AFRO-NETS> Realistic priorities for AIDS

[Claudio Schuftan <aviva@netnam.vn>]

Realistic priorities for AIDS
-----------------------------

Breaking the silence: Setting realistic priorities for AIDS control 
in developing countries

Martha Ainsworth* and Waranya Teokul**

*  Development Research Group, The World Bank, Washington, D.C., USA
** National Economic and Social Development Board, Office of the 
   Prime Minister, Bangkok, Thailand

Lancet 2000; 356: 55-60

The opinions expressed in this article are those of the authors and 
do not necessarily represent official policy of their respective in-
stitutions.

The AIDS pandemic is a human tragedy that is threatening development 
in the poorest countries. There is no cure or vaccine, but the tools 
to control the epidemic already exist. Nevertheless, there are few 
examples of national AIDS control programmes that have had an impact 
on the epidemic. We (an economist and a planner) attribute this to 
the reluctant of governments to confront AIDS and a failure to pri-
oritise activities in the face of severe financial and administrative 
constraints. When implementation capacity is weak, expanding the num-
ber of activities may not improve programme effectiveness. Rather, by 
implementing a smaller, core set of the most cost-effectiveness ac-
tivities on a national scale, policy makers could have a huge effect 
on the overall epidemic in a sustained way and provide a foundation 
for expansion. We propose three core priorities for AIDS control in 
poor countries for prevention, treatment, and mitigation of the im-
pact.
 
AIDS is continuing to ravage the poorest countries on the planet. The 
grim statistics are by now familiar: 95% of the 34 million people in-
fected with HIV-1 worldwide live in less-developed countries; more 
than two-thirds are in Sub-Saharan Africa.1 More than 15,000 people 
become HIV-1 infected every day. HIV-1 is now the single largest in-
fectious killer and fourth leading cause of death in the world.2 It 
is striking the most productive part of the population in the poorest 
countries-prime aged adults-robbing economies of scarce skills, fami-
lies of their breadwinners, and children of their parents. Life ex-
pectancy in developing countries rose from 40 to 63 between 1950 and 
1990, the result of enormous investments by families, governments, 
and the international community in improving the quality of life.3 
Now a single fatal infectious disease has erased these gains in the 
hardest-hit countries: life expectancy in countries like C=F4te 
d'Ivoire and Zimbabwe is 10-20 years shorter than it would have been 
without AIDS. 3 There is still no cure and a preventive vaccine is at 
least a decade away from being realized.

There is some cause for hope, however. First, we already have the 
tools to prevent HIV infection and AIDS. Increased condom use, treat-
ment of sexually transmitted disease (STDs), reduction in the number 
of sexual partners, safe injecting behavior, and drugs to prevent 
mother-to-child transmission have all been shown to be effective in 
preventing HIV/AIDS in pilot projects, controlled trials, or national 
programmes in developing countries.4 Second, nearly half of the 4.8 
billion people in developing countries live in areas where HIV infec-
tion is not yet widespread, even among people most exposed to con-
tracting HIV.3,5 If policymakers act early, billions of people in de-
veloping countries can be spared the ravages of the epidemic. Even in 
countries where HIV has infected those at highest risk, its spread 
into the general population can be dramatically slowed through sys-
tematic national programmes focused on preventing transmission among 
those with the most partners. 6,7

Nevertheless, there are remarkably few policy success stories on a 
national scale. Thailand is the clearest case: after an intense na-
tional campaign to raise condom use in commercial sex, the condom use 
rate for brothel-based sex workers reached more than 90%, STD cases 
declined precipitously, and HIV prevalence among army conscripts 
dropped by more than half. 6,8 Infection rates among pregnant women 
have since declined, although are still high at 1-2% 9, and these ac-
complishments seem mostly sustained throughout the East Asian finan-
cial crisis.10 In Uganda, HIV prevalence has declined among pregnant 
women 11 and young people 12, 13 who are delaying sexual activity 11. 
However, it is difficult to attribute either of these outcomes to 
public policy. 3,14,15 The decline in prevalence may be due to 
heightened mortality among HIV-positive individuals or the natural 
evolution of human behavior faced with a generation of high mortality 
associated with sexual behavior. 16

