Teacher's Corner: An R Shiny App for Sensitivity Analysis for Latent Growth Curve Mediation.

Mechanisms of behavior change are the processes through which interventions are hypothesized to cause changes in outcomes. Latent growth curve mediation models (LGCMM) are recommended for investigating the mechanisms of behavior change because LGCMM models establish temporal precedence of change from the mediator to the outcome variable. The Correlated Augmented Mediation Sensitivity Analyses (CAMSA) App implements sensitivity analysis for LGCMM models to evaluate if a mediating path (mechanism) is robust to potential confounding variables. The CAMSA approach is described and applied to simulated data, and data from a research study exploring a mechanism of change in the treatment of substance use disorder.

A permutation test for assessing the presence of individual differences in treatment effects.

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.

Latent Class Mediation: A Comparison of Six Approaches.

Latent class mediation modeling is designed to estimate the mediation effect when both the mediator and the outcome are latent class variables. We suggest using an adjusted one-step approach in which the latent class models for the mediator and the outcome are estimated first to decide on the number of classes, then the latent class models and the mediation model are jointly estimated. We present both an empirical demonstration and a simulation study to compare the performance of this one-step approach to a standard three-step approach with modal assignment (modal) and four different modern three-step approaches. Results from the study indicate that unadjusted modal, which ignores the classification errors of the latent class models, produced biased mediation effects. On the other hand, the adjusted one-step approach and the modern three-step approaches performed well with respect to bias for estimating mediation effects, regardless of measurement quality (i.e., model entropy) and latent class size. Among the three-step approaches we investigated, the maximum likelihood method with modal assignment and the BCH correction with robust standard error estimators are good alternatives to the adjusted one-step approach, given their unbiased standard error estimations.

The (Lack of) Replication of Self-Reported Mindfulness as a Mechanism of Change in Mindfulness-Based Relapse Prevention for Substance Use Disorders.

The development and evaluation of mindfulness-based interventions for a variety of psychological and medical disorders has grown exponentially over the past 20 years. Yet, calls for increasing the rigor of mindfulness research and recognition of the difficulties of conducting research on the topic of mindfulness have also increased. One of the major difficulties is the measurement of mindfulness, with varying definitions across studies and ambiguity with respect to the meaning of mindfulness. There is also concern about the reproducibility of findings given few attempts at replication. The current secondary analysis addressed the issue of reproducibility and robustness of the construct of self-reported mindfulness across two separate randomized clinical trials of mindfulness-based relapse prevention (MBRP), as an aftercare treatment for substance use disorder. Specifically, we tested the robustness of our previously published findings, which identified a latent construct of mindfulness as a significant mediator of the effect of MBRP on reducing craving following treatment. First, we attempted to replicate the findings in a separate randomized clinical trial of MBRP. Second, we conducted sensitivity analyses to test the assumption of the no-omitted confounder bias in a mediation model. The effect of MBRP on self-reported mindfulness and overall mediation effect failed to replicate in a new sample. The effect of self-reported mindfulness in predicting craving following treatment did replicate and was robust to the no-omitted confounder bias. The results of this work shine a light on the difficulties in the measurement of mindfulness and the importance of examining the robustness of findings.

An evaluation of the bootstrap for model validation in mixture models.

Bootstrapping has been used as a diagnostic tool for validating model results for a wide array of statistical models. Here we evaluate the use of the non-parametric bootstrap for model validation in mixture models. We show that the bootstrap is problematic for validating the results of class enumeration and demonstrating the stability of parameter estimates in both finite mixture and regression mixture models. In only 44% of simulations did bootstrapping detect the correct number of classes in at least 90% of the bootstrap samples for a finite mixture model without any model violations. For regression mixture models and cases with violated model assumptions, the performance was even worse. Consequently, we cannot recommend the non-parametric bootstrap for validating mixture models. The cause of the problem is that when resampling is used influential individual observations have a high likelihood of being sampled many times. The presence of multiple replications of even moderately extreme observations is shown to lead to additional latent classes being extracted. To verify that these replications cause the problems we show that leave-k-out cross-validation where sub-samples taken without replacement does not suffer from the same problem.

Longitudinal relationships between self-concept for physical activity and neighborhood social life as predictors of physical activity among older African American adults.

