Young adults, ages 20–35, experience the greatest rate of weight gain, averaging 1 to 2 pounds per year [1, 2]. Over time, this weight gain is associated with a worsening in cardiovascular risk factors and an increase in the prevalence of metabolic syndrome [3, 4]. To date, there have been few large trials designed to test ways to prevent weight gain in this age group, and the results have been disappointing. The present paper describes two new approaches to weight gain prevention and the design of a multi-site randomized controlled clinical trial that is underway to examine the efficacy of these approaches.
Weight gain in young adults
A number of studies have documented significant weight gain in young adults. In the Coronary Artery Risk Development in Young Adults (CARDIA) study, individuals aged 18–30 gained approximately 15 kg over 15 years or 1 kg/year . A study of over 8500 young women, aged 18–23, found that 41% gained more than 5% over their baseline weight over 4 years . Getting married, pregnancies, and entering the work force have all been related to weight gain in this age group [5, 6]. Moreover, the weight gained by young adults has adverse health consequences. In the CARDIA study, only 16.3% of young adults maintained a stable BMI over 15 years of follow-up, but those individuals who remained weight stable had essentially unchanged levels of all of the components of the metabolic syndrome, regardless of their initial body mass index, age, race, or gender. In contrast, those who gained weight had worsening in cardiovascular risk factors and increased prevalence of the metabolic syndrome [1, 7]. Weight gained during young adulthood has also been associated with increased risk of coronary heart disease events  and a variety of other diseases, including postmenopausal breast cancer, kidney stones, gout, hypertension and type 2 diabetes [9–12]. These studies suggest that preventing weight gain in young adults would decrease the risk of cardiovascular disease and improve overall health.
Prior trials for weight gain prevention in young adults
To date, there have been few randomized trials testing interventions specifically designed to target weight gain prevention and the two largest, longest trials had limited effects. In Pound Of Prevention (POP) , weight gain over 3 years was examined in a no-contact control group compared with a group given education through monthly newsletters and a group given the same education plus incentives for participation. None of the interventions were successful in reducing the average magnitude of weight gained over 3 years (1.8 kg in control; 1.6 kg in education and 1.5 kg in education plus incentive). The other large prevention trial by Levine et al.  compared an in person approach, a correspondence program and a no treatment control in a sample of 284 female participants, aged 25 – 44 with a BMI of 21 – 30 kg/m2 followed for 3 years. Mean weight changes over 3 years did not differ significantly between conditions (+0.7 kg, +0.3 kg, and −0.6 kg for the control, correspondence, and in-person conditions, respectively). Clearly new approaches to weight gain prevention for young adults are needed. Recognizing this need, National Heart, Lung, and Blood (NHLBI) has funded several trials on this topic and developed the Early Adult Reduction of weight through LifestYle intervention (EARLY) consortium (Lytle L, Svetkey LP, Patrick K, Belle SH, Fernandez ID, Jakicic J, Johnson KC, Olson C, Tate DF, Wing RR: The EARLY trials: a consortium of studies targeting weight control in young adults, submitted).
Novel interventions for weight gain prevention
One way to prevent weight gain is to engage in a process of self-regulation of behavior, an approach we tested successfully in the prevention of weight regain  but has not yet been tested for weight gain prevention. Self-regulation involves having a goal, having access to information about whether the goal is being achieved, and if not, taking steps to restore equilibrium. For example, applying self-regulation to type 1 diabetes, an individual must be knowledgeable about the level of blood glucose they are trying to achieve, they must monitor their glucose to see if there are discrepancies between their goal and their actual blood sugar, and then if there are discrepancies, they must adjust their diet, exercise, or insulin dose to reduce the discrepancy. Within the area of weight gain prevention, the individual has a goal of maintaining their current weight. Information about deviations from this goal is best provided by frequent self-weighing. Although scales are not perfect, they provide more accurate and immediate feedback than other indicators such as noticing if one’s clothes are too tight. If discrepancies are noted, the individual must change their behavior to reduce the discrepancies. It remains unclear, however, what type of behavior changes the individual should make to reduce this discrepancy. Two different approaches have been suggested—a “Small Change” and a “Large Change” approach.
