1
|
Kasza J, Bowden R, Forbes AB. Information content of stepped wedge designs with unequal cluster-period sizes in linear mixed models: Informing incomplete designs. Stat Med 2021; 40:1736-1751. [PMID: 33438255 DOI: 10.1002/sim.8867] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/07/2020] [Revised: 12/13/2020] [Accepted: 12/16/2020] [Indexed: 12/30/2022]
Abstract
In practice, stepped wedge trials frequently include clusters of differing sizes. However, investigations into the theoretical aspects of stepped wedge designs have, until recently, typically assumed equal numbers of subjects in each cluster and in each period. The information content of the cluster-period cells, clusters, and periods of stepped wedge designs has previously been investigated assuming equal cluster-period sizes, and has shown that incomplete stepped wedge designs may be efficient alternatives to the full stepped wedge. How this changes when cluster-period sizes are not equal is unknown, and we investigate this here. Working within the linear mixed model framework, we show that the information contributed by design components (clusters, sequences, and periods) does depend on the sizes of each cluster-period. Using a particular trial that assessed the impact of an individual education intervention on log-length of stay in rehabilitation units, we demonstrate how strongly the efficiency of incomplete designs depends on which cells are excluded: smaller incomplete designs may be more powerful than alternative incomplete designs that include a greater total number of participants. This also serves to demonstrate how the pattern of information content can be used to inform a set of incomplete designs to be considered as alternatives to the complete stepped wedge design. Our theoretical results for the information content can be extended to a broad class of longitudinal (ie, multiple period) cluster randomized trial designs.
Collapse
Affiliation(s)
- Jessica Kasza
- School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
| | - Rhys Bowden
- School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
| | - Andrew B Forbes
- School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia
| |
Collapse
|
2
|
Kasza J, Hooper R, Copas A, Forbes AB. Sample size and power calculations for open cohort longitudinal cluster randomized trials. Stat Med 2020; 39:1871-1883. [PMID: 32133688 PMCID: PMC7217159 DOI: 10.1002/sim.8519] [Citation(s) in RCA: 20] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/06/2019] [Revised: 01/15/2020] [Accepted: 02/17/2020] [Indexed: 01/24/2023]
Abstract
When calculating sample size or power for stepped wedge or other types of longitudinal cluster randomized trials, it is critical that the planned sampling structure be accurately specified. One common assumption is that participants will provide measurements in each trial period, that is, a closed cohort, and another is that each participant provides only one measurement during the course of the trial. However some studies have an "open cohort" sampling structure, where participants may provide measurements in variable numbers of periods. To date, sample size calculations for longitudinal cluster randomized trials have not accommodated open cohorts. Feldman and McKinlay (1994) provided some guidance, stating that the participant-level autocorrelation could be varied to account for the degree of overlap in different periods of the study, but did not indicate precisely how to do so. We present sample size and power formulas that allow for open cohorts and discuss the impact of the degree of "openness" on sample size and power. We consider designs where the number of participants in each cluster will be maintained throughout the trial, but individual participants may provide differing numbers of measurements. Our results are a unification of closed cohort and repeated cross-sectional sample results of Hooper et al (2016), and indicate precisely how participant autocorrelation of Feldman and McKinlay should be varied to account for an open cohort sampling structure. We discuss different types of open cohort sampling schemes and how open cohort sampling structure impacts on power in the presence of decaying within-cluster correlations and autoregressive participant-level errors.
