|Home | About | Journals | Submit | Contact Us | Français|
Expanded access to antiretroviral therapy (ART) offers opportunities to strengthen HIV prevention in resource-limited settings. We invited 27 ART programmes from urban settings in Africa, Asia and South America to participate in a survey, with the aim to examine what preventive services had been integrated in ART programmes. Twenty-two programmes participated; 8 (36%) from South Africa, 2 from Brazil, 2 from Zambia and 1 each from Argentina, India, Thailand, Botswana, Ivory Coast, Malawi, Morocco, Uganda and Zimbabwe. Twenty-one sites (96%) provided health education and social support, and 18 (82%) provided HIV testing and counselling. All sites encouraged disclosure of HIV infection to spouses and partners, but only 11 (50%) had a protocol for partner notification. Twenty-one sites (96%) supplied male condoms, 7 (32%) female condoms and 20 (91%) provided prophylactic ART for the prevention of mother-to-child transmission. Seven sites (33%) regularly screened for sexually transmitted infections (STI). Twelve sites (55%) were involved in activities aimed at women or adolescents, and 10 sites (46%) in activities aimed at serodiscordant couples. Stigma and discrimination, gender roles and funding constraints were perceived as the main obstacles to effective prevention in ART programmes. We conclude that preventive services in ART programmes in lower income countries focus on health education and the provision of social support and male condoms. Strategies that might be equally or more important in this setting, including partner notification, prompt diagnosis and treatment of STI, and reduction of stigma in the community, have not been implemented widely.
HIV prevention efforts in high-income and low-income settings have traditionally focused on reducing the risk of HIV infection among HIV-negative individuals or in populations with unknown serostatus, and concerns about stigma and negative social outcomes related to knowledge of HIV status have been paramount (Bunnell, 2006). HIV testing services were not widely available, particularly in resource-limited settings. This situation has changed in recent years with the scaling up of antiretroviral therapy (ART) and HIV testing in low- and middle-income countries. Close to 3 million people living with HIV/AIDS were estimated to be receiving treatment at the end of 2007 (World Health Organization, 2008c). Adherence to ART (Mills et al., 2006) and immunological and virological response (Braitstein et al., 2006; Keiser et al., 2008b) in resource-limited countries equal or exceed that in high-income settings, and a decline in mortality at the population level has recently been observed in Malawi and South Africa (Jahn et al., 2008; Egger & Boulle, 2008).
The scale-up of ART in resource-limited settings is a major achievement, which is extending the lives of many people with HIV/AIDS. It is however clear that “we cannot treat our way out of this epidemic”: in the absence of reductions in HIV incidence, ART programmes will not be able to keep up with the increasing number of people in urgent need of treatment (De Cock, 2006). The Global HIV Prevention Working Group, a panel of public health experts, clinicians, researchers and people affected by HIV/AIDS, the World Health Organization (WHO) and other agencies have argued that expanded access to ART offers new opportunities to strengthen HIV prevention (Gayle & Lange, 2004; World Health Organization, 2008b). It is unclear to what extent preventive services have been integrated in ART programmes in resource-limited settings. We addressed this question in a survey of treatment sites in Africa, Asia and South America.
We conducted an electronic survey of the 27 programmes that provide ART to adult patients in 13 countries in Africa, Asia and Latin America and participate in the ART in Lower Income Countries (ART-LINC) collaboration, which is part of the International epidemiological Databases to Evaluate AIDS (IeDEA). The collaboration has been described in detail elsewhere (Dabis et al., 2005; Keiser et al., 2008a). Briefly, it was set up in 2003 with the aims of defining the prognosis of HIV-infected patients treated with ART in resource-limited settings and to compare the experience between different settings, delivery modes and types of monitoring. Potentially eligible sites were identified in 2003 by screening abstracts of international conferences, performing MEDLINE searches, and through personal contacts. A total of 31 centres were contacted. To be eligible, centres needed to prospectively collect information on patients treated with ART. The ethics committees or institutional review boards of all participating sites approved participation in ART-LINC of IeDEA.
