Abstract
Objective:
Characterize young adults who experience significant weight gains (>10%) over 3 years in a weight gain prevention program.
Method:
Secondary data analysis from Study of Novel Approaches to Weight Gain Prevention (SNAP), a randomized trial comparing two self-regulation interventions and a control arm in young adults (18–35 years; BMI 21–30.9 kg/m2). Large Gainers (>10%; N=48), Small Gainers (2.6% to 9.9%; N=147) and Weight Stable participants ( ±2.5%; N=143) were compared on dimensions affecting weight gain.
Results:
Differences in weight gain among the three groups were significant by 1 year and subsequently increased. Those who became Large Gainers were heavier at baseline, further below their highest weight, and reported more weight cycling than Weight Stable, with Small Gainers intermediate. Neither study arm nor pregnancy explained weight change differences among the 3 groups. Large Gainers reported more depressive symptoms than Weight Stable at 1 and 2 years. Large Gainers were less likely to weigh themselves at least weekly at 4 months, before differences in weight gain emerged, and at 1 and 2 years.
Conclusion:
Large Gainers (representing almost 10% of participants) could be identified early by greater weight issues at baseline and lower use of weight gain prevention strategies.
Keywords: Weight gain, prevention, psychosocial variables, weight variability
INTRODUCTION
It is well recognized that there is heterogeneity among participants in the weight losses achieved in a behavioral weight loss program.1,2 A similar heterogeneity exists within weight gain prevention interventions,3 but has received less attention. Although weight loss and weight gain prevention programs have similar content, participants in weight gain prevention programs are typically younger, since weight gain occurs most commonly in young adults 4,5 and have lower body weight. Variables associated with their weight gain have been identified in observational studies, 6–9 but we are not aware of any studies examining the characteristics of young adults who experience significant weight gains while participating in a weight gain prevention program.
We recently reported positive results of two weight control interventions based on self-regulation relative to a control condition over an average of 3 years of follow-up.10 Although on average those who received the interventions gained less weight than the control group, the degree of heterogeneity in weight change from baseline to 3 years was striking, ranging from a 20.3% loss to a 47.1% gain. By understanding characteristics of the subgroup of participants who experienced the poorest treatment response in this study (those who gained the most weight) and the variables associated with their poor outcomes, we hope that more personalized approaches to weight gain prevention for young adults can be developed.
Based on prior research, we considered four categories of variables that might distinguish those who will subsequently gain weight vs. those who will remain weight stable. First, we considered the possibility that those who will gain weight are predisposed to a heavier BMI, as suggested by already being heavier at enrollment, history of weight loss followed by regain (i.e. weight cycling), and/or currently being further below their highest weight ever, predictors of weight gain in prior studies of young adults. 6,11,12 A second possible factor relates to pregnancy and child bearing, a common cause of weight gain in this age group.6,9,13 We also considered psychosocial predictors,14–17 in particular, stress and depressive symptoms at baseline and during the trial. Finally, young adults are at risk of decreasing their physical activity level and consuming a poorer quality diet.7,8,18 We hypothesized that Weight Stable and Large Gainers would differ in these behaviors and in use of weight control strategies and that Small Gainers would be intermediate.
METHODS
Subjects.
A total of 599 participants (22% men and 27% racial/ethnic minority) entered the Study of Novel Approaches to Weight Gain Prevention (SNAP) study. Participants were recruited at two clinical centers (University of North Carolina and The Miriam Hospital, Providence, RI) between 2010–2012 and all completed written informed consent. Eligibility criteria for SNAP have been described previously10 and included 18–35 years of age (Mean ± SD, 28.2±4.4, BMI of 21.0 to 30.9 (25.4±2.6), English speaking, and internet access. Although the primary results of SNAP were based on weight changes from baseline to an average follow-up of 3 years (ranging from 2 to 4 years), we continued to try to follow all participants and base our analyses in this paper on weight changes in 507 participants (85% of the original sample) who were followed from baseline to 3 years, omitting the 18 women who were pregnant within the 6 months prior to the three-year assessment.
Design.
Participants were randomly assigned to one of three conditions: self-regulation with Large Changes, self-regulation with Small Changes, and Control. Both self-regulation interventions stressed daily self-weighing, with the goal of remaining below baseline weight. Large Changes participants were taught to lose 5–10 pounds during the initial 4 months to provide a buffer against expected weight gains. The Small Changes group was taught to make daily changes in their eating and physical activity to produce approximately a 200 kcal/day change. Both groups attended separate weekly meetings for 8 weeks, monthly meetings for the next 2 months, and then received the rest of the intervention online. The Control group attended one face-to-face meeting where they were introduced to both Large and Small Changes strategies and encouraged to choose their preferred approach. The interventions and control conditions have been described in detail previously19 and the primary results of the trial have been reported.10 Mean weight gains averaged across the 3 years of follow-up were 0.26 kg, −0.56 kg, and −2.37 kg in Control, Small Changes, and Large Changes, with significant differences among all three groups. De-identified data will be available on the NHLBI Data Repository (https://biolincc.nhlbi.nih.gov/home/) by January 2020.
