Supplement 1

Systematic review of stepped wedge cluster randomized trials shows design is particularly used to evaluate interventions during routine implementation (Protocol).

INTRODUCTION

A previous systematic review of the stepped wedge trial design by Brown and Lilford1 concluded that the design offers a number of scientific advantages and advocated it for use in intervention evaluations. This review1 included 12 studies evaluating a wide range of interventions. IT also included both randomized and non-randomized studies, both individual and cluster allocations and was limited to the health sector. In this present review we will expand the search to include non-health care trials as these often have lessons that can influence the design of health care trials. In contrast to the previous review1 the present review will focus on randomized studies, and cluster allocations.

AIM:

To describe the extent to which the stepped wedge CRCT design has been used to measure the effectiveness of interventions.

OBJECTIVES:

The review explores

  1. The areas of research in which the stepped wedge CRCT design has been used
  2. The motivation for employing the design
  3. The general characteristics of the stepped wedge design
  4. Methods of data analysis
  5. The quality of reporting in CRCT using the stepped wedge design.

METHODS:

Literature Search

Search of electronic databases

The following sources will be searched: Medline and Medline InProcess, Embase, PsycInfo,HMIC, CINAHL, Cochrane Library, Web of Knowledge, Current Controlled Trials Register and Google Search Engine.

The following phrases will be used: stepped wedge, step wedge, experimentally staged introduction, delayed intervention, one directional cross over design and the fifteen possible combinations of incremental/phased/staggered/ step wise/ delayed and recruitment/introduction/ implementation.

Multiple baseline design will also be used as a search phrase as some trials using this design are actually stepped wedge trials.

Searching other resources

Articles will also be identified from the reference lists of the retrieved studies, protocols and reviews.Authors of the retrieved studies, protocols and reviews will also be contacted for any potential missing studies, including unpublished studies.

The search will be limited to English language only and no publication date limits will be applied.

Study Inclusion and Exclusion Criteria

Inclusion Criteria

  • Papers and protocols of cluster randomised original research studies that used/ plan to use stepped wedge design approach from all fields of research.
  • English language

Exclusion Criteria

  • Studies retrospectively analysed as stepped wedge design when the study was not originally designed as stepped wedge.
  • Where original papers for studies included in the first review as protocols has now been published, the published papers will be considered.

Selection of Studies for inclusion in the review

Two authors (NM and MM) will read all titles/and or abstracts resulting from the search process and eliminate any obviously irrelevant studies. Full copies of the remaining potentially relevant studies will be obtained. NM and MM will independently classify these as clearly relevant that is, meet all inclusion criteria therefore include, or clearly irrelevant/ meet the exclusion criteria therefore exclude, or insufficient information to make a decision. Where there is insufficient information, contact will be made with the authors for further information to aid the decision process. Differences in opinion will be resolved through consensus or referral to a third author (DT or CB).

Data extraction and management

Two authors (NM and MM) will independently extract data from publishedsources using a data extraction form.The extracted information will include study identifiers (lead author, year of publication), areas of research in which the design has been used (subject areas, disease/domain, nature of intervention, setting and country), motivation for employing the stepped wedge design, general characteristics of the stepped wedge design (level of stepping, number of steps, number of clusters, total number of participants, time period between steps), methods of data analysis (primary outcome, comparisons and statistical methods of data analysis) and selected items on quality of reporting. The data extraction form will be piloted on 3 papers and relevant adjustments made. For studies where abstracts or protocol summaries only are available, authors will be contacted for full reports. Where additional information is required, this will be obtained from the study authors. Where differences in data extracted occur this will be resolved through discussion. Where decisions cannot be resolved through discussion, these cases will be referred to a third author (DT or CB).

REFERENCES

1.Brown CA, Lilford RJ. The stepped wedge trial design: a systematic review. BMC Med Res Methodol. 2006; 6:54

Page 1 of 2