Power in Bayesian Mediation Analysis for Small Sample Research

Microbial Dysbiosis Linked to Metabolic Dysfunction-Associated Fatty Liver Disease in Asians: Prevotella copri Promotes Lipopolysaccharide Biosynthesis and Network Instability in the Prevotella Enterotype

Metabolic dysfunction-associated fatty liver disease (MAFLD), formerly known as non-alcoholic fatty liver disease (NAFLD), is characterized by hepatic fat accumulation by metabolic dysfunction. The rising prevalence of MAFLD, especially among Asians, may be associated with changes in gut microbiota. We investigated gut microbiota characteristics and potential mechanisms leading to MAFLD development according to enterotypes. Case-control studies examining the gut microbiota composition between MAFLD and non-MAFLD participants were searched in public databases until July 2023. Gut microbiota was categorized into two enterotypes by principal component analysis. According to the enterotypes, LEfSe, ALDEx2, XGBoost, and DCiPatho were utilized to identify differential abundances and pathogenic microbes in the gut between the MAFLD and non-MAFLD groups. We analyzed microbial community networks with the SprCC module and predicted microbial functions. In the Prevotella enterotype (ET-P), 98.6% of Asians and 65.1% of Caucasians were associated with MAFLD (p = 0.049). MAFLD incidence was correlated with enterotype, age, obesity, and ethnicity (p < 0.05). Asian MAFLD patients exhibited decreased Firmicutes and Akkermansia muciniphila and increased Bacteroidetes and P. copri. The pathogenicity scores were 0.006 for A. muciniphila and 0.868 for P. copri. The Asian MAFLD group showed decreased stability and complexity in the gut microbiota network. Metagenome function analysis revealed higher fructose metabolism and lipopolysaccharide (LPS) biosynthesis and lower animal proteins and α-linolenic acid metabolism in Asians with MAFLD compared with the non-MAFLD group. LPS biosynthesis was positively correlated with P. copri (p < 0.05). In conclusion, P. copri emerged as a potential microbial biomarker for MAFLD. These findings enhance our understanding of the pathological mechanisms of MAFLD mediated through the gut microbiota, providing insights for future interventions.

Multilevel mediation analysis in R: A comparison of bootstrap and Bayesian approaches

Mediation analysis in repeated measures studies can shed light on the mechanisms through which experimental manipulations change the outcome variable. However, the literature on interval estimation for the indirect effect in the 1-1-1 single mediator model is sparse. Most simulation studies to date evaluating mediation analysis in multilevel data considered scenarios that do not match the expected numbers of level 1 and level 2 units typically encountered in experimental studies, and no study to date has compared resampling and Bayesian methods for constructing intervals for the indirect effect in this context. We conducted a simulation study to compare statistical properties of interval estimates of the indirect effect obtained using four bootstrap and two Bayesian methods in the 1-1-1 mediation model with and without random effects. Bayesian credibility intervals had coverage closest to the nominal value and no instances of excessive Type I error rates, but lower power than resampling methods. Findings indicated that the pattern of performance for resampling methods often depended on the presence of random effects. We provide suggestions for selecting an interval estimator for the indirect effect depending on the most important statistical property for a given study, as well as code in R for implementing all methods evaluated in the simulation study. Findings and code from this project will hopefully support the use of mediation analysis in experimental research with repeated measures.

Reconsidering the failure model: Using a genetically controlled design to assess the spread of problems from reactive aggression to internalizing symptoms through peer rejection across the primary school years

According to the failure model (Patterson & Capaldi, 1990), peer rejection is the intermediary link between problem behaviors and internalizing symptoms. The present study tested the model with 464 monozygotic and same-sex dizygotic twin pairs (234 female, 230 male dyads). Teacher-reported reactive aggression and internalizing symptoms, and peer-reported peer rejection were collected at ages 6, 7, and 10 (from 2001 to 2008). Support for the failure model emerged in conventional non-genetically controlled analyses, but not twin-difference score analyses (which remove shared environmental and genetic contributions). Univariate biometric models attributed minimal variance in failure model variables to shared environmental factors, suggesting that genetic factors play an important unacknowledged role in developmental pathways historically ascribed to nonshared experiences in the failure model.