Collapse
Affiliation(s)
- Jessica Kasza
- School of Public Health and Preventive MedicineMonash UniversityMelbourneVictoriaAustralia
| | - Richard Hooper
- Centre for Primary Care and Public HealthQueen Mary University of LondonLondonUK
| | - Andrew Copas
- MRC Clinical Trials UnitUniversity College LondonLondonUK
| | - Andrew B. Forbes
- School of Public Health and Preventive MedicineMonash UniversityMelbourneVictoriaAustralia
| |
Collapse
|
3
|
van Oostveen RB, Romero-Palacios A, Whitlock R, Lee SF, Connolly S, Carignan A, Mazer CD, Loeb M, Mertz D. Prevention of Infections in Cardiac Surgery study (PICS): study protocol for a pragmatic cluster-randomized factorial crossover pilot trial. Trials 2018; 19:688. [PMID: 30558680 PMCID: PMC6296086 DOI: 10.1186/s13063-018-3080-y] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/04/2018] [Accepted: 11/26/2018] [Indexed: 11/26/2022] Open
Abstract
Background A wide range of prophylactic antibiotic regimens are used for patients undergoing open-heart cardiac surgery. This reflects clinical equipoise in choice and duration of antibiotic agents. Although individual-level randomized control trials (RCT) are considered the gold standard when evaluating the efficacy of an intervention, this approach is highly resource intensive and a cluster RCT can be more appropriate for testing clinical effectiveness in a real-world setting. Methods/design We are conducting a factorial cluster-randomized crossover pilot trial in cardiac surgery patients to evaluate the feasibility of this design for a definite trial to evaluate the optimal duration and choice of perioperative antibiotic prophylaxis. Specifically, we will evaluate: (a) the non-inferiority of a single preoperative dose compared to prolonged prophylaxis and (b) the potential superiority of adding vancomycin to routine cefazolin in terms of preventing deep and organ/space sternal surgical site infections (s-SSIs). There are four strategies: (i) short-term cefazolin, (ii) long-term cefazolin, (iii) short-term cefazolin + vancomycin, and (iv) long-term cefazolin + vancomycin. These strategies are delivered in a different order in each health-care center participating in the trial. The centers are randomized to an order, and the current strategy becomes the standard operating procedure in that center during the study. The three feasibility outcomes include: (1) the proportion of patients receiving preoperative, intra-operative, and postoperative antibiotics according to the study protocol, (2) the proportion of completed follow-up assessments, and (3) a full and final assessment of the incidence of s-SSIs by the outcome adjudication committee. Discussion We believe that a cluster-randomized factorial crossover trial is an effective and feasible design for these research questions, allowing an evaluation of the clinical effectiveness in a real-world setting. A waiver of individual informed consent was considered appropriate by the research ethics boards in each participating site in Canada as long as an information letter with an opt-out option was provided. However, a waiver of consent was not approved at two sites in Germany and Switzerland, respectively. Trial registration Clinicaltrials.gov, NCT02285140. Registered on 15 October 2015. Electronic supplementary material The online version of this article (10.1186/s13063-018-3080-y) contains supplementary material, which is available to authorized users.
Collapse
Affiliation(s)
- Rachel B van Oostveen
- Population Health Research Institute (PHRI), Hamilton Health Sciences, Hamilton, ON, Canada
| | | | - Richard Whitlock
- Population Health Research Institute (PHRI), Hamilton Health Sciences, Hamilton, ON, Canada.,McMaster University, Hamilton, ON, Canada
| | - Shun Fu Lee
- Population Health Research Institute (PHRI), Hamilton Health Sciences, Hamilton, ON, Canada
| | - Stuart Connolly
- Population Health Research Institute (PHRI), Hamilton Health Sciences, Hamilton, ON, Canada.,McMaster University, Hamilton, ON, Canada
| | - Alex Carignan
- Department of Microbiology and Infectious Diseases, Université de Sherbrooke, Sherbrooke, QC, Canada.,Centre de recherche du Centre hospitalier universitaire de Sherbrooke, Sherbrooke, QC, Canada
| | - C David Mazer
- Li Ka Shing Knowledge Institute of St. Michael's Hospital, University of Toronto, Toronto, ON, Canada
| | - Mark Loeb
- McMaster University, Hamilton, ON, Canada
| | - Dominik Mertz
- Population Health Research Institute (PHRI), Hamilton Health Sciences, Hamilton, ON, Canada. .,McMaster University, Hamilton, ON, Canada. .,Juravinski Hospital and Cancer Center, 711 Concession Street, Section M, Level 1, Room 3, Hamilton, ON, L8V 1C3, Canada.
| |
Collapse
|