Comparing the effect of a consult model versus an integrated palliative care and medical oncology co-rounding model on health care utilization in an acute hospital – an open-label stepped-wedge cluster-randomized trial

Yang, Grace Meijuan; Zhou, Siqin; Xu, Zhenhe; Goh, S.S.L.; Zhu, Xiaodong; Chong, Dawn Q.; Tan, Daniel S.W.; Kanesvaran, Ravindran; Yee, Alethea; Neo, Patricia Soek Hui; Cheung, Yin‐Bun

doi:10.1177/02692163211022957

Cited by 12 publications

(24 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…12,15,28 The SW-CRT of palliative care models that motivated this work did not include the unexposed cluster in its original analysis due to the lack of methodological guidance at the time. 32 In this article, we demonstrated that the inclusion of unexposed clusters in a fixed effects LSDV model is a viable strategy for improving the precision of the intervention effect estimate in a SW-CRT design. We mathematically proved this in Section 2.…”

Section: Discussionmentioning

confidence: 99%

“…Unexposed clusters (clusters never exposed to intervention) are seen in SW-CRTs in real-world research. 25,32 For example, SW-CRTs that are terminated early can result in clusters that are never exposed to intervention. SW-CRTs with unexposed clusters somewhat resemble the "optimal design" discussed by Thompson et al 36 and Girling and Hemming.…”

Section: Discussionmentioning

confidence: 99%

“…Alternatively, in the palliative care trial case study, reorganization of hospital management introduced an additional fifth cluster that ran concurrently alongside the other study clusters to serve the existing target population and reduce the patient load of some of the original clusters. 32 There was no expansion of study population and no change to the case-mix. In this case, we consider it appropriate to include the fifth cluster as an unexposed cluster in our fixed effects analysis.…”

Section: Discussionmentioning

confidence: 99%

“…We re-analyzed data from a recent SW-CRT comparing a standard palliative care delivery model (control) against a novel co-rounding model (intervention) for hospital inpatients with cancer. 32 In the standard care model, oncologists conducted the daily ward rounds and referred patients to the palliative care department if considered appropriate. In the novel co-rounding model, oncologists and palliative care specialists jointly conducted the ward rounds and initiated palliative care as per their consensus.…”

Section: Case Studymentioning

confidence: 99%

“…The original publication on the efficacy of the novel co-rounding model used a fixed effects LSDV model to account for the cluster effects and control for confounding. 32 The fixed effects LSDV model was used due to the difficulty in controlling for confounding (eg, the clusters differed not only in patient characteristics but also in unmeasured clinician characteristics) and concerns on the robustness of applying the mixed effects model to a SW-CRT with such a small number of clusters. 13,20 The analysis did not include the unexposed cluster because it was not part of the original trial plan and it was not yet methodologically clear what the implications of including an unexposed cluster in a fixed effects LSDV model were.…”

Section: Case Studymentioning

confidence: 99%

See 4 more Smart Citations

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

Lee

Yang

et al. 2022

Statistics in Medicine

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionmentioning

confidence: 99%