Statistical analysis of group-administered intervention data: reanalysis of two randomized trials

Effectiveness of rational emotive behavior therapy in reducing depression among undergraduate medical students

BACKGROUND

Studies reporting the incidence of depression among medical students have been on the increase. This research sought to determine whether rational emotive behavior therapy (REBT) can be applied to help manage depression in Nigerian undergraduate medical students.

METHODS

A randomized pretest/posttest control group design was used in this study. Ninety medical students with depression participated in the study and were assisted using the REBT depression manual. Using a mixed-model repeated measures analysis of variance, the researchers examined the intervention data.

RESULTS

The depressive symptoms and its associated irrational beliefs among medical students in the treatment arm were significantly altered by REBT intervention at posttest and this positive outcome was sustained at follow-up in contrast to the control arm.

CONCLUSION

REBT intervention significantly improves medical students' ability to overcome depression and irrational beliefs. Similar studies could be conducted in a variety of academic settings where these students can be found to expand the findings of this study.

Comparison of Prior Setting Methods for Multilevel Model Effect Estimation Based on Small Sample Imbalanced Nested Data in Bayesian Framework

In the fields of education and psychology, nested data with small samples and imbalances are very common. Bauer et al. (2008) first proposed adjusting the traditional multilevel model to analyze the small sample imbalanced nested data (SSIND). In terms of parameter estimation, the Bayesian method shows the possibility of providing unbiased estimation when the sample size is small. This study proposes that the Bayesian method should be used to analyze the SSIND. This study explores the performance of different treatment effects and nesting effects estimation methods in the multilevel model based on the Bayesian method that performs well in the case of small samples, to provide an appropriate and scientific method reference for the subsequent analysis of the model. Two prior setting methods are compared for multilevel model effect estimation based on a small sample of imbalanced nested data in the Bayesian framework. Two prior setting methods are gamma prior setting method and uniform prior setting method. The research results show that when the treatment condition ICC is small (0.05), the bias and RMSE values of the parameter estimation by the gamma prior setting method are larger and the performance is unstable, while the bias and RMSE values of the parameter estimation by the uniform prior setting method are smaller and the performance is relatively stable, so the uniform prior setting method is recommended; when the treatment condition ICC is large (0.15), the bias and RMSE values of the parameter estimation by the uniform prior setting method are larger and the performance is unstable, while the bias and RMSE values of the parameter estimation by the gamma prior setting method are smaller and the performance is relatively stable, so the gamma prior setting method is recommended; when the treatment condition ICC is between 0.05 and 0.15, both prior setting methods have similar effects. Furthermore, when the number of treatment groups is small (8), the gamma prior setting method is recommended; when the number of treatment groups is large (16), the uniform prior setting method is recommended; when the number of treatment groups is between 8 and 16, both prior setting methods have similar effects. Summarily, when we choose which prior setting method to use for the SSIND, we must consider the interaction between the ICC and the number of treatment groups.

Evaluating Coaching Intervention for Financial Risk Perception and Credit Risk Management in a Nigerian Sample

There is no doubt that a negative perception of financial risk and a lack of credit risk management adversely impact business growth and business owners’ wellbeing. Past studies suggest that most Nigerian traders have poor risk perceptions and manage risk poorly. A business coaching program within rational-emotive behavior therapy framework (REBT-based business coaching) was evaluated in order to determine its effects on financial risk perception and credit risk management among Nigerian traders. This study used an open-label parallel randomized control design. This 8-weeks coaching program had 60 traders in the REBT-based business coaching group and 59 traders in the control group. The mixed-model repeated measures ANOVA was utilized for analysis of the study data. Results show that relative to a control group, traders’ financial risk perception [F(1.09,127.15) = 637.29, p < 0.001, ω2 = 0.69] and credit risk management [F(1.55,181.22) = 795.51, p < 0.001, ω2 = 0.80] significantly improved after participating in REBT-based business coaching program. This study shows that REBT-based business coaching program is integral to increasing financial risk perception and credit risk management among Nigerian traders. This study contributes to the advancement of business coaching program within the rational-emotive behavior therapy framework for market traders, and its application in similar situations. The study considered the benefits of business coaching program for market traders from a developing country, which is a rarely studied group. It is highly recommended that professionals study the relationship between REBT-based business coaching and economic decision-making within various organizational contexts.

