Background: Monitoring and evaluation of antiretroviral treatment (ART) scale-up has been challenging in resource-limited settings. We describe an innovative cell-phone-based and internet-based reporting system (TRACnet) utilized in Rwanda.
Methods: From January 2004 to June 30, 2010, all health facilities with ART services submitted standardized monthly aggregate reports of key indicators. National cohort data were analyzed to examine trends in characteristics of patients initiating ART and cumulative cohort outcomes. Estimates of HIV-infected patients eligible for ART were obtained from Joint United Nations Program on HIV/AIDS (Estimation and Projection Package-Spectrum, 2010).
Results: By June 30, 2010, 295 (65%) of 451 health centers, District and referral hospitals provided ART services; of these, 255 (86%) were located outside Kigali, the capital. Cell phone–based and internet-based reporting was used by 253 (86%) and 42 (14%), respectively. As of June 30, 2010, 83,041 patients were alive on ART, 6171 (6%) had died, and 9621 (10%) were lost-to-follow-up. Of those alive on ART, 7111 (8.6%) were children, 50,971 (61.4%) were female, and 1823 (2.2%) were on a second-line regimen. The proportion of all patients initiating ART at World Health Organization clinical stages 3 and 4 declined from 65% in 2005 to 27% in 2010. National ART coverage of eligible patients increased from 13% in 2005 to 79% in 2010.
Conclusions: Rwanda has successfully expanded ART access and achieved high national ART coverage among eligible patients. TRACnet captured essential data about the ART program during rapid scale-up. Cell phone-based and internet-based reporting may be useful for monitoring and evaluation of similar public health initiatives in other resource-limited settings.
*TRAC Plus-Center for Treatment and Research on HIV/AIDS, Malaria, Tuberculosis, and Other Epidemic Diseases
†Division of Global HIV/AIDS, Center for Global Health, U.S. Centers for Disease Control and Prevention
‡Ministry of Health, Kigali, Rwanda
§Institute of Human Virology and Division of Infectious Diseases, University of Maryland School of Medicine, Baltimore, MD
Correspondence to: David J. Riedel, MD, Institute of Human Virology and Division of Infectious Diseases, University of Maryland School of Medicine, 725 W. Lombard Street, N552, Baltimore, MD 21201 (e-mail: firstname.lastname@example.org).
Supported by funds from the President's Emergency Plan for AIDS Relief through Cooperative Agreement Numbers PS001847 and PS002048 from the Centers for Disease Control and Prevention, awarded to Voxiva and TRAC Plus, respectively; and through Grant Number U51HA00002521 from the US Health Resources Services Administration under a subcontract awarded to AIDSRelief/University of Maryland. TRACnet was developed by TRAC Plus with Voxiva, Inc. Support for TRACnet was provided from the Centers for Disease Control and Prevention and US Health Resources Services Administration through funds from the President's Emergency Plan for AIDS Relief.
Presented in part at the 5th International Conference for Research and Exchange on HIV/AIDS in Kigali, Rwanda in October 2009.
Author contributions—study conception: S.N. and D.J.R. Data acquisition: R.H., S.C., J.P.N., R.M. Data analysis and interpretation: all authors. Drafting the article: S.N., D.J.R. Revisions: all authors. Approval of the final article: all authors.
The contents of the publication are solely the responsibility of the authors and do not necessarily represent the official views of Centers for Disease Control and Prevention or HRSA.
The authors have no conflicts of interest to disclose.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.jaids.com).
Received April 22, 2011
Accepted October 14, 2011
In recent years, unprecedented progress has been made in expanding access to antiretroviral therapy (ART) in low-income and middle-income countries and in reducing morbidity and mortality among people living with HIV/AIDS.1–8 When the “3 by 5” initiative was launched by the World Health Organization (WHO) in 2003, it was estimated that only 400,000 people in low-income and middle-income countries were receiving ART.9 Since then, increasing political commitment, resource mobilization, and efforts by multiple stakeholders have resulted in a massive increase in the number of people receiving ART in these countries. However, by the end of 2009, only 5.25 million people were receiving ART,10 and global coverage of ART remained limited, reaching only 36% of the estimated 15 million people eligible at the end of 2009.11 During this period of rapid ART scale-up, effective monitoring and evaluation of national programs has been a particular challenge.
