Skip Navigation LinksHome > April 15, 2012 - Volume 59 - Issue 5 > Commentary on Greig et al, Similar Mortality and Reduced Los...
Text sizing:
JAIDS Journal of Acquired Immune Deficiency Syndromes:
doi: 10.1097/QAI.0b013e31824c1985

Commentary on Greig et al, Similar Mortality and Reduced Loss to Follow-Up in Integrated Compared With Vertical Programs Providing Antiretroviral Treatment in Sub-Saharan Africa

Spira, Thomas J. MD; Ellerbrock, Tedd V. MD

Free Access
Article Outline
Collapse Box

Author Information

HIV Care and Treatment Branch, Division of Global HIV/AIDS, Center for Global Health, Centers for Disease Control and Prevention, Atlanta, GA

The authors have no funding or conflicts of interest to disclose.

Received January 19, 2012

Accepted January 19, 2012

In the article by Greig et al1 from Médecins sans Frontières, the authors compared data from their supported HIV/AIDS care and treatment programs at 17 sites in 9 countries in sub-Saharan Africa. Initially, their support began as vertical programs, but later they changed to an integrated model that incorporated HIV treatment into general health care services. In this retrospective study (2003–2010), they compared a number of clinical indicators of antiretroviral treatment (ART) outcome, although controlling for a variety of potential confounders. However, a major potential confounder, CD4 count at initiation of ART, was not included due to 30% with missing data. The study included 14,124 patients at 7 vertical program sites and 1279 at 10 integrated sites. All of the integrated sites were rural, whereas 4 of 7 vertical sites were urban. Programs were standardized across both program types. Training and advisory staff were similar for both. A standardized electronic database containing routinely collected data facilitated this study. Follow-up data were censored at 30 months on ART to make the follow-up time equal between the 2 programs.

Although the authors used data collected from a large number of sites in multiple countries, they examined only retrospective data and were unable to address specific country contexts and the relative costs of the 2 approaches. However, the study results add useful information to the small body of studies that evaluate the utility of integrated programs compared with vertical programs.

There has been much debate over the merits of vertical versus integrated health programs in resource-limited settings. Nevertheless, the fourth of the 5 foundations of the World Health Organization Joint United Nations Programme on HIV/AIDS' (WHO/UNAIDS') initiative that aims to catalyze the next phase of HIV treatment scale-up, Treatment 2.0, is “Adapt Delivery Systems.”2 The 2020 goal is “HIV care and treatment programs are decentralized and appropriately integrated with other HIV and non-HIV health services, with increased community engagement in service delivery and improved retention in care.” This goal assumes the benefits of integration, despite limited evidence to support this strategy.

Many articles on service integration lack a clear definition of what is meant by integration of services. In the area of HIV/AIDS care and treatment, integration of services often means the combined provision of HIV/AIDS care and treatment services in tuberculosis, sexually transmitted diseases, and antenatal care/maternal–child health clinics, and/or substance abuse programs, or vice versa. Alternatively, integration of services may mean providing the gamut of HIV/AIDS services in a primary health clinic (horizontal program) rather than in a specialized clinic (vertical program). Service integration may also mean the addition of non–HIV-related services in HIV/AIDS clinics.

A 2010 systemic review by Atun et al3 of the evidence on integration of target health interventions into health systems, which focused on randomized or cluster randomized trials, before and after evaluations, interrupted time series, and program evaluations without controls in 6 programmatic areas (neglected tropical disease, nutrition, immunization, child health and development, family planning services, and HIV/AIDS) found only 55 studies for review, only 1 of which was in the area of HIV/AIDS. The authors defined integration as the extent, pattern, and rate of adoption, and eventual assimilation of health interventions into critical health systems functions. These functions included stewardship and governance, financing, planning, service delivery, monitoring and evaluation, and demand generation. Based on their review, they concluded that programs were rarely fully vertical or fully integrated. Also, studies usually addressed only limited numbers of the functions listed above. Because of the paucity of evidence, they advocated for investment in studies with robust design, comparable control and intervention groups where possible, valid and reliable outcomes, and analysis of costs. Because of the varied contexts, they also suggest the need for country case studies, which examine such health interventions to better understand the extent and nature of interventions and the reasons for the designs that emerge. The use of a common methodology would facilitate comparisons among countries and programs.

A 2011 Cochrane review by Dudley and Garner4 of the evidence on the effectiveness of integration strategies at the point of delivery, including delivery of tuberculosis, HIV/AIDS, and reproductive health programs found only 5 randomized trials and 4 controlled before and after studies. The interventions were complex. The review concluded that there was some evidence that adding on services (or linkages) may improve utilization and outputs of health care delivery. However, there was no evidence that a fuller form of integration improves health care delivery or health status. Evidence suggested that full integration probably decreases the knowledge and utilization of specific services and may not result in any improvements in health status.