BayesGmed: An R-package for Bayesian causal mediation analysis

BACKGROUND

The past decade has seen an explosion of research in causal mediation analysis. However, most analytic tools developed so far rely on frequentist methods which may not be robust in the case of small sample sizes. In this paper, we propose a Bayesian approach for causal mediation analysis based on Bayesian g-formula, which will overcome the limitations of the frequentist methods.

METHODS

We created BayesGmed, an R-package for fitting Bayesian mediation models in R. The application of the methodology (and software tool) is demonstrated by a secondary analysis of data collected as part of the MUSICIAN study, a randomised controlled trial of remotely delivered cognitive behavioural therapy (tCBT) for people with chronic pain. We tested the hypothesis that the effect of tCBT would be mediated by improvements in active coping, passive coping, fear of movement and sleep problems. We then demonstrate the use of informative priors to conduct probabilistic sensitivity analysis around violations of causal identification assumptions.

RESULT

The analysis of MUSICIAN data shows that tCBT has better-improved patients' self-perceived change in health status compared to treatment as usual (TAU). The adjusted log-odds of tCBT compared to TAU range from 1.491 (95% CI: 0.452-2.612) when adjusted for sleep problems to 2.264 (95% CI: 1.063-3.610) when adjusted for fear of movement. Higher scores of fear of movement (log-odds, -0.141 [95% CI: -0.245, -0.048]), passive coping (log-odds, -0.217 [95% CI: -0.351, -0.104]), and sleep problem (log-odds, -0.179 [95% CI: -0.291, -0.078]) leads to lower odds of a positive self-perceived change in health status. The result of BayesGmed, however, shows that none of the mediated effects are statistically significant. We compared BayesGmed with the mediation R- package, and the results were comparable. Finally, our sensitivity analysis using the BayesGmed tool shows that the direct and total effect of tCBT persists even for a large departure in the assumption of no unmeasured confounding.

CONCLUSION

This paper comprehensively overviews causal mediation analysis and provides an open-source software package to fit Bayesian causal mediation models.

Unpacking the Relationship between Fear Motives and Self-Control Strategies among Managers: The Mediating Role of Intrusive Thoughts

In motive research, only a handful of studies have appeared on the correlates and antecedents of fear motives. In this research, we contribute to both research and practice by investigating the relationships between fear motives, intrusive thoughts, self-control strategies, and positive affect. We propose that fear motives, similar to trait anxiety, are positively associated with intrusive thoughts and that intrusive thoughts, in return, are negatively related to the frequency with which individuals employ self-control strategies. Finally, we propose that the frequency of self-control strategies is positively associated with positive affect. To test these, two field studies with managers (Study 1: N = 100 and Study 2: N = 80) were conducted. Bayesian mediation analyses showed that in both Study 1 and Study 2, fear motives were positively associated with intrusive thoughts, and intrusive thoughts were negatively related to self-control strategies. Additionally, in line with predictions, intrusive thoughts mediated the fear motives-self-control strategies relation. Finally, in Study 2, we found self-regulation strategies to be significantly and positively associated with positive affect. The theoretical and practical implications of the study are discussed.

Applying Monte Carlo Simulations to a Small Data Analysis of a Case of Economic Growth in COVID-19 Times

Studies on the going-on COVID-19 pandemic face small sample issues. In this context, Bayesian estimation is considered a viable alternative to frequentist estimation. Demonstrating the Bayesian approach's advantage in dealing with this problem, our research conducted a case study concerning ASEAN economic growth during the COVID-19 pandemic. By using Monte Carlo standard errors and interval hypothesis testing to check parameter bias within a Bayesian MCMC simulation study, the author obtained significant conclusions as follows: first, in insufficient sample sizes, in contrast to frequentist estimation, the Bayesian framework can offer meaningful results, that is, expansionary monetary and contractionary fiscal policies are positively associated with economic growth; second, in the face of a small sample, by incorporating more information into prior distributions for the model parameters, Bayesian Monte Carlo simulations perform so far better than naïve Bayesian and frequentist estimation; third, in case of a correctly specified prior, the inferences are robust to different prior specifications. The author strongly recommends applying specific informative priors to Bayesian analyses, particularly in small sample investigations.

