The number of people on antiretroviral therapy (ART) in middle and lower-income countries increased to 1 million by June 2005 , as WHO/UNAIDS ‘3 by 5’, the Presidential Emergency Program for AIDS Relief (PEPFAR) and other multilateral and bilateral programs [2–5] have enabled countries to scale-up HIV treatment and care. As part of the ‘Three Ones’ initiative, countries are encouraged to develop and implement comprehensive integrated national HIV programs . An important part of such national programs is the development and implementation of national guidelines for the use of antiretroviral therapy (ART).
The World Health Organization (WHO), in conjunction with institutional partners, has developed global recommendations for scaling up antiretroviral therapy in resource-limited settings using a public health approach. These have followed the principles of standardizing and simplifying ART to make it widely accessible but also ensuring that an effective and appropriate standard of care can be maintained. The first recommendations were developed in 2001 and revised in 2003. The recommendations cover when to start ART, preferred antiretroviral drugs regimens for first and second-line therapy, monitoring the management of people on ART and related topics .
To effectively implement ART services at country level, simplified and standardized national ART guidelines need to be developed and should be part of a comprehensive national HIV program. Such guidelines should be widely distributed, implemented in both public and private health sectors and regularly reviewed. The 2003 WHO ART guidelines are being revised, reflecting the rapid experience and new knowledge gained as ART is being scaled up in resource limited countries. As part of this process a questionnaire on national ART guidelines was sent to a number of resource-limited HIV high burden countries. The aim of this questionnaire was to investigate the existence of national antiretroviral (ART) guidelines, to describe their main recommendations in WHO ‘3 by 5’ focus countries and to compare national ART guidelines of these countries with 2003 WHO Adult ART guidelines.
Questionnaires were sent to 43 countries. These 43 countries were identified by WHO as requiring special attention for developing HIV therapeutic and preventive health services, because of their high HIV burden or because of their strategic importance in the various regions in terms of being able to scale-up HIV therapeutic and preventive health services . The 43 countries had a combined population of 3.4 billion and their population varied in size from Djibouti, with an estimated population of 700 000, to China with an estimated population of 1.3 billion. The HIV-infected population of these countries, which was estimated to need ART, was 3 545 800 in 2004, of whom an estimated 322 283 (9%) were on ART during that year .
Both English and French questionnaires were developed, piloted and sent to WHO staff in the respective countries. They were completed either by WHO staff or often in conjunction with members from the respective national AIDS programs or ministries of health. The completion of 16 (44%) questionnaires was performed by members of ministries of health, national AIDS programs or other relevant health professionals responsible for ART scale-up in countries. The remaining 20 (56%) questionnaires were completed by in-country WHO staff in conjunction with relevant members of ministries of health, national AIDS programs or other responsible for ART scale-up in the particular countries.
The questionnaire first established whether or not countries had developed national adult ART guidelines. If they had, informants were asked to complete the questionnaire, which consisted of 25 structured and open-ended questions, some of which were nested questions (Appendix 1 of the questionnaire). Topics covered were: initiation of ART; selection and management of first-line ART; selection and management of second line ART; monitoring treatment response and toxicity of ART; dissemination policy of national guidelines; provision of first and second-line ART recommendations for special situations. In addition open-ended questions enabled informants to write additional comments not covered in the questions.
Of all questions, 20 were identified to be directly related to the core WHO recommendations on scaling up HIV treatment with ART based on the current 2003 ART guidelines . A scoring system was developed after deliberation between the authors for these questions to measure the concordance between country responses and the WHO ART recommendations. The questions which were included to arrive at this score are indicated in Appendix 1 of the questionnaire and the weighted concordance score was derived based on responses as expected from the WHO guidelines. A greater weight was given to responses, which tallied strongly with the WHO recommendations. The weights given to responses ranged from 0 to 3 while missing responses were given a score of 0.