Given that we have the tools, why haven't national AIDS control pro-
grammes been able to show more success? And how might the response be 
made more effective? We offer the perspective of an economist and a 
planner, each with more than a decade of practical and/or research 
experience in HIV/AIDS policy issues in developing countries in Af-
rica, Asia, and Latin America.


What accounts for the lack of action?

We believe that two principal factors are to blame-first, the reluc-
tance of national governments to take responsibility for preventing 
HIV infection; and second, a failure by both national governments and 
international agencies to set realistic priorities that can have an 
effect on the overall epidemic in countries with scarce resources and 
weak implementation capacity.


The reluctance of governments to confront AIDS

The human suffering and effect of AIDS on poverty and health systems 
alone should be adequate motivation for governments to act. But even 
more critical, there are certain actions that are essential to stop-
ping the epidemic that only governments can assure. These include 
providing 'public goods' such as information, epidemiological and be-
havioral surveillance, analysis of the cost-effectiveness of preven-
tion and treatment, and the sectoral coordination for an effective 
response.3 The private sector will not produce public goods because 
people can enjoy the benefits without having to pay for them. Govern-
ment is also in a unique position to raise the incentives for those 
who engage in risky behavior to adopt safer behavior than they would 
on their own, through subsidized condoms, improve health information, 
STD treatment, or needle-exchange programmes, for example. Individual 
risky behavior raises the infection level in the whole population, 
putting others indirectly at risk-a 'negative externality' of high 
risk behavior.17 Civil society, non-governmental organizations 
(NGOs), and the private sector can all be involved in the response to 
AIDS, but without government involvement in these two areas, the epi-
demic cannot be beaten.

Nevertheless, governments are reluctant to act. AIDS prevention 
treads on sensitive topics that neither the government nor the public 
are eager to discuss: sexual behavior, marital fidelity, prostitu-
tion, sexual orientation, and injecting drug use. Every country has 
engaged in denial; leaders typically assert that the moral values of 
their society would not permit transmission of an agent such as HIV 
that is associated with risky sexual behavior, homosexuality, or in-
jecting drug use. But behaviors that spread HIV exist in every coun-
try. Denial merely delays the response and worsens the epidemic.

Even when denial is conquered, governments can hesitate to undertake 
the interventions most likely to have an impact. The key to stopping 
the AIDS epidemic is to bring the reproductive rate of HIV (the aver-
age number of susceptible people infected by an infected person over 
his/her lifetime) to below one.18,19 When that happens, the epidemic 
will eventually die out. The most efficient way to constrain the 
spread of HIV in the whole population is to prevent transmission 
among those for whom the reproductive rate is very high-i.e., those 
with the most sexual partners or who share unsterilized injecting 
equipment.3,17 However, there is no political constituency for pro-
grammes perceived to be helping people at the margins of society. 
Once the general population feels at risk, there is political pres-
sure for other programmes that are thought to benefit the average 
citizen, like safe blood supply and treatment to prevent mother to 
child transmission. These programmes have important benefits but vir-
tually no effect on the course of an epidemic fueled by sexual trans-
mission.

The long latent period of HIV infection contributes to delayed ac-
tion. Governments are reluctant to act until many people are visibly 
affected, since there are many other urgent health problems. Bangla-
desh, for example, has fewer than 200 persons known to be living with 
HIV (although estimates are that 21,000 people are infected), but 
620,000 people had tuberculosis as recently as 1997.20 It is far less 
controversial for government to act when large numbers of people fall 
ill. Treating programmes do not require public discussion of taboo 
subjects, and when people get sick there is a constituency that will 
demand treatment. But by then it is too late to prevent an epidemic. 
Both early and late in an AIDS epidemic, the constituency for preven-
tion tends to be weak.