In analyses conducted in 2018–2019, we compared the 48 Large Gainers (gain >10%), 149 Small Gainers (gain between 2.6% and 9.9%), and 143 participants who remained Weight Stable (± 2.5 %), representing 9%, 29% and 28% of the sample, respectively. We selected 10% as a large weight gain since this represents a clinically significant weight gain.20 Since there is no generally accepted criterion of weight stability over 3 years, we a priori selected ±2.5% (or approximately a 2 kg weight gain from baseline over the 3 years) as the criterion for stability since this represented a weight change of less than 1% (or 1 kg) per year. Although losing weight from baseline to 3 years would also be considered successful outcome, we focus on those who remained weight stable since the goal of the intervention was to prevent weight gain. Large weight losers will be examined in future manuscripts.
Assessments:
Assessments were completed at baseline, 4 months, and 1, 2 and 3 years by research staff who were masked to treatment assignment. Detailed descriptions of the assessments and the effects of treatment assignment on these measures have been published previously.10,19 Here we compared Large Gainers, Small Gainers, and Weight Stable participants on the following measures:
Anthropometric Measures.
Weight was measured in light clothes, without shoes, on a calibrated scale. Height was determined with a wall stadiometer. Body Mass Index (kg/m2) was then calculated.
Baseline Demographics and Weight History.
In addition to questions regarding race/ethnicity, age, gender, participants were asked several questions regarding their weight history, including their highest weight ever (excluding pregnancy); the difference between their highest weight and their current weight to provide a measure of “weight suppression” and their history of weight cycling, calculated by multiplying and then summing the number times (0, 1–2, 3–4 etc.) they have lost specific number of pounds (5–9 lbs.; 10–19 lbs;20–29 lbs., etc.) This measure does not provide data on weight regains.
Pregnancy.
At each assessment visit, female participants were asked about recent pregnancies. If a pregnancy was reported, the intervention was suspended and no assessments were conducted during or for 6 months after the pregnancy.
Psychosocial Variables.
Participants completed the 20-item Center for Epidemiological Studies Depression Scale (CES-D), 21 the 4-item Cohen Perceived Stress Scale22 and the life events questionnaire21–24 from the Coronary Artery Risk Development in Young (CARDIA) Study.23 They also completed the Eating Inventory,24 assessing dietary restraint, disinhibition, and hunger.
Dietary Intake.
Diet was assessed only at baseline, month 4 and 2 years with a Block Food Frequency Questionnaire.25 We used the baseline and 2-year data to determine changes in overall energy intake, percent of calories from macronutrients, and the Healthy Eating Index (HEI), which measures dietary quality based on adherence to the US Dietary Guidelines.26 HEI scores range from 0 to 100 with higher scores representing healthier diets.
Physical Activity.
Participants wore the SenseWear Pro Armband (Body Media, Pittsburgh, PA) at all but the year 3 assessment to provide an objective measure of physical activity.27 Participants were instructed to wear the device during all waking hours (except swimming and bathing) for a full week; monitoring for at least 10 hours per day on at least 4 days was required for analysis.
Self-weighing.
Participants reported on their frequency of self-weighing. Those who reported weighing themselves daily or several times per week were considered “frequent self-weighers”. This criterion was used in a prior analysis of SNAP data.28
Statistical Analysis.
Data were analyzed as a post hoc observational analysis. Analysis of variance (ANOVA) for continuous measures and Fisher’s Exact Tests for categorical data were used to were used to assess differences among Large Gainers, Small Gainers, and Weight Stable participants in single variable comparisons. Repeated measures analyses of variance, adjusting for baseline levels, were used to assess differences among the three groups in psychosocial and behavioral measures. Generalized estimating equations were used to compare groups on categorical measures over time. Significance was assumed in the overall test if p<0.05; if the overall p-value was significant, pairwise comparisons were assessed using p<0.0167 (0.05/3) to adjust for the 3 tests performed. SAS (version 9.4, Cary, NC, USA) was used for all statistical analyses.
RESULTS
Of 507 participants, 167 (33%) lost >2.5%, 143 (28%) remained Weight Stable (± 2.5%), 149 (29%) gained 2.5–9.9%, and 43 (9%) gained >10% over the three years. Subsequent analyses compare Large Gainers (>10%), Small Gainers (2.6%–9.9%), and Weight Stable (±2.5%) participants. Although Large and Small Gainers and Weight Stable groups had similar percent weight changes during the first 4 months of the program (−1.16 ± 2.1%, −2.37 ± 3.1%, and −1.76 ± 3.2%, respectively), by year 1, differences in weight gain were already significant (+2.94 ± 6.3%, +0.27 ± 4.9%, and −1.90 ± 3.5%, respectively) and further increased by year 3 (16.3 ±7.5% vs 5.6±2.1% vs. −0.07±1.4%) (See Table 1 / Figure 1).
