Table 2.

Methodology and Intervention design of Level III studies (K=13)

Author	Randomization Method (if any)	Methods of Identifying Index	Method of Identifying social network members	Network Member Involvement	Comparison/Control Group
Broadhead, RS et al. 43	None	Recruited through street outreach, word of mouth, ads and referrals from community agencies.	Peer educators identified and recruited network members who were previous sexual and/or drug using partners	Indexes and recruited network members became part of the study population	Yes. Comparison group is TOI
Heckathorn, DD et al. 44	None	Recruited through street outreach, word of mouth, ads and referrals from community agencies.	Peers educators are motivated to recruit other drug users via a coupon system.	Indexes and recruited network members became part of the study population	Yes. Comparison group is TOI
Servegev, B et al. 46	None	Recruited through street outreach, word of mouth, ads and referrals from community agencies	Peers educators are motivated to recruit other drug users via a coupon system.	Indexes and recruited network members became part of the study population	No
Broadhead, RS et al. 49	None	There were no ‘indexes’. All participants were recruited through referrals from local health care providers and peers	From the study population, project staff identified HAs and peers and then assigned one HA to a peer. No two participants played both roles for one another.	All participants became part of the study. There was no distinction between index and network members	No
Latkin, CA et al. 54	Randomly assigned in a ratio of 2:1 to the intervention or control condition respectively	Recruited through targeted outreach. Recruitment areas in Baltimore City were identified through ethnographic observations, focus groups, and geographical coding of drug- related arrests in Baltimore in the prior 3 years of the study	Index participants recruited network members	Indexes were asked to recruit a maximum of two network members for assessment at baseline and follow-ups. The network members did not receive the intervention	Yes. Control group with no intervention
Broadhead, RS et al. 58	None	Indexes were referred to the project by local narcologists or physicians	Index participants recruited a maximum of 3 IDU peers from their community	Indexes and recruited network members became part of the study population	Yes. Standard PDI
Booth, RE et al. 66	None	Peer leaders were recruited through street outreach by former IDUs	Index participants recruited a maximum of 3 IDU peers from their injecting network	Received communication from peer leaders, completed baseline and follow-up surveys. Network members did not receive the intervention	Yes. Individual based intervention
Deering, KN et al. 67	None	There were no indexes. All participants were recruited into the PDI through referral by an HIV specialist, family care physician, or other health provider, friend, or by self-referral	Project staff allocated the pairing of a health advocate- peer dyad	All participants were part of the study population	No
Sherman, SG et al. 68	Random allocation to two arms	Index participants were recruited based on an extensive 18- month formative, ethnographic research stage prior to the RCT	Index participants enrolled at least one of their sex or drug network members in the study within 45 days of screening.	Received communication/conversation from peer leaders, completed baseline and follow-up surveys. Network members did not receive the intervention	Yes. Comparison group is Life- Skill curriculum
Booth, RE et al. 73	None	Peer educators and those in the individual intervention were recruited by recovering drug users serving as outreach workers	Index participants recruited a maximum of 3 members of their injecting network.	Received communication from peer educators, completed baseline and follow-up surveys	Yes. Comparison group is Individual Intervention.
Li, J. et al. 78	None	Index participants or PHAs were recruited by outreach workers’ judgment, based on their familiarity with participants, about the candidates’ links in the drug-using community	Indexes were asked to refer 2 – 3 drug using peers (injection or non-injection heroin or cocaine/crack users)	Received intervention from PHAs, recruited in the study and completed baseline and follow-up social network egocentric surveys	No
Smyrnov, P et al. 80	None	The health educators (HEs) chose active IDUs to serve as indexes, known to be knowledgeable of the local drug- using community	Index participants recruited a maximum of 3 members of their injecting network.	Indexes and recruited network members became part of the study population	Yes. Comparison group is TOI
Latkin, CA et al. 87	Yes. Indexes were randomized into a peer education intervention or control condition	Index participants were identified through a community-based recruitment, which included ethnography and outreach in zip codes with high rates of HIV/AIDS cases based on data from the Philadelphia Department of Public Health. Outreach workers disseminated verbal and written information about the study	Peer educators identified and recruited network members from their sexual and or drug sharing networks	Received communication from peer educators, completed baseline and follow-up surveys. Social network members did not receive the intervention	Yes. Control group with no intervention

Notes: PDI = Peer driven intervention; IDU = Injecting drug users; Peers are same as Indexes; TOI = traditional outreach intervention