Failure to prioritize in resource-scarce settings

Resources are very scarce in poor countries-both financial and admin-
istrative-and AIDS is not the only problem that policymakers face. 
Consider the case of Uganda, with per capita GNP of $310 in 1998. 
More than half of the population was below the national poverty line 
and 37% lived on less than $1/day in the mid 1990s.20 Annual public 
and private health expenditure per caput during the past decade aver-
aged only $14. There are fewer than 5 physicians per 100,000 popula-
tion. Two-thirds of the population lacks access to safe water. Only 
38% of pregnancies are attended by a health professional. More than a 
quarter of school-aged children are not enrolled in primary school 
and more than 85% are not enrolled in secondary school. In addition, 
almost 10% of the adult population is infected with HIV. The cost of 
treating a single AIDS patient in the poorest African countries in 
1995 was more than the annual cost of educating ten primary stu-
dents.3 In the face of resource scarcity and urgent social needs, 
governments and donors need to prioritize.

In contrast, government AIDS control strategy documents typically em-
brace everything that might be done to fight AIDS. These strategies 
are reinforced by technical best practices on specific interventions 
and by international donors, which may be able to support only inter-
ventions that address their institutional mandates. Objectives are 
expressed in terms of programmes components, not outcome. The pro-
gramme elements are typically not ranked in terms of their effective-
ness in preventing the overall epidemic, given their costs. The re-
sult is that many activities and pilot projects are launched by gov-
ernment, donors, and NGOs but very few are implemented on a scale 
that would register an impact on the overall epidemic and the activi-
ties selected are not necessarily those that would yield the greatest 
impact. Furthermore, in the absence of any sense of priority, the ac-
tivities that tend to get done are those with political support and 
those that are the least controversial. Those likely to have the 
largest impact on the epidemic-prevention among those most likely to 
pass HIV to others-are the lowest on the political agenda. The fail-
ure to prioritize has resulted in a lack of focus on specific objec-
tives and a lack of results.


The expanded, multisectoral response

The international community has responded to the lack of success by 
launching what has been variously termed as an "expanded", "intensi-
fied", and "multisectoral" response. 21, 22 This strategy aims to 
generate greater political commitment by national governments, to mo-
bilize more resources from within and outside countries, and to rep-
licate on a national scale a more comprehensive programme that in-
cludes an increased number of interventions targeted to virtually all 
groups in society. The expanded response emphasizes prevention as 
well as treatment, policies and programmes to mitigate the impact of 
AIDS, and policies that will change the societal factors that influ-
ence vulnerability to HIV in the long run. 'Priority' is used to de-
scribe essential components of an AIDS programme from a technical 
point of view, not a ranking of components according to their rela-
tive costs and impacts.

To the extent that this approach succeeds in building the political 
consensus to act among key decision-makers (particularly on preven-
tion, for which incentives to act are always the weakest) could re-
move one of the main constraints to effective programmes. An increase 
in resources for AIDS programmes will also help to ease the budget 
constraint, if the absorptive capacity (e.g., the ability of pro-
grammes to expand rapidly when trained manpower is scarce) exists. 
However, implementation capacity is very difficult to affect in the 
short term. The multisectoral response rightly recognizes the role of 
social and contextual factors that condition individual decisions, 
but can only be addressed in the long term. Mobilisation of many dif-
ferent segments of society will be necessary to have an impact in the 
hardest hit countries. Policies regarding fair and humane treatment 
and care for those who are infected are central to strategies to cope 
with the epidemic and reinforce prevention.