Rwanda, a country with a population of about 9.3 million,12 has an HIV prevalence of approximately 3% among adults 15–49 with an estimated 150,000 to 200,000 HIV-infected persons in 2010.13 Universal access, that is, provision of ART to all patients who meet the eligibility criteria, continues to be a central priority of the Government of Rwanda (GoR) and the Ministry of Health. With funding from the US President's Emergency Plan for AIDS Relief, the Global Fund to fight AIDS, Tuberculosis, and Malaria, and other sources, the GoR has launched a massive effort to control the HIV epidemic. The GoR response to HIV/AIDS has emphasized innovative technologies and programs to improve the care and treatment of people living with HIV/AIDS and the general population14; the Center for Treatment and Research on AIDS, Malaria, Tuberculosis, and Other Epidemic Diseases (TRAC Plus) has been charged with leading the national response.
TRACnet, a national electronic cell phone-based and web-based monitoring and evaluation system for both pre-ART HIV care and ART services, has facilitated effective planning and expansion of treatment access. The TRACnet system was developed to overcome the common limitations of paper-based reporting systems although supporting roll-out and monitoring of the life-saving intervention—ART. Here, we present the first report using the novel TRACnet reporting system on the scale-up of HIV care and treatment (HCT) services in Rwanda for the period 2004–2010.
National ART Initiation Criteria and Regimens
The first national ART guidelines for clinical providers were distributed in 2004–2005. At that time, adults and children were eligible for ART if they were as follows: (1) HIV infected and classified according to WHO clinical stage 4, irrespective of CD4 cell count; (2) WHO clinical stage 3 with a CD4 cell count <350 cells per cubic millimeter; or (3) WHO clinical stage 1 or 2 with a CD4 cell count <200 cells per cubic millimeter. Other suggested criteria for initiation of ART included participation in adherence education and identification of a treatment supporter. The recommended first-line regimen in 2004 was stavudine or zidovudine with lamivudine (3TC) and either nevirapine or efavirenz.
During 2007–2008, eligibility criteria for ART initiation were modified. The threshold CD4 count for initiation of ART was increased from 200 to 350 cells per cubic millimeter, regardless of the WHO stage (1, 2, or 3); WHO stage 4 remained an indication for ART initiation regardless of CD4 count. Additionally, zidovudine with 3TC became the preferred first-line nucleoside reverse transcriptase inhibitors backbone combined with nevirapine or efavirenz. In 2009, national guidelines were again revised to recommend tenofovir as the preferred first-line nucleotide reverse transcriptase inhibitor in combination with 3TC and either nevirapine or efavirenz. A second-line ART regimen of zidovudine, 3TC, and lopinavir/ritonavir is currently used. Since 2004, ART has been provided free of charge to all patients eligible to receive it by these criteria.
Routine Clinical and Laboratory Monitoring
At the time of ART initiation, routine baseline clinical and laboratory assessment of HIV-infected persons in Rwanda include weight, CD4 cell count, and WHO clinical staging. ART patients receive monthly clinical follow-up for the first 6 months, followed by routine clinical and CD4 evaluations at 6-month intervals. Pharmacy refills, including weight measurements, continue on a monthly basis. Before 2009, cotrimoxazole (CTX) prophylaxis was routinely provided to any patient with a CD4 cell count <350 cells per cubic millimeter; a diagnosis of active tuberculosis disease; or WHO clinical stage 4 criteria. During 2009, all patients with documented HIV infection were considered eligible for CTX regardless of CD4 cell count. A patient was defined as lost-to-follow-up (LTFU) if his last contact with the health facility for any reason (clinical follow-up, pharmacy refill, or laboratory monitoring) was more than 3 months for ART patients or more than 6 months for pre-ART patients, if not known to have died or transferred out. Deaths were recorded by each facility as they became known during hospitalization or from home visits after being identified as LTFU.
Routine Data Collection Tools and TRACnet Reporting
Standardized pre-ART and ART registers, medical records (the “dossier vert”), and pharmacy files are used by all sites providing HCT. Each patient enrolled in HCT services is given a unique alphanumeric TRACnet identification code. On a monthly basis, every HCT site aggregates data for select indicators and submits a report into the TRACnet system. Indicators include number of patients on ART by gender, age category (adult vs. pediatric), and regimen (first vs. second line); number of patients on CTX prophylaxis; patients' WHO stage at ART initiation; death; and LTFU. Reporting is done either by cellular phone via interactive voice response or internet depending on availability. TRACnet technology was developed by Voxiva, in collaboration with the Ministry of Health and the US Centers for Disease Control and Prevention, with President's Emergency Plan for AIDS Relief funding.
Data reporting involves dialing a toll-free number or logging onto a bilingual (English and French) website. To date, 1057 TRACnet users in 295 health facilities have been trained to submit TRACnet reports. From 2004 to present, cross-sectional baseline data for all patients enrolling in HCT services were routinely reported in the TRACnet system. Data were then compiled and entered in a central automated database managed by TRAC Plus in Kigali. For the purpose of this report, reported data from January 2004 through June 30, 2010, were used. Baseline sociodemographic and clinical data were not complete for the period of January to October 2004. All data entry via cell phone or Internet is subject to data validation rules, built-in consistency checks. For example, if the number of new ART patients is 5, then the total number of WHO stages should also be 5; if these numbers do not match, a report cannot be submitted. From 2004 to 2007, data quality assurance measures were performed in an ad hoc manner and then routinely from 2007 to present. Each monthly report was reviewed at the national level for completeness and consistency.
After data entry and cleaning, statistical analysis was performed using STATA 10.0 IC software (StataCorp, College Station, TX). Analyses included frequencies and trend analyses of select indicators using χ2 tests. The cumulative cohort of patients currently on ART for each year between 2004 and 2010 was constructed by summing the totals of patients currently on ART with those reported to have died or LTFU during the reporting period. As reported data on deaths and LTFU were based on cumulative cohort, not longitudinal patient-level outcomes, only absolute numbers of deaths and LTFU are reported. Patients who transferred out during any monthly reporting period are only reported as “currently on ART” at a single health facility at any given time, thus double counting was usually avoided. If a patient unofficially transferred to a second site, both sites 1 and 2 could count the patient alive on ART for up to 90 days maximum when site 1 would then consider them LTFU. In this way, atmost 2–3 months of double attribution could occur before the system autocorrects by subtracting the transferred subject as LTFU after 90 days at site 1. They would thus be removed them from the cumulative cohort of persons considered to remain alive on ART at any given point in time. The absolute level of double attribution would be quite small at any given point in time because total LTFU rate is relatively low.
This study analyzed routine data collected by the national HIV care program. TRAC Plus considered this activity routine program evaluation; the Rwandan National Ethics Committee concurred and provided an exemption from formal ethical review. TRACnet data collection was approved by the Centers for Disease Control and Prevention as program evaluation and not research.
Health Facilities Providing ART
The number of health facilities offering ART in Rwanda increased from 24 in 2004 to 295 by the end of June 2010 (Fig. 1; see Supplemental Digital Content 1, http://links.lww.com/QAI/A236). All provinces, including Kigali City, had a marked increase in the number of ART sites (Fig. 1). Of health facilities providing ART in June 2010, 250 (84.7%) were primary health centers (includes 11 private clinics and 11 prisons), 40 (13.6%) were district hospitals, and 5 (1.7%) were referral hospitals. Expansion was greatest at the health center level with an increase in sites offering ART of more than 5-fold from 2005 (49–250).
ART Patient Enrollment
For adult patients, the average rate of ART initiation per month increased from 160 patients per month in 2004 to 1354 in 2010 (Fig. 2). For pediatric patients, the average monthly rate of ART initiation increased from 12 patients per month in 2004 to 145 in 2010. The overall rate of ART initiation among all patients in 2004 was 98 and 74 per month in and outside Kigali, respectively. By 2010, the overall rates of ART initiation had increased to 331 and 1168 per month inside and outside Kigali, respectively.