Bayesian Mediation Analysis with Power Prior Distributions

Bayesian methods are often suggested as a solution for issues encountered in small sample research, however, Bayesian methods often require informative priors to outperform classical methods in these settings. Specifying accurate priors with respect to the true value of the parameter of interest is challenging and inaccurate informative priors can have detrimental effects on conclusions from the statistical analysis. This paper proposes an objective procedure for creating informative priors for mediation analysis based on a historical data set; the only requirements for implementing the procedure are that the data from the current study constitute a representative sample from the population of interest, and that the historical and current data sets contain measures of the same covariates and independent variable, mediator, and outcome. The simulation study findings show that the proposed method leads to appropriate amount of borrowing from the historical data set, which leads to increases in precision and power when the historical data and current data are exchangeable, and does not induce bias when the historical and current studies are not exchangeable. The proposed method is illustrated using data from the project PROsetta Stone, and we provide rstan code for implementing the proposed method.

Methods for Modeling Autocorrelation and Handling Missing Data in Mediation Analysis in Single Case Experimental Designs (SCEDs)

Single-Case Experimental Designs (SCEDs) are increasingly recognized as a valuable alternative to group designs. Mediation analysis is useful in SCEDs contexts because it informs researchers about the underlying mechanism through which an intervention influences the outcome. However, methods for conducting mediation analysis in SCEDs have only recently been proposed. Furthermore, repeated measures of a target behavior present the challenges of autocorrelation and missing data. This paper aims to extend methods for estimating indirect effects in piecewise regression analysis in SCEDs by (1) evaluating three methods for modeling autocorrelation, namely, Newey-West (NW) estimation, feasible generalized least squares (FGLS) estimation, and explicit modeling of an autoregressive structure of order one (AR(1)) in the error terms and (2) evaluating multiple imputation in the presence of data that are missing completely at random. FGLS and AR(1) outperformed NW and OLS estimation in terms of efficiency, Type I error rates, and coverage, while OLS was superior to the methods in terms of power for larger samples. The performance of all methods is consistent across 0% and 20% missing data conditions. 50% missing data led to unsatisfactory power and biased estimates. In light of these findings, we provide recommendations for applied researchers.

Sample Size Requirements for Simple and Complex Mediation Models

Mediation models have been widely used in many disciplines to better understand the underlying processes between independent and dependent variables. Despite their popularity and importance, the appropriate sample sizes for estimating those models are not well known. Although several approaches (such as Monte Carlo methods) exist, applied researchers tend to use insufficient sample sizes to estimate their models of interest, which might result in unstable and inaccurate estimation of the model parameters including mediation effects. In the present study, sample size requirements were investigated for four frequently used mediation models: one simple mediation model and three complex mediation models. For each model, path and structural equation modeling approaches were examined, and partial and complete mediation conditions were considered. Both the percentile bootstrap method and the multivariate delta method were compared for testing mediation effects. A series of Monte Carlo simulations was conducted under various simulation conditions, including those concerning the level of effect sizes, the number of indicators, the magnitude of factor loadings, and the proportion of missing data. The results not only present practical and general guidelines for substantive researchers to determine minimum required sample sizes but also improve understanding of which factors are related to sample size requirements in mediation models.

Cardiorespiratory fitness and lifestyle on severe COVID-19 risk in 279,455 adults: a case control study

Background

The impact of cardiorespiratory fitness (CRF) and other lifestyle-related factors on severe COVID-19 risk is understudied. The present study aims to investigate lifestyle-related and socioeconomic factors as possible predictors of COVID-19, with special focus on CRF, and to further study whether these factors may attenuate obesity- and hypertension-related risks, as well as mediate associations between socioeconomic factors and severe COVID-19 risk.

Methods

Out of initially 407,131 participants who participated in nationwide occupational health service screening between 1992 and 2020, n = 857 cases (70% men, mean age 49.9 years) of severe COVID-19 were identified. CRF was estimated using a sub-maximum cycle test, and other lifestyle variables were self-reported. Analyses were performed including both unmatched, n = 278,598, and sex-and age-matched, n = 3426, controls. Severe COVID-19 included hospitalization, intensive care or death due to COVID-19.