The study was discussed with the Steering Committee of ART-LINC, and all collaborators were invited to comment on draft questionnaires. Item generation involved the compilation of a comprehensive list of relevant recommendations on HIV-prevention in HIV-infected people (“positive prevention”). We performed searches in Medline (Pubmed version) without language restrictions to identify relevant articles in May 2007and updated searches in July 2008. We checked the citations to five key articles (Rao et al., 1991; Allen et al., 1992; Oxman et al., 1994; Marks et al., 1999; Kalichman et al., 2001) in the Science Citation Index (ISI Web of Science). We also checked the reference lists of relevant articles, contacted experts and incorporated suggestions proposed by the steering committee of the collaboration. We then selected the items we considered relevant to the settings represented in ART-LINC, formulated questions, and organised them in different sections. We provided space to allow respondents to comment. The questionnaire was pilot tested and further changes were made.
The final version consisted of 140 closed-ended and 54 open-ended questions relating to the site, the person completing the questionnaire and the prevention services and other relevant activities implemented by the programme. The content of the questionnaire, which is available on request, is summarised in Table I. The questionnaire was implemented in September 2008 in DataCol (Data Collector), a web-based tool developed by WHO to create electronic surveys and forms (World Health Organization, 2008a). The data are stored in a WHO-maintained database that is regularly backed up. Respondents downloaded the questionnaire as an executable file, completed it on-line or off-line, and sent it back to the data centre at University of Bern.
Answers to open-ended questions were summarised and coded where appropriate. We used univariable, unweighted logistic regression analysis to explore whether programme characteristics influenced the availability of preventive services. We considered the type of programme (government versus non-governmental organizations or private clinic), access to ART (free of charge versus fee-for-service) and work load (scale-up cohorts with rapid increase in patient numbers in recent years versus cohorts with stable patient numbers). Dependent variables were whether or not the programme had a protocol for partner notification, whether or not it routinely screened for sexually transmitted infections (STI) in asymptomatic patients and whether dedicated services for adolescents or serodiscordant couples were available. Most countries contributed one site only, which precluded analyses of country characteristics. Results are presented as number of sites (%) or odds ratios (ORs) with 95% confidence intervals (CIs). All analyses were done with STATA version 10 (Stata Corp; College Station, TX).
Twenty-two of 27 sites returned the questionnaire (response rate 82%). The questionnaire was completed by the clinical director in 15 sites (68%), by researchers in five sites (23%) and by clinic managers in two sites (9%). Nine respondents were female (41%) and thirteen were male (59%).
The characteristics of sites are summarised in Table II. Eight sites (36%) were located in South Africa, two in Brazil, two in Zambia and one each in Argentina, India, Thailand, Botswana, Ivory Coast, Malawi, Morocco, Uganda and Zimbabwe. The occupational programme of a brewery company included five countries (Nigeria, Republic of Congo, Democratic Republic of Congo, Rwanda, Burundi). This programme was standardised across sites and was therefore treated as a single site. All 22 sites were located in urban areas, 19 were specialised ART clinics and three sites (two occupational clinics and a private clinic) also functioned as general practices. Sixteen programmes (73%) were public and government-funded and six were run by non-governmental organisations. Sixteen sites (73%) provided ART free of charge (Table II). The five sites that did not participate included one smaller research cohort in Senegal and four scale-up cohorts in Botswana, Kenya, South Africa and Zambia.
Health education was provided by 21 of 22 sites (96%). Table III details the topics covered: most programmes addressed the routes of transmission, prevention of sexual transmission by male condoms and prevention of mother to child transmission (PMTCT). Other topics, for example female condoms, were covered by fewer programmes. Health education was delivered through talks in the waiting area (17 sites, 81%), posters (15 sites, 71%) and flyers (13 sites, 62%). Twenty-one sites (96%) provided male condoms and seven (32%) female condoms. Practical instructions on the correct use of male condoms were provided by 12 sites (57%), and on the use of female condoms by 7 sites (33%). Condoms were generally available free of charge. Twelve sites (55%) provided injectable or oral contraceptives. All sites provided post exposure prophylaxis (PEP) for staff, for example after needle stick injuries. No site provided clean needles or syringes for intravenous drug users. Twenty-one sites (96%) tested for STI but only 12 of these (57%) screened asymptomatic patients at the first visit and seven (33%) screened at regular intervals. Twenty sites (91%) screened routinely for syphilis in pregnant women. Point of care testing for syphilis was available in 15 sites (68%) and same day treatment in nine of these 15 sites (60%).
Ten sites (46%) were involved in activities to reduce poverty in HIV-infected patients, including, for example, projects to generate income or support to access social security grants. One programme was involved in activities to reduce stigma associated with HIV in the community, through marches, radio talks, theatre or the distribution of T-shirts printed with the words ‘HIV-positive’.
All sites encouraged patients to disclose their HIV test result to partners and spouses. Twenty-one sites (96%) provided psychosocial support. This included counselling (21 sites, 100%), referral to self-help groups (17 sites, 81%) and drop-in centres (4 sites, 19%). Issues addressed included financial problems, stress management, coping strategies and the strengthening of self-confidence. Eleven sites (52%) addressed gender-based violence. Counselling and testing was available on site in 18 (82%) programmes. This included voluntary counselling and testing (VCT) at 16 of 18 sites (89%), couple counselling and testing in 15 sites (83%) and routine, opt-out, testing in pregnant women at 11 sites (61%). Eleven sites (50%) had a protocol for managing partner notification of HIV-infected people; most of them used the patient referral method, where patients themselves inform sexual partners and ask them to attend a clinic for counselling and testing. One site indicated that they used provider referral where the health professional contacts the partner on behalf of the patient, and two sites reported that they would inform partners if the patient had not done so after a certain period of time.
Twenty sites (91%) provided ART to prevent mother-to-child transmission (Table IV). Regimens for the mother varied widely. Nine sites (41%) provided triple combination ART based on NVP or efavirenz. Other regimens included zidovudine (AZT) from 28 weeks and single dose nevirapine (NVP), dual therapy with AZT and lamivudine (3TC) and single dose NVP in all women or in women with CD4 cell counts above 200 cells/μL, Similarly, in newborns, regimens included single dose NVP, single dose NVP and AZT for one week or one month. A minority of sites (8, 36%) recommended elective Caesarean section.
Most programmes (19, 86%) recommended exclusive replacement feeding to prevent MTCT but over half of them (11 of 19, 58%) also recommended exclusive breast feeding either followed by rapid weaning or complementary (mixed) feeding (Table IV). Eleven of the 19 programmes (58%) provided formula milk free of charge. With the exception of three sites in South Africa and Botswana, 12 of the 15 programmes in sub-Saharan Africa (80%) recommended six months of exclusive breastfeeding followed by complementary feeding (5 sites) or rapid weaning (7 sites). Figure 1 shows the number of sites involved in preventive services targeted at different groups. In addition to PMTCT, 12 sites (55%) were involved in activities aimed at women or adolescents, and 10 sites (46%) worked with serodiscordant couples. Few sites provided preventive services to other groups.
There were no statistically significant associations with programme characteristics. There was some evidence that partner notification protocols were less likely to be in place in programmes not funded through government (OR 0.39, 95% CI 0.06–2.8), in programmes that charged patients for ART (0.12, 95% CI 0.01–1.29) and in scale-up programmes (0.66, 95% CI 0.11–4.0). Regular screening for STI tended to be less likely in scale-up cohorts compared to other programmes (OR 0.30, 95% CI 0.03–3.3).
In an open-ended question respondents were asked to describe the main problems in implementing preventive services in their programme. The “background of stigma, discrimination, blame” (female respondent from Thailand) was considered a major problem in implementing prevention services and similar comments were made by six other respondents from Argentina, India, South Africa, and Zambia. Another recurring theme was the very high patient load “so there is often inadequate time for counselling” (female respondent from Zimbabwe) and financial constraints, “which prevent programmes from hiring additional staff” (male respondent from Côte d’Ivoire).