Cumulative data for adult (age ≥ 15 years) patients alive and on ART by year are displayed in Table 1. Each year, 88%–91% of patients receiving ART were adults. The majority of patients were female; from 2004 to 2010, 62%–65% of patients were female. Characteristics of children (age < 15 years) receiving ART are also displayed in Table 1. Among children, the percentage of males and females initiating ART each year was nearly equal (∼50%). As of June 30, 2010, 97.8% of patients were taking first-line regimens, and only 2.2% of patients were reported taking second-line regimens. Of the cumulative cohort of 98,833 patients who had ever started ART from January 1, 2005, to June 30, 2010, 83,041 (84%) were alive on ART; 9621 (10%) were LTFU, and 6171 (6%) were reported to have died.
WHO Staging at ART Initiation
The proportion of patients initiating ART that were asymptomatic (WHO clinical stage 1) increased from 10% in 2005 to 46% in 2010 (P for trend < 0.0001). Conversely, the proportion of new ART patients with stage 3/4 disease decreased from 64% in 2005 to 24% in 2010 (P for trend <0.0001; Fig. 3A). Similar trends were noted for children (Fig. 3B), with an increasing percentage of patients initiating ART at WHO stage 1 (P < 0.0001 for the trend); however, a similar percentage of children were still initiating ART at WHO stage 4 in 2010 as in 2005.
Using Estimation and Projection Package/Spectrum modeling, the most recent national estimate of the total number of people infected with HIV in Rwanda in 2010 ranged from 149,490 to 198,620 (median: 174,090).13 Based on the actual number of patients on ART in Rwanda and the projected estimates of patients eligible for treatment, the percentage of patients on treatment (national coverage) increased substantially each year from 2005 to 2010 (Fig. 4). At the end of 2004 (based on the CD4 <200 cells/mm3 guideline), only 13% of eligible patients were receiving treatment (using the upper bound estimate; 17% using the median). However, by the end of 2007, more than 80% of patients eligible for ART (based on CD4 < 200 cells/mm3) were receiving it (using the upper bound estimate; 95% using the median). When the national eligibility criteria were changed in 2007 from CD4 <200 cells per cubic millimeter to CD4 <350 cells per cubic millimeter, the percentage was reduced; however, by mid-2010, an estimated 79% of eligible patients were receiving treatment (using the upper bound estimate; 90% using the median).
Data Completeness and Monitoring
By June 30, 2010, 42 (14%) health facilities providing ART services were routinely reporting monthly indicator data to TRAC Plus by Internet, whereas 253 (86%) were reporting by cell phone. During quarter 2 (April to June) 2010, 99% of health facilities reporting in TRACnet submitted reports on time (within 30 days). An external, national data audit performed in 2009 indicated that TRACnet data were consistent with primary data sources.
Rwanda is the only country in sub-Saharan Africa to have employed a cell phone–based and web-based monitoring and evaluation system of a national ART program. This report demonstrates that Rwanda has been successful at increasing national ART coverage among eligible patients during a period of rapid scale-up. Furthermore, TRACnet has provided essential information to program planners to facilitate this broad expansion and decentralization of ART services. Similar to other sub-Saharan African ART programs, a majority of patients were adults, female, and on first-line regimens. Encouragingly, from 2005 to 2010, more patients initiated ART at earlier WHO clinical stages, which is consistent with national and international guidelines. National ART coverage is substantial and, according to Joint United Nations Program on HIV/AIDS,10 one of the highest in the region, despite the higher CD4 eligibility threshold compared with those used in other countries. Last, the program results reported through TRACnet showed relatively low attrition rates (death and LTFU).