Results

Patients with more severe COVID-19 had significantly lower CRF, higher BMI, a greater presence of comorbidities and were more often daily smokers. In matched analyses, there was a graded decrease in odds for severe COVID-19 with each ml in CRF (OR = 0.98, 95% CI 0.970 to 0.998), and a two-fold increase in odds between the lowest and highest (< 32 vs. ≥ 46 ml·min⁻¹·kg⁻¹) CRF group. Higher BMI (per unit increase, OR = 1.09, 1.06 to 1.12), larger waist circumference (per cm, OR = 1.04, 1.02 to 1.06), daily smoking (OR = 0.60, 0.41 to 0.89) and high overall stress (OR = 1.36, 1.001 to 1.84) also remained significantly associated with severe COVID-19 risk. Obesity- and blood pressure-related risks were attenuated by adjustment for CRF and lifestyle variables. Mediation through CRF, BMI and smoking accounted for 9% to 54% of the associations between low education, low income and blue collar/low skilled occupations and severe COVID-19 risk. The results were consistent using either matched or unmatched controls.

Conclusions

Both lifestyle-related and socioeconomic factors were associated with risk of severe COVID-19. However, higher CRF attenuated the risk associated with obesity and high blood pressure, and mediated the risk associated with various socioeconomic factors. This emphasises the importance of interventions to maintain or increase CRF in the general population to strengthen the resilience to severe COVID-19, especially in high-risk individuals.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12966-021-01198-5.

The effects of combination levodopa-ropinirole on cognitive improvement and treatment outcome in individuals with cocaine use disorder: A bayesian mediation analysis

BACKGROUND

Chronic cocaine users show impairments in cognitive processes associated with dopamine (DA) circuitry. Medications aimed at bolstering cognitive functions via DA modulation might enhance treatment outcome.

METHODS

The trial used a double-blind, double-dummy, parallel-group design with four treatment arms comparing placebo (PLC) to levodopa/carbidopa 800 mg/200 mg alone (LR0), levodopa plus extended release (XR) ropinirole 2 mg (LR2) or XR ropinirole 4 mg (LR4). Adults (n = 110) with cocaine use disorder attended thrice weekly clinic visits for 10 weeks. Potential cognitive mediators assessed at week 5 consisted of measures of decision-making (Iowa Gambling Task, Risky Decision-Making Task), attention/impulsivity (Immediate Memory Task), motivation (Progressive Ratio task), and cognitive control (Cocaine Stoop task). The primary outcome measure was the treatment effectiveness score (TES) calculated as the number of cocaine-negative urines collected from weeks 6-10.

RESULTS

Bayesian mediation examined indirect and total effects of the relationships between each active treatment (compared to PLC) and TES. Total (direct) effects were supported for LR0 and LR2, but not for LR4. Indirect effects were tested for each mediator. Notably, 22.3 % and 35.4 % of the total effects of LR0 and LR2 on TES were mediated by changes in attention/impulsivity.

CONCLUSIONS

The hypothesized mediation effect was strongest for levodopa plus 2 mg ropinirole, indicating that this DA medication combination predicted change (improvement) in attention/impulsivity, which in turn predicted change (reduction) in cocaine use. This finding provides modest support for cognitive enhancement as a target for medications to treat cocaine use disorder.

A multilevel Bayesian framework for predicting municipal waste generation rates

Prediction of waste production is an essential part of the design and planning of waste management systems. The quality and applicability of such predictions depend heavily on model assumptions and the structure of the collected data. Ordinarily, municipal waste generation data are organized in hierarchical structures with municipal or county levels, and multilevel models can be used to generalize linear regression by directly incorporating the structure into the model. However, small amounts of data can limit the applicability of multilevel models and provide biased estimates. To cope with this problem, Bayesian estimation is often recommended as an alternative to frequentist estimation, such as least squares or maximum likelihood estimation. This paper proposes a multilevel framework under a Bayesian approach to model municipal waste generation with hierarchical data structures. Using a real-world dataset of municipal waste generation in Denmark, the predictive accuracy of multilevel models is compared to aggregated and disaggregated Bayesian models using socio-economic external variables. Results show that Bayesian multilevel models outperform the other models in prediction accuracy, based on the leave-one-out information criterion. A comparison of the Bayesian approach with its frequentist alternative shows that the Bayesian model is more conservative in coefficient estimation, with estimates shrinking to the grand mean and broader credible intervals, in contrast with narrower confidence intervals produced by the frequentist models.