Table 1.
Weight Change from Baseline to 3 Years for Small & Large Gainers and Stable Participants
| Large Gainers | Small Gainers | Weight Stable | ||
|---|---|---|---|---|
| N: | 48 | 149 | 143 | |
| WT Change (% of initial weight) | ||||
| Mean | 16.6% | 5.6% | −0.08% | |
| SD | 7.9 | 2.1 | 1.40 | |
| Median | 14.2% | 5.2% | −0.11% | |
| Range | 10 – 47% | 2.6 – 9.9% | −2.4 – +2.4% | |
| WT Change (kg) | ||||
| Mean | 11.87 kg | 3.96 kg | −0.04 kg | |
| SD | 6.33 kg | 1.64 kg | 1.01 kg | |
| Median | 9.55 kg | 3.55 kg | −0.10 kg | |
| Range | 6.0 – 32.7 kg | 1.6 – 8.2 kg | 2.17 – 2.00 kg |
Figure 1:
Percent Weight Change of Large Gainers, Small Gainers and Weight Stable Participants at 4 months and 1,2,3 years
Demographic Variables.
Table 2 compares the three groups on baseline demographic variables. As shown, the groups did not significantly differ in gender, race/ethnicity, or age or in the proportion who had been randomly assigned to Control, Small Changes, and Large Changes arms. However, there were significant difference among the three groups on several weight history variables. Those who would subsequently gain large or small amounts of weight were heavier at baseline than weight stable participants. Large Gainers reported the highest weight cycling levels, and currently were further below their highest-ever weight (suggesting greater weight suppression) than either of the other groups. After adjusting for BMI, the difference in weight cycling was no longer significant (p=0.086), but the difference from highest weight ever remained significant (p=0.013), with Large Gainers significantly different from the other two categories in pairwise comparisons.
Table 2.
Baseline Characteristics of Large Gainers, Small Gainers, and Weight Stable Participants
| Large Gainers (+10%) | Small Gainers (+2.6 – 9.9%) | WT Stable (±2.5%) | P-Value | ||
|---|---|---|---|---|---|
| N | 48 | 149 | 143 | ||
| % WT Gain (BL-3) | 16.6 ± 7.9 | 5.6 ± 2.1 | −.08± 1.40 | ||
| Age | 27.5 ± 4.5 | 27.5 ± 4.5 | 27.2 ± 4.4 | .64 | |
| BMI | 26.1 ± 2.4b | 25.5 ± 2.6b | 24.6 ± 2.6a | .0009 | |
| Weight (kg) | 70.4 ± 8.9 | 70.9 ± 10.0 | 70.4 ± 11.2 | .91 | |
| Height (cm) | 164.3 ± 7.7 | 166.5 ± 8.3 | 168.7 ± 8.6 | .004 | |
| Gender (% Female) | 90% | 77% | 78% | .15 | |
| Race | |||||
| White | 71% | 75% | 69% | .53 | |
| African American | 10% | 11% | 14% | ||
| Asian/Pacific Islander | 10% | 7% | 8% | ||
| Hispanic | 8% | 7% | 3% | ||
| Study Arm | |||||
| Control | 19 (40%) | 61 (41%) | 44 (31%) |  |
.20 |
| Large | 11 (23%) | 46 (31%) | 54 (38%) | ||
| Small | 18 (37%) | 42 (28%) | 45 (31%) | ||
| Weight Cycling | 31.2 ± 33.1b | 20.9 ±25.8a | 18.8±25.1a | .020 | |
| Weight Suppression (kgs below highest weight) | 7.8±8.8b | 5.0±5.0a | 5.1 ±5.2a | .008 | |
Groups within rows with different superscripts (i.e., a, b) differ significantly from each other in post-hoc tests with Bonferroni corrections (p<0.016).
Pregnancy.
One possible explanation for greater weight gain in the gainers could be pregnancy. Although a somewhat greater proportion of women in the Large Gain group reported a pregnancy over the 3 years of follow-up (20% ,12% and 7% of Large Gainers, Small Gainers and Weight Stable, respectively, p=.074), weight gain of those who did or did not have a pregnancy did not differ within the weight gain groups: Large Gainers [14.0% (3.4) vs 16.8 (8.2); p=0.39]; Small Gainers [5.8% (2.2) vs 5.9% (2.2); p=0.86] or Weight Stable [ −0.3% (1.1) vs −0.1% (1.4; (p=0.71].
Psychological Characteristics.
Those who would subsequently be Large Gainers, Small Gainers or Weight Stable did not differ significantly at baseline on depressive symptoms, restraint, or disinhibition (Table 3). However, the perceived stress scale approached significance (p=0.055) with greater perceived stress in those who would later be Large Gainers. In addition, CES-D, restraint, and disinhibition all differed among the three weight change groups over time. Post hoc tests showed that Large Gainers reported more depressive symptoms, relative to the Weight Stable group at Years 1 and 2; p= 0.002 and 00.005 (p=0.018 at year 3). Large Gainers reported significantly lower levels of dietary restraint than either Small Gainers and Weight Stable at 1 year (p=0.004 and 0.006, respectively); differences in disinhibition among groups were not significant using the adjusted p-value for pair-wise comparisons.