Currently the message being given to the public is that daily small changes in eating and exercise behavior will prevent weight gain. The America on the Move Foundation (https://aom3.americaonthemove.org/) encourages Americans to take small, simple lifestyle changes - versus dramatic changes - to ensure effective long-term weight control. This message is based on the fact that the average weight gain with aging is about 1 kg per year; therefore, a decrease of 10–15 kcal/day (or 20–30 kcal/day based on the estimated 50% energy cost to storing excess energy), should be sufficient to prevent weight gain. Using these calculations, Hill and colleagues  have suggested that if we could modify energy imbalance by 100 kcal/day through small changes in eating and/or physical activity, we could prevent weight gain in 90% of the U.S. adult population. Behavioral theory also suggests that small changes (i.e. gradual shaping of new behaviors with small incremental changes toward a goal) should be easier to initiate and maintain than larger behavior changes since they represent less drastic modifications in behavior . Although several recent studies have provided empirical support for this approach, these studies have been short in duration (13–24 weeks) and the recommended changes (e.g. consuming cereal for 2 meals per day) could be considered relatively substantial in nature, as opposed to representing small and easily integrated changes based on one’s current lifestyle, as suggested by the AOM campaign.
Another approach to weight gain prevention is to produce initial weight loss as a buffer against the expected weight gain, which we refer to as the “Large Change” approach. There is stronger empirical evidence for this approach, coming from the Women’s Healthy Lifestyle Project (WHLP), the only study that has actually succeeded in preventing weight gain and the worsening in CVD risk factors over a period of 5 years . In this study conducted with women (BMI of 20 to 34) during the menopausal transition, the intervention group was encouraged to lose 5–15 pounds as a means of counteracting the weight gains that is expected with aging. The intervention group lost a mean of 0.09 kg over the 5-year intervention whereas the assessment only group gained 2.4 kg. The intervention also reduced the worsening in cholesterol during this time period. A large behavior change approach is also supported by a secondary analysis of data from the POP study. Jeffery, McGuire and French  found that although weight loss was not targeted, 9.3% of their study population lost >5% (mean= 6.4 kg) in Year 1; these individuals were the only group that was below baseline at year 3 (−2.6 kg). Similarly in the Levine  study, 70% of the participants who lost > 2.3 kg from baseline to 1 year were still below their baseline weight at year 3; among those who lost 0.9–2.3 kg at 1 year, 60% were still below baseline at year 3. However, among those who were weight stable from baseline to year 1 (± 0.9 kg) only 35% were still below baseline at 3 years. To date, there have been no prior trials comparing the small and large changes approach and examining their efficacy for preventing weight gain in young adults.
The Study of Novel Approaches to Weight Gain Prevention (SNAP) is a two site, randomized clinical trial funded by the National Heart Lung and Blood Institute. The two clinical sites are at The Miriam Hospital (R. Wing, PI) and The University of North Carolina at Chapel Hill (D. Tate, PI). The Coordinating Center is at Wake Forest School of Medicine (M. Espeland, PI). SNAP is comparing the efficacy of self-regulation plus small behavior changes intervention, a self-regulation plus large behavior changes intervention, and a minimal treatment control condition in preventing weight gain in 600 young adults age 18 – 35 over an average planned follow-up of 3 years.
The primary hypothesis is that the magnitude of weight gain across 3 years will differ among the three groups. Specific a priori hypotheses are that the magnitude of weight gain across the 3 years will be lowest in the self-regulation plus large behavior changes intervention, followed by the self-regulation plus small behavior changes intervention, and greatest in the control condition.
Secondary aims compare the 3 conditions on mean weight gain at 2 years, on the proportion who gain <1 pound or >1 pound at 3 years, and on the proportion who become obese at 3 years. The three groups are also compared on changes in behavior (e.g. diet, physical activity, disordered eating behaviors, use of healthy and unhealthy weight control practices), psychosocial measures (restraint, depression), and changes in CVD risk factors (including blood pressure, lipids, insulin sensitivity, and waist circumference). The study will also examine demographic and psychological variables that may predict weight change over the average follow-up of 3 years and/or moderate the effects of the interventions, including initial BMI, ethnicity, age, scores on the Eating Inventory, and treatment preference and examine potential mediators of the effect of the interventions, including changes in diet, physical activity, restraint, self efficacy and frequency of self-weighing.