Randomization is an important tool used to establish causal inferences in studies designed to further our understanding of questions related to obesity and nutrition. To take advantage of the inferences afforded by randomization, scientific standards must be upheld during the planning, execution, analysis, and reporting of such studies. We discuss ten errors in randomized experiments from real-world examples from the literature and outline best practices for their avoidance. These ten errors include: representing nonrandom allocation as random, failing to adequately conceal allocation, not accounting for changing allocation ratios, replacing subjects in nonrandom ways, failing to account for non-independence, drawing inferences by comparing statistical significance from within-group comparisons instead of between-groups, pooling data and breaking the randomized design, failing to account for missing data, failing to report sufficient information to understand study methods, and failing to frame the causal question as testing the randomized assignment per se. We hope that these examples will aid researchers, reviewers, journal editors, and other readers to endeavor to a high standard of scientific rigor in randomized experiments within obesity and nutrition research.

Objectives: Regarding weight management, changes in exercise behavior can also influence nutrition behavior by application of self-regulatory psychological resources across behaviors (transfer effect). This study aimed to determine: (1) if changes in exercise frequency in young adulthood predict simultaneous changes in fruit/vegetable intake (transfer as co-occurrence); and (2) if exercise frequency affects future fruit/vegetable intake (transfer as carry-over). Methods: 6244 respondents of the National Longitudinal Survey of Youth 1997 were followed at ages 18–22 (Time-1), 23–27 (Time-2), and 27–31 (Time-3). Repeated measures analysis of variance and hierarchical multiple regression determined if the change in exercise frequency between Time-1 and Time-2 was associated with simultaneous and sequential changes in fruit/vegetable intake frequency, controlling for sex, race/ethnicity, education, income, body mass index, and baseline fruit/vegetable intake. Results: Only 9% continued exercising for 30 minutes more than 5 days/week, while 15% transitioned to adequate exercise and another 15% transitioned to inadequate exercise; for both fruits and vegetables, intake of once per day or more increased with age. Males were more likely to exercise adequately and females to consume fruits/vegetables adequately. Exercise frequency transition was linearly associated with concurrent fruit/vegetable intake during Time-1 and Time-2. The highest increase in mean fruit/vegetable intake occurred for participants who transitioned from inadequate to adequate exercise. A significant Time-2 exercise frequency effect on Time-3 fruit/vegetable intake emerged, after accounting for baseline intake. Increase in Time-2 exercise by one day/week resulted in increased Time-3 fruit and vegetable intakes by 0.17 and 0.13 times/week, respectively. Conclusion: Transfer effects, although usually discussed in interventions, may also be applicable to voluntary behavior change processes. Newly engaging in and continuing exercise behavior over time may establish exercise habits that facilitate improved fruit/vegetable consumption. Interventions that facilitate transferring resources across behaviors likely will enhance this effect.

Background: The accuracy and reliability of self-reported height and weight among adolescents in the process of calculating BMI is usually subject to bias. The aim of this study was to determine whether over- and under-reporting of self-reported height and weight existed among US high school students by weight category; if so, to examine anthropometric, behavioral, and demographic factors associated with over- and under-reporting. Methods: Data were retrieved from the National Youth Physical Activity and Nutrition Study, 2010, a nationally representative sample (7160 students, grades 9–12). Analysis of variance was performed to determine any significant difference between weight categories in misreporting. Discriminant function analysis and sequential logistic regression were executed to detect behavioral and demographic predictors of reporting accuracy, respectively. Results: The mean over-reporting of height and under-reporting of weight were 1.1 cm and 1.020 kg, respectively, which underestimated BMI and BMI percentile by 0.671 and 2.734, respectively. Use of self-reported height and weight for BMI calculation overestimated prevalence of healthy weight by 3.8% and underestimated prevalence of obesity by 4.1%. Underweight students under-reported height and over-reported weight, whereas overweight and obese students over-reported height and under-reported weight. Reporting accuracy of females was significantly higher. Weight loss behaviors, both healthy and unhealthy, were associated with BMI underestimation, whereas fast foods and screen time were associated with overestimation. Conclusion: Whenever possible, measuring height and weight is essential. However, because many studies must rely on self-reported values alone, additional research should examine the relationships between misreport of anthropometric data and lifestyle features in diverse adolescent samples to better interpret self-reported anthropometric data.

Background: Schools increasingly prioritize social-emotional competence and bullying and cyberbullying prevention, so the development of novel, low-cost, and high-yield programs addressing these topics is important. Further, rigorous assessment of interventions prior to widespread dissemination is crucial. Objective: This study assesses the effectiveness and implementation fidelity of the ACT Out! Social Issue Theater program, a 1-hour psychodramatic intervention by professional actors; it also measures students’ receptiveness to the intervention. Methods: This study is a 2-arm cluster randomized control trial with 1:1 allocation that randomized either to the ACT Out! intervention or control (treatment as usual) at the classroom level (n=76 classrooms in 12 schools across 5 counties in Indiana, comprised of 1571 students at pretest in fourth, seventh, and tenth grades). The primary outcomes were self-reported social-emotional competence, bullying perpetration, and bullying victimization; the secondary outcomes were receptiveness to the intervention, implementation fidelity (independent observer observation), and prespecified subanalyses of social-emotional competence for seventh- and tenth-grade students. All outcomes were collected at baseline and 2-week posttest, with planned 3-months posttest data collection prevented due to the COVID-19 pandemic. Results: Intervention fidelity was uniformly excellent (>96% adherence), and students were highly receptive to the program. However, trial results did not support the hypothesis that the intervention would increase participants’ social-emotional competence. The intervention’s impact on bullying was complicated to interpret and included some evidence of small interaction effects (reduced cyberbullying victimization and increased physical bullying perpetration). Additionally, pooled within-group reductions were also observed and discussed but were not appropriate for causal attribution. Conclusions: This study found no superiority for a 1-hour ACT Out! intervention compared to treatment as usual for social-emotional competence or offline bullying, but some evidence of a small effect for cyberbullying. On the basis of these results and the within-group effects, as a next step, we encourage research into whether the ACT Out! intervention may engender a bystander effect not amenable to randomization by classroom. Therefore, we recommend a larger trial of the ACT Out! intervention that focuses specifically on cyberbullying, measures bystander behavior, is randomized by school, and is controlled for extant bullying prevention efforts at each school. Trial Registration: Clinicaltrials.gov NCT04097496; https://clinicaltrials.gov/ct2/show/NCT04097496 International Registered Report Identifier (IRRID): RR2-10.2196/17900