Estimating the Effects of a Complex, Multidimensional Moderator: An Example of Latent Class Moderation to Examine Differential Intervention Effects of Substance Use Services

Improvements in substance use disorder recovery may be achieved by recognizing that effective interventions do not work equally well for all individuals. Heterogeneity of intervention effects is traditionally examined as a function of a single variable, such as gender or baseline severity. However, responsiveness to an intervention is likely a result of multiple, intersecting factors. Latent class moderation enables the examination of heterogeneity in intervention effects across subgroups characterized by profiles of characteristics. This study analyzed data from adolescents (aged 13 to 18 years old) who needed treatment for cannabis use (n = 14,854) and participated in the Global Appraisal of Individual Needs to evaluate differential effects of substance use services on cannabis use outcomes. We demonstrate an adjusted three-step approach using weights that account for measurement error; sample codes in Mplus and Latent Gold are provided and data are publicly available. Indicators of the latent class moderator comprised six contextual (e.g., recovery environment risk) and individual (e.g., internal mental distress) risk factors. The latent class moderator comprised four subgroups: low risk (21.1%), social risk (21.1%), environmental risk (12.5%), and mixed risk (45.2%). Limited moderation of associations between level of care and any past 90-day cannabis use were observed. In predicting number of cannabis use-days, compared to individuals with low risk, those with environmental risk showed improved outcomes from intensive outpatient care whereas individuals with social risk and mixed risk showed improved outcomes from residential care (all compared to early intervention/outpatient care). Latent class moderation holds potential to elucidate heterogeneity in intervention effectiveness that otherwise may go undetected.

Differential effects of colorectal cancer screening across sociodemographic groups in Denmark: a register-based study

INTRODUCTION

Colorectal cancer (CRC) does not affect different sociodemographic groups uniformly. CRC screening programmes could seek to reduce this inequality; however, the screening programmes themselves might be subject to differential participation across sociodemographic groups. This study investigates the sociodemographic inequality at all steps in Denmark's nationwide CRC screening programme: screening participation, faecal immunochemical test (FIT) results, colonoscopy compliance, CRC diagnosis, and cancer stage.

MATERIAL AND METHODS

This cohort study includes all first-time invitees from the beginning of the Danish population-based CRC screening programme from 1 March 2014 to 31 December 2017.

RESULTS

Sixty-four percent of the invitees participated in the screening programme, and of those 7% were FIT-positive. After being invited to further diagnostic procedures, 90% responded to the invitation, and among those 5% were CRC-positive. Among those diagnosed with CRC, 9% were stage IV. Through multivariable analyses, we identified sociodemographic inequalities in all steps of the screening programme from returning a stool sample to being diagnosed with CRC. Specifically, we identified inequalities across sex, age, migration status, relationship status, the screening status of one's partner, education, income, and use of health services. Women were more likely to participate compared to men (RR = 1.13; 95% CI: 1.12-1.13), but had a lower risk of a positive FIT result (RR = 0.67; 95% CI: 0.66-0.68) and of a CRC diagnosis (RR = 0.88; 95% CI: 0.82-0.93) compared to men. The likelihood of participating as well as the risk of positive FIT results and CRC diagnosis increased with age.

CONCLUSION

All steps of the screening programme were subject to sociodemographic inequalities. Interventions are needed to target groups identified as having lower uptake as well as high-risk of being FIT- and/or CRC-positive. These groups include males, individuals aged 60+ years and individuals who do not visit their GP regularly.

Parent- and therapist-rated treatment satisfaction following routine child cognitive-behavioral therapy

This observational study examined treatment satisfaction (TS) following routine outpatient cognitive-behavioral therapy (CBT) in a large sample of children (n = 795; aged 6 to 10 years). TS was investigated in parent and therapist rating. Means, standard deviations and inter-rater correlations were calculated to investigate TS. Regression analysis was conducted to examine potential correlates of TS (patient-related variables, mental disorder characteristics, socio-demographic factors and treatment variables). High TS in parent and therapist rating was found, with therapists showing a lower degree of TS than parents (completely or predominantly satisfied: parent rating 94.1%, therapist rating 69.5%). A statistically significant, moderate inter-rater correlation was found. Regression analysis explained 21.8% of the variance in parent rating and 57.2% in therapist rating. Most of the TS variance was explained by mental disorder characteristics (parent-rated symptoms and therapist-rated global impairment at treatment end) and by treatment variables (especially the therapist-rated cooperation of parents and patients), whereas socio-demographic and patient-related variables did not show any relevant associations with TS. Based on these results, to optimize TS, therapists should concentrate on establishing a sustainable cooperation of parents and children during therapy, and work to achieve a low global impairment at treatment end.

Differential relationships between waist circumference and cardiorespiratory fitness among people with and without type 2 diabetes

Adults with type 2 diabetes mellitus tend to exhibit an increased level of central adiposity, augmenting their risk of further non-communicable diseases (NCDs). Importantly, consistent evidence demonstrates a significant, negative association between cardiorespiratory fitness (CRF) and waist circumference (WC). However, no previous studies have investigated differences in these CRF-related reductions in WC between adults with and without diabetes. This study used data from the Aerobic Center for Longitudinal Studies, conducted between 1970 and 2006 among predominately Non-Hispanic White, middle-to-upper class adults in Texas. Quantile regression models were used to estimate CRF-related differences in WC between persons with and without diabetes. Age, height, smoking status and birth cohort served as covariates. The analytic sample included 45901 adults. Significantly larger reductions in WC were observed among adults with diabetes as compared to without diabetes across all WC percentiles. Among males, high CRF levels were associated with significant reductions, as compared to their low-fit counterparts, in WC as large as 21.9 cm for adults without diabetes and as large as 27 cm for adults with diabetes. Among females, high CRF levels were associated with significant reductions, as compared to their low-fit counterparts, in WC as large as 22.3 and 30.0 cm for adults without and with diabetes, respectively. This study demonstrated that higher CRF is associated with significant reductions in WC, with greater magnitudes found among adults with diabetes, especially among the most centrally obese, highlighting the necessity of exercise prescription in this clinical population potentially leading to lower risks of future NCDs.

Evaluating the implementation of Check & Connect in various school settings: Is intervention fidelity necessarily associated with positive outcomes?

There are numerous school dropout prevention programs. However, few of them have undergone a rigorous implementation evaluation to understand their effects. This research presents two studies that evaluated the intervention fidelity and differential effects of Check & Connect (C&C), a targeted school dropout prevention program aimed at promoting student engagement and achievement. A total of 145 elementary school students (Study 1) and 200 secondary school students (Study 2) from two French-Canadian school boards (regional districts grouping elementary and secondary schools) received the C&C intervention for two years. In both studies, a clinical monitoring form was used to compare the intervention fidelity of each program component and active ingredient with what was initially planned. The relation between intervention fidelity and the effects of C&C on student engagement and achievement was analyzed using multiple linear regressions. Overall, the results show that intervention fidelity varies across elementary and secondary schools from one component to another and from one site to another. Furthermore, the association between the fidelity of each component and positive outcomes varies, depending on the implementation site. This evaluation supports the relevance of every component of C&C to favor engagement and academic achievement among at-risk elementary and secondary school students, while suggesting that the importance of certain program components may vary, depending on contextual influences on implementation and outcomes.

Current approaches used in epidemiologic studies to examine short-term multipollutant air pollution exposures

PURPOSE

Air pollution epidemiology traditionally focuses on the relationship between individual air pollutants and health outcomes (e.g., mortality). To account for potential copollutant confounding, individual pollutant associations are often estimated by adjusting or controlling for other pollutants in the mixture. Recently, the need to characterize the relationship between health outcomes and the larger multipollutant mixture has been emphasized in an attempt to better protect public health and inform more sustainable air quality management decisions.

METHODS

New and innovative statistical methods to examine multipollutant exposures were identified through a broad literature search, with a specific focus on those statistical approaches currently used in epidemiologic studies of short-term exposures to criteria air pollutants (i.e., particulate matter, carbon monoxide, sulfur dioxide, nitrogen dioxide, and ozone).

RESULTS

Five broad classes of statistical approaches were identified for examining associations between short-term multipollutant exposures and health outcomes, specifically additive main effects, effect measure modification, unsupervised dimension reduction, supervised dimension reduction, and nonparametric methods. These approaches are characterized including advantages and limitations in different epidemiologic scenarios.

