In 2003, Malawi, a resource-limited country in sub-Saharan Africa with a population of 13.6 million, had an adult HIV seroprevalence of 14% with approximately 1 million persons living with HIV/AIDS.1 In 2004, an estimated 170,000 persons were in need of antiretroviral treatment (ART). National scale-up of free ART, funded primarily by the Global Fund to Fight AIDS, Tuberculosis, and Malaria, began in June 2004.2 Influenced by the success of the tuberculosis (TB) directly observed therapy-short-course program, which is based on case finding, standardized treatment regimens, and simplified monitoring and evaluation of patient outcomes, and spurred by a dire shortage of human resources in the health sector, Malawi adopted a public health approach to rapidly scale up HIV treatment services.2-7
This approach, which is characterized by simplified clinical decision making, standardized ART regimens, nonphysician care, limited laboratory support, and centralized monitoring and evaluation, resulted in more than 141,000 persons being initiated on ART in the public sector by December 31, 2007.8 Those initiated on ART represent about two thirds of patients estimated to be eligible for ART (A. D. Harries, MD, written personal communication 2008). Although a few facilities in Malawi receive external technical and financial support from international organizations, such as Médecins Sans Frontieres,9 government health care workers are the sole providers in most ART clinics. As of December 31, 2007, 45 private sector sites had initiated ART, based on the national clinical guidelines and monitoring and evaluation tools, for more than 5400 patients.
The World Health Organization (WHO) has recommended a public health approach to scale up ART services in resource-limited countries. However, trends in national ART program outcomes have not been previously reported from sub-Saharan Africa. We examined trends in 4 measures of program quality, including treatment access, key early clinical outcomes, such as mortality, program retention, and 6- and 12-month survival probability during a 27-month period of the ART program in Malawi to determine whether the public health approach to rapid scale-up of ART has been successful.
The National ART Guidelines
In Malawi, the first-line ART regimen is a fixed-dose combination tablet of stavudine, lamivudine, and nevirapine that is taken twice daily.10 Adults and children older than 18 months are eligible for ART if they are HIV-positive and are in WHO clinical stage III or IV or stage I or II and have a CD4 T-lymphocyte count <250 cells per cubic millimeter for adults or below the threshold value for children.10,11 Children less than 18 months are eligible for ART based on HIV seropositivity, WHO stage, and clinical condition.
The National Monitoring and Evaluation System
In ART clinics, adults and children are followed monthly and vital data are recorded at each visit by use of standardized monitoring tools.4,6,7 The principal ART monitoring tools are a patient register and a patient master card. The registers and master cards are kept in the ART clinics, and the patient master cards are maintained in an ordered sequence inside polypropylene sheet protectors in hardcover ring binders. Patient information captured by these tools includes a unique, site-specific ART registration number, age, sex, occupation (a proxy for socioeconomic status), HIV test date and place, reason for starting ART, key opportunistic infections, and the drug regimen. When patients return for follow-up each month, the patient master cards are completed and standardized clinical outcomes, including alive on ART, dead, default (lost to follow-up for more than 90 days), transferred out, and stopped treatment, are recorded. Cumulative and quarterly cohort reports, generated from nonoverlapping intervals, are based on recorded data, which are reviewed for accuracy during quarterly supervision visits to all ART sites. The number of patients in each quarterly cohort includes double-counting of patients who transferred, estimated to be about 10% of all patients registered.9 Patients who initiated ART before October 1, 2004, were retrospectively included in the cumulative and initial 12-month longitudinal cohorts.
Patients newly initiated on ART during the quarter are followed over time, and survival outcomes are assessed at 6 and 12 months. For example, the first longitudinal cohort started ART between April 1 and June 30, 2004, and 6- and 12-month outcomes were censored on December 31, 2004, and June 30, 2005, respectively. Six- and 12-month patient outcome data were not censored during the first (October 1 to December 31, 2004) and second (January 1 to March 31, 2005) quarters because the national monitoring and evaluation system was not in place for patients initiating ART before October 1, 2004. Some ART clinics have active patient tracking systems for default patients but most do not because of limited resources. Each patient keeps an ART identity card, which contains his or her unique ART registration number and other related information, including ART drug regimen and initiation date.