In addition to overall concordance score of each country, scores were grouped into items. The total scores from the responses for each country were taken as a fraction of the observed concordance score relative to the expected concordance score. These were expressed as a number up to 100. So the minimum possible concordance score could be 0 or no concordance, while the maximum score could be 100 indicating perfect concordance with WHO ART recommendations. Median and inter-quartile ranges (IQR) of concordance scores are presented. Comparisons are based on when the national guidelines were last reviewed, regional responses, as well as comparisons of combined questions between criteria for starting ART, first-line therapy, second-line regimens and laboratory investigations.
Associations between individual weighted country scores and a number of relevant economic and epidemiological country parameters were assessed using Pearson's correlation coefficient and tested using one way analysis of variance test for trend to investigate their statistical significance. These country parameters included: HIV prevalence, estimated number of people needing HIV treatment, estimated number of people on ART, percentage of estimated number on ART of those estimated number of people needing HIV treatment, gross domestic product (GDP) per capita, health expenditure per capita and health expenditure as a percentage of GDP. Information for these country parameters were obtained from WHO/UNAIDS . Between-group comparisons in median scores were assessed using the Mann–Whitney U test while the combined question comparisons between different components within country responses were analysed using the paired t-test.
Of the 43 questionnaires sent, 39 (91%) were returned. Three countries did not complete the questionnaire, including two without national ART guidelines, and these three were excluded from the analyses. Of the remaining 36, three countries (8%) had developed national ART guidelines, but had not developed and published a comprehensive national HIV treatment and care plan.
Sixteen (44%) countries had distributed their ART guidelines to ART prescribers in both public and private sectors, 10 (28%) only to those working in the public sector, while three (8%) countries had only sent them to ART prescribers who had asked for them. Seven (19%) countries had not distributed them or not answered the question.
Starting antiretroviral therapy
The guidelines of 16 (44%) countries recommended using WHO clinical staging criteria combined with CD4 cell count or total lymphocyte count (TLC). Twelve (33%) countries recommended using WHO clinical staging criteria and CD4 cell count, whereas four (11%) recommended starting ART based only on CD4 cell counts (Fig. 1). Fifteen (42%) counties reported starting ART if the CD4 cell count was less than 200 cells/μl but considered ART if the CD4 cell count was less than 350 cells/μl, 14 (39%) recommended to start only if the CD4 cell count was less than 200 cells/μl and 3 (8%) recommended to start ART only if the CD4 cell count was less than 350 cells/μl. In terms of TLC, 17 (47%) countries recommended to start ART at less than 1200 cells/μl in WHO Stage II and regardless of TLC in Stages III or IV. Thirty one (86%) countries reported that that viral load was not recommended as a starting criterion for ART.
Thirty five (97%) countries reported that the national ART guidelines recommended at least one standard first-line regimen. For 24 (69%) countries, stavudine + lamivudine + nevirapine combination was one of their preferred first-line regimen, zidovudine + lamivudine + nevirapine was one of the preferred regimen for 19 (54%) countries, zidovudine + lamivudine + efavirenz in 13 (37%) countries and stavudine + lamivudine + efavirenz one of the preferred first-line regimen in nine (26%) countries (Fig. 2). In terms of alternative first-line regimens, 17 (47%) countries recommended stavudine + lamivudine + efavirenz, 13 (36%) recommended zidovudine + lamivudine + efavirenz, nine (25%) zidovudine + lamivudine + nevirapine and six (17%) recommended stavudine + lamivudine + nevirapine.
Twenty seven (75%) countries reported that they recommended the use of fixed dose combinations, 14 (52%) of which recommended the combination of stavudine + lamivudine + nevirapine and 11 (41%) recommended zidovudine + lamivudine + nevirapine. Thirty three (92%) countries reported that the national ART guidelines clearly defined alternatives for substituting first-line drugs, if specific toxities arose or in case of contra-indications
Thirty three (92%) of the 36 countries indicated that they recommended second-line regimens. Of these 14 (42%) countries indicated that abacavir + didanosine + lopinavir/ritonavir was one of the preferred choices of second-line therapy (Fig. 3). The second most common reported combination was abacavir + didanosine + nelfinavir in 10 (30%), followed by abacavir + didanosine + saquinavir/ritonavir in nine (27%) countries. Tenofovir + didanosine + lopinavir/ritonavir was the preferred choice in eight (24%) countries, and tenofovir + didanosine + nelfinavir in seven (21%) countries.
