Table 1

The definitions of eligible study designs

Study designs	Definition	Notes
RCT or randomised trial	‘An experimental study in which people are allocated to different interventions using methods that are random’.	We will include both individual and cluster randomised trials. In individual randomised trials, the randomisation is done at the individual levels, while in cluster randomised trials, the randomisation is based on cluster or groups of individuals. We will also consider factorial design trials where multiple interventions are studied in the same trial.
NRCT or non-randomised trial	‘An experimental study in which people are allocated to different interventions using methods that are not random’.	We will avoid using the term quasiexperimental studies as it means differently by different authors. We will exclude the experimental studies where there was no control group.
CBA study	‘A study in which observations are made before and after the implementation of an intervention, both in a group that receives the intervention and in a control group that does not’.	We will require two minimum criteria for the inclusion of CBA studies. Data collection: we will include CBA studies if the data for the intervention and control groups were collected prospectively in the same time frame. Choice of control: we will include CBA studies that include a control at a second site to avoid contamination of the intervention to the control group if the settings and populations are the same for the intervention and control groups.

The definitions of study designs were adopted from The Cochrane Effective Practice and Organization of Care group.33
CBA, controlled before and after; NRCT, non-randomised controlled trial; RCT, randomised controlled trial.