Preventive services for HIV-positive people were available in most of the ART programmes surveyed in this study. The programmes focused on health education aimed at changing behaviour of patients and the provision of social support, male condoms and of interventions to prevent transmission from mothers to their newborns and infants. Fewer sites had protocols for partner notification or offered specific interventions for discordant couples, adolescents and other groups at high risk. Also, only a minority of sites regularly screened for STI and only one site was involved in activities to reduce the stigma and discrimination associated with HIV infection.
Our survey predates the publication of WHO’s recommendations on essential prevention interventions (World Health Organization, 2008b) but it covered many of the recommended interventions and may thus serve as a baseline to assess progress with integrating interventions in ART programmes. Our study has several limitations, however. The clinics and programmes that participate in ART-LINC of IeDEA are not representative of all sites providing ART in the countries studied, being mainly urban and having electronic databases for data capture (Forster et al., 2008). The provision of preventive services, or the lack thereof, was not, however, a criterion for inclusion in the ART-LINC collaboration. Also, many of the countries that are scaling up ART were represented in the survey. The data were self-reported and results might have been affected by social desirability bias. This seems unlikely considering that sites have been collaborating for several years and helped design the study. Finally, the quality of preventive services was not assessed.
Although VCT was provided by most sites, only a minority of programmes reported specific interventions targeted at discordant couples. Discordant couples represent an important target for HIV-1 prevention, particularly in sub-Saharan Africa, where most transmissions occur among discordant couples who are unaware of their serostatus (Gayle & Lange, 2004; Piot et al., 2008). ART programmes are in an ideal position to identify discordant couples and deliver interventions. A trial in Kenya, Tanzania and Trinidad, which randomly assigned couples to VCT or basic health education, found that couples assigned to counselling and testing reduced unprotected intercourse more than couples assigned to health education (The Voluntary HIV-1 Counseling and Testing Efficacy Study Group, 2000).
Providing advice on disclosing to sexual partners is part of the testing and counselling process. There is therefore substantial overlap with partner notification, in which the partners of people with STI are informed of their exposure (Hogben et al., 2007; Mathews et al., 2001), but only half of programmes had protocols for partner notification. Partner notification for HIV infection has also been difficult to implement in industrialised countries, where it is sometimes viewed as a coercive public health intervention that conflicts with protecting the individual’s right to privacy, especially before the widespread introduction of ART (Golden et al., 2008). The diagnosed person might experience negative events including relationship breakdown or physical violence (Grinstead et al., 2001; Hogben et al, 2007). The terminology of personalised risk reduction in a client-centred approach might have facilitated the acceptance of VCT services for HIV infection. The benefits of involving patients with other STI in informing their partners are becoming increasingly recognised (Trelle et al., 2007). Sharing successful approaches to VCT and partner notification interventions may thus contribute to improving the control of all STI. Clearly, either intervention should only be offered if the support needed to help newly diagnosed people, and particularly women, to cope with their disease is also available. It is reassuring that most sites provided such support.
Few sites offered services that covered several or all components of PMTCT, including the prevention of infection in HIV-negative women, of unwanted pregnancies, and of transmission to the offspring during pregnancy, birth and breastfeeding. Only 12 sites (55%) provided contraception other than condoms. The prevention of unwanted pregnancies as an important primary prevention strategy does not appear to have been widely implemented. Also, recommendations and drug regimens for the prevention of transmission to neonates and infants varied widely. This may be a reflection of the complexity of current guidelines (World Health Organization, 2006) but some of the variation must be of concern: not all sites systematically provided suppressive triple combination ART to pregnant women with indications for ART, a strategy that is safe, highly effective and unequivocally recommended by WHO (World Health Organization, 2006).
