University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Peer-Driven HIV Interventions for Drug Injectors in Russia: First Year Impact Results of a Field Experiment
Global Health Sciences Literature Digest
Published November 16, 2006
Journal Article

Broadheada RS, Volkanevsky VL, Rydanovab T, Ryabkovab M, Borcha C, van Hulsta Y, Fullertona A, Sergeyevc B, Heckathornd DD. Peer-Driven HIV Interventions for Drug Injectors in Russia: First Year Impact Results of a Field Experiment. Intrnl Jrnl Drug Policy. 2006 Sept;17(5):379-92.


To determine how different reimbursement strategies for participant referral can impact recruitment and educational effectiveness of peer-driven interventions (PDI) for injection drug users (IDUs), in Russia.


This paper initially describes a pilot PDI performed between 1996-1998 in Yaroslavl, Russia, the results of which prompted the present study. In the pilot program, IDUs recruited community members, who in turn were asked to recruit other peer IDUs (a method now called respondent-driven sampling). Health educators were responsible for ensuring follow-up. Follow-up was only 34% at three months; reductions in sharing of drug paraphernalia were present but small; and knowledge remained poor. As a result, the authors modified the PDI strategy to determine whether they could enhance education at initial recruitment, improve follow-up rates, and reduce recruitment costs.

Study Design

Quasi-experimental two-arm intervention conducted between February 2003 and August 2004 in which a standard peer-driven intervention (standard PDI) was compared to a simplified PDI.


Two cities (Bragino and Rybinsk) in Yaroslavl Province, western Russia, approximately 60 miles apart. These cities are the most affected by HIV in the province and the most similar demographically.


A total of 857 participants were peer-recruited into the two intervention arms: 493 recruits participated in the standard PDI, and 364 recruits participated in simplified PDI. All were IDUs, and most were males in their teens and twenties.


Both intervention arms used the following components of a peer-driven intervention model. Current IDUs were asked to: 1) educate peers in the field about HIV prevention; 2) recruit peers to the study site where they could be enrolled, undergo health and risk assessment, and receive prevention services, including HIV testing and counseling, needle exchange, and education; and 3) distribute harm-reduction materials in the field, such as condoms, bleach, clean paraphernalia, bandages, sterile cotton pads, and informational brochures. Once participants ("recruits") were enrolled at the study site and had completed the baseline survey, they were asked in turn to become "recruiters," that is, to educate other peer IDUs and send them to the study site. New recruiters were given "crib" cards reminding them of the eight information points about which they were to educate peers in the field, and on which new participants would be subsequently tested. Six months after baseline, participants were "re-recruited" for follow-up, educated by peers on additional HIV prevention information on which they would be re-tested.

In both the standard and simplified PDI, participants were paid for completing the survey. Those who became recruiters were also given monetary rewards based on how well their referred peers performed on the knowledge test, how many were women, and whether verifiable follow-up information was collected. The primary difference between the two intervention arms was that in the standard PDI, recruiters received additional reimbursement for each IDU they referred to the study.

Primary Outcome

1) The number of IDUs recruited; 2) effectiveness in educating recruits about HIV prevention, measured by a knowledge test; 3) re-recruitment and follow-up 6 months post-baseline; 4) effectiveness in educating follow-up recruits, as measured by a second knowledge test; 5) self-reported drug-and sexual risk behavior; 6) differences in costs between the two intervention strategies.


The costs associated with paying recruiters in the simplified PDI were 50% less than the standard PDI. Over 18 months, the standard PDI recruited 129 more IDUs (493 vs. 364). There was no difference in six-month follow-up rates between the two arms: 74% follow-up in the standard and 72% in the simplified intervention. However, participants in the simplified PDI scored better on knowledge than those in the standard PDI. Between baseline and follow-up, there was a similar reduction in injection frequency, the sharing of paraphernalia, and rates of unprotected sex in the two arms.


The authors concluded that by offering different rewards, intervention projects get what they pay for. The more participants are rewarded for educating vs. recruiting peers, the more effort they will put into educating them; while the more participants are paid for recruiting, the more persons they will refer. In comparison to the pilot study, both interventions demonstrated improved follow-up rates, which could be attributed to the fact that peers, rather than health educators, facilitated follow-up.

Quality Rating

The quality of this quasi-experimental study was not graded. Although the authors made an effort to implement the interventions in comparable cities, the differences between the cities may have confounded the differential recruitment rates. The quality of this study was also limited in that the only difference between the two intervention arms was the incentives provided to recruiters, which is unlikely to have impacted most of the outcomes of interest. The primary difference was a change in education score, which did not translate into a change in behavior.

In Context

Developed in the US in the 1990s, the PDI referral model has been successfully used in a variety of settings to recruit IDUs. Although the results of the 1996-1998 PDI pilot study in Yaroslavl demonstrated the feasibility of using a standard PDI recruitment and education model for use in Russia,(1) the literature regarding the use of peer-driven recruitment among hard-to-reach populations in developing countries is just beginning to be published.(2) This study provides interesting and fairly detailed information about how such programs are implemented.

Programmatic Implications

Among injection drug users, peer-driven interventions can be successfully employed to recruit and educate other IDUs about HIV prevention. However, the results of this study indicate that there were no differences between the two interventions with regard to reduction in risk behavior. Thus, although exploring different incentive schemes may increase the number of participants, it may not impact viral transmission. Exploration of factors that can reduce risky sex and sharing of paraphernalia, rather than simply improving the ability to perform well on a written test, is still needed. Efforts should be made to improve the method and quality of peer education.


  1. Sergeyev, B. et al. HIV prevention in Yaroslavl, Russia: A peer-driven intervention and needle exchange. Journal for Drug Issues, 29(4), 777-804.
  2. Stormer A, Tun W, Guli L, Harxhi A, Bodanovskaia Z, Yakovleva A, Rusakova M, Levina O, Bani R, Rjepaj K, Bino S. An Analysis of Respondent Driven Sampling with Injection Drug Users (IDU) in Albania and the Russian Federation. J Urban Health. 2006 Oct 31; [Epub ahead of print].