In terms of alternative choices, tenofovir + didanosine + nelfinavir and abacavir + didanosine + nelfinavir were the most commonly recommended in six (18%) and five (15%) countries respectively. Eighteen (55%) countries reported that the national guidelines defined second-line alternatives for substitution in case of specific toxicities or contra-indications to second-line drugs.
Monitoring responses to antiretroviral therapy: criteria of treatment failure
Of the 36 countries, 33 (92%) responded that they did have specific recommendations for monitoring of ART. Of these, 26 (79%) recommended CD4 cell count plus some other indicator, including TLC, viral load or hemoglobin, five (15%) recommended CD4 cell count only, while two (6%) countries recommended TLC with viral load and hemoglobin (Fig. 4). Of the 36 respondents, 25 (69%) reported that they had clearly established criteria of first-line treatment failure, with clinical and immunological criteria being the preferential options.
Concordance between national and WHO antiretroviral therapy guidelines
For all 20 individual items, the median concordance score between country and WHO guidelines was 65 (IQR, 56 to 76). The country-specific WHO concordance scores were found to be significantly and inversely correlated with health expenditure per capita (r = −0.43, P = 0.009) and GDP per capita (r = −0.37, P < 0.03). None of the other country parameters investigated were significantly correlated with country-specific WHO concordance scores.
When comparing the scores between the various sub-groups, countries that revised their guidelines after publication of the current 2003 WHO guidelines had a median concordance score that was 27% higher than for countries with earlier revisions: 56 (IQR, 50 to 65) versus 71 (IQR, 59 to 76), (P = 0.175). Furthermore, the median concordance score of the countries from the WHO African Region was 15% greater than the median score from countries in the other WHO regions: 68 (IQR, 62 to 76) versus 59 (IQR, 52 to 74), respectively (P = 0.243)
Questions were combined into four groups: criteria for starting ART; first-line therapy; second-line regimens; and laboratory investigations. The median concordance score for the criteria for starting first-line therapy was 100 (IQR, 67 to 100). For first-line therapy, the median score was 70 (IQR, 60 to 80), while for second-line the median concordance score was 45 (IQR, 27 to 55). The median concordance score for laboratory investigations was 80 (IQR, 80 to 100) (Fig. 5). The concordance score for starting ART criteria was significantly higher than either of those for first-line regiments (P = 0.001), second-line regimens (P < 0.001) or for laboratory investigations (P = 0.056).
This analysis was limited to 43 WHO ‘3 by 5’ focus countries and did not involve all middle and lower-income countries currently scaling up HIV treatment and care. It is the intention of WHO that those countries not covered by this survey, will be surveyed in the near future. This evaluation was focused on the development of national guidelines. It did not consider their effective implementation and use of the guidelines at health facility level, and substantive differences may well exist between the development of national guidelines and program implementation resulting in actual clinical practice. The relationship between existing national ART guidelines, efforts to support implementation of ART, actual prescribing practices and the outcome and impact of their implementation needs to be urgently investigated and documented within countries.
The demonstrated presence of strong concordance between many of the national and WHO guidelines is not necessarily indicative of how the national guidelines were arrived at and the role of the WHO guidelines in this process. However, it does suggest that the WHO guidelines have guided or informed the development of the majority of national recommendations. This was particularly relevant for those countries with lower health expenditure and GDP per capita. Within a slightly different context, the practical utility of WHO guidelines at country level was recently again demonstrated in Cambodia .
