Dataset Information

A structural mean model to allow for noncompliance in a randomized trial comparing 2 active treatments.

ABSTRACT: We propose a structural mean modeling approach to obtain compliance-adjusted estimates for treatment effects in a randomized-controlled trial comparing 2 active treatments. The model relates an individual's observed outcome to his or her counterfactual untreated outcome through the observed receipt of active treatments. Our proposed estimation procedure exploits baseline covariates that predict compliance levels on each arm. We give a closed-form estimator which allows for differential and unexplained selectivity (i.e. noncausal compliance-outcome association due to unobserved confounding) as well as a nonparametric error distribution. In a simple linear model for a 2-arm trial, we show that the distinct causal parameters are identified unless covariate-specific expected compliance levels are proportional on both treatment arms. In the latter case, only a linear contrast between the 2 treatment effects is estimable and may well be of key interest. We demonstrate the method in a clinical trial comparing 2 antidepressants.

SUBMITTER: Fischer K

PROVIDER: S-EPMC3062146 | biostudies-literature | 2011 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A structural mean model to allow for noncompliance in a randomized trial comparing 2 active treatments.

Fischer Krista K Goetghebeur Els E Vrijens Bernard B White Ian R IR

Biostatistics (Oxford, England) 20100830 2

We propose a structural mean modeling approach to obtain compliance-adjusted estimates for treatment effects in a randomized-controlled trial comparing 2 active treatments. The model relates an individual's observed outcome to his or her counterfactual untreated outcome through the observed receipt of active treatments. Our proposed estimation procedure exploits baseline covariates that predict compliance levels on each arm. We give a closed-form estimator which allows for differential and unexp ...[more]

PMID: 20805286

Similar Datasets

Project description:BackgroundMindfulness Based Stress Reduction (MBSR) is a secular form of meditation training. The vast majority of the extant literature investigating the health effects of mindfulness interventions relies on wait-list control comparisons. Previous studies have found that meditation training over several months is associated with improvements in cognitive control and attention.Methodology/principal findingsWe used a visual continuous performance task (CPT) to test the effects of eight weeks of mindfulness training on sustained attention by comparing MBSR to the Health Enhancement Program (HEP), a structurally equivalent, active control condition in a randomized, longitudinal design (ClinicalTrials.gov, NCT01301105) focusing on a non-clinical population typical of MBSR participants. Researchers were blind to group assignment. 63 community participants were randomized to either MBSR (n = 31) or HEP (n = 32). CPT analyses were conducted on 29 MBSR participants and 25 HEP participants. We predicted that MBSR would improve visual discrimination ability and sustained attention over time on the CPT compared to HEP, with more home practice associated with greater improvements. Our hypotheses were not confirmed but we did find some evidence for improved visual discrimination similar to effects in partial replication of other research. Our study had sufficient power to demonstrate that intervention groups do not differ in their improvement over time in sustained attention performance. One of our primary predictions concerning the effects of intervention on attentional fatigue was significant but not interpretable.ConclusionsAttentional sensitivity is not affected by mindfulness practice as taught in MBSR, but it is unclear whether mindfulness might positively affect another aspect of attention, vigilance. These results also highlight the relevant procedural modifications required by future research to correctly investigate the role of sustained attention in similar samples.Trial registrationClinicalTrials.gov, NCT01301105.

Dataset Information

A structural mean model to allow for noncompliance in a randomized trial comparing 2 active treatments.

Publications

A structural mean model to allow for noncompliance in a randomized trial comparing 2 active treatments.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets