OBJECTIVES: The clusters participating in cluster randomized trials are not necessarily representative of the target population of trial-eligible clusters where the experimental treatments might be applied. To the extent that participating and non-participating clusters differ in terms of covariates that are modifiers of the treatment effect, the average treatment effect in the trial may not apply to the entire target population. We describe analyses that extend (generalize or transport) causal inferences from cluster randomized trials to a target population of clusters, under a general nonparametric model that allows for arbitrary within-cluster dependence.

METHODS: We consider study designs where a subset of a cohort of clusters are invited and agree to participate in a randomized trial with cluster-level treatment assignment (i.e., nested trial designs). Treatment and outcome data need only be collected from clusters participating in the trial, but data on baseline covariates must be collected from the entire cohort. We propose doubly robust estimators of potential outcome means in the target population that exploit individual- level data on covariates and outcomes to improve efficiency and are appropriate for use with machine learning. We illustrate the methods using a 2x2 factorial cluster randomized trial of influenza vaccination strategies in 818 nursing homes that were nested in a cohort of 4,475 trial- eligible Medicare-certified nursing homes.

RESULTS: We used machine learning methods to estimate the conditional probability of participation and the conditional probability of death, to obtain the potential outcome means for the vaccination strategies in the target population of trial-eligible nursing homes. Point estimates from generalizability analyses using different methods were similar, suggesting that the trial-only estimates were applicable to the broader population of trial-eligible nursing homes for the outcome of all-cause death.

CONCLUSIONS: When cluster randomized trials are embedded in large health-care systems, investigators can use the methods to extend causal inferences.