0.38
$20,000–$39,000	3.75**	1.62–8.68	0.43	1.77	0.83–3.77	0.39
$40,000–$49,000	1.60	0.58–4.39	0.52	0.94	0.38–2.31	0.46
$50,000–$59,000	1.96	0.73–5.26	0.51	0.91	0.36–2.32	0.48
$60,000–$69,000	1.13	1.15–8.53	0.51	1.14	0.42–3.13	0.52
$70,000–$79,000	1.36	0.33–5.64	0.72	1.42	0.47–4.30	0.56
SM microaggressions	1.88**	1.26–2.83	0.21	1.39	0.99–2.11	0.21
SM internalized stigma	1.19	0.77–1.82	0.22	1.16	0.76–1.77	0.21
LGBT victimization	1.14	0.50–2.58	0.42	1.21	0.55–2.62	0.40

Reference group: Dependent variable (normal weight), race (White), sexual identity (gay/lesbian), and gender identity (cisgender female). All results are from single-predictor multinomial logistic regressions.

^* p < 0.05; ^**p < 0.01; ^***p < 0.001.

CI, confidence interval; SE, standard error.

Results from the multivariate model are given in Table 3. When controlling for all other potential risk factors, Black and Latinx SM-AFAB youth were still at higher risk for obesity, but not for overweight. More differences emerged by sexual identity: Bisexual, pansexual, and queer individuals were at greater risk for obesity than gay/lesbian participants. LGBT victimization was negatively associated with obesity (OR = 0.18, p < 0.01).

Table 3.

Associations of Demographic Characteristics and Minority Stressors with Weight Status

Odds ratios	Obesity (N = 81)	95% CI	SE	Overweight (N = 91)	95% CI	SE
Race
Black	4.55***	1.95–10.63	0.43	1.96	0.91–4.02	0.38
Latinx	3.09**	1.34–7.17	0.43	1.23	0.96–4.02	0.37
Other	1.89	0.68–4.87	0.50	1.31	0.53–2.84	0.43
Sexual identity
Bisexual	2.77*	1.17–5.55	0.44	2.61*	1.22–5.57	0.38
Queer	3.36*	1.08–10.43	0.58	1.86	0.66–5.24	0.53
Pansexual	3.48**	1.36–8.90	0.48	1.69	0.69–4.12	0.46
Other	1.71	0.50–5.92	0.63	1.72	0.58–5.11	0.56
Gender identity
Transgender identity	1.21	0.44–3.33	0.52	0.57	0.20–1.67	0.55
Nonbinary/GNC	0.69	0.30–1.56	0.42	0.97	0.47–2.00	0.37
Household income
<$20,000	2.61*	1.02–6.68	0.48	1.52	0.67–3.45	0.42
$20,000–$39,000	2.70*	1.08–6.73	0.47	1.56	0.69–3.51	0.42
$40,000–$49,000	1.03	0.34–3.10	0.56	0.88	0.34–2.28	0.49
$50,000–$59,000	1.39	0.48–3.94	0.54	0.77	0.29–2.06	0.50
$60,000–$69,000	1.45	1.14–10.41	0.56	1.35	0.47–3.90	0.54
$70,000–$79,000	1.34	0.30–6.03	0.77	1.50	0.48–4.72	0.59
SM microaggressions	2.75***	1.51–5.00	0.30	1.56	0.88–2.75	0.29
SM internalized stigma	1.28	0.77–2.12	0.36	1.29	0.71–1.79	0.24
LGBT victimization	0.18**	0.52–0.63	0.64	0.63	0.21–1.88	0.56

Reference group: Dependent variable (normal weight), race (White), sexual identity (gay/lesbian), and gender identity (cisgender female). All results are from a multivariate regression in which race/ethnicity, sexual identity, gender identity, household income, microaggressions, SM internalized stigma, and LGBT victimization were simultaneously entered as predictors.

^* p < 0.05; ^**p < 0.01; ^***p < 0.001.

Discussion

Study findings highlight the high rates of obesity in SM-AFAB youth (age 16–20 years); 25% were overweight and 22% were obese. The rates are higher than the 15% obesity rate observed in representative samples of the general population of women aged 18–24 years,³⁹ reinforcing previous evidence that SM-AFAB youth face disparities in obesity⁴ and underscoring the need for targeted obesity prevention and intervention efforts for SM-AFAB youth.

Findings also illuminate how SM-AFAB youth with multiple marginalized identities may be at increased risk for obesity. Consistent with racial disparities in the general population⁴⁰ and adult SM women,¹⁶ obesity was more prevalent among Black and Latinx than among White SM-AFAB youth. Though further research is needed to explore the mechanisms behind these racial disparities, SM-AFAB youth of color may be more vulnerable to obesity than White SM-AFAB youth due to racial stigma, which has been associated with weight gain among people of color.⁴¹

That is, sexual minority people of color (SM-POC) may be at “double jeopardy” for obesity, as they are for other adverse health outcomes, due to compounding negative effects of race- and SM-based stigma.⁴² These findings highlight the importance of attending to the intersection of individuals' multiple social identities in obesity research.⁴²

Racial disparities in poverty⁴³ may also contribute to higher rates of obesity in Black and Latinx versus White SM-AFAB youth. In line with evidence that low income is associated with obesity among adult SM women and the general population,^15,16 we observed higher rates of obesity among SM-AFAB youth with household incomes <$40,000 than those with higher incomes.