Our study is relevant to the debate on whether infants of HIV-infected women in resource-limited settings should be breastfed or formula fed (Mofenson & McIntyre, 2000; Coutsoudis et al., 2002). HIV transmission to infants occurs through breastfeeding but exclusive breastfeeding for the first few months of life is associated with lower transmission rates than is complementary feeding (Iliff et al., 2005; Coovadia et al., 2007). Formula feeding substantially increases the risk of death due to diarrhoea, acute respiratory infections and other infectious diseases (WHO Collaborative Study Team, 2000). The updated WHO guidelines (World Health Organization, 2007) therefore stress that although formula feeding may be the right choice for some HIV-infected women, exclusive breastfeeding for the first six months is the gold standard for the babies of HIV-infected mothers who live in resource-poor settings and cannot supply replacement feeding safely (Coutsoudis et al., 2008). In keeping with these guidelines recommendations and practices differed across sites, and differed within the same country, for example in South Africa, where large disparities in socio-economic status and infant mortality rates exist. It nevertheless appears that the updated guidelines may not yet have been widely implemented: nine sites (41%) promoted rapid weaning, which is not recommended by the guidelines (World Health Organization, 2007).
Our respondents stressed that stigma and discrimination at the level of the community, high patient load and financial constraints were important obstacles for the implementation of preventive services in ART programmes. The Global HIV Prevention Working Group (Gayle & Lange, 2004) and other agencies have argued that staff of ART programmes should be trained to deliver prevention services. In particular, ART programmes should actively address and reduce discrimination of HIV-infected people in the communities they serve. Piot and colleagues called for a coalition of youth, women’s, and religious organisations as well as business leaders and HIV/AIDS activists to create a newly energised HIV prevention movement (Piot et al, 2008). By making effective treatment and appropriate prevention services available, ART programmes can make important contributions to both prevention and the expansion of access to ART. Treatment programmes should not operate in isolation but need to be part of, and supported by, the prevention constituency.
We are grateful to all the sites that participated in this survey: Fundación Huesped, Buenos Aires, Argentina: Independent Surgery, Gaborone, Botswana; Service des Maladies Infectieuses, Casablanca, Maroc; McCord Hospital, University of Kwazulu Natal, South Africa; MTCT+ Network, Cato Manor, South Africa; Y R Gaitonde Centre for AIDS Research and Education (YRG Care), Chennai, India; Helen Joseph Hospital Themba Lethu Clinic, Johannesburg, South Africa; Perinatal HIV Research Unit, Soweto, South Africa; CorpMed Medical Centre, Lusaka, Zambia; Centre de Prise en Charge, de Recherche et de Formation sur le VIH/SIDA (CEPREF), Abidjan, Côte d’Ivoire; Immune Suppression Syndrome (ISS) Clinic, Mbarara, Uganda; Gugulethu ART Programme, Cape Town, South Africa; Masiphumalele ART Programme, Cape Town, South Africa; Connaught Clinic, Harare, Zimbabwe; Tygerberg Hospital, Cape Town, South Africa; Heineken Medical Services ART Programme, Amsterdam, The Netherlands; South Brazil HIV Cohort (SOBRHIV), Hospital de Clinicas, Porto Alegre, Brazil; MTCT+ Network, Lusaka, Zambia; Thai Red Cross AIDS Research Centre, Bangkok, Thailand; Rio HIV Cohort, Rio de Janeiro, Brazil; Khayelitsha ART Programme, Cape Town, South Africa; Lighthouse Trust Clinic, Lilongwe, Malawi. We thank Kamal El Filali, Janet Giddy, Gupreet Kindra, Pappie Louis Majuba, Tim Meade, Eugène Messou, Nicholas Musinguzi, Catherine Orrell, Margaret Pascoe, Elizabeth Stringer, Somsri Tantipaibulvut, Gilles Van Cutsem and Ralf Weigel for completing the questionnaire.
The ART-LINC collaboration of the International epidemiological Databases to Evaluate AIDS (IeDEA) is funded by the US National Institutes of Health (Office of AIDS Research and National Institute of Allergy and Infectious Diseases) and the French Agence Nationale de Recherches sur le Sida (ANRS).
Conflict of Interest
We declare that we have no conflict of interest.