Country response to HIV/AIDS: National Health
Accounts on HIV/AIDS in Brazil, Guatemala, Honduras,
Mexico and Uruguay
Jose ´-Antonio Izazola-Liceaa, Carlos Avila-Figueroab, Daniel Ara ´nc,
Sergio Piolad, Rodulio Perdomoe, Patricia Hernandezf,
Jorge A. Saavedra-Lopezgand Ricardo Valladares-Cardonah
National expenditures on HIV/AIDS were estimated as summary indicators to assess
the country’s response to HIV/AIDS. The methodology is based on a matrix system
describing the level and flow of health expenditures on HIV/AIDS: an adaptation of
the National Health Accounts methods. The expenditures were classified by source
(public, private, international), by the use of funds (prevention, care), by object, and
by type of provider institution. The results are reported in US$ using the official
exchange rate for the year of estimation. For international comparisons monetary units
were adjusted by the purchasing power parity (US$PPP). National HIV/AIDS total
expenditures were: Guatemala US$PPP29.5 million, Uruguay US$PPP32.5 million,
Mexico US$PPP257 million, and Brazil US$PPP587.4 million during 1998, and
Honduras US$PPP33.9 million for 1999. The total HIV/AIDS expenditures per capita
for 1998 were: Brazil US$2.69, Mexico US$1.25, Guatemala US$1.08, Uruguay
US$6.63, and Honduras US$3.6 for 1999. The 1998 distribution of the total HIV/AIDS
expenditures in prevention and care were, respectively, Brazil 10 and 80%, Guatema-
la 15 and 70%, Mexico 29 and 66%, Uruguay 36 and 51%, and Honduras 28 and
65% for 1999. The share of total expenditures on antiretroviral drugs ranged from 52%
in Guatemala to 75% in Brazil, even when the estimated coverage of antiretroviral
therapy was close to 10% in Guatemala and universal in Brazil. The estimated flow
from international sources per capita in 1998 was Uruguay US$0.03, Brazil US$0.24,
Guatemala US$0.11, Mexico US$0.01, and Honduras US$1.04 in 1999. The data
allow international comparisons and provide critical information to improve equity
and efficiency in the allocation of scarce resources. The National HIV/AIDS Accounts
also constitute a powerful tool to describe the country’s response to HIV/AIDS.
& 2002 Lippincott Williams & Wilkins
AIDS 2002, 16 (suppl 3):S66–S75
Keywords: costs, financing, expenditures, resource allocation, country response,
National Health Accounts, HIV/AIDS, Latin America
From theaFundacio ´n Mexicana para la Salud, A.C. (FUNSALUD), Regional AIDS Initiative for Latin America and the Caribbean
(SIDALAC), Me ´xico, DF Me ´xico;bSecretarı ´a de Salud, Me ´xico, DF Me ´xico;cCentro de Estudios de Economı ´a y Salud (CEES),
Montevideo, