Population-Level Reduction in Adult Mortality after Extension of Free Anti-Retroviral Therapy Provision into Rural Areas in Northern Malawi

Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, United Kingdom.
PLoS ONE (Impact Factor: 3.23). 10/2010; 5(10):e13499. DOI: 10.1371/journal.pone.0013499
Source: PubMed


Four studies from sub-Saharan Africa have found a substantial population-level effect of ART provision on adult mortality. It is important to see if the impact changes with time since the start of treatment scale-up, and as treatment moves to smaller clinics.
During 2002-4 a demographic surveillance site (DSS) was established in Karonga district, northern Malawi. Information on births and deaths is collected monthly, with verbal autopsies conducted for all deaths; migrations are updated annually. We analysed mortality trends by comparing three time periods: pre-ART roll-out in the district (August 2002-June 2005), ART period 1 (July 2005-September 2006) when ART was available only in a town 70 km away, and ART period 2 (October 2006-September 2008), when ART was available at a clinic within the DSS area. HIV prevalence and ART uptake were estimated from a sero-survey conducted in 2007/2008. The all-cause mortality rate among 15-59 year olds was 10.2 per 1000 person-years in the pre-ART period (288 deaths/28285 person-years). It fell by 16% in ART period 1 and by 32% in ART period 2 (95% CI 18%-43%), compared with the pre-ART period. The AIDS mortality rate fell from 6.4 to 4.6 to 2.7 per 1000 person-years in the pre-ART period, period 1 and period 2 respectively (rate ratio for period 2 = 0.43, 95% CI 0.33-0.56). There was little change in non-AIDS mortality. Treatment coverage among individuals eligible to start ART was around 70% in 2008.
ART can have a dramatic effect on mortality in a resource-constrained setting in Africa, at least in the early years of treatment provision. Our findings support the decentralised delivery of ART from peripheral health centres with unsophisticated facilities. Continued funding to maintain and further scale-up treatment provision will bring large benefits in terms of saving lives.