Wireless Measurement of Sympathetic Arousal During in vivo Occupational Therapy Sessions

Purpose

One goal of occupational therapists working with children who have sensory processing challenges is the regulation of arousal. Regulation strategies have not been evaluated using an empirical measure of physiological arousal.

Objective

To establish the feasibility of using an objective physiologic measure of sympathetic arousal in therapeutic settings and explore the relation between therapeutic activities and sympathetic arousal. To evaluate changes in electrodermal activity (EDA) during occupational therapy sessions.

Methods

Twenty-two children identified with sensory modulation dysfunction (SMD) wore a wireless EDA sensor during 50 min occupational therapy sessions (n = 77 sessions).

Results

All children were able to wear the sensor on the lower calf without being distracted by the device. The five insights below are based on a comparison of EDA recordings in relation to therapists' reflections describing how sympathetic arousal might correspond to therapeutic activities.

Conclusion

Objective physiological assessment of a child's sympathetic arousal during therapy is possible using a wireless EDA measurement system. Changes in EDA may correspond directly with therapeutic activities. The article provides a foundation for designing future therapeutic studies that include continuous measures of EDA.

Effects of a cognitive ergonomics workplace intervention (CogErg) on cognitive strain and well-being: a cluster-randomized controlled trial. A study protocol

Background

Cognitively straining conditions such as disruptions, interruptions, and information overload are related to impaired task performance and diminished well-being at work. It is therefore essential that we reduce their harmful consequences to individual employees and organizations. Our intervention study implements practices for managing the cognitive strain typical to office work tasks and working conditions in offices. We will examine the effects of a cognitive ergonomics intervention on working conditions, workflow, well-being, and productivity.

Methods/design

The study is a stratified cluster randomized trial. The clusters are work units, for example, teams or offices. The four participating organizations entered a total of 36 clusters, and we invited all 1169 knowledge employees of these units to participate. We randomly allocated the clusters into an intervention group (cognitive ergonomics) or an active control group (recovery supporting). We invited an additional 471 participants to join a passive control group only for baseline and follow-up measurements, with no intervention.

The study consists of a baseline survey and interviews and observations at the workplace, followed by an intervention. It starts with a workshop defining the specific actions for the intervention implementation stage, during which we send task reminder questionnaires to all employees to support behaviour change at the individual and team levels. The primary outcome measure is perceived frequency of cognitive strain from working conditions; the secondary outcome measures include subjective cognitive load, well-being, workflow/productivity, and cognitive stress symptoms. Process evaluation uses the quantitative and qualitative data obtained during the implementation and evaluation phases. The baseline measurements, intervention phase, and end-of-treatment measurements are now complete, and follow-up will continue until November 2019.

Discussion

There is a need to expand the research of cognitive strain, which poses a considerable risk to work performance and employee well-being in cognitively demanding tasks. Our study will provide new information about factors that contribute to such strain. Most importantly, the results will show which evidence-based cognitive ergonomic practices support work performance in knowledge work, and the project will provide concrete examples of how to improve at work.

Trial registration

ClinicalTrials.gov, NCT03573674. Registered 29 June 2018.

Meta-analysis of the effects of cognitive-behavioral therapy for binge-eating-type disorders on abstinence rates in nonrandomized effectiveness studies: Comparable outcomes to randomized, controlled trials?

OBJECTIVE

The efficacy of cognitive-behavioral therapy (CBT) for eating disorders is well-established. The extent to which CBT tested in controlled research settings generalizes to real-world circumstances is unknown. We conducted a meta-analysis of nonrandomized studies of CBT for eating disorders, with three aims: (a) to estimate the prevalence of patients who achieve binge-purge abstinence after CBT in routine practice; (b) to compare these estimates with those derived from two recent meta-analyses of randomized controlled trials (RCTs) of CBT for bulimia nervosa (BN) and binge-eating disorder (BED); (c) to examine whether the degree of clinical representativeness of studies was associated with effect sizes.