Table 3.
Psychosocial Variables in Large Gainers, Small Gainers and Weight Stable Participants (adjusted for baseline level, time, and randomization group)
| Baseline | 1 Year | 2 Year | 3 Year | Group | |||||
|---|---|---|---|---|---|---|---|---|---|
| Mean1 | SD | LS Mean | SE | LS Mean | SE | LS Mean | SE | P-Value | |
| CES-D | |||||||||
| Weight Stable | 5.6 | 4.8 | 6.3a | 0.5 | 6.5a | 0.5 | 6.3 | 0.6 | 0.0029 |
| Small Gainers | 5.3 | 4.2 | 7.1a,b | 0.5 | 7.0a,b | 0.5 | 8.1 | 0.5 | |
| Large Gainers | 5.6 | 4.3 | 9.3b | 0.8 | 9.5b | 0.9 | 8.9 | 1.0 | |
| Restraint | |||||||||
| Weight Stable | 10.6 | 1.9 | 11.4a | 0.2 | 11.2 | 0.2 | 11.2 | 0.2 | 0.0069 |
| Small Gainers | 10.4 | 1.9 | 11.3a | 0.2 | 11.1 | 0.2 | 11.2 | 0.2 | |
| Large Gainers | 10.5 | 1.8 | 10.4b | 0.3 | 10.8 | 0.3 | 10.4 | 0.3 | |
| Disinhibition | |||||||||
| Weight Stable | 9.2 | 2.5 | 9.1 | 0.2 | 0.2 | 0.2 | 9.1 | 0.2 | 0.048 |
| Small Gainers | 9.4 | 2.2 | 9.1 | 0.2 | 9.1 | 0.2 | 9.4 | 0.2 | |
| Large Gainers | 9.4 | 2.1 | 9.8 | 0.3 | 9.8 | 0.3 | 9.7 | 0.3 | |
| # of Life Events | |||||||||
| Weight Stable | 5.4 | 3.3 | 5.1 | 0.3 | 5.3 | 0.3 | 5.9 | 0.5 | 0.95 |
| Small Gainers | 5.6 | 4.1 | 4.7 | 0.3 | 5.5 | 0.3 | 6.0 | 0.3 | |
| Large Gainers | 6.6 | 4.3 | 4.7 | 0.5 | 6.3 | 0.5 | 5.8 | 0.5 | |
| Perceived Stress | |||||||||
| Weight Stable | 4.4 | 2.5 | 4.9 | 0.2 | 4.9 | 0.2 | 4.8 | 0.2 | 0.11 |
| Small Gainers | 4.1 | 2.8 | 5.3 | 0.2 | 5.4 | 0.2 | 5.2 | 0.2 | |
| Large Gainers | 5.2 | 2.5 | 5.9 | 0.4 | 6.0 | 0.5 | 5.0 | 0.4 | |
Ranges of questionnaires as follows: CES-D: 0–60; Restraint: 0–21; Disinhibition: 0–16; Perceived Stress: 0–16. The Life Events Questionnaire includes 67 yes/no items but many are inconsistent with each other. In this study, the total number of life events ranged from 0–24.
For each questionnaire, groups within columns with different superscripts (i.e., a, b) differ significantly from each other in post-hoc tests with Bonferroni corrections (p<0.016).
Adherence to Weight related behaviors.
There were no differences between gainers and stable in dietary intake at baseline (Table 4). However, at 2 years, the three groups differed in the HEI (60.5, 62.1 and 64.0 for Large Changes, Small Changes, and Weight Stable groups, respectively; p=0.015), but none of the pair-wise tests were significant.
Table 4.
