Effectiveness of Behavior Change Communications for Reducing Transmission Risks Among People Living with HIV in 6 Countries in Central America

This region-wide Central American study aims to estimate prevalence of HIV-related risks (sexual behaviour, HIV disclosure, number of sex partners, and violence) and factors associated with these risks, as well as evaluate a behaviour change communications (BCC) programme intended to reach people living with HIV (PLHIV) in 6 countries:  Belize, Guatemala, Panama, El Salvador, Honduras, Costa Rica, and Nicaragua. (Footnotes are removed throughout.)

The Pan American Social Marketing Organization (PASMO), a member of Population Services International (PSI) network, has been implementing an intervention programme among key populations, including men who have sex with men (MSM), transgender women (TW), female sex workers (FSWs), men at risk, and PLHIV since 2010, aligned with the Positive Health Dignity Prevention (PHDP) framework. For PLHIV, "the minimum package includes: (a) participation in at least three behavior change communications (BCC) interventions, including peer-led interpersonal communication (IPC) or online outreach with topics such as safer sex practices, condom access, adherence to ART [antiretroviral therapy], and nutrition; (b) referrals to screening and treatment of STIs [sexually transmitted infections], treatment for opportunistic infections, and access to ART programs; and (c) referrals to structural and complementary services such as family planning, support groups, legal support, and treatment for alcohol and drug use." 

Through a convenience sampling strategy, PLHIV were identified at clinics and within organisations in which staff were sensitised to the study criteria and conditions and invited to participate in late 2012. Face-to-face interviews were used to survey participants using a survey instrument adapted from the AIDS Indicator Survey and the Global Network of PLHIV Survey (GNP+). "Key outcomes for this paper were determined by programmatic objectives, including increased condom use at last sex, HIV status disclosure to sex partner, ART adherence, and current participation in a support group....Exposure to PASMO’s behavior change communications intervention was measured in 3 variables: (1) exposure to 3 TV advertisements and (2) exposure to at least one IPC session via face-to-face chat with an outreach worker or peer educator in the past 12 months. Exposure to the third variable, Internet chat, was low and thus was not included in the analysis." Statistical matching as a quasi-experimental method with coarsened exact matching (CEM) to create statistically equivalent groups was used with multiple logistic regression controlling for potential covariates, performed using the CEM-matched sample.

Results related to communication include:

• HIV-related transmission risks were associated, for women, with gender-based violence (GBV) and less likelihood of condom use, and, for men, of multiple partners and low disclosure of status. "Regardless of on-going interventions targeting health care providers who provide services for PLHIV and stigmatized populations, a small proportion of PLHIV still reported being discriminated against by a health care provider. In the context of the PHDP framework, policies and programs should continue to ensure equal access to non-discriminatory health services."
• Factors associated with condom use included, as stated earlier, GBV. "Interventions to raise awareness about this issue and to reduce violent acts are critical for the successful implementation of any HIV intervention in Central America....The study also supports the importance of health seeking behaviors in reducing HIV risk among PLHIV. Currently on ART, participation in a self-help group, and regular monitoring of CD4 [(cluster of differentiation 4), a glycoprotein found on the surface of immune cells] increased condom use."
• Factors associated with HIV-status disclosure included gender (heterosexual and homosexual men were less likely than women to disclose). "In addition, our analysis suggests that a set of risk behaviors might be nested within one another. In particular, we found PLHIV with multiple sex partners were less likely to disclose HIV status; and those who did not disclose were less likely to use condoms....These findings imply that knowing one’s partner’s HIV status is vital and sero-sorting might be an effective HIV prevention strategy."
• Effects of PASMO’s Combination Prevention Program, after 2 years of implementation, were found to have moderate reach among PLHIV in the 6 countries. Among those participating, condom use was high, most were adhering to ART and monitoring CD4, with a third participating in a self-help group.  An intervention channels analysis found: "mass media significantly increased the odds of condom use by 1.8 times, HIV disclosure by 1.5 times, and participation in a self-help group by 1.4 times. IPC has made positive impacts on condom use and participation in self-help groups by an even larger magnitude compared to mass media intervention messaging....Exposure to IPC increased odds of condom use by 2.7 and participation in self-help groups by 4.4 times.... We did not find effects of either mass media or IPC on ART adherence; and IPC also had no effect on HIV disclosure. This is probably because the messaging around ART adherence was not easily communicated by non-health communications channels, or perhaps it may take more time to have an effect. We also found no effect of both mass media and IPC on number of sex partners....The findings suggest that behavioral interventions in the larger context of a combination prevention program have positive effects on reducing HIV transmission risk among PLHIV. This approach ensures that intervention programs targeting PLHIV address their comprehensive needs, including access to care and treatment, legal support, social and emotional support, and counseling on gender-based violence. We hypothesize that the synergistic overlap of multi-component programming, like PASMO’s Combination Prevention Program might have contributed to the impact."