Table 1.

Gatekeepers and gatekeeper roles in diverse CRTs in health research

Setting	Country	Cluster type	Level of intervention	Individual consent reported	Gatekeeper	Gatekeeper role	Ref.
Athletic organizations	Norway	Sports clubs	Cluster	No	Coaches	Agreement to participatea	[12]
	Netherlands	Geographic regions	Individual	No	Coaches	Consentb	[13]
	Canada	Athletic teams	Individual	Yes	Head athletic therapist, trainer or sports medicine physician	Identification of cluster members	[14]
Communities	USA	Cities	Cluster	No	Community medical leaders	Permission to randomize	[15]
	USA	Rural communities	Cluster	Yes	Community advisory board	Cluster consultation	[16]
	Bulgaria	Social circles	Individual	Yes	Index memberc	Identification of cluster members	[17]
	Gambia	Geographic areas/districts	Individual	Yes	Community leaders	Cluster permission	[18]
	Tanzania	Residential areas (Bazoli)	Cluster	Yes	Tanzania Institute for Medical Research	Protocol approval	[19]
	India	Villages	Individual	Yes	Local health system and community leaders	Cluster consultation; agreement to participatea	[20]
	Uganda	Villages	Individual	Yes	Local government authorities and local leaders	Protocol approval	[21]
Health centers	USA	Health centers	Individual	Yes	Practicing internists	Identification of cluster members; permission to approach cluster members	[22]
	UK	Primary-care practices	Cluster/ Individual	Yes	Primary-care trust administrative authority	Protocol approval	[23]
	UK	Primary-care practice locations	Cluster	Yes	General practitioners	Identification of cluster members; permission to approach cluster members	[24]
	UK	Midwives	Individual	Yes	NHS trustsd	Agreement to participatea	[25]
Nursing homes	Canada	Nursing homes	Individual	Yes	Management	Not applicablee	[26]
	Australia	Nursing homes	Individual	Yes	Director of nursing	Not applicablef	[27]
	UK	Nursing homes	Individual	Yes	Management	Agreement to participatea	[28]
	New Zealand	Wards within nursing homes	Individual	Yes	Senior management	Consentb	[29]
Schools	Germany	Classrooms	Cluster	Yes	Responsible local governments	Agreement to participatea	[30]
	Belgium	Classrooms	Cluster	No	Principals	Organizational permission	[31]
	USA	Classrooms	Cluster	No	School districtg	Protocol approval	[32]
	UK	Teachers	Individual	Yes	Principals	Organizational permission	[33]
	Canada	Schools	Cluster	No	District principals	Proxy consent for cluster members	[34]
Work sites	Sweden	Work sites	Individual	No	Managers and human resources	Identification of cluster members; permission to randomize	[35]
	USA	Work sites	Cluster	No	Employee advisory board	Cluster consultation	[36]
	USA	Fire stations	Individual	Yes	Department chiefs/ union representatives	Identification of cluster members; organizational permission	[37]
	USA	Pools	Individual	Yes	Pool managers, community advisors, recreation leaders, pool directors	Cluster consultation	[38]

^aSeveral studies report ‘agreement to participate’ or ‘permission to conduct the study’ although it remains unclear if these are cluster permission, protocol approval, or another gatekeeper role.

^bStudies that report ‘consent’ from a potential gatekeeper have not specified if this is cluster permission or proxy consent for cluster members.

^cIndex members were defined the ‘leaders of Roma (gypsy) men’s social networks’.

^dMidwives may also have acted as gatekeepers as ‘all participating midwives were given detailed training on the procedure to identify, recruit, and obtain written informed consent from participants’.

^eGatekeeper involvement was indicated by the statement that ‘[n]ursing homes withdrew…based on a decision by the nursing home management’, although a specific role was not given.

^fResearchers reported that directors of nursing ‘were given the opportunity to participate in the study’ without further clarification.

^gThis study also used a joint staff service committee (principal, vice principal, and faculty) that was informed and provided support for the study. The committee was not identified as a gatekeeper.