Our concerns

First, even with more resources and with acknowledgment of the impact 
and potential role of different sectors in a response, there is still 
a need to prioritize. The expanded response calls for more interven-
tions to more constituencies, on a national scale, involving sectors 
with even more limited knowledge and capacity to act, which often 
have limited ability to effectively realise their existing mandate. 
23 There is the real risk that by stretching implementation capacity 
even thinner in the most affected countries, even less will get done. 
AIDS mortality may strike every sector of the economy; however, this 
does not necessarily imply that adding AIDS prevention and mitigation 
to every ministry's programme will be a cost-effective way of reduc-
ing the epidemic.

Second, the expanded multisectoral strategy is defined by the actors 
and scope of the programmes, not by concrete, achievable outcomes. 
One of the important lessons from Thailand's experience is that a 
clear objective was set-to lower transmission via commercial sex-and 
the necessary ministries, programmes, and policies were mobilised to 
achieve that objective. 6,8 This included most importantly the '100% 
condom programme', but also a massive campaign to make condom use so-
cially acceptable. It was supported by monitoring activities in STD 
clinics and in behavioral and HIV surveillance among sex workers and 
their clients. The Thai Ministry of Public Health, Ministry of Inte-
rior and Office of the Prime Minister were the key public agencies 
involved. The principle of focusing government efforts on a smaller 
set of achievable outcomes that would have the greatest effect then 
sustaining those results as the programme expands is an important 
lesson for AIDS control efforts in countries with fewer resources, 
like those in Sub-Saharan Africa, Cambodia, India, and Haiti.

Third, in pursuing a multisectoral response, there is also the danger 
that the core activity that is necessary to stop the epidemic-
sustained behavior change among those most likely to contract and 
spread HIV-will continue to be shortchanged for the same reasons it 
was before. There is little political constituency for people at the 
margins of society, and by expanding the involvement of civil society 
even further, the demands will increase for programmes that do not 
shut down the engine of the epidemic. Ensuring that individuals most 
likely to spread HIV adopt safer behaviour, through whatever combina-
tion of direct or indirect interventions, by use of NGOs or others, 
is a core responsibility of government. Nationwide achievement of 
this objective should be a core responsibility of national AIDS pro-
grammes.

Fourth, because the implementation capacity of government is so 
stretched already, expanded multisectoral responses are likely to 
dramatically increase the responsibilities of NGOs and the need for 
coordination. However, in many instances, the tendency of interna-
tional agencies to work through NGOs has the effect of marginalizing 
the government.15 NGOs have a crucial role in the effective delivery 
of AIDS prevention and care, particularly to marginalized groups who 
may actually fear contact with government. However, this does not ab-
solve government of its responsibility to produce those public goods 
essential to disease control, to coordinate and monitor the response, 
to ensure that the NGOs receiving public subsidies are fully quali-
fied and evaluated, and to ensure that the objectives are met in a 
cost-effective way. Strengthening these functions is critical for 
both limited and expanded responses.

Finally, this strategy is proposed for all countries, regardless of 
the stage of the epidemic or the available resources. But we doubt 
that the same strategy is equally appropriate for Bangladesh as for 
Uganda, where 10% of the population is infected, or as for Thailand, 
with ten times the income per capita.


Setting realistic priorities

To improve the performance of national AIDS control programmes, poli-
cymakers need to build on a smaller core set of objectives, defined 
in terms of measurable outcomes and impact, and identify the most 
cost-effective set of activities and actors to meet them on a na-
tional scale. Focusing resources on scaling up interventions to meet 
these core objectives may not completely halt the spread of HIV in 
hard-hit areas, but has the potential to have a huge impact on the 
overall epidemic, in terms of sustained prevention, treatment, and 
mitigation, in a sustained way with the limited resources available. 
This is not to say that a programme would be limited to these core 
objectives (in some middle-income countries they have already been 
met), but as fragile AIDS control programmes expand, these are the 
core set of objectives that are a foundation for expansion. The first 
priority for resources would be to ensure that these core functions 
are met.