Behavioral Variables in Large Gainers, Small Gainers and Weight Stable Participants (adjusted for baseline level, time, and randomization group)
| Baseline | 1 Year | 2 Year | Group | ||||
|---|---|---|---|---|---|---|---|
| Mean | SD | LS Mean | SE | LS Mean | SE | P-value | |
| Diet Variables | |||||||
| Calories / day | |||||||
| Weight Stable | 1630 | 591 | 1469 | 35 | 0.095 | ||
| Small Gainers | 1652 | 615 | 1505 | 34 | |||
| Large Gainers | 1769 | 751 | 1634 | 67 | |||
| Fat % | |||||||
| Weight Stable | 35.9 | 5.4 | 36.9 | 0.5 | 0.55 | ||
| Small Gainers | 35.1 | 5.8 | 36.9 | 0.5 | |||
| Large Gainers | 34.7 | 5.7 | 35.7 | 1.0 | |||
| Healthy Eating Index | |||||||
| Weight Stable | 60.6 | 11.0 | 64.0 | 0.7 | 0.047 | ||
| Small Gainers | 60.2 | 11.0 | 62.1 | 0.7 | |||
| Large Gainers | 58.6 | 12.3 | 60.5 | 1.4 | |||
| Physical Activity Variables | |||||||
| Sedentary (% of day) | |||||||
| Weight Stable | 64.4 | 9.2 | 64.6 | 0.7 | 64.0 | 0.7 | 0.44 |
| Small Gainers | 64.8 | 10.4 | 63.7 | 0.7 | 64.1 | 0.7 | |
| Large Gainers | 60.4 | 10.9 | 65.4 | 1.3 | 65.6 | 1.4 | |
| Light Activity (% of day) | |||||||
| Weight Stable | 25.2 | 7.6 | 24.1 | 0.6 | 24.7 | 0.6 | 0.21 |
| Small Gainers | 24.5 | 7.4 | 25.6 | 0.6 | 25.4 | 0.6 | |
| Large Gainers | 29.3 | 7.2 | 24.9 | 1.1 | 24.4 | 1.1 | |
| Moderate Activity (% of day) | |||||||
| Weight Stable | 9.7 | 4.3 | 10.1 | 0.4 | 10.7 | 0.4 | 0.038 |
| Small Gainers | 10.2 | 6.0 | 10.0 | 0.4 | 9.9 | 0.4 | |
| Large Gainers | 9.6 | 6.5 | 8.6 | 0.8 | 8.5 | 0.8 | |
| Vigorous Activity (% of day) | |||||||
| Weight Stable | 0.7 | 0.9 | 1.2 | 0.3 | 0.7 | 0.2 | 0.51 |
| Small Gainers | 0.6 | 0.9 | 0.7 | 0.3 | 0.7 | 0.2 | |
| Large Gainers | 0.6 | 0.9 | 0.8 | 0.5 | 1.1 | 0.3 | |
| Minutes/day of Moderate or Vigorous Activity | |||||||
| Weight Stable | 88.9 | 42.8 | 93.2 | 4.3 | 92.0 | 3.9 | 0.051 |
| Small Gainers | 91.1 | 54.6 | 88.2 | 4.3 | 86.1 | 3.9 | |
| Large Gainers | 83.7 | 57.5 | 75.2 | 7.9 | 77.5 | 7.5 | |
| Minutes/day of Moderate Vigorous Activity in 10-minute bouts | |||||||
| Weight Stable | 36.9 | 31.3 | 40.3 | 3.0 | 42.3 | 3.1 | 0.039 |
| Small Gainers | 40.7 | 40.7 | 39.9 | 3.0 | 39.2 | 3.0 | |
| Large Gainers | 34.7 | 42.4 | 27.8 | 5.5 | 28.9 | 5.8 | |
| Percent reporting frequent self-weighing * | |||||||
| Weight Stable | 40% | 69%a | 68%a | 0.0002 | |||
| Small Gainers | 26% | 57%a,b | 46%b | ||||
| Large Gainers | 38% | 45%b | 49%b | ||||
Includes participants reporting weighing themselves several times per week or ≥ each day. The full list of response options was: ≥ each day; several times per week; once per week; less than once per week; less than once per month; never.
The allocation of time in sedentary and light activity differed among the three groups at baseline (Table 4). Adjusting for baseline differences, there were also significant differences over the three years of follow-up among the 3 groups in percent of the day in moderate activity, and minutes of MVPA, both total and within 10-minute bouts. Although the Large Gainers reported the lowest levels of physical activity, none of the pairwise contrasts related to physical activity were significant.
Analyses adjusting for self-weighing at baseline showed that the three groups differed in self-weighing frequency at all subsequent time points (p=0.021, 0.003 and 0.053 at years 1, 2 and 3). Pair-wise differences indicated that those who remained Weight Stable were more likely to be frequent self-weighers at Year 1 than Large Gainers (p=0.0024) and at year 2 than both of the Gainer groups (both p =0.002). This difference is shown in Figure 2.
Figure 2:
Percent of Large Gainers, Small Gainers and Weight Stable Participants that Report Frequent Self-Weighing (≥ each day or several times per week)
Differences at 4 months.
Finally, when significant differences were seen at 1 year, we also compared the groups at 4 months, i.e. before the differences in weight gain were observed, to examine which variables preceded the weight gain. At 4 months, the proportion reporting frequent self-weighing already differed significantly for Large Gainers vs Weight Stable subjects (86% vs. 70%, p<.001) as did dietary restraint (11.6 vs 10.8, p=0.014). When the Small Gainers were included in the analysis, the 3-group differences were no longer significant (p=0.083 for three group comparison of self-weighing and p=0.06 for restraint), but the Small Gainers were intermediate (80% frequent self-weighers and restraint scores of 11.3). There were no other significant differences at 4 months.