In July 2006, the Ministry of Health initiated quarterly assessments of data quality. High performing ART facilities are given certificates of excellence. For the following analyses, patient retention in Malawi was defined as whether a patient remains alive on ART or has been officially transferred out after a specified follow-up interval, consistent with published standards.12
Mixed-effects logistic regression models with random effects on the intercept to account for clustering of patients within facilities were used to analyze both linear trends in aggregate facility-level quarterly cohort data, including baseline patient demographic and clinical characteristics and end-of-quarter outcomes and nonlinear trends in 6- and 12-month longitudinal cohort survival data. The nonlinear trend analysis was selected to estimate survival probability because the median time trend was nonlinear. Data were analyzed by use of PROC GLIMMIX (SAS Institute, Cary, NC).13 Aggregate response variables were specified as sample proportions, using events/trials syntax, where events are the number of patients experiencing or reporting an outcome and trials are the total number of patients.14 In the multivariate analysis, region (north, central, and south), patient enrollment target (low: 25 new patients on ART per month, medium: 50 new patients per month, and high: 150 new patients per month),8 facility type (governmental or nongovernmental), and quality of routine data collection (excellent or standard) were controlled for and adjusted odds ratios (AORs) provided.
Survival probabilities were calculated by use of the actuarial method.14 Patients who defaulted, transferred out, or stopped treatment during a 6- or 12-month period were considered at risk for half of the interval, 3 and 6 months, respectively. To model trends in these survival probabilities by use of the events/trials syntax, each survival probability was converted to an aggregate sample proportion by multiplying the survival probability by the total number assessed at the beginning of the interval minus half of those who defaulted, stopped treatment, or transferred out. This product gave the number of events or rather the number alive at the end of the interval, after adjusting for censoring. Trials were based on the above denominator. Reported P values were based on 2-sided tests.
The evaluation was exempted from ethical review by the National Health Sciences Research Committee in Malawi and the US Centers for Disease Control and Prevention because this study was designed for program monitoring and evaluation.
The total number of HIV-infected patients registered for ART at public clinics in Malawi by December 31, 2006, was 81,821 (Table 1). Of these patients, 57,356 (70%) were known to be on ART and 7364 (9%) had transferred to another site, thus nationally 79% of patients remained on treatment. Eleven percent of patients had died, 9% were lost to follow-up, and 1% stopped treatment.
Standardized, routine, national ART data collection began on October 1, 2004, at 23 public sector ART sites. During the period, October 1, 2004, and December 31, 2006, a total of 72,666 patients were registered at 104 ART sites and were included in the following analysis.
Access to Treatment
From quarter 1 (October 1 to December 31, 2004) through quarter 9 (October 1 to December 31, 2006), percentages of patients initiating ART who were women were 60%-63%, whereas percentages of children and subsistence farmers increased from 5.5% to 9.0% (P < 0.0001 for trend), and from 23% to 32% (P < 0.0001 for trend), respectively (Table 2). The percentage of patients who were started on ART because they were in WHO clinical stage IV decreased from 26% to 21% during this period (P < 0.001 for trends), whereas the trend in the percentage of patients who initiated ART who were in stage I or II with CD4 cell count <250 cells per microliter increased from quarter 1 to quarter 9 (P < 0.001 for trends). The estimated percentage of patients starting ART who had active TB disease or a prior history of TB within 2 years varied between 10% and 18%.
Early Clinical Outcomes
During October 2004 to December 2006, by the end of the quarter of treatment initiation (average duration of treatment = 6 weeks), the percentage of those known to have died was between 3% and 5% and the percentage of patients known to be alive and on ART was between 93% and 96%, whereas the percentage of patients transferred out increased from <1% to 2% (P < 0.0001 for trend) (Table 3). Of those patients alive at the end of the quarter, the percentage of patients who remained on the first-line regimen was about 99% and the percentage of patients whose end-of-quarter functional status was reported as ambulatory was from 96% to 99%, with 89%-94% of patients reportedly working. Patient adherence ≥95% was between 93% and 97% for the 47,405 (65%) who had their pill counts assessed.
Longitudinal cohort data at 6 and 12 months were available for 7 of 9 quarterly cohorts (Table 4). The trend in 6- and 12-month survival probability estimates followed U-shaped patterns. Six-month survival probability decreased from 87% in quarter 1 to 85% in quarter 4 and then increased to 88% in quarter 7 (P < 0.05 linear; P < 0.01 quadratic) (Fig. 1). Twelve-month survival probability decreased from 84% in quarter 1 to a nadir of 81% in quarter 5 and increased to 82% in quarter 7 (P < 0.01 linear; P < 0.01 quadratic).
Analysis of Facility Characteristics
In multivariate analysis, location in the southern region [AOR 0.74; confidence interval (CI): 0.60 to 0.91], low patient enrollment (AOR 0.72; CI: 0.54 to 0.94), and being a government site (AOR 0.65; CI: 0.52 to 0.82) were facility-level characteristics associated with lower 6-month survival probability, although not having received a certificate of excellence for routine data collection was of borderline significance (AOR 0.83; CI: 0.69 to 1.01), with central region, high patient enrollment, nongovernmental site, and receipt of a certificate of excellence for routine data collection as referent groups, respectively. Low ART patient enrollment and governmental status remained statistically significantly associated with lower 12-month survival probabilities, although not having received a certificate of excellence for routine data collection became significant (AOR 0.75; CI: 0.60 to 0.93), and region was no longer statistically significant.
ART is being rapidly scaled up in sub-Saharan Africa. Although the expansion of treatment services has been substantial, many millions of eligible patients remain untreated.15 To achieve universal ART coverage in Africa, new service delivery models are needed, especially those that facilitate increased access to treatment while ensuring quality. The public health approach to ART, promoted by the WHO and based on the principles of simplified clinical decision making, standardized ART regimens, nonphysician care, limited laboratory support, and centralized monitoring and evaluation, is regarded as a potentially important model; however, patient outcome data from national ART programs that use this approach are limited.3
Although ART patient outcomes from centers of excellence with private technical support, for example, from international organizations, have been highly encouraging, these sites are usually not nationally representative.9,16-23 Malawi, an impoverished country fighting a generalized HIV epidemic with limited health infrastructure and human resources, has used a public health approach to scale up ART.2 This approach, which is the result of difficult decisions requiring national consensus, is designed to balance goals of achieving maximal and equitable access to ART and ensuring high-quality care.2,3 For example, patients on ART do not receive laboratory monitoring, such as hemoglobin and hematocrit, electrolytes, renal and liver function, and CD4+ cell count testing. And clinical officers, not physicians, provide the majority of HIV care and treatment services in Malawi, a task-shifting strategy viewed as critical to ART scale-up.
This report describes 4 measures of success of national rapid scale-up of ART in Malawi: treatment access, key early clinical outcomes, such as mortality, program retention, and 6- and 12-month survival probability. Our findings demonstrate that, during the first 27 months of scale-up, Malawi maintained or improved the quality of national ART services, according to these parameters. Specifically, Malawi increased the proportion of those starting ART who were children and subsistence farmers, who represent the rural poor, from 5% to 9% and from 23% to 32%, respectively. Additionally, the proportion of patients starting ART without advanced disease progression by WHO clinical staging almost doubled, and the percentage of patients starting ART who had active TB disease or a prior history of TB within 2 years was between 10% and 18%.
Overall, 11% of patients were known to have died, whereas between 3% and 5% of patients died during the quarter of ART initiation. Retention on ART at 6 and 12 months was 80% and 75%, respectively, comparable to mean patient retention from select, subnational ART cohorts in sub-Saharan Africa reported in a recent systematic review.13 Documented mortality at 6 and 12 months was 11% and 13%, respectively, whereas the proportion lost to follow-up was 8% at 6 months and 11% at 12 months. Some patients who are lost to follow-up presumably represent undocumented deaths. A study of patients who were lost to follow-up in northern Malawi found that 50% were actually undocumented deaths.24 If this finding is generalizable to the national ART program, true mortality at 6 months would be about 15% at 6 months and 18.5% at 12 months.
Estimated ART patient survival probability was 85%-88% at 6 months and 81%-84% at 12 months. Because 88% of patients initiated on ART were in WHO clinical stage III or IV, and data from Malawi indicate that an estimated 75% of untreated patients in these advanced stages will die within 12 months, many thousands of lives are being saved by use of ART each year.25 Recent data from Malawi suggest that the introduction of ART reduced overall population-level mortality by 10% in 15- to 59-year-olds in this period.26
Although there seems to have been an initial decline in estimated 6- and 12-month survival during the first few quarters of national scale-up, likely coinciding with the early extension of ART services beyond established centers of excellence, recent trends in estimated survival have increased, suggesting that national stakeholders have managed to keep pace with this expansion and improve quality. Features of the public health approach that have facilitated rapid scale-up of ART services in Malawi include the following: (1) simplified national ART patient monitoring guidelines; (2) standardized ART regimens; (3) a standardized intensive clinical training program, based on the national ART guidelines, and preservice clinical mentoring at experienced ART sites; (4) a structured system of ART site accreditation, before initiating treatment at a facility; (5) supervision and data quality assurance during quarterly site visits to all ART sites by the HIV Unit of the Ministry of Health and its partners; and (6) a system of supplying drugs to facilities based on Ministry assessments of patient enrollment; and (7) no ART drug stockouts.27
This report profiles the first 27 months of ART scale-up in Malawi, within the “honeymoon” period for first-line antiretroviral drugs. Acquired and transmitted drug resistance remain important concerns, although a recent survey showed low or no prevalence of transmitted drug-resistant mutations.28 Medication adherence, initially presumed to be suboptimal in resource-limited settings, has been considerably better than expected.19,29 The national ART patient survival estimates and retention compare favorably with data from centers of excellence in resource-limited settings.9,12,21-23,30 Importantly, patients are progressively being initiated on ART with lower clinical staging, which is likely to further increase survival.18 However, selected groups of sites with lower patient survival probabilities, such as those identified in this study that have fewer ART patients or are located in the southern region, may need additional support, ideally provided on a regional basis by established centers of excellence.
Ongoing scale-up toward the goal of universal access to ART will be difficult. Many medium- and large-burden ART sites are currently at or near peak patient carrying capacity. Reliable monitoring and evaluation, essential to rapid scale-up, will become increasingly challenging and will require innovative approaches to ensure data quality.29,31-33 Further expansion of ART coverage will require diligence and innovation, including decentralization of services to primary health centers, further task shifting of responsibilities to lower cadres of care providers, and development of novel data collection tools.
Nevertheless, important non-ART interventions, such as co-trimoxazole prophylaxis and insecticide-treated bed nets, are being scaled up, which should have further beneficial effects on patient survival.34-37 Additionally, although 10,639 (23%) of the estimated 47,094 new TB patients who were HIV infected were initiated on ART during the evaluation period, improved linkage between TB and ART programs is essential.8 Importantly, substantial numbers of health care workers, critical to the national ART scale-up effort, are accessing ART, with good outcomes.38 Finally, although recent data suggest that national HIV prevalence may be declining, prevalence in rural areas, where the vast majority of Malawians live, seems to be increasing.39 It is important that access to HIV care and treatment services continues to expand and be maximally utilized to enhance national HIV prevention efforts, such as with “prevention for positives” interventions.40
Limitations of these analyses were that data were quarterly, aggregate, and facility level, not patient level, thus restricting the type of analysis. Also, some ART facilities had missing data, particularly early on in the scale-up, which may have biased the results. Additionally, the numbers of patients in quarterly cohorts included double-counting of transfers, estimated at 10% of total registered, which may have led to erroneous estimation of survival probability.
In summary, Malawi has vigorously responded to a nationwide HIV/AIDS epidemic by use of a public health approach to ART scale-up. By ensuring treatment access for vulnerable populations, maintaining low early mortality, and increasing survival, Malawi's national ART program has performed well and had substantial societal impact. This approach represents a relevant paradigm for scale-up of ART in other resource-limited settings which continue to face stark choices between slow scale-up of more robust, resource-intensive service delivery models, as global resources are mobilized, and rapid scale-up with minimal supporting services.
AIDS Commission. HIV and Syphilis Sero-Survey and National HIV Prevalence Estimates Report 2005
. The Republic of Malawi, Africa: Ministry of Health; December 2005.
2. Harries AD, Schouten EJ, Libamba E. Scaling up antiretroviral treatment
in resource poor settings. Lancet
3. Gilks CF, Crowley S, Ekpini R, et al. The WHO public-health approach to antiretroviral treatment
against HIV in resource-limited settings. Lancet
4. Libamba E, Makombe S, Harries AD, et al. Scaling up antiretroviral therapy in Africa: learning from tuberculosis control programmes-the case of Malawi. Int J Tuberc Lung Dis
5. Ministry of Health, Republic of Malawi. Human Resources in the Health Sector: Toward a Solution
. Lilongwe, Malawi: Ministry of Health; 2004.
6. Harries AD, Gomani P, Teck R, et al. Monitoring the response to antiretroviral therapy in resource-poor settings: the Malawi model. Trans R Soc Trop Med Hyg
7. Harries AD, Libamba E, Schouten EJ, et al. Expanding antiretroviral therapy in Malawi: drawing on the country's experience with tuberculosis. BMJ
8. Ministry of Health, Republic of Malawi. ART in the Public Sector in Malawi: Results up to 31st December 2007
. Lilongwe, Malawi: Ministry of Health; 2008.
9. Ferradini L, Jeannin A, Pinoges L, et al. Scaling up of highly active antiretroviral therapy in a rural district in Malawi: an effectiveness assessment. Lancet
10. Ministry of Health, Republic of Malawi. Treatment of AIDS. Guidelines for the Use of Antiretroviral Therapy in Malawi
. 2nd ed. Lilongwe, Malawi: Ministry of Health; 2006.
11. World Health Organization. Antiretroviral Therapyof HIV Infection in Infants and Children in Resource-Limited Settings: Towards Universal Access
. Geneva, Switzerland: World Health Organization; 2006.
12. Rosen S, Fox MP, Gill CJ. Patient retention in antiretroviral therapy programs in sub-Saharan Africa
: a systematic review. PLoS Med
. 2007;4(10):e298. doi:10.1371/journal.pmed.0040298.
13. The Glimmix Procedure, November 2005. Available at: http://support.sas.com/rnd/app/papers/glimmix.pdf
. Accessed April 24, 2007.
14. Chiang CL. Medical follow-up studies. In: The Life Table and Its Applications
. Malabar, FL: Robert E. Krieger; 1984:221-244.
15. Report on the Global AIDS Epidemic 2006
. Geneva, Switzerland: The Joint United Nations Programme on HIV/AIDS; 2006.