What would be on this short list of "core" activities? In addition to 
the main activity of providing coordination, monitoring, and other 
public goods, we would like to highlight only three. Only the first 
of these is relevant to countries with nascent epidemics; for coun-
tries with concentrated or generalized epidemics, all three will be 
important. 


Ensuring behavior change among those with the riskiest behavior

Pilot projects have demonstrated the effectiveness of reducing trans-
mission among those with the riskiest behavior in preventing many 
secondary infections 24-26 and Thailand has shown the effectiveness 
of this approach on a national scale.6-8 Epidemiological models have 
shown that even in a generalized AIDS epidemic like the one in so 
many sub-Saharan countries, this strategy is key to lowering preva-
lence in the whole population. 3,27,28 Yet no government in Africa 
has systematically attempted to reduce high-risk behaviors on the 
scale achieved in Thailand. The distribution of high-risk behavior 
will be different in other countries, but it still should be possible 
to systematically map the areas and populations at greatest risk of 
spreading HIV and to focus resources on behavior change first in 
those areas and populations. This is already under discussion in 
Bangladesh and Burkina Faso, and Cambodia is seeking to implement its 
own 100% condom programme in commercial sex. Government may not have 
a comparative advantage at providing some services to hard-to-access 
populations, but it can subsidize others to ensure that they are de-
livered. An effective response may require both direct and indirect 
approaches-e.g., peer-based prevention, condom promotion, and STD 
treatment for sex workers and their clients, reinforced through pub-
lic campaigns to popularize condoms.

Ensuring universal access to treatment for opportunistic infections
The lowest income countries are extremely constrained in the avail-
ability of finance medical care. The countries classified by the 
World Bank as 'low-income', with average gross national product per 
caput of $520 per year, spend on the order of $23 per person per year 
on health care, including both government and private spending.20 Al-
though there is no cure for AIDS, there are inexpensive treatments 
for palliative care, opportunistic infections, and STDs that are af-
fordable in low-income countries (Table 1). Making these cost-
effective drugs widely available in the public and private health 
systems would substantially improve the quality of life of those with 
AIDS, at relatively low cost. In fact, many households would be will-
ing to buy these drugs if they were available, but in many cases they 
are not. Availability of essential drugs is still not assured in many 
countries, and the detection and cure rates for tuberculosis, the 
most common opportunistic infection of AIDS, are still often low (Ta-
ble 2). Universal access to cost-effective drugs for palliative care 
and treatment of opportunistic infections should be a second 'core' 
responsibility of national AIDS programmes. Access to essential drugs 
for the treatment of opportunistic infections is the foundation on 
which additional interventions for treatment and care can be built.

Highly active antiretroviral therapy is not affordable on a large 
scale in less-developed countries either privately or by government, 
29 nor is it a technology that most poor people could adhere to or 
that existing health systems in poor countries are equipped to sup-
port.30 Even if the drugs were cheap, this type of therapy would 
probably not be cost-effective as implemented for most patients com-
pared to treatment of opportunistic infections because of problems 
with non-compliance and erratic supplies of drugs. Weak compliance 
can generate the spread of drug-resistant strains of HIV, to the det-
riment of future AIDS patients. The use of public funds to subsidize 
the treatment of patients most able to comply-who are better educated 
and have access to better health care-would be a highly inequitable, 
and would shift health resources from the poor to those who are not 
poor.


Integrating AIDS into poverty alleviation strategies

The relation between AIDS and poverty is complex.31,32 However, among 
those affected by AIDS, those who are poor have the fewest resources 
to cope and non-poor families can become poor as a result of AIDS 
illness and death. AIDS will unquestionably worsen poverty, and pov-
erty alleviation is a core responsibility of government.
Most less-developed countries do have poverty reduction policies in 
place: to raise incomes through economic growth; to improve the ac-
cess of poor people to productive assets (e.g., land, equipment, work 
animals, schooling, and training), and social and technical services; 
to increase their schooling; and through various 'safety nets' that 
help people recover in times of crisis. These programmes suffer from 
the same resource constraints as AIDS programmes-not just financial, 
but in terms of implementational capacity. What should be the appro-
priate strategy for mitigating the impact of AIDS on poverty? Who 
should be targeted by antipoverty programmes?