DISCUSSION
Despite being enrolled in a weight gain prevention program, almost 10% of these young adults gained at least 10% of their body weight over 3 years. This group is of concern since weight gains of this magnitude have negative effects on changes in cardiovascular risk factors.4 Prior studies have shown that individuals who experience early and rapid weight gain during young adulthood are more likely to be on a steeper weight gain trajectory and at greater risk of obesity-related conditions.29 Likewise, weight gains between ages 18–35 years are more strongly associated with mortality than weight gains at older ages.30
An additional 30% had small weight gains (2.6% to 9.9%) over the 3 years. As expected, this group fell between the other two groups on many of the outcomes under investigation. When significant differences among the three weight change groups were detected, Large Gainers and Weight Stable were most likely to differ from each other. Small Gainers also differed from Weight Stable on several dimensions (higher BMI at baseline, higher CES-D and lower dietary restraint at year 1, and less frequent self-weighing at 2 years), but differed from Large Gainers only in weight suppression.
We considered four possible explanations for the large weight gain—1) that they are predisposed to be heavier; 2) that they are more likely to have had a pregnancy; 3) that they are dealing with more depressive symptoms, and 4) that they are less adherent to the behavioral strategies recommended for weight gain prevention.
A stronger predisposition to obesity among Large Gainers is suggested by their higher BMI at baseline and higher weight at their heaviest previous weight. Although they also reported a greater history of weight cycling, this difference did not persist after adjustment for BMI. . These predictors, especially the higher baseline BMI, have been noted previously.11,12 Lowe et al (2015) found that weight variability over a 6-month period predicted weight gain over 2 years in college females. Weight suppression has also been predictive of weight gain.31,32
In contrast, we found little support for the hypothesis that the weight gain resulted from pregnancies. Although a somewhat greater percent of women in the Large Gainers group had a pregnancy over the 3 years, within Large Gainers those who became pregnant and those who did not had similar weight gain as did those who became pregnant or not within the other two groups. These data suggest that something other than their pregnancy was contributing to the difference in weight gain among the 3 groups. However, we cannot rule out that Large Gainers with pregnancies might not have gained weight if pregnancy did not occur. Observational studies have shown complex relationships between pregnancy and weight gain in young adults.13
There was some support for the hypothesis that the difference in weight gain between Large and Small Gainers and Weight Stable group might relate to their level of depressive symptoms during the intervention. However, since these differences were not seen at 4 months, prior to the weight gain, they may have resulted from (rather than leading to) the weight gain. In addition, while statistically significant, the differences in depression were not clinically significant. There was some evidence of differences in baseline level of perceived stress, but again these differences were not clinically meaningful. Prior studies examining the association between depression in adolescence and subsequent weight gain have yielded inconsistent results.7,15
Weight gain appeared to be related to lower self-reported use of the behavioral recommendations for weight gain prevention. This is seen most clearly in the proportion that reported frequent self-weighing. At baseline 26–47% reported frequent self-weighing, but this increased during the study to almost 70% in Weight Stable participants, whereas it remained around 45% within the Large Gainers. Interestingly, Small Gainers initially had increases in frequent self-weighing, but the increase was not sustained at 2 years. Dietary restraint also differed among the three groups, with Large Gainers reporting lower levels than either of the other groups. Significant differences in frequency of self-weighing and dietary restraint were seen at month 4, suggesting that these differences in adherence preceded the weight gain. The three groups also differed in HEI and in minutes of moderate and vigorous physical activity bout-related activity at 2 years. However, none of the pair-wise comparisons were significant and these differences were not seen at month 4.
In conclusion, large weight gains (>10% of initial weight) occur in almost 10% of young adults who are participating in a weight gain prevention intervention study. These participants may be identified early by their baseline weight and weight history and by their lower probability of frequent self-weighing. Using a more personalized approach to weight gain prevention, in which those young adults who are at risk of gaining significant amounts of weight are provided with a more intensive intervention, may help to advance the field of weight gain prevention in this vulnerable population.
Bullet Points.
What is known about this subject:
Young adults are at increased risk of weight gain, leading to greater future health problems.
Observational studies have identified demographic and behavioral factors associated with weight gain in young adults and with weight loss in obesity treatments, but it is not known whether there are similar predictors of weight gain during a prevention program.
Identifying variables that are associated with poor outcomes (large weight gains) during these programs could permit more personalized approaches to weight gain prevention.
What this study adds:
We found that almost 10% of young adults gained >10% of their body weight over 3 years while enrolled in a weight gain prevention study.
Large Gainers (>10%; N=43) could be identified at baseline by their prior history of greater weight issues (higher current BMI, greater history of weight cycling and being further below their highest weight ever) than Weight Stable (±2.5%; N=143); Small Gainers (2.6–9.9%; N=147) were intermediate.
At month 4 (prior to the time when significant differences in weight change emerged), Large Gainers reported being less likely than Weight Stable to weigh themselves frequently (daily or several times per week).