METHOD

Twenty-seven studies, mainly involving BN, were included. Pooled event rates were calculated using random effects models.

RESULTS

The percentage of treatment completers who achieved abstinence at post-treatment was 42.1% (95% CI = 34.7-50.0). The intention-to-treat (ITT) estimate was lower (34.6% [95% CI = 29.3-40.4]). However, abstinence rates varied across diagnoses, such that the completer and ITT analysis abstinence estimates were larger for BED samples (completer = 50.2%, 95% CI = 29.4-70.9; ITT = 47.2%, 95% CI = 29.8-65.2) than for BN (completer = 37.4%, 95% CI = 29.1-46.5; ITT = 29.8%, 95% CI = 24.9-35.3) and atypical eating disorder samples (completer = 37.8%, 95% CI = 20.2-59.3; ITT = 28.8%, 95% CI = 18.2-42.4). No relationship between the degree of clinical representativeness and the effect size was observed, and our estimates were highly comparable to those observed in recent meta-analyses of RCTs.

DISCUSSION

Findings suggest that CBT for eating disorder can be effectively delivered in real-world settings. This study provides evidence for the generalizability of CBT from controlled research settings to routine clinical services.

Recommendations for the analysis of individually randomised controlled trials with clustering in one arm - a case of continuous outcomes

Background

In an individually randomised controlled trial where the treatment is delivered by a health professional it seems likely that the effectiveness of the treatment, independent of any treatment effect, could depend on the skill, training or even enthusiasm of the health professional delivering it. This may then lead to a potential clustering of the outcomes for patients treated by the same health professional, but similar clustering may not occur in the control arm. Using four case studies, we aim to provide practical guidance and recommendations for the analysis of trials with some element of clustering in one arm.

Methods

Five approaches to the analysis of outcomes from an individually randomised controlled trial with clustering in one arm are identified in the literature. Some of these methods are applied to four case studies of completed randomised controlled trials with clustering in one arm with sample sizes ranging from 56 to 539. Results are obtained using the statistical packages R and Stata and summarised using a forest plot.

Results

The intra-cluster correlation coefficient (ICC) for each of the case studies was small (<0.05) indicating little dependence on the outcomes related to cluster allocations. All models fitted produced similar results, including the simplest approach of ignoring clustering for the case studies considered.

Conclusions

A partially clustered approach, modelling the clustering in just one arm, most accurately represents the trial design and provides valid results. Modelling homogeneous variances between the clustered and unclustered arm is adequate in scenarios similar to the case studies considered. We recommend treating each participant in the unclustered arm as a single cluster. This approach is simple to implement in R and Stata and is recommended for the analysis of trials with clustering in one arm only. However, the case studies considered had small ICC values, limiting the generalisability of these results.

Electronic supplementary material

The online version of this article (doi:10.1186/s12874-016-0249-5) contains supplementary material, which is available to authorized users.

The Influence of Retention, Turnover, and Alliance on Process and Outcomes in Rolling Group Psychotherapy for Cocaine Disorder

A longitudinal repeated measures design was used to investigate the influence of treatment retention, and membership turnover on individual outcomes, and the relationship between individual-level and group-level therapeutic alliance in rolling group psychotherapy for individuals with cocaine abuse disorder. Eighteen patients were studied; data were collected prior to therapy and then monthly for 24 months. Patient outcomes and process were assessed. Improvements in symptoms and psychological functioning were associated with retention in psychotherapy. Membership turnover had a negative influence on symptoms. Development of an individual alliance was affected by the development of group-level therapeutic alliance, but not by time spent in therapy.

