**1. Introduction**

The high prevalence of obesity among youth is one of the most concerning public health issues in both developed and developing countries [1]. Most overweight or obese children live in developing countries, where the rate of increase is over 30% higher than in more developed countries [1]. In the past 40 years, the number of obese children and adolescents (aged five to 19 years) has increased from 11 million in 1975 to 124 million in 2016 [2]. If current trends continue, by 2022, there will be more obese children and adolescents worldwide than moderately or severely underweight children [2]. Childhood and adolescent overweight and obesity are associated with increased risk of subsequent diabetes, stroke, coronary heart disease, hypertension, functional disability, as well as premature adult mortality and morbidity [3,4]. In the United States (US), an estimated 34.5% of adolescents aged 12–19 years were overweight or obese from 2011–2012, and of these, 16.9% were obese [5]. This number rose to 20.6% in 2015–2016 [6]. Disparities in obesity and overweight exist across racial and ethnic groups in the US, with African American and Mexican-American adolescents ranking highest in prevalence [7,8]. During 2011–2012, the prevalence of overweight and obesity was 39.8% and 38.1% among African American and Latino adolescents, respectively, followed by non-Latino white (31.2%) and Asian (24.6%) adolescents [5]. From 2013–2016, the prevalence of obesity among youth of Mexican origin aged 12–19 years was 26.6%, as compared to 17.2% among non-Latino whites [9]. Studies have also shown that US-born Mexicans are significantly more likely to be overweight or obese than Mexican immigrants [10,11]. In Mexico, the prevalence of obesity and overweight among adolescents aged 12–19 years was 36.3% in 2016 [12].

Addressing obesity is complex, due to its multi-causal nature that includes various psychosocial, behavioral, and socio-environmental factors. Previous studies have found an association between a range of psychosocial factors and increased obesity risk, such as body size dissatisfaction and self-perception of overweight, because they may promote unhealthy weight control behaviors [13–15]. Other psychosocial factors that have been examined include exposure to adverse life events and the influence of the family and peer environment, which may be associated with a greater risk of childhood overweight/obesity [16]. Studies have also found that depressive symptoms are a risk factor for obesity because binge eating may be used as a coping mechanism [14,17]. Obese youth report having a lower quality of life (QOL) [18–20], which improves when they lose weight [21].

Unhealthy weight control and dietary restraint behaviors have been found to predict the onset of obesity [14,15,17,22]. Studies also show that prioritizing healthy eating may protect youth from becoming overweight or obese [23,24], whereas prioritizing physical activity appears to be less protective [23]. Various studies have demonstrated a negative association between breakfast consumption and an increase in body mass index (BMI) [14,22,25–27], which could be due to its association with favorable nutrient intake, improved food choices, and higher physical activity levels [26,28]. The protective effect of physical activity has also been observed in both cross-sectional and longitudinal studies [14,22]. However, the relationship between fast food consumption and obesity has not been established conclusively in the literature. Some studies have shown an inverse association between fast food consumption and obesity [14,27], while others report that fast food consumption is a predictor of weight gain [28,29].

Several socio-environmental factors have also been associated with risk of obesity in adolescents. Parents who have unhealthy lifestyles are more likely to have children who become overweight or obese [30–32]. Conversely, positive parental influence regarding healthy diet and frequent physical activity have been associated with reductions in BMI among overweight and obese adolescents [33]. Studies also report that increased availability of healthy food at home encourages healthy eating in adolescents and is protective against overweight and obesity [34,35], while parental obesity is associated with an increased risk of adolescent and ensuing adult obesity [17]. However, Haines et al. found that the availability of healthy food at home and perceived parental obesity did not predict onset of obesity [14]. Parental concern regarding their child's weight has been positively associated with their child being overweight or obese [14,36]. Parental concern may lead to parental pressure to lose weight and encouragemen<sup>t</sup> of restrictive feeding practices, which could lead to weight gain [14,36]. However, parents who reported being concerned about their child's weight were more likely to improve the family's diet, limit child screen time, and attempt to increase their child's physical activity levels [36].

Although there is no individual factor that causes obesity, most research to date has lacked an integrated approach to examine the factors that may be contributing to the high rates of overweight and obesity among youth [17]. An exception would be a study by Haines et al., which looked at the effects of personal, behavioral, and socio-environmental factors on risk of overweight in an ethnically diverse population in Minnesota [14]. To the best of our knowledge, the present study is the first to compare the effects of multiple domains on overweight or obesity risk among a bi-national, ethnically diverse sample of youth. The objective of this study was to identify risk and protective factors for overweight or obesity within the following three domains: Psychosocial, behavioral, and socio-environmental, in a sample of African American, Caucasian, and Latino youths in the US, and Mexican youths in Mexico, and determine differences by sex.

### **2. Research Methods and Procedures**

### *2.1. Study Population and Data Collection Procedures*

US participants were recruited from community centers, schools, clinics, and youth programs in Seattle, Washington and Los Angeles, California (*n* = 452). A convenience sample of youth was also recruited from the main Mexican Institute of Social Security (IMSS, as per its Spanish abbreviation) hospital in Cuernavaca, Morelos (*n* = 181). Study flyers were posted in various areas of the IMSS clinic, and potential participants were also informed of the study by staff during their visit to the primary care clinics. All individuals who expressed an interest in the study were contacted by a study recruiter who conducted a telephone interview with the primary caregivers of the potential participants to determine eligibility. Participants had to be African American, Caucasian, or Latino, and between the age of 11–18 years. Youths who met study inclusion criteria of age, 5th grade reading ability, and no serious physical or mental illness diagnosis were informed that participation in the study would involve completing a 40-min questionnaire and having their weight, height, and waist circumference measured. All study participants were enrolled between 2006 and 2008, and informed consent was obtained from each participant and a parent or guardian prior to their inclusion in the study. Further details regarding study design, methodology, and baseline participant characteristics are specified elsewhere [19,37,38]. The Institutional Review Boards of the University of Washington, the University of California, Los Angeles, and the Mexican Institute of Social Security approved all study materials including the study questionnaire, protocol, and consent forms (Seattle Children's Hospital IRB approval number: 11916; IMSS IRB approval number: R-2007-1701-13; UCLA IRB approval number: G06-09-094-01).

Study participants completed a self-administered questionnaire that included the 21-item youth quality of life weight-specific measure (YQOL-W), a generic youth quality of life Instrument (YQOL-R), as well as measures of perceived general health, physical function, body shape satisfaction, and symptoms of depression. The Spanish versions of these measures have been used extensively and validated in other research studies [21,38–44]. All study materials were designed to be readable and understandable at a 5th grade level.
