University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Home > Global Health Literature Digest > Population-Level
Population-level reduction in adult mortality after extension of free antiretroviral therapy provision into rural areas in northern Malawi
Global Health Sciences Literature Digest
Published March 14, 2011
Journal Article

Floyd S, Molesworth A, Dube A, Banda E, Jahn A, Mwafulirwa C, Ngwira B, Branson K, Crampin AC, Zaba B, Glynn JR, French N. Population-level reduction in adult mortality after extension of free antiretroviral therapy provision into rural areas in northern Malawi. PLoS One. 2010; 5:Number, e13499.

In Context

The authors state that data from northern Malawi indicate that 63% of deaths were attributed to HIV/AIDS during the period 2002 to 2005, prior to initiation of this study. The findings in this paper are in line with other studies of the population-level impact of antiretroviral therapy (ART) on AIDS mortality: a study in Ethiopia found a 50% reduction in AIDS mortality during the first two years of a free ART program.(1) A study in a southern district of Malawi found a 37% reduction in deaths with treatment coverage of 80%.(2) Data have also been published from survival and retention in care from national-level ART programs in southern and East Africa, including Botswana(3) Zambia,(4) and Rwanda,(5) which showed ART retention rates of 76 to 86%.


To evaluate changes in all-cause and AIDS mortality during the first three years of ART roll-out compared to the pre-ART time period.


The rural Karonga district in northern Malawi; rural residents of southern Karonga district, aged 15 to 59 years.


The article presents data from an ongoing demographic surveillance site (DSS) and a cross-sectional population-based HIV sero-survey. During 2002 to 2004, a DSS was established in the southern part of the district. Information on deaths was obtained monthly, and in-and out-migration data were updated yearly. Verbal autopsies were performed 6 to 8 weeks after death using a standardized World Health Organization (WHO) tool. Data from 2002-2008 were analyzed, and focused on 15-59 year-olds. Three time periods were specified: pre-ART (August 2002 to September 2006); ART period 1 (July 2005 to September 2006); ART period 2 (October 2006 to September 2008). In ART period 1, ART was only available in a clinic 70km from the DSS area; in period 2, ART was available from a clinic within the DSS area. Approximately half of the population in this area lived near a tarmac road ('rural roadside' population), with the remainder living more remotely ('rural remote' population). Between September 2007 and October 2008, an HIV serosurvey was conducted in the DSS area, with testing offered to all those15 years of age or older. Questions on whether persons had ever taken or were currently on ART were asked. Estimation of ART need was based on a Weibull regression model fitted to age-specific survival patterns of 196 HIV-positive persons from a retrospective cohort study conducted from 1998 to 2000. These data were then applied to models of individuals who tested HIV positive during the 2007-2008 sero-survey to calculate their predicted probability of death within three years; the assumption was made that during the three years prior to death, persons are in need of ART. Retention in care was estimated by analyzing data of all newly registered HIV positive patients at the ART clinic in the DSS area from January 2008.


During the DSS study period (2002 to 2008), there were 655 deaths in 77,179 person-years (py) of follow-up or 8.5 per1000 py; cause of death was not specified for 40 persons (6%). All-cause mortality rates dropped during the three time periods, from 10.2 (pre-ART), to 8.5 (period 1) to 6.9 (period 2) per 1000 py (p<0.001). Mortality reduction occurred from age 25 years onward. The probability of surviving to age 60 was 57% in period 2, compared to 46% in the pre-ART period. Mortality reduction was larger in the 'rural roadside' area (13.2 to 7.8 per 1000 py) compared to the 'remote' areas (6.9 to 6.0 per 1000 py). AIDS also mortality fell dramatically during ART roll-out (6.4, 4.6 and 2.7 per 1000 py in successive time periods), while non-AIDS mortality remained fairly constant. Of the 14,951 persons in the area at the time of the 2007-2008 sero-survey, 67% (10,059) agreed to respond to survey questions and have blood testing; 8% (805) were found to be HIV positive, and of those, 24% (190 of 805) reported having taken ART. The Weibull regression model predicted that 31% were in need of ART in 2008; therefore, uptake among those in need was 77%. Uptake was higher among those in the 'rural roadside' than the 'rural remote' areas, and higher among women than men. An estimated additional 165 persons (355 total) were estimated to be taking ART, based on survey responses From January 2008 to June 2009, 194 persons started ART at the clinic within the DSS area; retention at 12 months was 91%.


In this rural area of Malawi, AIDS mortality fell by 57% during the second and third years of ART rollout. Uptake of ART was fairly high (estimated at 77%). The authors conclude that this occurred with the adoption of a public health approach to ART delivery, allowing a rapid increase in treatment coverage and accessibility of care.(6)

Study Quality

This was a good quality study, based on available demographic and sero-surveillance data. Conclusions were drawn from population-based surveys, rather than cohort data. Due to the limitations of such data, ART uptake was estimated and number of persons in need was modeled.

Programmatic Implications

This study provides additional support for the feasibility of rolling out ART in rural areas. AIDS mortality dropped, ART uptake was fairly high, and 12-month retention on treatment was also good. The fact that mortality-reduction was greater when a clinic opened within the study area, and was also higher among those in close proximity to a road (and therefore transportation to services), indicates that accessibility remains very important to a program's success. Although the authors indicate that this study supports the feasibility of small clinics providing HIV treatment in rural areas, the means of service delivery and/or support were not described in this article. Although these reductions in mortality are promising, it is not known what would happen as patients develop treatment failure, and require more sophisticated clinical management (e.g. with laboratory monitoring, not easily accessible in such rural areas).


  1. Reniers G, Araya T, Davey G, Nagelkerke N, Berhane Y, et al. (2009) Steep declines in population-level AIDS mortality following the introduction of antiretroviral therapy in Addis Ababa, Ethiopia. AIDS 23: 511-518.
  2. Mwagomba B, Zachariah R, Massaquoi M, Misindi D, Manzi M, et al. (2010) Mortality reduction associated with HIV/AIDS care and antiretroviral treatment in rural Malawi: evidence from registers, coffin sales and funerals. PLoS One 5: e10452.
  3. Bussmann H, Wester CW, Ndwapi N, Grundmann N, Gaolathe T, et al. (2008) Five-year outcomes of initial patients treated in Botswana's National Antiretroviral Treatment Program. Aids 22: 2303-2311.
  4. Stringer JS, Zulu I, Levy J, Stringer EM, Mwango A, et al. (2006) Rapid scale-up of antiretroviral therapy at primary care sites in Zambia: feasibility and early outcomes. JAMA 296: 782-793.
  5. Lowrance DW, Ndamage F, Kayirangwa E, Ndagije F, Lo W, et al. (2009) Adult clinical and immunologic outcomes of the national antiretroviral treatment program in Rwanda during 2004-2005. J Acquir Immune Defic Syndr 52: 49-55.
  6. Lowrance DW, Makombe S, Harries AD, Shiraishi RW, Hochgesang M, et al. (2008) A public health approach to rapid scale-up of antiretroviral treatment in Malawi during 2004-2006. J Acquir Immune Defic Syndr 49: 287-293.