Contamination fear in the time of the COVID-19 pandemic: A moderated mediation quasi-experimental model of the effect of disgust on outgroup bias towards diaspora

Disgust sensitivity plays a key role in generating and maintaining outgroup biases. To test our hypotheses, we used a quasi-experimental between-subjects design, in which participants were randomly assigned to a disgust induction condition (N = 102) or a non-induction neutral group (N = 92). The induction scenario featured the return of the diaspora to their home country due to COVID-19 concerns. In one scenario, the diaspora lied about the country they arrived from, and in the other, there was no moral transgression. We hypothesized that the effect of disgust sensitivity on dehumanization and aggressive tendencies passed through contamination fear and the moderated mediation model indicated that this indirect effect was stronger for participants in the disgust-induction than in the non-induction group. This effect was found for biological dehumanization and passive aggression outcomes, both related to outgroup bias. Consistent with the role of disgust as a disease-avoidance mechanism, our results suggest that disgust could facilitate stronger outgroup bias in the context of a high health threat, such as the COVID-19 pandemic.

Mediation analysis for mixture Cox proportional hazards cure models

Mediation analysis aims to decompose a total effect into specific pathways and investigate the underlying causal mechanism. Although existing methods have been developed to conduct mediation analysis in the context of survival models, none of these methods accommodates the existence of a substantial proportion of subjects who never experience the event of interest, even if the follow-up is sufficiently long. In this study, we consider mediation analysis for the mixture of Cox proportional hazards cure models that cope with the cure fraction problem. Path-specific effects on restricted mean survival time and survival probability are assessed by introducing a partially latent group indicator and applying the mediation formula approach in a three-stage mediation framework. A Bayesian approach with P-splines for approximating the baseline hazard function is developed to conduct analysis. The satisfactory performance of the proposed method is verified through simulation studies. An application of the Alzheimer's disease (AD) neuroimaging initiative dataset investigates the causal effects of APOE-ϵ4 allele on AD progression.

Research using population-based administration data integrated with longitudinal data in child protection settings: A systematic review

INTRODUCTION

Over the past decade there has been a marked growth in the use of linked population administrative data for child protection research. This is the first systematic review of studies to report on research design and statistical methods used where population-based administrative data is integrated with longitudinal data in child protection settings.

METHODS

The systematic review was conducted according to Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) statement. The electronic databases Medline (Ovid), PsycINFO, Embase, ERIC, and CINAHL were systematically searched in November 2019 to identify all the relevant studies. The protocol for this review was registered and published with Open Science Framework (Registration DOI: 10.17605/OSF.IO/96PX8).

RESULTS

The review identified 30 studies reporting on child maltreatment, mental health, drug and alcohol abuse and education. The quality of almost all studies was strong, however the studies rated poorly on the reporting of data linkage methods. The statistical analysis methods described failed to take into account mediating factors which may have an indirect effect on the outcomes of interest and there was lack of utilisation of multi-level analysis.

CONCLUSION

We recommend reporting of data linkage processes through following recommended and standardised data linkage processes, which can be achieved through greater co-ordination among data providers and researchers.

Different Roles of Prior Distributions in the Single Mediator Model with Latent Variables

In manifest variable models, Bayesian methods for mediation analysis can have better statistical properties than commonly used frequentist methods. However, with latent variables, Bayesian mediation analysis with diffuse priors can yield worse statistical properties than frequentist methods, and no study to date has evaluated the impact of informative priors on statistical properties of point and interval summaries of the mediated effect. This article describes the first examination of using fully conjugate and informative (accurate and inaccurate) priors in Bayesian mediation analysis with latent variables. Results suggest that fully conjugate priors and informative priors with the same relative prior sample sizes have notably different effects at N = 200 and 400, than at N = 50 and 100. Consequences of a small amount of inaccuracy in priors for loadings can be alleviated by making the prior less informative, whereas the same is not always true of inaccuracy in priors for structural paths. Finally, the consequences of using informative priors depend on the inferential goals of the analysis: inaccurate priors are more detrimental for accurately estimating the mediated effect than for evaluating whether the mediated effect is nonzero. Recommendations are provided about when to gainfully employ Bayesian mediation analysis with latent variables.

Does self-compassion help to deal with dietary lapses among overweight and obese adults who pursue weight-loss goals?