One of the primary objectives of the international response to HIV/AIDS is to ensure universal ART access to all those who meet country-specific eligibility criteria for treatment.15 The GoR, with financial assistance from various global initiatives, has substantially increased the number of patients on ART. In 2005, with eligibility criteria centered on a CD4 threshold of less than 200 cells per cubic millimeter or WHO stage 4 criteria, less than 15% of eligible patients were receiving treatment. By early 2007, more than 80% of eligible patients were receiving ART using the CD4 threshold of less than 200 cells per cubic millimeter. Rwanda raised the CD4 threshold to 350 cells per cubic millimeter in late 2007, and thus increased the estimated number of ART-eligible patients from 51,440 in 2007 to 87,410 in 2008, an almost 70% increase based primarily on the new CD4 threshold (Fig. 4). Despite this major policy change, national ART coverage continued to expand, and the program has reached an estimated 79% of those eligible by mid-2010. Overall, in sub-Saharan Africa, only 37% of patients eligible for ART are receiving it.10 Rwanda ranks as one of only a few countries in sub-Saharan Africa to achieve such high ART coverage rates.11
When Rwanda began offering widespread ART in 2004, it was similar to other programs described in sub-Saharan Africa,2–6 where the vast majority of patients were WHO stage 3 and 4 at the time of ART initiation. For example, in 2005, 65% of patients in Rwanda were WHO stage 3 or 4 at ART initiation; this percentage was similar in 2004 to that in Malawi (82%),2 Zambia (73%),3 and Ivory Coast (81%).6 Because earlier treatment usually yields improved treatment outcomes,16–18 one marker of overall program quality is that more people are being tested, entered into the HCT services earlier, and started on ART before reaching WHO stage 3 and 4 disease. In Rwanda, we observed a very significant shift from 2005, when approximately two-thirds of patients initiated ART at WHO stages 3 and 4, to 2010, when nearly three-fourths of patients were initiating ART at WHO stages 1 and 2 (P for trend < 0.0001). This shift marks significant progress being made at the country level. A similar but less dramatic shift was recently reported from South Africa, where 50% of patients in 2002/2003 were WHO stage 4 when initiating therapy compared with 27.7% stage 4 by 2007.19
The WHO stage and/or CD4 count at which patients initiate ART is an important predictor of outcomes. Data from varied cohorts have shown that survival and immunologic outcomes are closely correlated to WHO stage and/or CD4 count at treatment initiation.16,20–23 Excellent 6-month and 12-month retention and immunologic outcomes during the first 2 years of rapid scale-up have been previously reported in Rwanda.7 With the shift from ART initiation at WHO stages 3 and 4 to initiation at WHO stages 1 and 2, Rwanda will expect to see continued improvement in programmatic and immunologic outcomes. As reported through TRACnet, the cumulative cohort retention, mortality, and LTFU, less robust but important indicators of program quality, compare favorably with those of other large national ART cohorts reported in the region.
TRACnet technology facilitates electronic transmission of information from clinical providers using mobile phones and internet connections. This innovation has facilitated standardized data collection and analysis and has improved data management systems, which became quick and easy, even for the most remote health facilities. Improving information exchange between health facilities and the national level has reinforced the accountability of HIV services and health facilities. The system has helped to overcome some and has the potential to address other key challenges in health information systems for resource-limited settings, including late reports, incomplete data, and lack of feedback from the central level to peripheral levels. The TRACnet system has enhanced the quantity of information obtained from health facilities and has facilitated its use to inform the national program. For instance, TRACnet data are utilized by the national program in ART quantification and procurement, making the system quite valuable for HIV decision makers.
One limitation of TRACnet analyses is that reported data are aggregated at the health facility level. Thus, it is not possible to describe specific patient-level demographic and clinical characteristics, outcomes, or events. The accuracy of TRACnet data has not been well assessed to date, though an external national data audit performed in 2009 indicated that TRACnet data were consistent with primary data sources, Additionally, comparisons with other national HIV/AIDS information systems, such as performance-based financing, suggest consistently lower reporting of key indicators through TRACnet, such as the number of patients on ART (unpublished data, TRAC Plus). Such a bias would tend to produce an underestimate of national ART coverage rates by TRACnet; with performance-based financing, the built-in financial incentive to both health care personnel and their facilities tends to lead toward higher reporting levels. However, we feel that these aggregate data provide a realistic overview of the national program. Additionally, a second national ART program evaluation is now under way, which will allow for comparison with routinely reported TRACnet data. Evaluation of program quality via TRACnet data is also limited by a lack of longitudinal patient follow-up data, such as CD4 cell count, because this type of reporting is logistically complex and time consuming. Cumulative levels of ART attrition were relatively low in this analysis, but these figures may be subject to underreporting, particularly the mortality figures, which may not reflect nondocumented deaths.
