IntroductionDifferences in dietary intake and physical activity may explain the higher prevalence of obesity among adolescents living in rural versus urban settings. The objective of this cross-sectional secondary analysis was to compare baseline dietary intake and physical activity of adolescents by rurality.MethodsWe analyzed data on 940 adolescents who participated in ACTION PAC (Adolescents Committed to Improvement of Nutrition and Physical Activity), an obesity prevention and management intervention trial conducted from 2014 through 2017 in 8 public high schools in the southwestern United States. Dietary intake was assessed with the Block Food Screener, and participants completed an exercise log and wore an accelerometer to provide data on physical activity. We compared data by rural–urban commuting area (RUCA) codes and log population density by using multilevel models, with students nested within zip code and repeated measures for accelerometer analysis.ResultsAfter adjusting for socioeconomic status and ethnicity, accelerometer data indicated that moderate-to-vigorous physical activity was 8.17 min/d (P = .02) higher and sedentary time was 20.42 min/d (P = .02) lower in moderately urban areas than in the urban reference area. Each 1-unit increase in log population density was associated with higher reported intake of whole grains (0.02 ounce equivalents, P = .03), potatoes (0.01 cup equivalents, P = .02), and added sugar (0.37 tsp, P = .02) after adjusting for socioeconomic status and ethnicity.ConclusionDifferences in reported dietary intake and physical activity level by measures of rurality were small and inconsistent in direction to explain the disparities observed in rural versus urban areas.

Objective: Numerous behavioral treatments for alcohol use disorder (AUD) are effective, but there are substantial individual differences in treatment response. This study examines the potential use of new methods for personalized medicine to test for individual differences in the effects of cognitive behavioral therapy (CBT) versus motivational enhancement therapy (MET) and to provide predictions of which will work best for individuals with AUD. We highlight both the potential contribution and the limitations of these methods. Method: We performed secondary analyses of abstinence among 1,144 participants with AUD participating in either outpatient or aftercare treatment who were randomized to receive either CBT or MET in Project MATCH. We first obtained predicted individual treatment effects (PITEs), as a function of 19 baseline client characteristics identified a priori by MATCH investigators. Then, we tested for the significance of individual differences and examined the predicted individual differences in abstinence 1 year following treatment. Predictive intervals were estimated for each individual to determine if they were 80% more likely to achieve abstinence in one treatment versus the other. Results: Results indicated that individual differences in the likelihood of abstinence at 1 year following treatment were significant for those in the outpatient sample, but not for those in the aftercare sample. Individual predictive intervals showed that 37% had a better chance of abstinence with CBT than MET, and 16% had a better chance of abstinence with MET. Obtaining predictions for a new individual is demonstrated. Conclusions: Personalized medicine methods, and PITE in particular, have the potential to identify individuals most likely to benefit from one versus another intervention. New personalized medicine methods play an important role in putting together differential effects due to previously identified variables into one prediction designed to be useful to clinicians and clients choosing between treatment options.

An important goal of personalized medicine is to identify heterogeneity in treatment effects and then use that heterogeneity to target the intervention to those most likely to benefit. Heterogeneity is assessed using the predicted individual treatment effects framework, and a permutation test is proposed to establish if significant heterogeneity is present given the covariates and predictive model or algorithm used for predicted individual treatment effects. We first show evidence for heterogeneity in the effects of treatment across an illustrative example data set. We then use simulations with two different predictive methods (linear regression model and Random Forests) to show that the permutation test has adequate type-I error control. Next, we use an example dataset as the basis for simulations to demonstrate the ability of the permutation test to find heterogeneity in treatment effects for a predicted individual treatment effects estimate as a function of both effect size and sample size. We find that the proposed test has good power for detecting heterogeneity in treatment effects when the heterogeneity was due primarily to a single predictor, or when it was spread across the predictors. Power was found to be greater for predictions from a linear model than from random forests. This non-parametric permutation test can be used to test for significant differences across individuals in predicted individual treatment effects obtained with a given set of covariates using any predictive method with no additional assumptions.

Background: Reports of physical activity (PA) measured via wrist-worn accelerometers in adolescents are limited. This study describes PA levels in adolescents at baseline of an obesity prevention and weight management trial. Methods: Adolescents (n = 930) at 8 high schools wore an accelerometer for 7 days, with average acceleration values of <50 mg, >150 mg, and >500 mg categorized as sedentary, moderate, and vigorous PA, respectively. In a 3-level mixed-effects generalized linear model, PA was regressed on sex, weight status, and day of week. Daily PA was nested within students, and students within schools, with random effects included for both. Results: Adolescents accumulated a median of 40 minutes daily of moderate to vigorous PA (MVPA). MVPA was significantly different for teens with obesity versus teens with normal weight (−5.4 min/d, P = .03); boys versus girls (16.3 min/d, P < .001); and Sundays versus midweek (−16.6 min/d, P < .001). Average sedentary time increased on weekends (Saturday: 19.1 min/d, P < .001; Sunday: 44.8 min, P < .001) relative to midweek but did not differ by sex or weight status. Conclusions: Interventions to increase PA in adolescents may benefit from focusing on increasing weekend PA and increasing MVPA in girls.

We integrate literature on gender and adolescent friendships to examine the association between adolescent dating violence victimization (ADVV) and relationship dissolution. In particular, we test whether ADVV increases the hazard of relationship dissolution among adolescent romances, and whether a number of friendship dynamics alter the association between ADVV and relationship dissolution. Using discrete time event history models from 5,787 romantically involved youth from the National Longitudinal Study of Adolescent to Adult Health (Add Health), results indicated, on average, ADVV was not associated with the hazard of relationship dissolution for girls or boys. However, the positive effect of ADVV was stronger for girls who did not withdraw from their friendships over the course of their romantic relationships. This study highlights the importance of peer groups and gender in shaping youths’ decisions to exit abusive relationships.