16. Wools-Kaloustian K, Kimaiyo S, Diero L, et al. Viability and effectiveness of large-scale HIV treatment initiatives in sub-Saharan Africa
: experience from western Kenya. AIDS
17. Stringer JSA, Zulu I, Levy J, et al. Rapid scale-up of antiretroviral therapy at primary care sites in Zambia. JAMA
18. ART-LINC Collaboration and ART Cohort Collaboration Groups. Mortality of HIV-infected patients in the first year of antiretroviral therapy: comparison between low-income and high-income countries. Lancet
19. Severe P, Leger P, Macarthur C, et al. Antiretroviral therapy in a thousand patients with AIDS in Haiti. N Engl J Med
20. Madec Y, Laureillard D, Pinoges L, et al. Response to highly active antiretroviral therapy among severely immuno-compromised HIV-infected patients in Cambodia. AIDS
21. Coetzee D, Hildebrand K, Boulle A, et al. Outcomes
after two year of providing antiretroviral treatment
in Khayelitsha, South Africa. AIDS
22. O'Brien DP, Saufageot D, Zachariah R, et al. In resource-limited settings good early outcomes
can be achieved in children using adult fixed-dose combination antiretroviral therapy. AIDS
23. Nyandiko WM, Ayaya S, Nabakwe E, et al. Outcomes
of HIV-infected orphaned and non-orphaned children on antiretroviral therapy in western Kenya. J Acquir Immune Defic Syndr
24. Yu JK, Chen SC, Wang KY, et al. True outcomes
for patients on antiretroviral treatment
who are lost to follow-up in Malawi. Bull World Health Organ
25. van Oosterhout J, Laufer M, Graham S, et al. A community-based study of the incidence of trimethoprim-sulfamethoxazole-preventable infections in Malawian adults living with HIV. J Acquir Immune Defic Syndr
26. Jahn A, Floyd S, Crampin AC, et al. Population-level effect of HIV on adult mortality and early evidence of reversal after introduction of antiretroviral therapy in Malawi. Lancet
27. Harries AD, Schouten EJ, Makombe SD, et al. Ensuring uninterrupted supplies of antiretroviral drugs in resource-poor settings: an example from Malawi. Bull World Health Organ
28. Kamoto K, Aberle-Grasse J, et al. Malawi HIV Drug Resistance Threshold Baseline Survey Results. Antiviral Therapy
. 2008;Suppl 2:69-75.
29. Montaner JSG, Hogg R, Wood E, et al. The case for expanding access to highly active antiretroviral therapy to curb the growth of the HIV epidemic. Lancet
30. Lawn SD, Myer L, Orrell C, et al. Early mortality among adults accessing a community-based antiretroviral service in South Africa: implications for programme design. AIDS
31. Lowrance D, Filler S, Makombe S, et al. Assessment of a national
monitoring and evaluation for rapid expansion of antiretroviral treatment
in Malawi. Trop Med Int Health
32. Makombe SD, Hochgesang M, Jahn A, et al. Assessing the quality of data aggregated by antiretroviral treatment
clinics in Malawi. Bull World Health Organ
33. Boerma JT, Stanecki KA, Newell M-L, et al. Monitoring of scale-up of antiretroviral therapy programmes: methods to estimate coverage. Bull World Health Organ
34. Ministry of Health, Republic of Malawi. Cotrimoxazole Preventive Therapy for HIV-Positive Persons in Malawi
. Lilongwe, Malawi: Ministry of Health; 2006.
35. Mermin J, Ekwaru JP, Liechty CA, et al. Effect of co-trimoxazole prophylaxis, antiretroviral therapy, and insecticide-treated bed nets on the frequency of malaria in HIV-1-infected adults in Uganda: a prospective cohort study. Lancet
36. Lowrance D, Makombe S, Harries AD, et al. Lower early mortality rates among patients receiving antiretroviral treatment
at clinics providing cotrimoxazole prophylaxis in Malawi. J Acquir Immune Defic Syndr
37. Hawley WA, Phillips-Howard PA, ter Kuile FO, et al. Community-wide effects of permethrin-treated bednets on child mortality and malaria morbidity in western Kenya. Am J Trop Med Hyg
38. Makombe S, Jahn A, Tweya H, et al. A national
survey of the impact of national
scale-up of antiretroviral therapy on health-care workers in Malawi: effects on human resources and survival
. Bull World Health Organ
39. Ministry of Health, Republic of Malawi. Triangulation Workshop: Findings, Conclusions, and Recommendations
. Lilongwe, Malawi: Ministry of Health; 2006.
40. Bunnell R, Mermin J, De Cock KM. HIV prevention for a threatened continent: implementing positive prevention in Africa. JAMA