In summary, TRACnet data indicate that the Rwanda national ART program has successfully scaled-up ART services and achieved high national ART coverage. Cell phone–based and internet-based reporting of key HCT indicators have facilitated rapid reporting of national ART scale-up in Rwanda and may be useful in other resource-limited settings.
2. Ferradini L, Jeannin A, Pinoges L, et al.. Scaling up of highly active antiretroviral therapy in a rural district of Malawi: an effectiveness assessment. Lancet. 2006;367:1335–1342.
3. Stringer JS, Zulu I, Levy J, et al.. Rapid scale-up of antiretroviral therapy at primary care sites in Zambia: feasibility and early outcomes. JAMA. 2006;296:782–793.
4. Severe P, Leger P, Charles M, et al.. Antiretroviral therapy in a thousand patients with AIDS in Haiti. N Engl J Med. 2005;353:2325–2334.
5. Lowrance DW, Makombe S, Harries AD, et al.. A public health approach to rapid scale-up of antiretroviral treatment in Malawi during 2004-2006. J Acquir Immune Defic Syndr. 2008;49:287–293.
6. Toure S, Kouadio B, Seyler C, et al.. Rapid scaling-up of antiretroviral therapy in 10,000 adults in Cote d'Ivoire: 2-year outcomes and determinants. AIDS. 2008;22:873–882.
7. Lowrance DW, Ndamage F, Kayirangwa E, et al.. Adult clinical and immunologic outcomes of the national antiretroviral treatment program in Rwanda during 2004-2005. J Acquir Immune Defic Syndr. 2009;52:49–55.
8. 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. 2008;371:1603–1611.
11. World Health Organization. Towards universal access: scaling up priority HIV/AIDS interventions in the health sector—progress report. Available at: http://www.who.int/hiv/pub/tuapr_2009_en.pdf
. Accessed October 1, 2009.
12. Ministry of Health (Rwanda), National Institute of Statistics of Rwanda, and ICF Macro. Rwanda Interim Demographic and Health Survey 2007–2008. Calverton, MD: MOH, NISR, and ICF Macro; 2009.
13. Mwumvaneza M, Hinda R, Nyankesha E, et al.. HIV and AIDS in Rwanda: 2010 Epidemiologic Update. Kigali, Rwanda: CNLS (National AIDS Committee) and TRACPlus/MOH; 2010.
14. Logie DE, Rowson M, Ndagije F. Innovations in Rwanda's health system: looking to the future. Lancet. 2008;372:256–261.
16. Moore RD, Keruly JC. CD4+ cell count 6 years after commencement of highly active antiretroviral therapy in persons with sustained virologic suppression. Clin Infect Dis. 2007;44:441–446.
17. Kitahata MM, Gange SJ, Abraham AG, et al.. Effect of early versus deferred antiretroviral therapy for HIV on survival. N Engl J Med. 2009;360:1815–1826.
18. Sterne JA, May M, Costagliola D, et al.. Timing of initiation of antiretroviral therapy in AIDS-free HIV-1-infected patients: a collaborative analysis of 18 HIV cohort studies. Lancet. 2009;373:1352–1363.
19. Cornell M, Grimsrud A, Fairall L, et al.. Temporal changes in programme outcomes among adult patients initiating antiretroviral therapy across South Africa, 2002–2007. AIDS. 2010;24:2263–2270.
20. Nash D, Katyal M, Brinkhof MW, et al.. Long-term immunologic response to antiretroviral therapy in low-income countries: a collaborative analysis of prospective studies. AIDS. 2008;22:2291–2302.
21. Lawn SD, Little F, Bekker LG, et al.. Changing mortality risk associated with CD4 cell response to antiretroviral therapy in South Africa. AIDS. 2009;23:335–342.
22. Zhang F, Dou Z, Ma Y, et al.. Five-year outcomes of the China National Free Antiretroviral Treatment Program. Ann Intern Med. 2009;151:241–251, W-252.
23. Severe P, Juste MA, Ambroise A, et al.. Early versus standard antiretroviral therapy for HIV-infected adults in Haiti. N Engl J Med. 2010;363:257–265.