Some recent studies, not included in the previous reviews, are of interest. Pfeiffer et al5 describe integration efforts in Mozambique. In 2005, Mozambique began a process of decentralization of the ART and HIV care program from a “vertical” approach, primarily at large sites in urban areas, to smaller integrated sites in rural areas. A description of this process and its effects in 2 provinces, Sofala and Manica, provides some insights into the implementation of this process at the provincial level. Integration was defined as colocation of different services within the same facility, even if separately staffed; training of personnel to provide multiple services; provision of tools, processes, and training to better link separate services; strengthening of linkages, referral and follow-up between facility levels; and harmonization of logistics systems, such as data collection, drug and material distribution, transport, and supervision across services. The change in care and treatment provision resulted in an increase in the percentage of local qualifying patients initiating on ART and shortened the length of time from registration to initiation. There was also a decrease in the loss to follow-up from referrals of HIV-positive women from prevention of mother-to-child transmission services to ART services from 70% at vertical sites to 25% at integrated sites. Workforce shortages were reported as the greatest single challenge to successful integration. There was also a concern that the addition of HIV care tasks to overburdened staff may affect the quality of care.

Topp et al6 have reported on the integration of vertical ART and outpatient department (OPD) services in the low-resource and high HIV-prevalence setting of Lusaka, Zambia. Integration involved the shared use of space and staffing, standardization of medical records, introduction of routine provider-initiated HIV testing and counseling, and modification of patient flow patterns. Integration resulted in increased HIV case finding and entry into HIV care, a reduction both patient and staff perceptions of stigma associated with HIV care and treatment, an increase in OPD patients receiving prescreening collection of vital signs, and improved staff communication and teamwork. However, it also resulted in an overall increase in patient waiting times for both OPD and ART patients. The waiting time increase in the OPD patient-waiting times was thought to be due to the reintroduction of the collection of vital signs and the addition of provider-initiated HIV testing and counseling. Another negative or unintended outcome was some ART patients' perception that the integrated service provided a less secure environment in which to share experiences with fellow patients.

In a recent article, Grépin7 discussed the negatives associated with vertical programs which include distortion of local health priorities, undermining country ownership, and potential for negative spillovers on health systems. Vertical programs may also not be sustainable in the long run. However, the integration approach, that of integrating additional health services into existing HIV infrastructure, might undermine the HIV programs themselves because of decreased focus on HIV-specific outcomes and/or decreased funding for these services. Grépin6 suggests that international donors and health system planners carefully consider whether the benefits outweigh the potential costs of these well-intentioned integration efforts.

There is clearly a need for more rigorous studies of different strategies to assess integration impact over a wider range of services and settings. Because of varying country contexts, obtaining country-specific data is also important. As noted above, a common robust methodological framework would facilitate comparison across countries. These studies should include economic evaluation and the views of clients. which would influence their uptake of integration strategies at the point of delivery. Ideally, studies should also be of sufficient duration to address long-term consequences on quality of care and sustainability. Informed evidence-based decision-making on this issue is needed and will have major impact on future program design and implementation. With future funding for strengthening health care in resource-limited countries likely to be limited, we need to ensure that programs are efficient and provide quality care to patients.

Back to Top | Article Outline


1. Greig J, O'Brien D, Ford N, et al.. Similar mortality and reduced loss to follow-up in integrated compared with vertical programs providing antiretroviral treatment in sub-Saharan Africa. J Acquir Immune Defic Syndr. 2012;59:e92–e98.

2. WHO. The Treatment 2.0 Framework for Action; Catalyzing the Next Phase of Treatment, Care and Support. Geneva, Switzerland: WHO; 2011. Available at: February 17, 2012.

3. Atun R, de Jongh T, Secci F, et al.. A systematic review of the evidence on integration of targeted health interventions into health systems. Health Policy Plan. 2010;25:1–14. Available at: Accessed February 17, 2012.

4. Dudley L, Garner P. Strategies for integrating primary health services in low- and middle-income countries at the point of delivery. Cochrane Database Syst Rev. 2011. CD003318. doi:10, 1002/14651858. Available at: Accessed February 17, 2012.

5. Pfeiffer J, Montoya P, Baptista AJ, et al.. Integration of HIV/AIDS services into African primary health care: lessons learned for health system strengthening in Mozambique—a case study. J Int AIDS Soc. 2010;13:3–11. Available at: Accessed February 17, 2012.

6. Topp SM, Chipukuma JM, Giganti M, et al.. Strengthening health systems at facility-level: feasibility of integrating antiretroviral therapy into primary health care services in Lusaka, Zambia. PLoS ONE. 2010;57e11522. doi:10.1371/journal.pone.0011522. Available at: Accessed February 17, 2012.

7. Grépin KA. Leveraging HIV programs to deliver an integrated package of health services: some words of caution. J Acquir Immune Defic Syndr. 2011;57:S77–S79. Available at: Accessed February 17, 2012.

© 2012 Lippincott Williams & Wilkins, Inc.