A randomization strategy for a cluster-randomized controlled trial with variable operating room availability

Area of Science:

• Medical research methodology
• Pulmonology
• Clinical trial design

Background:

• Comparing diagnostic bronchoscopy platforms requires robust trial designs.
• Resource limitations in operating rooms (ORs) and equipment can challenge patient randomization.
• Cluster randomized controlled trials (cRCTs) are suitable for pragmatic comparisons but need careful allocation strategies.

Purpose of the Study:

• To compare clinical outcomes of two diagnostic bronchoscopy approaches (Platform A and Platform B).
• To develop and evaluate an effective randomization strategy for a cluster randomized controlled trial (cRCT) given resource constraints.
• To minimize patient assignment imbalance between the two platforms.

Main Methods:

• A single-center, open-label, pragmatic, non-inferiority cRCT was designed.
• Clusters were defined as operating room (OR) days.
• A permuted-block randomization method, stratified by OR availability, was simulated and evaluated for assignment balance.

Main Results:

• Stratification by OR availability significantly reduced assignment imbalance (mean imbalance = -0.006, SD = 0.690) compared to stratification by weekdays (mean imbalance = 0.044, SD = 1.110).
• The proposed method effectively minimized the difference in patient enrollment between Platform A and Platform B.
• The strategy optimized the allocation of limited OR and bronchoscopy platform resources.

Conclusions:

• A permuted-block randomization method stratified by OR availability is an effective strategy for cRCTs with resource limitations.
• This innovative design improves the balance of patient assignments in diagnostic bronchoscopy trials.
• The findings support optimized resource allocation and reliable clinical outcome comparisons in pragmatic trials.