Cluster randomized trial designs for modeling time-varying intervention effects.

Stepped-wedge cluster randomized trials (SW-CRTs) are typically analyzed assuming a constant intervention effect. In practice, the intervention effect may vary as a function of exposure time, leading to biased results. The estimation of time-on-intervention (TOI) effects specifies separate discrete intervention effects for each elapsed period of exposure time since the intervention was first introduced. It has been demonstrated to produce results with minimum bias and nominal coverage probabilities in the analysis of SW-CRTs. Due to the design's staggered crossover, TOI effect variances are heteroskedastic in a SW-CRT. Accordingly, we hypothesize that alternative CRT designs will be more efficient at modeling certain TOI effects. We derive and compare the variance estimators of TOI effects between a SW-CRT, parallel CRT (P-CRT), parallel CRT with baseline (PB-CRT), and novel parallel CRT with baseline and an all-exposed period (PBAE-CRT). We also prove that the time-averaged TOI effect variance and point estimators are identical to that of the constant intervention effect in both P-CRTs and PB-CRTs. We then use data collected from a hospital disinvestment study to simulate and compare the differences in TOI effect estimates between the different CRT designs. Our results reveal that the SW-CRT has the most efficient estimator for the early TOI effect, whereas the PB-CRT typically has the most efficient estimator for the long-term and time-averaged TOI effects. Overall, the PB-CRT with TOI effects can be a more appropriate choice of CRT design for modeling intervention effects that vary by exposure time.

Estimation and reduction of bias in self-controlled case series with non-rare event dependent outcomes and heterogeneous populations.

The self-controlled case series (SCCS) is a commonly adopted study design in the assessment of vaccine and drug safety. Recurrent event data collected from SCCS studies are typically analyzed using the conditional Poisson model which assumes event times are independent within-cases. This assumption is violated in the presence of event dependence, where the occurrence of an event influences the probability and timing of subsequent events. When event dependence is suspected in an SCCS study, the standard recommendation is to include only the first event from each case in the analysis. However, first event analysis can still yield biased estimates of the exposure relative incidence if the outcome event is not rare. We first demonstrate that the bias in first event analysis can be even higher than previously assumed when subpopulations with different baseline incidence rates are present and describe an improved method for estimating this bias. Subsequently, we propose a novel partitioned analysis method and demonstrate how it can reduce this bias. We provide a recommendation to guide the number of partitions to use with the partitioned analysis, illustrate this recommendation with an example SCCS study of the association between beta-blockers and acute myocardial infarction, and compare the partitioned analysis against other SCCS analysis methods by simulation.

Inclusion of unexposed clusters improves the precision of fixed effects analysis of stepped-wedge cluster randomized trials.

Stepped-wedge cluster randomized trials (SW-CRTs) are typically analyzed using mixed effects models. The fixed effects model is a useful alternative that controls for all time-invariant cluster-level confounders and has proper control of type I error when the number of clusters is small. In principle, all clusters in SW-CRTs are designed to eventually receive the intervention, but in real-world research, some trials can end with unexposed clusters (clusters that never received the intervention), such as when a trial is terminated early based on interim analysis results. Typically, unexposed clusters are expected to contribute no information to the fixed effects intervention effect estimator and are excluded from fixed effects analyses. In this article we mathematically prove that inclusion of unexposed clusters improves the precision of the fixed effects least squares dummy variable (LSDV) intervention effect estimator, re-analyze data from a recent SW-CRT of a novel palliative care intervention containing an unexposed cluster, and evaluate the methods by simulation. We found that including unexposed clusters improves the precision of the fixed effects LSDV intervention effect estimator in both real and simulated datasets. Our simulations also reveal an increase in power and decrease in root mean square error. These improvements are present even if the assumptions of constant residual variance and period effects are violated. In the case that a SW-CRT concludes with unexposed clusters, these unexposed clusters can be included in the fixed effects LSDV analysis to improve precision, power, and root mean square error.