Objectives

Self‐compassion can facilitate self‐improvement motivation. We examined the effects of self‐compassion in response to dietary lapses on outcomes relevant to weight‐loss strivings using a longitudinal design. The indirect effects of self‐compassion via guilt and shame were also explored.

Design

An Ecological Momentary Assessment methodology was employed with a sample of adults who were overweight or obese attempting to lose weight via dietary restriction (N = 56; M_age = 34.88; SD = 13.93; M_BMI = 32.50; SD = 6.88) and who responded to brief surveys sent to their mobile phones twice daily for two weeks.

Methods

Dietary temptations and lapses were assessed at each diary entry, and self‐compassion in response to dietary lapses, intention to continue dieting, weight‐loss‐related self‐efficacy, negative reactions to the lapse, and self‐conscious emotions were surveyed on occasions when participants reported having experienced a dietary lapse. The participants were also weighed in a laboratory prior to the EMA phase and via self‐report straight after the EMA phase. Weight was measured again in the laboratory 12 weeks after the EMA period.

Results

Bayesian multilevel path analyses showed that self‐compassion did not predict weight loss. However, at the within‐person level, self‐compassion was positively related to intentions and self‐efficacy to continue dieting, and negatively related to negative affective reactions to the lapses. Guilt mediated the associations of self‐compassion with intention, self‐efficacy, and negative reactions.

Conclusion

Self‐compassion may be a powerful internal resource to cultivate when dieters experience inevitable setbacks during weight‐loss strivings which could facilitate weight‐loss perseverance.

Subjective suffering in patients with low back pain

AIM

Low back pain (LPB) is a prevalent rheumatic and musculoskeletal disease and patients often experience physical discomfort and functional limitations. This study aimed to evaluate how subjective suffering relates to illness perceptions, pain intensity and psychological morbidity as well as to understand the mediator (direct and indirect) effects on functional disability.

METHODS

A cross-sectional study with 125 patients who received physiotherapy and answered the following instruments: Illness Perception Questionnaire-Revised; Illness Subjective Suffering Inventory; Hospital Anxiety and Depression Scale and the Oswestry LBP Incapacity Questionnaire.

RESULTS

In the structural equation modeling analysis, subjective suffering totally mediated the relationship between both emotional response (estimate = 0.068, 95% highest posterior density interval [HPDI] = [0.017, 0.133]), consequences (estimate = 0.118, 95% HPDI = [0.035, 0.211]) and functional disability. The relationship between coherence and disability was partially mediated by subjective suffering (estimate = -0.067, 95% HPDI = [-0.134, -0.016]). Only identity and pain intensity showed direct effects on functional disability.

CONCLUSION

Intervention in patients with LBP should focus on subjective suffering to reduce functional disability, providing information and promoting the acceptance of disease and pain.

The Paradoxical Effect of Inclusive Leadership on Subordinates' Creativity

Previous research about inclusive leadership and creativity has produced contradictory results. The present study tried to explain the contradictory findings based on the antecedent-benefit-cost framework (ABC). We found that inclusive leadership promoted subordinates' creativity by enhancing subordinates' psychological safety but discouraged subordinates' creativity by reducing challenge-related stress. The present study illustrated the complex mediating mechanism of inclusive leadership's impact on creativity, presenting a complementary explanation of the conflicting relationships between inclusive leadership and creativity. In addition, we validated the ABC framework.

Association Between Screen Time, Fast Foods, Sugar-Sweetened Beverages and Depressive Symptoms in Chinese Adolescents

OBJECTIVE

Although previous studies have shown that screen time (ST), fast foods (FFs) and sugar-sweetened beverages (SSBs) consumption are associated with depressive symptoms in adolescents, research on these associations in Chinese adolescents is scarce. This study aimed to examine the association between ST, FFs, SSBs and depressive symptoms in Chinese adolescents, and explore the mediating effects of FFs and SSBs in the association between ST and depressive symptoms.

METHODS

This school-based nationwide survey was carried out among 14,500 students in four provinces of China. The Children's Depression Inventory was used to assess the participants' depressive symptoms. ST, FFs and SSBs consumption was measured by a self-reported questionnaire. The Bayesian multiple mediation model was used to analyze the mediation effect.

