Multiple health behaviors: Patterns and correlates of diet and exercise in a Hispanic college sample

Abstract

Obesity rates are alarming in various ethnocultural groups, particularly in Hispanics. With Hispanics being the fastest growing group to enter college, the aims of the current study were to examine patterns and correlates of exercise and dietary behaviors in Hispanic college students. Data were collected from 693 Hispanic undergraduates who enrolled online and received course extra credit for participation. Individuals completed questionnaires assessing constructs of the transtheoretical model for three health behaviors (exercise, dietary fat, and fruit/vegetable stages of change) along with demographic, psychosocial, and acculturation measures. Less than 1% of students had 0 obesity-relevant risks, while 68% indicated 2 or more risks. Only 2% of the sample met fruit and vegetable guidelines. Lower income was associated with greater obesity-relevant risks, while stress coping ability was associated with fewer such risks. Findings indicate specific obesity risk behaviors in Hispanic college students and suggest demographic and psychosocial targets for prevention and intervention according to stage of change.

Introduction

Over two-thirds of Americans are overweight or obese (Flegal, Carroll, Ogden, & Curtin, 2010). Clinical guidelines define adult overweight and obesity as a body mass index (BMI) of 25–29.9 kg/m² and ≥ 30 kg/m², respectively (National Heart, Lung, & Blood Institute, 1998). Obesity is a health concern as it is associated with a number of preventable or reversible medical complications, including diabetes, hypertension, elevated cholesterol, and heart disease (National Heart, Lung, & Blood Institute, 1998).

In the United States, 76.9% of Hispanics are overweight or obese (Flegal et al., 2010). Specifically, Mexican American adolescents – of whom 22% are obese and 40% are overweight (Hedley et al., 2004) – are at greater risk for obesity than non-Hispanic White and Black adolescents (Flegal et al., 2010), necessitating an understanding of characteristics associated with overweight in relatively younger Hispanic populations. Moreover, as Hispanics are less likely to seek evidenced-based treatment for weight loss (Tsai et al., 2009), novel interventions that are empirically derived seem warranted in order to garner interest and participation of this particular population.

Known correlates of obesity in Hispanics include poor diet (Allen et al., 2007), acculturation (Hubert, Snider, & Winkleby, 2005), low physical activity, income, and education levels (Bowie, Juon, Cho, & Rodriguez, 2007). Perceived stress and smoking have also been associated with unhealthy eating behaviors in Hispanics (Jenkins, Rew, & Sternglanz, 2005), meriting the investigation of the above factors as potential correlates.

Given that significant increases in weight occur between 18 and 29 years (Mokdad et al., 2003), an examination of activity and eating behaviors in college students is warranted for the development of prevention and intervention efforts, as college is a salient transition period for health behavior change in young adults (Harris, Gordon-Larson, Chantala, & Udry, 2006). In fact, only 37% of Hispanic college students regularly engage in moderate physical activity, while less than one-fourth consume at least five fruits and vegetables daily (Centers for Disease Control & Prevention, 2010).

Research suggests that unhealthy behaviors co-occur, with potentially multiplicative health consequences (Berrigan, Dodd, Troiano, Krebs-Smith, & Barbash, 2003). The most common pattern of health behaviors in U.S. adults involves a lack of adherence to exercise, dietary fat, and fruit and vegetable recommendations (Berrigan et al., 2003). Thus, the rationale underlying research on multiple health behavior change (MHBC) is one of maximizing health benefits and reducing healthcare costs (Prochaska, Spring, & Nigg, 2008).

The transtheoretical model (TTM) is one avenue for tailoring MHBC interventions, as it addresses psychosocial and motivational mediators of behavior change (Prochaska & DiClemente, 1983). The TTM conceptualizes temporal behavior change along a cyclical continuum of readiness in five stages (Prochaska & DiClemente, 1983). In precontemplation, individuals are not considering change within the next 6 months, while individuals in the contemplation stage plan to alter behavior within 6 months; preparation individuals have either taken moderate action or intend to act within 1 month. Adequate behavior modifications from zero to 6 months comprise the action stage, while the maintenance stage entails consistent changed behavior for more than 6 months (Prochaska & DiClemente, 1983).

TTM constructs have been successfully applied to single health behaviors, including smoking (DiClemente et al., 1991), physical activity (Marshall & Biddle, 2001), and diet (Park et al., 2008). The model's efficacy has been demonstrated in two MHBC interventions on smoking, dietary fat, and sun exposure (Prochaska et al., 2004, Prochaska et al., 2005). Nevertheless, ethnic college groups are not widely studied in the TTM literature on MHBC. An expanded understanding of multiple health behaviors in these populations may inform prevention, intervention, and health policy efforts that target specific behaviors at the college level.

One pilot study among college students demonstrated significant increases in physical activity and consumption of healthy fats after a brief image-based intervention (Werch et al., 2007), though similar studies among Hispanic counterparts have yet to be performed.

As such, the assessment of TTM constructs for activity and dietary behavior change will likely promote more targeted MHBC intervention in the context of obesity risk. Authors examined in Hispanic college students the extent of risk in lack of adherence to physical activity, dietary fat, and 5-A-Day fruit and vegetable recommendations. The co-occurrence of lack of adherence and its correlates was also assessed. Potential sociodemographic correlates of risk included age, sex, income, and acculturation, meaning in this case the degree to which Hispanics have adopted U. S. culture. Health-related correlates included smoking status and BMI. Psychosocial correlates included the level of perceived stress and ability to cope with stress. Individuals in this study were hypothesized to report being in earlier stages of change for the three behaviors; risk was hypothesized to correlate with high acculturation, low income, high BMI, smoking, and greater perceived stress.