Repairing the efficiency loss due to varying cluster sizes in two-level two-armed randomized trials with heterogeneous clustering

In two-armed trials with clustered observations the arms may differ in terms of (i) the intraclass correlation, (ii) the outcome variance, (iii) the average cluster size, and (iv) the number of clusters. For a linear mixed model analysis of the treatment effect, this paper examines the expected efficiency loss due to varying cluster sizes based upon the asymptotic relative efficiency of varying versus constant cluster sizes. Simple, but nearly cost-optimal, correction factors are derived for the numbers of clusters to repair this efficiency loss. In an extensive Monte Carlo simulation, the accuracy of the asymptotic relative efficiency and its Taylor approximation are examined for small sample sizes. Practical guidelines are derived to correct the numbers of clusters calculated under constant cluster sizes (within each treatment) when planning a study. Because of the variety of simulation conditions, these guidelines can be considered conservative but safe in many realistic situations. Copyright © 2016 John Wiley & Sons, Ltd.

Partially nested designs in psychotherapy trials: A review of modeling developments

OBJECTIVE

Individually-randomized psychotherapy trials are often partially nested. For instance, individuals assigned to a treatment arm may be clustered into therapy groups for purposes of treatment administration, whereas individuals assigned to a wait-list control are unclustered. The past several years have seen rapid expansion and investigation of methods for analyzing partially nested data. Yet partial nesting often remains ignored in psychotherapy trials.

METHODS

This review integrates and disseminates developments in the analysis of partially nested data that are particularly relevant for psychotherapy researchers.

RESULTS

First, we differentiate among alternative partially nested designs. Then, we present adaptations of multilevel model specifications that accommodate each design. Next, we address how moderation by treatment as well as mediation of the treatment effect can be investigated in partially nested designs. Model fitting results, annotated software syntax, and illustrative data sets are provided and key methodological issues are discussed.

CONCLUSIONS

We emphasize that cluster-level variability in the treatment arm need not be considered a nuisance; it can be modeled to yield insights about the treatment process.

Sample size calculation for treatment effects in randomized trials with fixed cluster sizes and heterogeneous intraclass correlations and variances

When comparing two different kinds of group therapy or two individual treatments where patients within each arm are nested within care providers, clustering of observations may occur in both arms. The arms may differ in terms of (a) the intraclass correlation, (b) the outcome variance, (c) the cluster size, and (d) the number of clusters, and there may be some ideal group size or ideal caseload in case of care providers, fixing the cluster size. For this case, optimal cluster numbers are derived for a linear mixed model analysis of the treatment effect under cost constraints as well as under power constraints. To account for uncertain prior knowledge on relevant model parameters, also maximin sample sizes are given. Formulas for sample size calculation are derived, based on the standard normal as the asymptotic distribution of the test statistic. For small sample sizes, an extensive numerical evaluation shows that in a two-tailed test employing restricted maximum likelihood estimation, a safe correction for both 80% and 90% power, is to add three clusters to each arm for a 5% type I error rate and four clusters to each arm for a 1% type I error rate.

Reducing interparental conflict among parents in contentious child custody disputes: an initial investigation of the Working Together Program

Interparental conflict is one of the primary risk factors for negative outcomes for children whose parents separate, and it is likely to be high while parents are separating. Results are mixed regarding the effectiveness of interventions aimed at reducing interparental conflict. This study examined co-parents who were court-ordered to attend a 12-hr co-parenting intervention and completed pre-postassessments (n = 20) and 2-month follow-up assessments (n = 17). The results demonstrated increases in co-parents' relationship functioning and confidence in co-parenting. Both men and women reported decreased amounts of conflict in the presence of their children; however, only women reported decreases in general negative communication with the co-parent. These changes were generally maintained at a 2-month follow-up assessment. These findings suggest that interventions for high-conflict co-parents may increase their ability to work cooperatively.

Report on a Systematic Review of the Efficacy and Clinical Effectiveness of Group Analysis and Analytic/Dynamic Group Psychotherapy

This article reports on a systematic review to assess the efficacy and clinical effectiveness of group analysis and analytic/dynamic (A/D) group psychotherapy. Systematic literature searches were undertaken; 34 primary studies and 19 reviews met the inclusion criteria and their data was extracted. Of the primary studies, the randomized controlled trials provided evidence for the efficacy and clinical effectiveness of group therapy approaches in a range of clinical problems, but not for specific benefits of any particular theoretical approach. The controlled studies and observational studies gave support for the use of group psychotherapy in a variety of conditions.

