Statistical analysis and handling of missing data in cluster randomised trials: protocol for a systematic review

The West Midlands ActiVe lifestyle and healthy Eating in School children (WAVES) study: a cluster randomised controlled trial testing the clinical effectiveness and cost-effectiveness of a multifaceted obesity prevention intervention programme targeted at children aged 6-7 years

BACKGROUND

Systematic reviews suggest that school-based interventions can be effective in preventing childhood obesity, but better-designed trials are needed that consider costs, process, equity, potential harms and longer-term outcomes.

OBJECTIVE

To assess the clinical effectiveness and cost-effectiveness of the WAVES (West Midlands ActiVe lifestyle and healthy Eating in School children) study intervention, compared with usual practice, in preventing obesity among primary school children.

DESIGN

A cluster randomised controlled trial, split across two groups, which were randomised using a blocked balancing algorithm. Schools/participants could not be blinded to trial arm. Measurement staff were blind to allocation arm as far as possible.

SETTING

Primary schools, West Midlands, UK.

PARTICIPANTS

Schools within a 35-mile radius of the study centre and all year 1 pupils (aged 5-6 years) were eligible. Schools with a higher proportion of pupils from minority ethnic populations were oversampled to enable subgroup analyses.

INTERVENTIONS

The 12-month intervention encouraged healthy eating/physical activity (PA) by (1) helping teachers to provide 30 minutes of additional daily PA, (2) promoting 'Villa Vitality' (interactive healthy lifestyles learning, in an inspirational setting), (3) running school-based healthy cooking skills/education workshops for parents and children and (4) highlighting information to families with regard to local PA opportunities.

MAIN OUTCOME MEASURES

The primary outcomes were the difference in body mass index z-scores (BMI-zs) between arms (adjusted for baseline body mass index) at 3 and 18 months post intervention (clinical outcome), and cost per quality-adjusted life-year (QALY) (cost-effectiveness outcome). The secondary outcomes were further anthropometric, dietary, PA and psychological measurements, and the difference in BMI-z between arms at 27 months post intervention in a subset of schools.

RESULTS

Two groups of schools were randomised: 27 in 2011 (n = 650 pupils) [group 1 (G1)] and another 27 in 2012 (n = 817 pupils) [group 2 (G2)]. Primary outcome data were available at first follow-up (n = 1249 pupils) and second follow-up (n = 1145 pupils) from 53 schools. The mean difference (MD) in BMI-z between the control and intervention arms was -0.075 [95% confidence interval (CI) -0.183 to 0.033] and -0.027 (95% CI -0.137 to 0.083) at 3 and 18 months post intervention, respectively. The main analyses showed no evidence of between-arm differences for any secondary outcomes. Third follow-up included data on 467 pupils from 27 G1 schools, and showed a statistically significant difference in BMI-z (MD -0.20, 95% CI -0.40 to -0.01). The mean cost of the intervention was £266.35 per consented child (£155.53 per child receiving the intervention). The incremental cost-effectiveness ratio associated with the base case was £46,083 per QALY (best case £26,804 per QALY), suggesting that the intervention was not cost-effective.

LIMITATIONS

The presence of baseline primary outcome imbalance between the arms, and interschool variation in fidelity of intervention delivery.

CONCLUSIONS

The primary analyses show no evidence of clinical effectiveness or cost-effectiveness of the WAVES study intervention. A post hoc analysis, driven by findings at third follow-up, suggests a possible intervention effect, which could have been attenuated by baseline imbalances. There was no evidence of an intervention effect on measures of diet or PA and no evidence of harm.

FUTURE WORK

A realist evidence synthesis could provide insights into contextual factors and strategies for future interventions. School-based interventions need to be integrated within a wider societal framework and supported by upstream interventions.

TRIAL REGISTRATION

Current Controlled Trials ISRCTN97000586.

FUNDING

This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 22, No. 8. See the NIHR Journals Library website for further project information.

