1. A 2006 Reproductive Health Matters.
All rights reserved.
Reproductive Health Matters 2006;14(27):24–26
0968-8080/06 $ – see front matter
PII: S 0 9 6 8 - 8 0 8 0 ( 0 6 ) 2 7 2 4 3 - 4
www.rhm-elsevier.com www.rhmjournal.org.uk
Health System Strengthening and Scaling Up
Antiretroviral Therapy: The Need for
Context-Specific Delivery Models:
Comment on Schneider et al
Wim Van Damme,a Guy Kegelsb
a Senior Lecturer, Department of Public Health, Institute of Tropical Medicine, Antwerp,
Belgium. E-mail: wvdamme@itg.be
b Senior Lecturer, Department of Public Health, Institute of Tropical Medicine, Antwerp
N this issue Schneider and colleagues1 point
I
usefulness of this comparison is limited. Strate-
to some of the health system challenges for gies developed to assure treatment adherence over
scaling up antiretroviral (ARV) therapy to the six months, such as direct observation by health
majority of the people living with HIV and AIDS workers in TB-DOTS, may not be very inspiring
(PLWHAs) who will need it over the coming years. for assuring lifelong adherence. Indeed, only a
They assume that in low-income countries finan- few of the ARV delivery models documented build
cial resources and supplies of drugs do not con- directly on the TB-DOTS experience. Instead, the
stitute anymore the most important obstacles to process of ART as documented in pilot projects, as
Schneider et al1 explain, is often framed in patient-
widespread ARV use. Indeed many of the countries
most affected by HIV now have steeply increased centred and rights-based discourses around patient
budgets available for AIDS care, either from their empowerment and participation. Such an approach
government budgets or from sources such as is invariably labour-intensive in skilled personnel.
the Global Fund, PEPFAR and the World Bank. Whether such patient-centred approaches are fea-
Despite funding being available, scaling up ARV sible on a large scale in all of the countries hardest
therapy has been slower than hoped for. The WHO hit by HIV and AIDS, remains to be seen.
objective of treating three million people by the The challenges ahead differ between countries,
end of 2005 (‘‘3 by 5’’) has not been reached; in- as the data in Table 1 show. This table simply
stead, an estimated 1.6 million were put on ARVs.2 tabulates for a selection of countries the data pub-
Schneider et al1 identify the need for reorientation lished on WHO and UNAIDS websites3,4 on the
of service delivery towards chronic disease care, availability of medical doctors and nurses against
insufficient supply of human resources for health the number of PLWHAs in the country, and esti-
and existing service delivery cultures as key con- mates the number of PLWHAs per medical doctor
straints to scaling up. This seems a fair assessment of and per nurse. It is often estimated that some 20%
the situation in many countries of southern Africa. of PLWHAs are presently in need of ARVs. How-
Indeed the challenge is unprecedented. Health ever, after large-scale introduction of ARV therapy
systems that were mainly set up to deliver mater- these cumulative numbers will grow rapidly, and
nal and child health services and care for acute ultimately all PLWHAs will end up needing ARVs.
episodes of disease suddenly have to cater for It is striking to note that most countries praised
large numbers of PLWHAs, in need of life- for their performance in the ARV scale-up are
long chronic disease care. The closest analogy among those with the lowest numbers of PLWHAs
to ARV therapy in the health system in sub- per doctor (below the dotted line in the Table). This
Saharan Africa is tuberculosis (TB) care, but the is most obvious for Brazil, Thailand and Cambodia
24

2. W Van Damme, G Kegels / Reproductive Health Matters 2006;14(27):24–26
which have 2, 30 and 75 PLWHAs per doctor communities are busy coping with it. It thus
respectively. But also within sub-Saharan Africa seems likely that creative solutions are being
there is wide variety. The numbers of PLWHAs developed, probably not by academics, rather by
per doctor are well below 1,000 and those per field workers and local communities. As Schneider
et al1 describe, some pilot projects use patient-
nurse well below 100 in South Africa, Uganda and
Botswana. In other countries, these ratios are far centred models, heavily relying on qualified per-
higher. Malawi, Mozambique and Zimbabwe are sonnel. But other projects are rapidly delegating
the most extreme cases; they have more than ten most tasks to less qualified personnel, pushing
times more PLWHAs per doctor than South Africa. standardisation and simplification as far as pos-
They even have considerably more PLWHAs per sible. Others may be relying more on new cadres,
qualified nurse than South Africa or Cambodia such as lay providers and expert patients.