The review of reviews confirms that group therapies in general are more effective than wait list or standard care controls. Methodological challenges encountered and recommendations for further research are presented, along with a more detailed discussion of the implications of the review for the research community.

Grief and trauma intervention for children after disaster: exploring coping skills versus trauma narration

This study evaluated the differential effects of the Grief and Trauma Intervention (GTI) with coping skills and trauma narrative processing (CN) and coping skills only (C). Seventy African American children (6-12 years old) were randomly assigned to GTI-CN or GTI-C. Both treatments consisted of a manualized 11-session intervention and a parent meeting. Measures of trauma exposure, posttraumatic stress symptoms, depression, traumatic grief, global distress, social support, and parent reported behavioral problems were administered at pre, post, 3 and 12 months post intervention. In general, children in both treatment groups demonstrated significant improvements in distress related symptoms and social support, which, with the exception of externalizing symptoms for GTI-C, were maintained up to 12 months post intervention. Results suggest that building coping skills without the structured trauma narrative may be a viable intervention to achieve symptom relief in children experiencing trauma-related distress. However, it may be that highly distressed children experience more symptom relief with coping skills plus narrative processing than with coping skills alone. More research on the differential effects of coping skills and trauma narration on child distress and adaptive functioning outcomes is needed.

Evaluating models for partially clustered designs

Partially clustered designs, where clustering occurs in some conditions and not others, are common in psychology, particularly in prevention and intervention trials. This article reports results from a simulation comparing 5 approaches to analyzing partially clustered data, including Type I errors, parameter bias, efficiency, and power. Results indicate that multilevel models adapted for partially clustered data are relatively unbiased and efficient and consistently maintain the nominal Type I error rate when using appropriate degrees of freedom. To attain sufficient power in partially clustered designs, researchers should attend primarily to the number of clusters in the study. An illustration using data from a partially clustered eating disorder prevention trial is provided.

Evaluating models for partially clustered designs

Partially clustered designs, where clustering occurs in some conditions and not others, are common in psychology, particularly in prevention and intervention trials. This article reports results from a simulation comparing 5 approaches to analyzing partially clustered data, including Type I errors, parameter bias, efficiency, and power. Results indicate that multilevel models adapted for partially clustered data are relatively unbiased and efficient and consistently maintain the nominal Type I error rate when using appropriate degrees of freedom. To attain sufficient power in partially clustered designs, researchers should attend primarily to the number of clusters in the study. An illustration using data from a partially clustered eating disorder prevention trial is provided.

IS SUCCESS IN WEIGHT LOSS TREATMENT CONTAGIOUS (DO ATTENDANCE AND OUTCOMES CLUSTER WITHIN TREATMENT GROUPS)?

Data from a trial of a 26-session structured group behavioral weight loss intervention were used to examine how group attendance, weight loss, and changes in depression clustered within therapy groups. Participants were recruited via a population-based survey of female health plan members aged 40 to 65. The sample included 143 women attending 13 therapy groups. Average number of sessions attended was 15.9 (st. dev. 7.2) and average weight loss over 12 months was 4.24 kg (st. dev. 7.79 kg). In hierarchical (random effects) linear models predicting attendance and weight loss, attendance did cluster significantly within therapy groups (F=2.83, df=12, p=.002; ICC = .14) but weight loss (F=0.89, df=12, p=.56; ICC = .00) and change in depressive symptoms (F=0.25, df=12, p=.99; ICC = .00) did not. Accounting for baseline characteristics of group participants had no significant effect on these findings. Relatively small average weight loss in this sample may have limited our ability to detect clustering of weight loss within groups. We conclude that clinicians should consider addressing the effects of drop-out on other group members and that researchers should consider the impact of clustering in analyses of data regarding group treatments.