Relative to Weight Stable, Large Gainers also reported more depressive symptoms, lower dietary restraint, and less frequent self-weighing at later time points (1, 2, and/or 3 years).
Acknowledgements:
Data Sharing Plan
All de-identified data will be available from the NHLBI Data Repository (https://biolincc.nhlbi.nih.gov/home/) by January 2020. This will include the data, data dictionary, and informed consent documents. Data will be made available to others according to NHLBI policies.
*SNAP Research Group Members
The Study of Novel Approaches to Weight Gain Prevention (SNAP) Research Group members (listed alphabetically by site unless otherwise specified) were from the following clinical sites:
Miriam Hospital/Brown Medical School: Rena R. Wing, PhD (principal investigator); Elissa Jelalian, PhD (co-investigator); Erica Ferguson Robichaud, MSW, RD (program coordinator); Jose DaCruz; Kaitlyn Dahlborg, BS; Caitlin Egan, MS; Denise Fernandes Pierre, BS; Chelsea Pimentel, BA; Wilza Rodrigues, BS; Samantha Williams, MSc.
University of North Carolina at Chapel Hill: Deborah Tate, PhD (principal investigator); Kristen Polzien, PhD (program coordinator); Candice Alick, MS; Loneke Blackman, MS, RD; Molly Diamond, MPH; Karen E. Hatley, MPH; Brett Meager; Julianne Power, MS; Brooke Tompkins Nezami, MA; Paige Trexler; Carmina Valle, PhD, MPH.
Wake Forest School of Medicine: Mark A. Espeland, PhD (principal investigator); Judy L. Bahnson, BA, CCRP (co-investigator); Letitia H. Perdue, MS (program coordinator); Cheryl Bentley; Patty Davis, BS; Katelyn Garcia, MS; Rebecca H. Neiberg, MS; Julia Robertson, BS; Greg Russell, MS.
Consultant co-investigators were Cora E. Lewis, MD, MSPH, University of Alabama at Birmingham; Amy A. Gorin, PhD, University of Connecticut; Jessica G. LaRose, PhD, Virginia Commonwealth University School of Medicine.
Funding: National Institutes of Health grants U01HL090864, U01HL090875 and 5R01HL127341.
Dr. Tate serves on the Scientific Advisory board and receives grant funding from WW.
Footnotes
Trial Registration: clinicaltrials.gov Identifier:
Disclosures: None of the other authors report having any financial disclosures.
REFERENCES
- 1.Unick JL, Neiberg RH, Hogan PE, et al. Weight change in the first 2 months of a lifestyle intervention predicts weight changes 8 years later. Obesity (Silver Spring). 2015;23(7):1353–1356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Jakicic JM, Marcus BH, Lang W, Janney C. Effect of exercise on 24-month weight loss maintenance in overweight women. Arch Intern Med. 2008;168(14):1550–1559; discussion 1559–1560. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lytle LA, Laska MN, Linde JA, et al. Weight-Gain Reduction Among 2-Year College Students: The CHOICES RCT. Am J Prev Med. 2017;52(2):183–191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Truesdale KP, Stevens J, Lewis CE, Schreiner PJ, Loria CM, Cai J. Changes in risk factors for cardiovascular disease by baseline weight status in young adults who maintain or gain weight over 15 years: the CARDIA study. Int J Obes (Lond). 2006;30(9):1397–1407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Williamson DF, Kahn HS, Remington PL, Anda RF. The 10-year incidence of overweight and major weight gain in US adults. Arch Intern Med. 1990;150(3):665–672. [PubMed] [Google Scholar]
- 6.Ball K, Crawford D. An investigation of psychological, social and environmental correlates of obesity and weight gain in young women. Int J Obes (Lond). 2006;30(8):1240–1249. [DOI] [PubMed] [Google Scholar]
- 7.Viner RM, Cole TJ. Who changes body mass between adolescence and adulthood? Factors predicting change in BMI between 16 year and 30 years in the 1970 British Birth Cohort. Int J Obes (Lond). 2006;30(9):1368–1374. [DOI] [PubMed] [Google Scholar]
- 8.Pereira MA, Kartashov AI, Ebbeling CB, et al. Fast-food habits, weight gain, and insulin resistance (the CARDIA study): 15-year prospective analysis. Lancet. 2005;365(9453):36–42. [DOI] [PubMed] [Google Scholar]
- 9.Brown WJ, Trost SG. Life transitions and changing physical activity patterns in young women. Am J Prev Med. 2003;25(2):140–143. [DOI] [PubMed] [Google Scholar]
- 10.Wing RR, Tate DF, Espeland MA, et al. Innovative self-regulation strategies to reduce weight gain in young adults: the study of novel appproaches to weight gain prevention (SNAP) randomized clinical trial. JAMA Intern Med. 2016;176(6):755–762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ball K, Brown W, Crawford D. Who does not gain weight? Prevalence and predictors of weight maintenance in young women. Int J Obes Relat Metab Disord. 2002;26(12):1570–1578. [DOI] [PubMed] [Google Scholar]