In this sample of SM-AFAB youth, bisexual, pansexual, and queer participants had higher rates of obesity, and bisexual participants had higher rates of overweight than gay/lesbian participants. Though these findings contrast with previous studies in which lesbian adults were equally²⁴ or more²³ likely to be obese than bisexual women, they are consistent with a growing literature suggesting that bisexual and pansexual individuals are at higher risk than lesbian/gay individuals for myriad negative health outcomes.^22,44 Stigma against having attractions to more than one gender, present in both heterosexual and SM communities, may contribute to unhealthy weight-related behaviors as it does other health behaviors.⁴⁵

Weight status was unrelated to gender identity, in contrast to previous research.²¹ This null finding was not due to lower power associated with the relatively small number of transgender participants (n = 35); raw rates of obesity were actually highest among our cisgender female participants (though not significantly higher than other groups). The inconsistency in findings between samples may reflect how, among transgender men, youth are less likely than adults to be on testosterone treatment,⁴⁶ which is associated with increased BMI.⁴⁷

Results also suggest that some, but not all, minority stressors are associated with obesity. Supporting minority stress theory and its application to obesity among SM individuals,²⁸ participants who experienced more, versus fewer, anti-SM microaggressions had higher rates of obesity. Unexpectedly, victimization and internalized stigma were not associated with weight status and victimization was negatively associated with obesity in the multivariate model, likely due to its overlapping variance with microaggressions.

Though we expected all three minority stressors to be correlates of obesity, this pattern of findings may be understood within the framework of chronic stress theory, which highlights the particularly negative effects of stressors that occur repeatedly.⁴⁸ In our sample, participants experienced microaggressions, on average, 23–69 times in the past 6 months, compared with, on average, fewer than 10 victimization events.

Therefore, microaggressions may more often trigger a cycle in which instances of stigma increase weight, leading to further weight-related stigma, and then more weight gain.² Although it is important to not dismiss the harmfulness of anti-SM victimization, day-to-day microaggressions may represent a greater risk factor for obesity among SM-AFAB youth, echoing previous findings of other SM health outcomes (e.g., anxiety and depression).^37,49

Limitations

Several study limitations should be noted. First, BMI does not differentiate individuals with high bone or muscle mass from those with high body fat (the only ones at risk for obesity-related health concerns).³⁴ There is also evidence of racial bias in BMI, overestimating the prevalence of obesity among Black individuals while underestimating it among Latinx and Asian women.⁵⁰ However, BMI continues to be the standard for assessing obesity due to its ease of administration.

Second, we had no non-SM comparison group, prohibiting comparisons of obesity by sexual orientation. However, a substantial body of research demonstrates this disparity.⁵¹ The nonprobability sample, which reported relatively low levels of minority stress, may limit generalizability of findings. The small number of participants with particular sexual and gender identities did not provide power to explore differences among all identities and raises the possibility that findings for certain identities (e.g., transgender) were sample specific.

Third, because of the cross-sectional nature of these findings, they do not speak of direction of effects; for example, it is possible that higher weight individuals are more likely to be in low-income families because they are systematically paid less than other people rather than because poverty raises risk for obesity. Future prospective longitudinal research is needed to more confidently identify risk factors for obesity. Finally, public health concerns about obesity center on its contributions to serious disease and health conditions;¹ future research is needed to examine whether the factors associated with obesity risk in this sample raise the likelihood of these adverse health outcomes.

Clinical implications

Targeted obesity prevention and treatment programs should consider the unique needs, challenges, and strengths of SM-AFAB youth. Current findings suggest that such programs might benefit from teaching skills for responding to and coping with anti-SM microaggressions. Furthermore, because microaggressions are often perpetrated by friends or family, programs might offer safe spaces to escape these experiences or educate family and friends about the harmful effects of common, often unintended, microaggressions.

Because most stigma events experienced by overweight SM women involve being stigmatized based on both weight and sexual orientation,⁵² programs are likely to be most effective if they address the multiple forms of stigma overweight SM women face based on their intersecting social identities and weight status.

Furthermore, emphasizing healthy behaviors rather than weight loss may help avoid compounding SM stress with weight-related stigma.^53,54 Due to racism within SM communities, and both racism and homonegativity within the larger community,⁵⁵ SM people of color may lack social support to succeed in obesity prevention programs.⁵⁶ Therefore, prevention programs might include efforts to build social support for SM-POC, to maximize their chances of success.

Conclusion

This study highlights the risk for obesity among SM-AFAB youth, particularly for those who identify as a racial minority, those with low income, those attracted to more than one gender, and those who experience high levels of anti-SM microaggressions. Our study supports minority stress theory, which emphasizes how individuals with minoritized social identities are at increased risk for health problems due to societal stigma. SM-AFAB youth represent an underserved high-risk group who must not be ignored when addressing the obesity epidemic.

Disclaimer

The primary data collection organization for this secondary analysis article is the Institute for Sexual and Gender Minority Health and Wellbeing at Northwestern University.

Authors' Contributions

E.A.D. was primarily responsible for the article's hypothesis, statistical analyses, and main text. S.W. advised E.A.D. and provided detailed feedback on the text. M.E.N. provided feedback on the text and guidance of the sample. All coauthors reviewed and approved the article before submission.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

There was no direct research support for this study. Data collection was supported by a grant from the National Institute of Child Health and Human Development (R01 HD086170; principal investigator: S.W.).