The WHO recommendations for scaling up adult antiretroviral therapy in resource-limited settings used a standardized and simplified public health approach to make it widely accessible in resource-limited settings and ensure the maintenance of effective and appropriate standards of care. However, each country needs to develop their own simplified and standardized national ART guidelines, which form part of a comprehensive national HIV program, which are regularly reviewed and implemented in both public and private health sectors. In this preliminary analysis, while good concordance between national guidelines and WHO recommendations existed on first-line regimens, criteria for starting ART and routine laboratory monitoring, recommendations for second-line regimens need to be improved as less concordance in selection and management of second-line regimens was identified. This probably reflects the current realities regarding less experience with second-line line regimens in these settings, with their higher cost and reduced availabilities. These findings will be taken into consideration during the current review of the WHO ART guidelines but suggest that a public health approach towards managing diseases in middle and lower-income countries can successfully guide the development of national policies. However, the implementation of these national policies in clinical practice needs to be verified . National ART guidelines and actual prescription practices need to be compared, to ensure optimum effectiveness and efficiency in terms of outcomes and impact at individual and population levels respectively; a need that is reinforced by the recent commitment of the G8 and other countries towards ensuring universal access to HIV services .
We are grateful to members of ministries of health, national AIDS programs and other relevant health professionals in countries, the WHO ‘3 by 5’ or ‘National Program Officers’ who completed or reviewed the questionnaires, the WHO regional advisers and other regional WHO staff, and Parijat Baijal, Andrew Ball, Mazuwa Banda, Clem Chan-Kam, Micheline Diepart and Ousmane Diouf at WHO Geneva for their assistance and feedback in completing this survey and Catherine Hankins for commenting on earlier drafts.
1. WHO. Progress on Global Access to HIV Antiretroviral Therapy: an update on “3 by 5” June 2005http://www.who.int/3by5/fullreportJune2005.pdf
Accessed 1 September 2005.
2. USAIDS. The President's Emergency Plan for AIDS Relief: Fact Sheets
. Washington: US Department of State; 2005 http://www.usaid.gov/our_work/global_health/aids/pepfarfact.html
Accessed 23 September 2005.
3. Clinton Foundation Programs. HIV/AIDS Initiativehttp://www.clintonfoundation.org/programs-hs-ai.htm
Accessed 23 September 2005.
4. DIFD HIV/AIDS Strategy
May 2001:11-2 http://www.dfid.gov.uk/pubs/files/hiv-isp.pdf
Accessed 23 September 2005.
7. WHO. Scaling up Anti-retroviral Therapy in Resource-limited Settings: Treatment Guidelines for a Public Health Approach
. Geneva, Switzerland: WHO; December 2003 http://www.who.int/3by5/publications/documents/arv_guidelines/en/index.html
Accessed 1 September 2005.
8. WHO. Investing in a Comprehensive Health Sector Response to HIV/AIDS: Scaling up Treatment and Accelerating Prevention
. Geneva, Switzerland: WHO; June 2004 http://www.who.int/3by5/en/HIV_AIDSplan.pdf
Accessed 1 September 2005.
9. WHO. Country Profiles on Treatment Scale Up as of June 2005
. Geneva, Switzerland: WHO; June 2005 http://www.who.int/3by5/countryprofiles/en/index.html
Accessed 1 September 2005.
10. WHO. HIV/AIDS Country Information
Accessed 23 January 2006.
11. Thai S, De Munter P, Jacques G, Van den Ende J, Schrooten W, Lynen L. Accuracy of World Health Organisation's 2002 and 2003 criteria to start HAART in the absence of CD4 count.Proceedings of the IAS Conference
, Rio de Janeiro, 2005.[abstract WePe12.2C11].
12. Bautista-Arredondo S, Mane A, Bertozzi SM. Economic impact of antiretroviral therapy prescription decisions in the context of rapid scale-up of access to treatment: lessons from Mexico. AIDS 2006; 20:101–109.
13. Anonymous. The Post Gleneagles Communiqué G8 Gleneagles 2005
Accessed 17 October 2005.