DISCUSSION

By highlighting the characteristics of various studies in which multipollutant statistical methods have been used, this review provides epidemiologists and biostatisticians with a resource to aid in the selection of the most optimal statistical method to use when examining multipollutant exposures.

Impact of an equality constraint on the class-specific residual variances in regression mixtures: A Monte Carlo simulation study

Regression mixture models are a novel approach to modeling the heterogeneous effects of predictors on an outcome. In the model-building process, often residual variances are disregarded and simplifying assumptions are made without thorough examination of the consequences. In this simulation study, we investigated the impact of an equality constraint on the residual variances across latent classes. We examined the consequences of constraining the residual variances on class enumeration (finding the true number of latent classes) and on the parameter estimates, under a number of different simulation conditions meant to reflect the types of heterogeneity likely to exist in applied analyses. The results showed that bias in class enumeration increased as the difference in residual variances between the classes increased. Also, an inappropriate equality constraint on the residual variances greatly impacted on the estimated class sizes and showed the potential to greatly affect the parameter estimates in each class. These results suggest that it is important to make assumptions about residual variances with care and to carefully report what assumptions are made.

Differential segmentation responses to an alcohol social marketing program

OBJECTIVE

This study seeks to establish whether meaningful subgroups exist within a 14-16 year old adolescent population and if these segments respond differently to the Game On: Know Alcohol (GOKA) intervention, a school-based alcohol social marketing program.

METHODOLOGY

This study is part of a larger cluster randomized controlled evaluation of the GOKA program implemented in 14 schools in 2013/2014. TwoStep cluster analysis was conducted to segment 2,114 high school adolescents (14-16 years old) on the basis of 22 demographic, behavioral, and psychographic variables. Program effects on knowledge, attitudes, behavioral intentions, social norms, alcohol expectancies, and drinking refusal self-efficacy of identified segments were subsequently examined.

RESULTS

Three segments were identified: (1) Abstainers, (2) Bingers, and (3) Moderate Drinkers. Program effects varied significantly across segments. The strongest positive change effects post-participation were observed for Bingers, while mixed effects were evident for Moderate Drinkers and Abstainers.

CONCLUSIONS

These findings provide preliminary empirical evidence supporting the application of social marketing segmentation in alcohol education programs. Development of targeted programs that meet the unique needs of each of the three identified segments will extend the social marketing footprint in alcohol education.

Does Head Start differentially benefit children with risks targeted by the program's service model?

Data from the Head Start Impact Study (N = 3540) were used to test for differential benefits of Head Start after one program year and after kindergarten on pre-academic and behavior outcomes for children at risk in the domains targeted by the program's comprehensive services. Although random assignment to Head Start produced positive treatment main effects on children's pre-academic skills and behavior problems, residualized growth models showed that random assignment to Head Start did not differentially benefit the pre-academic skills of children with risk factors targeted by the Head Start service model. The models showed detrimental impacts of Head Start for maternal-reported behavior problems of high-risk children, but slightly more positive impacts for teacher-reported behavior. Policy implications for Head Start are discussed.

Population heterogeneity in the salience of multiple risk factors for adolescent delinquency

PURPOSE

To present mixture regression analysis as an alternative to more standard regression analysis for predicting adolescent delinquency. We demonstrate how mixture regression analysis allows for the identification of population subgroups defined by the salience of multiple risk factors.

METHODS

We identified population subgroups (i.e., latent classes) of individuals based on their coefficients in a regression model predicting adolescent delinquency from eight previously established risk indices drawn from the community, school, family, peer, and individual levels. The study included N = 37,763 10th-grade adolescents who participated in the Communities That Care Youth Survey. Standard, zero-inflated, and mixture Poisson and negative binomial regression models were considered.

RESULTS

Standard and mixture negative binomial regression models were selected as optimal. The five-class regression model was interpreted based on the class-specific regression coefficients, indicating that risk factors had varying salience across classes of adolescents.

CONCLUSIONS

Standard regression showed that all risk factors were significantly associated with delinquency. Mixture regression provided more nuanced information, suggesting a unique set of risk factors that were salient for different subgroups of adolescents. Implications for the design of subgroup-specific interventions are discussed.