- 12.Burke GL, Bild DE, Hilner JE, Folsom AR, Wagenknecht LE, Sidney S. Differences in weight gain in relation to race, gender, age and education in young adults: the CARDIA Study. Coronary Artery Risk Development in Young Adults. Ethn Health. 1996;1(4):327–335. [DOI] [PubMed] [Google Scholar]
- 13.Gunderson E, Murtaugh M, Lewis C, Quesenberry C, West DS, Sidney S. Excess gains in weight and waist circumference associated with childbearing: The Coronary Artery Risk Development in Young Adults Study (CARDIA). Int J Obes. 2004;28(4):525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Needham BL, Epel ES, Adler NE, Kiefe C. Trajectories of change in obesity and symptoms of depression: the CARDIA study. Am J Public Health. 2010;100(6):1040–1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Richardson LP, Davis R, Poulton R, et al. A Longitudinal Evaluation of Adolescent Depression and Adult Obesity. Arch Pediatr Adolesc Med. 2003;157(8):739–745. [DOI] [PubMed] [Google Scholar]
- 16.Korkeila M, Kaprio J, Rissanen A, Koskenvuo M, Sörensen T. Predictors of major weight gain in adult Finns: stress, life satisfaction and personality traits. Int J Obes. 1998;22(10):949. [DOI] [PubMed] [Google Scholar]
- 17.Gaysina D, Hotopf M, Richards M, Colman I, Kuh D, Hardy R. Symptoms of depression and anxiety, and change in body mass index from adolescence to adulthood: results from a British birth cohort. Psychol Med. 2011;41(1):175–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schmitz KH, Jacobs DR Jr., Leon AS, Schreiner PJ, Sternfeld B. Physical activity and body weight: associations over ten years in the CARDIA study. Coronary Artery Risk Development in Young Adults. Int J Obes Relat Metab Disord. 2000;24(11):1475–1487. [DOI] [PubMed] [Google Scholar]
- 19.Wing RR, Tate D, Espeland M, et al. Weight gain prevention in young adults: design of the study of novel approaches to weight gain prevention (SNAP) randomized controlled trial. BMC public health. 2013;13(1):300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Guidelines (2013) for managing overweight and obesity in adults. Preface to the Expert Panel Report (comprehensive version which includes systematic evidence review, evidence statements, and recommendations). Obesity (Silver Spring). 2014;22 Suppl 2:S40. [DOI] [PubMed] [Google Scholar]
- 21.Turvey CL, Wallace RB, Herzog R. A revised CES-D measure of depressive symptoms and a DSM-based measure of major depressive episodes in the elderly. Int Psychogeriatr. 1999;11(2):139–148. [DOI] [PubMed] [Google Scholar]
- 22.Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav. 1983;24(4):385–396. [PubMed] [Google Scholar]
- 23.Scherwitz L, Perkins L, Chesney M, Hughes G. Cook-Medley Hostility scale and subsets: relationship to demographic and psychosocial characteristics in young adults in the CARDIA study. Psychosom Med. 1991;53(1):36–49. [DOI] [PubMed] [Google Scholar]
- 24.Stunkard AJ, Messick S. Eating Inventory Manual. New York: Psychological Corporation; 1988. [Google Scholar]
- 25.Block G, Hartman AM, Dresser CM, Carroll MD, Gannon J, Gardner L. A data-based approach to diet questionnaire design and testing. Am J Epidemiol. 1986;124:453–469. [DOI] [PubMed] [Google Scholar]
- 26.Guenther PM, Casavale KO, Reedy J, et al. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet. 2013;113(4):569–580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Jakicic JM, Marcus M, Gallagher KI, et al. Evaluation of the SenseWear Pro Armband to assess energy expenditure during exercise. Med Sci Sports Exerc. 2004;36(5):897–904. [DOI] [PubMed] [Google Scholar]
- 28.Gorin AA, Gokee LaRose J, Espeland MA, et al. Eating pathology and psychological outcomes in young adults in self-regulation interventions using daily self-weighing. Health Psychol. 2019;38(2):143–150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Malhotra R, Østbye T, Riley CM, Finkelstein EA. Young adult weight trajectories through midlife by body mass category. Obesity. 2013;21(9):1923–1934. [DOI] [PubMed] [Google Scholar]
- 30.Adams KF, Leitzmann MF, Ballard-Barbash R, et al. Body mass and weight change in adults in relation to mortality risk. Am J Epidemiol. 2014;179(2):135–144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Lowe MR, Marti CN, Lesser EL, Stice E. Weight suppression uniquely predicts body fat gain in first-year female college students. Eat Behav. 2019;32:60–64. [DOI] [PubMed] [Google Scholar]
- 32.Lowe MR, Feig EH, Winter SR, Stice E. Short-term variability in body weight predicts long-term weight gain. Am J Clin Nutr. 2015;102(5):995–999. [DOI] [PMC free article] [PubMed] [Google Scholar]