Review of Recent Methodological Developments in Group-Randomized Trials: Part 2-Analysis

In 2004, Murray et al. reviewed methodological developments in the design and analysis of group-randomized trials (GRTs). We have updated that review with developments in analysis of the past 13 years, with a companion article to focus on developments in design. We discuss developments in the topics of the earlier review (e.g., methods for parallel-arm GRTs, individually randomized group-treatment trials, and missing data) and in new topics, including methods to account for multiple-level clustering and alternative estimation methods (e.g., augmented generalized estimating equations, targeted maximum likelihood, and quadratic inference functions). In addition, we describe developments in analysis of alternative group designs (including stepped-wedge GRTs, network-randomized trials, and pseudocluster randomized trials), which require clustering to be accounted for in their design and analysis.

Generalized estimating equations in cluster randomized trials with a small number of clusters: Review of practice and simulation study

Background/aims:

Generalized estimating equations are a common modeling approach used in cluster randomized trials to account for within-cluster correlation. It is well known that the sandwich variance estimator is biased when the number of clusters is small (≤40), resulting in an inflated type I error rate. Various bias correction methods have been proposed in the statistical literature, but how adequately they are utilized in current practice for cluster randomized trials is not clear. The aim of this study is to evaluate the use of generalized estimating equation bias correction methods in recently published cluster randomized trials and demonstrate the necessity of such methods when the number of clusters is small.

Methods:

Review of cluster randomized trials published between August 2013 and July 2014 and using generalized estimating equations for their primary analyses. Two independent reviewers collected data from each study using a standardized, pre-piloted data extraction template. A two-arm cluster randomized trial was simulated under various scenarios to show the potential effect of a small number of clusters on type I error rate when estimating the treatment effect. The nominal level was set at 0.05 for the simulation study.

Results:

Of the 51 included trials, 28 (54.9%) analyzed 40 or fewer clusters with a minimum of four total clusters. Of these 28 trials, only one trial used a bias correction method for generalized estimating equations. The simulation study showed that with four clusters, the type I error rate ranged between 0.43 and 0.47. Even though type I error rate moved closer to the nominal level as the number of clusters increases, it still ranged between 0.06 and 0.07 with 40 clusters.

Conclusions:

Our results showed that statistical issues arising from small number of clusters in generalized estimating equations is currently inadequately handled in cluster randomized trials. Potential for type I error inflation could be very high when the sandwich estimator is used without bias correction.

Statistical analysis and handling of missing data in cluster randomized trials: a systematic review

Background

Cluster randomized trials (CRTs) randomize participants in groups, rather than as individuals and are key tools used to assess interventions in health research where treatment contamination is likely or if individual randomization is not feasible. Two potential major pitfalls exist regarding CRTs, namely handling missing data and not accounting for clustering in the primary analysis. The aim of this review was to evaluate approaches for handling missing data and statistical analysis with respect to the primary outcome in CRTs.

Methods

We systematically searched for CRTs published between August 2013 and July 2014 using PubMed, Web of Science, and PsycINFO. For each trial, two independent reviewers assessed the extent of the missing data and method(s) used for handling missing data in the primary and sensitivity analyses. We evaluated the primary analysis and determined whether it was at the cluster or individual level.

Results

Of the 86 included CRTs, 80 (93 %) trials reported some missing outcome data. Of those reporting missing data, the median percent of individuals with a missing outcome was 19 % (range 0.5 to 90 %). The most common way to handle missing data in the primary analysis was complete case analysis (44, 55 %), whereas 18 (22 %) used mixed models, six (8 %) used single imputation, four (5 %) used unweighted generalized estimating equations, and two (2 %) used multiple imputation. Fourteen (16 %) trials reported a sensitivity analysis for missing data, but most assumed the same missing data mechanism as in the primary analysis. Overall, 67 (78 %) trials accounted for clustering in the primary analysis.

Conclusions

High rates of missing outcome data are present in the majority of CRTs, yet handling missing data in practice remains suboptimal. Researchers and applied statisticians should carry out appropriate missing data methods, which are valid under plausible assumptions in order to increase statistical power in trials and reduce the possibility of bias. Sensitivity analysis should be performed, with weakened assumptions regarding the missing data mechanism to explore the robustness of results reported in the primary analysis.

Electronic supplementary material

The online version of this article (doi:10.1186/s13063-016-1201-z) contains supplementary material, which is available to authorized users.