BACKGROUND: Engaging in regular physical activity (PA) as an older adult has been associated with numerous physical and mental health benefits. The aim of this study is to directly compare how individual-level cognitive factors (self-efficacy for PA, self-determined motivation for PA, self-concept for PA) and neighborhood perceptions of the social factors (neighborhood satisfaction, neighborhood social life) impact moderate-to-vigorous physical activity (MVPA) longitudinally among older African American adults. METHODS: Data were analyzed from a sub-set of older African American adults (N = 224, M age = 63.23 years, SD = 8.74, 63.23% female, M Body Mass Index = 32.01, SD = 7.52) enrolled in the Positive Action for Today's Health trial. MVPA was assessed using 7-day accelerometry-estimates and psychosocial data (self-efficacy for PA, self-determined motivation for PA, self-concept for PA, neighborhood satisfaction, neighborhood social life) were collected at baseline, 12-, 18-, and 24-months. RESULTS: Multilevel growth modeling was used to examine within- and between-person effects of individual-level cognitive and social environmental factors on MVPA. At the between-person level, self-concept (b = 0.872, SE = 0.239, p < 0.001), and neighborhood social life (b = 0.826, SE = 0.176, p < 0.001) predicted greater MVPA, whereas neighborhood satisfaction predicted lower MVPA (b = -0.422, SE = 0.172, p = 0.015). Among the between-person effects, only average social life was moderated by time (b = 0.361, SE = 0.147, p = 0.014), indicating that the impact of a relatively positive social life on MVPA increased across time. At the within-person level, positive increases in self-concept (b = 0.294, SE = 0.145, p = 0.043) and neighborhood social life (b = 0.270, SE = 0.113, p = 0.017) were associated with increased MVPA. CONCLUSIONS: These results suggest that people with a higher average self-concept for PA and a more positive social life engaged in greater average MVPA. Additionally, changes in perceptions of one's neighborhood social life and one's self-concept for PA were associated with greater MVPA over 2 years. These factors may be particularly relevant for future interventions targeting long-term change and maintenance of MVPA in older African Americans. TRIAL REGISTRATION: ClinicalTrials.Gov # NCT01025726 registered 1 December 2009.

Modeling predictors of latent classes in regression mixture models.

The purpose of the current study is to provide guidance on a process for including latent class predictors in regression mixture models. We first examine the performance of current practice for using the 1-step and 3-step approaches where the direct covariate effect on the outcome is omitted. None of the approaches show adequate estimates of model parameters. Given that the step-1 of the three-step approach shows adequate results in class enumeration, we suggest using an alternative approach: 1) decide the number of latent classes without predictors of latent classes and 2) bring the latent class predictors into the model with the inclusion of hypothesized direct covariates effects. Our simulations show that this approach leads to good estimates for all model parameters. The proposed approach is demonstrated by using empirical data to examine the differential effects of family resources on students' academic achievement outcome. Implications of the study are discussed.

Not quite normal: Consequences of violating the assumption of normality in regression mixture models.

Regression mixture models are a new approach for finding differential effects which have only recently begun to be used in applied research. This approach comes at the cost of the assumption that error terms are normally distributed within classes. The current study uses Monte Carlo simulations to explore the effects of relatively minor violations of this assumption, the use of an ordered polytomous outcome is then examined as an alternative which makes somewhat weaker assumptions, and finally both approaches are demonstrated with an applied example looking at differences in the effects of family management on the highly skewed outcome of drug use. Results show that violating the assumption of normal errors results in systematic bias in both latent class enumeration and parameter estimates. Additional classes which reflect violations of distributional assumptions are found. Under some conditions it is possible to come to conclusions that are consistent with the effects in the population, but when errors are skewed in both classes the results typically no longer reflect even the pattern of effects in the population. The polytomous regression model performs better under all scenarios examined and comes to reasonable results with the highly skewed outcome in the applied example. We recommend that careful evaluation of model sensitivity to distributional assumptions be the norm when conducting regression mixture models.

An overview of the "Positive Action for Today's Health" (PATH) trial for increasing walking in low income, ethnic minority communities.

BACKGROUND: Ethnic minorities and lower-income adults have among the highest rates of obesity and lowest levels of regular physical activity (PA). The Positive Action for Today's Health (PATH) trial compares three communities that are randomly assigned to different levels of an environmental intervention to improve safety and access for walking in low income communities. DESIGN AND SETTING: Three communities matched on census tract information (crime, PA, ethnic minorities, and income) were randomized to receive either: an intervention that combines a police-patrolled-walking program with social marketing strategies to promote PA, a police-patrolled-walking only intervention, or no-walking intervention (general health education only). Measures include PA (7-day accelerometer estimates), body composition, blood pressure, psychosocial measures, and perceptions of safety and access for PA at baseline, 6, 12, 18, and 24 months. INTERVENTION: The police-patrolled walking plus social marketing intervention targets increasing safety (training community leaders as walking captains, hiring off-duty police officers to patrol the walking trail, and containing stray dogs), increasing access for PA (marking a walking route), and utilizes a social marketing campaign that targets psychosocial and environmental mediators for increasing PA. MAIN HYPOTHESES/OUTCOMES: It is hypothesized that the police-patrolled walking plus social marketing intervention will result in greater increases in moderate-to-vigorous PA as compared to the police-patrolled-walking only or the general health intervention after 12 months and that this effect will be maintained at 18 and 24 months. CONCLUSIONS: Implications of this community-based trial are discussed.

Analysis strategies for a community trial to reduce adolescent ATOD use: a comparison of random coefficient and ANOVA/ANCOVA models.

The Community Youth Development Study (CYDS) will evaluate the Communities That Care (CTC) operating system for its effects on alcohol, tobacco, drug use, and other outcomes among adolescents resident in the 24 participating communities. The CYDS employs a combination of both cross-sectional and cohort designs. We use data from an earlier study that included the CYDS communities to estimate power for CYDS intervention effects given several analytic models that might be applied to the multiple baseline and follow-up surveys that define the CYDS cross-sectional design. We compare pre-post mixed-model ANCOVA models against random coefficients models, both in one- and two-stage versions. The two-stage pre-post mixed-model ANCOVA offers the best power for the primary outcomes and will provide adequate power for detection of modest but important intervention effects.