RESULTS

ST, FFs and SSBs, were more likely to be associated with depressive symptoms, and ORs (95%CI) was 1.075 (1.036-1.116), 1.062 (1.046-1.078) and 1.140 (1.115-1.166), after we adjusted for sociodemographic variables. Additionally, in Bayesian multiple mediation model, direct effect, mediating effect, total effect, the ratio of mediating effect to total effect was 0.125, 0.034, 0.159, and 0.214, respectively. All path coefficients of the three mediation paths are statistically significant (p < 0.05).

CONCLUSIONS

Our study demonstrates that ST, FFs and SSBs consumption are associated with depressive symptoms in Chinese adolescents. It is likely that FFs and SSBs partially mediate the association between ST and depressive symptoms by chain-mediating effects.

A Tutorial in Bayesian Mediation Analysis With Latent Variables

Abstract. Maximum Likelihood (ML) estimation is a common estimation method in Structural Equation Modeling (SEM), and parameters in such analyses are interpreted using frequentist terms and definition of probability. It is also possible, and sometimes more advantageous ( Lee & Song, 2004 ; Rindskopf, 2012 ), to fit structural equation models in the Bayesian framework ( Kaplan & Depaoli, 2012 ; Levy & Choi, 2013 ; Scheines, Hoijtink, & Boomsma, 1999 ). Bayesian mediation analysis has been described for manifest variable models ( Enders, Fairchild, & MacKinnon, 2013 ; Yuan & MacKinnon, 2009 ). This tutorial outlines considerations in the analysis and interpretation of results for the single mediator model with latent variables. The reader is guided through model specification, estimation, and the interpretations of results obtained using two kinds of diffuse priors and one set of informative priors. Recommendations are made for applied researchers and annotated syntax is provided in R2OpenBUGS and Mplus. The target audience for this article are researchers wanting to learn how to fit the single mediator model as a Bayesian SEM.

The Double-Edged Sword Effect of Abusive Supervision on Subordinates' Innovative Behavior

Existing studies on the relationship between abusive supervision and innovative behavior do not present a united picture. Drawing up the antecedent-benefit-cost framework and social cognitive theory, we tried to explain the contradictory relationships between them based on the mediating mechanism. Results showed that abusive supervision discouraged subordinates’ innovative behavior through reducing subordinates’ psychological safety but promoted subordinates’ innovative behavior through enhancing challenge-related stress.

Benchmark validation of statistical models: Application to mediation analysis of imagery and memory

This article describes benchmark validation, an approach to validating a statistical model. According to benchmark validation, a valid model generates estimates and research conclusions consistent with a known substantive effect. Three types of benchmark validation-(a) benchmark value, (b) benchmark estimate, and (c) benchmark effect-are described and illustrated with examples. Benchmark validation methods are especially useful for statistical models with assumptions that are untestable or very difficult to test. Benchmark effect validation methods were applied to evaluate statistical mediation analysis in eight studies using the established effect that increasing mental imagery improves recall of words. Statistical mediation analysis led to conclusions about mediation that were consistent with established theory that increased imagery leads to increased word recall. Benchmark validation based on established substantive theory is discussed as a general way to investigate characteristics of statistical models and a complement to mathematical proof and statistical simulation. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Parent and child neurocognitive functioning predict response to behavioral parent training for youth with ADHD

Parental cognitive functioning is thought to play a key role in parenting behavior and may inform response to behavioral intervention. This open-label pilot study examined the extent to which parent and child cognition impacted response to behavioral parent training for children with ADHD. Fifty-four participants (27 parent-child dyads; Mages = 10.6 and 45.2 for children and parents, respectively) completed tasks assessing visuospatial and phonological working memory, inhibitory control, and choice-reaction speed at pre-treatment. Drift diffusion modeling decomposed choice-reaction time data into indicators of processing speed (drift rate) and response caution (boundary separation). Parents completed a 10-week manualized behavioral parent training program. Primary outcomes were pre- and post-treatment child ADHD and conduct problem severity, and parent-reported relational frustration and parenting confidence. Bayesian multiple regressions assessed parent and child cognitive processes as predictors of post-treatment outcomes, controlling for pre-treatment behavior. Better child visuospatial and phonological WM and higher parental response caution were associated with greater reductions in inattention. For conduct problems, better parental self-regulation (stronger inhibitory control and greater response caution) predicted fewer post-treatment conduct problems. Higher parental response caution also predicted lower post-treatment relational frustration and higher parental confidence. Bayesian evidence supported no relation between parent and child cognitive functions and treatment-related changes in hyperactivity. This pilot study demonstrates that cognitive processes central to etiologic theories of ADHD and models of parenting behavior can be successfully integrated into treatment outcome research to inform which families are most likely to benefit from behavioral interventions. This study demonstrates the feasibility of bridging the translational research gap between basic and applied clinical science and facilitates research on the role of cognition in psychosocial interventions.

