A.  Quantification of the population uptake of antiretroviral treatment through primary care in rural South Africa

Research question
To quantify the population uptake (and factors associated with uptake) of antiretrovirals through a decentralised system of primary care in rural South Africa.

Fit with Africa Centre portfolio
The impact of ART on the surveillance population is one of the key components of the Africa Centre’s scientific strategy.

Data sources and methods
We use results from the Centre’s population-based HIV surveillance and a district-based ART programme in combination with geographical information systems (GIS) technology to calculate treatment uptake and explore geographical variation in uptake across the study area.  To assess the equity of treatment access, we also compare the characteristics of a sample of linked HIV positive patients on treatment against HIV infected individuals not on treatment.

Results
By the end of 2008, 6,354 patients were actively receiving treatment within the antiretroviral programme. The
mean uptake of ART treatment among all HIV positive resident adults (≥15 years) in the surveillance area was 21.0% (95%CI = 20.1 – 21.9) and differences in uptake by clinic catchment ranged from 18.3 to 27.7% with the highest uptake occurring in one of the more rural catchments. At a population level, ART uptake among HIV positive men (19.2%) was only slightly lower than women (21.8%, p-value of the difference =0.011). At an individual level, there was no evidence that an HIV positive individual’s likelihood of accessing treatment was associated with level of education, urban/rural locale or household assets, but despite the decentralization of services to primary care, treatment uptake was strongly negatively associated with distance from the nearest primary healthcare facility (aOR = 0.728 per square-root transformed km, 95 % CI 0.658 – 0.963, p=0.002).

Policy implications
We show that within the early stages of antiretroviral roll-out, a decentralized system of treatment delivery through primary care has been able to reach over 20% of HIV positive adults in a rural setting.  Whilst HIV positive men are slightly less likely to have accessed treatment (in comparison to women), the service does not favour sub-populations on the basis household wealth, education or urban/rural location. However, despite decentralized services, individuals living further away from clinics were substantially less likely to be on treatment. Sustainably achieving higher levels of treatment coverage is going to require new, creative strategies.

B.  Researching equity in access to health care (REACH)

Research questions

We partnered with three sites in South Africa (Agincourt, University of the Witwatersrand, University of Cape Town) and McMaster University in Canada in this study of equity in access to health care with the following objectives:

·         To describe and compare current patterns of public sector service utilisation for three tracer interventions (ART services, TB services, and maternal health services) in different settings and study populations and to generate information on system, provider, individual and population level factors that explain these patterns.

·         To explore the dimensions of access to public sector health services, evaluate inequalities in access to services for the tracer interventions and identify barriers to access (or the determinants of ‘failures of fit’) to these interventions among those who need them.

Fit

The study fits well with an emerging health systems research agenda at the Africa Centre and the broader scientific interest in the role of health services in meeting the most pressing health needs in the Africa Centre community.

Data sources and methods

We conducted 300 patient interviews and chart reviews on dimensions of access to care (accessibility, affordability, availability, and acceptability) in each of the three tracer interventions.  The Africa Centre data was pooled with the data collected at the other three South African sites.  Ongoing analyses include estimation of equity indices, regression analyses of determinants of access to care for the three tracer intervention, economic evaluations and modeling exercises to extrapolate findings to larger population.

Findings

The data collection has been completed recently.  We expect substantive study findings and first paper submissions by May 2010.

Policy implications

The REACH collaboration plans to use study findings

·         To engage with national policy makers, local decision makers, provider groups and civil society

in the development of evidence-informed policy options for promoting more equitable access to public sector health services and to evaluate the processes of research to practice engagement.