Download full-text


Available from: Basia Zaba, May 24, 2014
  • Source
    • "As none of the studies were randomized trials, none used any method of randomization or concealment of allocation. Only two studies included a form of control group (Bendavid and Bhattacharya 2009; Jia et al. 2010); in neither of these was the control group adequately matched to the intervention group, although the study by Bendavid and Bjattacharya attempted to Table 1 Summary of health outcomes and impact measures reported in included studies Indicator category Indicator Studies reporting indicator Service coverage ART coverage Floyd et al. (2010), Jahn et al. (2008) TB case detection and notification rates Jia et al. (2010), Yumo et al. (2011) % of households owning at least one (long lasting) insecticide treated net Akachi and Atun (2011), Chizema-Kawesha et al. (2010) and Steketee et al. (2008), Flaxman et al. (2010), Noor et al. (2007) % of pregnant women receiving at least two doses of Intermittent Preventive Therapy Chizema-Kawesha et al. (2010) and Steketee et al. (2008) Service utilization % of children under-5 years sleeping under an insecticide-treated bed net Noor et al. (2007), Flaxman et al. (2010), Steketee et al. (2008) ART adherence and loss to follow-up rates Wools-Kaloustian et al. (2009) Health impacts HIV infections averted Ng et al. (2011) HIV prevalence Bendavid and Bhattacharya (2009), Ng et al. (2011) Deaths due to HIV/AIDS Bendavid and Bhattacharya (2009), Floyd et al. (2010) TB treatment outcomes Yumo et al. (2011) Malaria cases Aregawi et al. (2011), Chizema-Kawesha et al. (2010) Anaemia cases Aregawi et al. (2011), Chizema-Kawesha et al. (2010) Malaria-attributed mortality Akachi and Atun (2011), Aregawi et al. (2011), Chizema-Kawesha et al. (2010) All-cause adult mortality Floyd et al. (2010) All-cause under-5 mortality Chizema-Kawesha et al. (2010) Lives saved Akachi and Atun (2011) Table 2 Summary and key findings of included studies "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Since 2002, development assistance for health has substantially increased, especially investments for HIV, tuberculosis (TB) and malaria control. We undertook a systematic review to assess and synthesize the existing evidence in the scientific literature on the health impacts of these investments. Methods and findings We systematically searched databases for peer-reviewed and grey literature, using tailored search strategies. We screened studies for study design and relevance, using predefined inclusion criteria, and selected those that enabled us to link health outcomes or impact to increased external funding. For all included studies, we recorded dataset and study characteristics, health outcomes and impacts. We analysed the data using a causal-chain framework to develop a narrative summary of the published evidence. Thirteen articles, representing 11 individual studies set in Africa and Asia reporting impacts on HIV, tuberculosis and malaria, met the inclusion criteria. Only two of these studies documented the entire causal-chain spanning from funding to programme scale-up, to outputs, outcomes and impacts. Nonetheless, overall we find a positive correlation between consecutive steps in the causal chain, suggesting that external funds for HIV, tuberculosis and malaria programmes contributed to improved health outcomes and impact. Conclusions Despite the large number of supported programmes worldwide and despite an abundance of published studies on HIV, TB and malaria control, we identified very few eligible studies that adequately demonstrated the full process by which external funding has been translated to health impact. Most of these studies did not move beyond demonstrating statistical association, as opposed to contribution or causation. We thus recommend that funding organizations and researchers increase the emphasis on ensuring data capture along the causal pathway to demonstrate effect and contribution of external financing. The findings of these comprehensive and rigorously conducted impact evaluations should also be made publicly accessible.
    Full-text · Article · Aug 2013 · Health Policy and Planning
  • Source
    • "These Global Health Initiatives developed guidance on ‘triangulating’ a combination of relevant existing data sources to understand national and local HIV epidemics and evaluate program outcomes and impact [10,11]. The triangulation approach was initially developed and tested in two countries with high HIV prevalence and relatively advanced program responses, Botswana [12–15] and Malawi [16–19]. Other examples of successful HIV data triangulation include Ukraine, Republic of Moldova and Estonia [20,21]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Swaziland's severe HIV epidemic inspired an early national response since the late 1980s, and regular reporting of program outcomes since the onset of a national antiretroviral treatment (ART) program in 2004. We assessed effectiveness outcomes and mortality trends in relation to ART, HIV testing and counseling (HTC), tuberculosis (TB) and prevention of mother to child transmission (PMTCT). Data triangulated include intervention coverage and outcomes according to program registries (2001-2010), hospital admissions and deaths disaggregated by age and sex (2001-2010) and population mortality estimates from the 1997 and 2007 censuses and the 2007 demographic and health survey. By 2010, ART reached 70% of the estimated number of people living with HIV/AIDS with CD4<350/mm(3), with progressively improving patient retention and survival. As of 2010, 88% of health facilities providing antenatal care offered comprehensive PMTCT services. The HTC program recorded a halving in the proportion of adults tested who were HIV-infected; similarly HIV infection rates among HIV-exposed babies halved from 2007 to 2010. Case fatality rates among hospital patients diagnosed with HIV/AIDS started to decrease from 2005-6 in adults and especially in children, contrasting with stable case fatality for other causes including TB. All-cause child in-patient case fatality rates started to decrease from 2005-6. TB case notifications as well as rates of HIV/TB co-infection among notified TB patients continued a steady increase through 2010, while coverage of HIV testing and CPT for co-infected patients increased to above 80%. Against a background of high, but stable HIV prevalence and decreasing HIV incidence, we documented early evidence of a mortality decline associated with the expanded national HIV response since 2004. Attribution of impact to specific interventions (versus natural epidemic dynamics) will require additional data from future household surveys, and improved routine (program, surveillance, and hospital) data at district level.
    Full-text · Article · Jul 2013 · PLoS ONE
  • Source
    • "Previous community-based studies with the same need to identify cluster residents while at a site distant from the household have used several other methods, including patient-carried referral cards and key-informant systems (9, 17). However, these can be resource intensive and subject to failure. "
    [Show abstract] [Hide abstract]
    ABSTRACT: A significant methodological challenge in implementing community-based cluster-randomized trials is how to accurately categorize cluster residency when data are collected at a site distant from households. This study set out to validate a map book system for use in urban slums with no municipal address systems, where classification has been shown to be inaccurate when address descriptions were used. Between April and July 2011, 28 noncontiguous clusters were demarcated in Blantyre, Malawi. In December 2011, antiretroviral therapy initiators were asked to identify themselves as cluster residents (yes/no and which cluster) by using map books. A random sample of antiretroviral therapy initiators was used to validate map book categorization against Global Positioning System coordinates taken from participants' households. Of the 202 antiretroviral therapy initiators, 48 (23.8%) were categorized with the map book system as in-cluster residents and 147 (72.8%) as out-of-cluster residents, and 7 (3.4%) were unsure. Agreement between map books and the Global Positioning System was 100% in the 20 adults selected for validation and was 95.0% (κ = 0.96, 95% confidence interval: 0.84, 1.00) in an additional 20 in-cluster residents (overall κ = 0.97, 95% confidence interval: 0.90, 1.00). With map books, cluster residents were classified rapidly and accurately. If validated elsewhere, this approach could be of widespread value in that it would enable accurate categorization without home visits.
    Full-text · Article · Apr 2013 · American journal of epidemiology
Show more