This is a topic about which amazingly little is known. Those con-
cerned about the economic effects of AIDS tend to focus on what 
should specifically be done to help patients with AIDS and their 
families without placing those effects in the context of overall pub-
lic policy on poverty.33 Should policy be focused on those affected 
by AIDS, the poor more generally, the poor affected by AIDS or some 
other category? What are the operational implications of these 
strategies? To take an example, in a country such as Cambodia with 
HIV prevalence of about 3%, where almost 36% of the population is in 
poverty, more than 2% is disabled and many adults and children are 
orphaned from war, 34 is it appropriate to channel the limited public 
safety net to households or children that are affected by AIDS?
The answer to these questions will be very country-specific and de-
pend on the stage of the epidemic, income levels, and the socioeco-
nomic status of those who are infected. A first step is to understand 
how existing antipoverty programmes and strategies are helping those 
affected by AIDS.35 But for countries with serious AIDS epidemics, 
the appropriate integration of AIDS control into national antipoverty 
programmes is a 'core' mitigation priority. Officials outside of the 
health ministry must act on this issue; however, a single appropriate 
solution is not apparent.


Priorities for the international community

The international community has the responsibility for ensuring the 
production of global public goods- i.e., knowledge and technology 
that are public goods on an international scale.3 Two key research 
priorities would be of enormous benefit to less-developed countries.
The first priority is the implementation of randomized controlled 
trials of behavioral interventions to prevent HIV. There have been 
large community trials on syndromic STD treatment in Mwanza, Tanza-
nia, 36 and mass STD treatment in Rakai, Uganda, 37 but we still do 
not know whether massive condom promotion would have had an even big-
ger effect on HIV infection and STDs at lower cost, or how the impact 
might have varied by the degree to which services are targeted to 
those with the riskiest behavior. Less-developed countries need ran-
domised controlled trials of the effectiveness of behavioral inter-
ventions that take into account the secondary infections averted and 
that include HIV and STD incidence as outcome measures.3, 38

The second priority is the assessment of cost-effectiveness of wider 
availability of drugs to treat opportunistic infections. Drug compa-
nies have a strong incentive to evaluate the effectiveness of highly 
active antiretroviral therapy in a handful of high-income patients in 
the poorest countries. But the effectiveness or impact of wide avail-
ability of drugs to treat common symptoms and opportunistic infec-
tions remains understudied.

Another international priority is to promote public-private partner-
ships to develop medical products for developing countries, including 
an HIV vaccine and vaginal microbicides. 39,40 Private firms do not 
have sufficient incentives to develop the technology, both because it 
is a global public good and because the main beneficiaries are people 
with low ability to pay.


Fewer priorities, greater impact

We know how to stop the AIDS pandemic in less-developed countries. We 
do not have to wait for a "magic bullet". Spreading resources across 
programmes in many sectors risks stretching the limited implementa-
tion capacity even more thinly with negligible or even negative im-
pact on the effectiveness of national programmes. By ensuring re-
sources for national implementation of a limited set of priorities, 
policymakers can have a much larger impact on HIV/AIDS prevention, 
treatment, and mitigation while building a foundation for implementa-
tion of more extensive measures as resources permit.
 
The tables and footnotes are too long to attach.

Courtesy of Jamie Uhrig.


--
Claudio Schuftan
mailto:aviva@netnam.vn

--
Send mail for the `AFRO-NETS' conference to `afro-nets@usa.healthnet.org'.
Mail administrative requests to `majordomo@usa.healthnet.org'.
For additional assistance, send mail to:  `owner-afro-nets@usa.healthnet.org'.