have PLWHAs per doctor. Relatively little is reported about this grass-
Given these human resource constraints, there roots reality. This may partly be because practical
are theoretically two ways forward: 1) rapidly issues of health services organisation are often
increasing the number of doctors and nurses avail- considered to be ‘‘local’’ and hence too context-
able for service delivery; or 2) adopting ARV deliv- bound to be of ‘‘scientific’’ interest. It may also
ery models that need fewer doctors and nurses. be that some actors try to conceal the reality, to
The first option has received quite some attention shield themselves from criticism. Indeed, these
lately,5 the second considerably less. We will focus new realities are likely to challenge the medical
here on the issue of ARV delivery models. and nursing professions’ established modes of
Our contention is that ARV delivery models operating and their related monopolies. These
will have to be context-specific. The countries at actors may even have made technical ‘‘choices’’ –
the top of the table, with over 2,000 PLWHAs such as to forego laboratory monitoring of patients
per doctor, may have to develop ARV delivery on ARVs – which may be judged as unaccept-
models that are quite different from Botswana able by certain physicians’ standards. But such
or South Africa, and certainly from Brazil, with approaches may well be the only feasible ones for
its two PLWHAs per doctor. the really significant ARV scale-up needed for the
This challenge to health systems is unprece- required impact. In high burden communities,
dented, but on the ground health services and mortality among young adults is so high that truly
25

3. W Van Damme, G Kegels / Reproductive Health Matters 2006;14(27):24–26
massive scaling up will be essential. Thus, there approaches. There are simply no precedents on a
may be a balance to strike between the physician’s similar scale in sub-Saharan Africa.
traditional individualised perspective (What is This prospect also points to the most important
best for the individual patient?) and the collective of all health systems challenges: how to sub-
perspective (How can we stop this community’s stantially decrease new HIV infections in high
social degradation?). prevalence countries. Despite some rhetoric on
Recently, some pilot projects in low-income the treatment–prevention synergy, till now little
countries have published the outcomes of their hard evidence is available on how the opportu-
patient cohorts.6,7 And the news is good: patients nities created by ARVs can be used to intensify
on ARVs are faring well, even in resource-poor and harness HIV prevention. Only in a few of the
countries, even PLWHAs who started ARV ther- high burden countries has HIV transmission sub-
apy late, with very low CD4 counts. Mortality stantially decreased. How this has happened
among PLWHAs on ARVs in these pilot projects remains controversial. Whether health systems
is around 10% in the first year, and much lower will be able to cope with the growing numbers of
afterwards. This good news comes at a price: the PLWHAs on ARVs in the longer term will criti-
caseload of patients on ARVs is likely to grow cally depend on decreased HIV transmission in
relentlessly, and far beyond the current esti- the short term. In the current state of affairs,
mates, which still consider that PLWHAs will every new HIV infection will be in need of ARV
only be on ARVs for an average of three years.8 therapy some ten years later, and will need to be
In the short term, the challenge was to put maintained on ARVs for many years. Countries
three million PLWHAs on ARVs by 2005. In the with a high burden of HIV and AIDS will need
long term, the challenge may well become to vastly strengthened health systems to do so, but
maintain 10–15 million PLWHAs on long-term the practical health service configurations able to
ARV therapy. To do so will require innovative cope with such a challenge still have to emerge.
References
1. Schneider H, Blaauw D, Gilson who.int/globalatlas/default. AIDS in Haiti. New England
L, et al. Health systems and aspN. Accessed 21 February Journal of Medicine 2005;353:
access to antiretroviral drugs for 2006. 2325–34.
HIV in southern Africa: service 4. UNAIDS. HIV data. At: bhttp:// 7. Coetzee D, Hildebrand K, Boulle
delivery and human resources www.unaids.org/en/Regions_ A, et al. Outcomes after two
challenges. Reproductive Health Countries/default.asp N. years of providing antiretroviral
Matters 2006;14(27):12–23. Accessed 21 February 2006. treatment in Khayelitsha, South
2. World Health Organization. The 5. Chen L, Evans T, Anand S, et al. Africa. AIDS 2004;18:887–95.
3 by 5 Initiative. At: bhttp:// Human resources for health: 8. Salomon JA, Hogan DR,
www.who.int/3by5/en/N. overcoming the crisis. Lancet Stover J, et al. Integrating HIV
Accessed 21 February 2006. 2004;364:1984–90. prevention and treatment:
3. World Health Organization. 6. Severe P, Leger P, Charles M, from slogans to impact. Public
Global Atlas of the Health et al. Antiretroviral therapy Library of Science Medicine
Workforce. At: bhttp://www. in a thousand patients with 2005;2:e16.
26