Evaluating the Impact of Physical Activity Apps and Wearables: Interdisciplinary Review

Background

Although many smartphone apps and wearables have been designed to improve physical activity, their rapidly evolving nature and complexity present challenges for evaluating their impact. Traditional methodologies, such as randomized controlled trials (RCTs), can be slow. To keep pace with rapid technological development, evaluations of mobile health technologies must be efficient. Rapid alternative research designs have been proposed, and efficient in-app data collection methods, including in-device sensors and device-generated logs, are available. Along with effectiveness, it is important to measure engagement (ie, users’ interaction and usage behavior) and acceptability (ie, users’ subjective perceptions and experiences) to help explain how and why apps and wearables work.

Objectives

This study aimed to (1) explore the extent to which evaluations of physical activity apps and wearables: employ rapid research designs; assess engagement, acceptability, as well as effectiveness; use efficient data collection methods; and (2) describe which dimensions of engagement and acceptability are assessed.

Method

An interdisciplinary scoping review using 8 databases from health and computing sciences. Included studies measured physical activity, and evaluated physical activity apps or wearables that provided sensor-based feedback. Results were analyzed using descriptive numerical summaries, chi-square testing, and qualitative thematic analysis.

Results

A total of 1829 abstracts were screened, and 858 articles read in full. Of 111 included studies, 61 (55.0%) were published between 2015 and 2017. Most (55.0%, 61/111) were RCTs, and only 2 studies (1.8%) used rapid research designs: 1 single-case design and 1 multiphase optimization strategy. Other research designs included 23 (22.5%) repeated measures designs, 11 (9.9%) nonrandomized group designs, 10 (9.0%) case studies, and 4 (3.6%) observational studies. Less than one-third of the studies (32.0%, 35/111) investigated effectiveness, engagement, and acceptability together. To measure physical activity, most studies (90.1%, 101/111) employed sensors (either in-device [67.6%, 75/111] or external [23.4%, 26/111]). RCTs were more likely to employ external sensors (accelerometers: P=.005). Studies that assessed engagement (52.3%, 58/111) mostly used device-generated logs (91%, 53/58) to measure the frequency, depth, and length of engagement. Studies that assessed acceptability (57.7%, 64/111) most often used questionnaires (64%, 42/64) and/or qualitative methods (53%, 34/64) to explore appreciation, perceived effectiveness and usefulness, satisfaction, intention to continue use, and social acceptability. Some studies (14.4%, 16/111) assessed dimensions more closely related to usability (ie, burden of sensor wear and use, interface complexity, and perceived technical performance).

Conclusions

The rapid increase of research into the impact of physical activity apps and wearables means that evaluation guidelines are urgently needed to promote efficiency through the use of rapid research designs, in-device sensors and user-logs to assess effectiveness, engagement, and acceptability. Screening articles was time-consuming because reporting across health and computing sciences lacked standardization. Reporting guidelines are therefore needed to facilitate the synthesis of evidence across disciplines.

A Tutorial in Bayesian Potential Outcomes Mediation Analysis

Statistical mediation analysis is used to investigate intermediate variables in the relation between independent and dependent variables. Causal interpretation of mediation analyses is challenging because randomization of subjects to levels of the independent variable does not rule out the possibility of unmeasured confounders of the mediator to outcome relation. Furthermore, commonly used frequentist methods for mediation analysis compute the probability of the data given the null hypothesis, which is not the probability of a hypothesis given the data as in Bayesian analysis. Under certain assumptions, applying the potential outcomes framework to mediation analysis allows for the computation of causal effects, and statistical mediation in the Bayesian framework gives indirect effects probabilistic interpretations. This tutorial combines causal inference and Bayesian methods for mediation analysis so the indirect and direct effects have both causal and probabilistic interpretations. Steps in Bayesian causal mediation analysis are shown in the application to an empirical example.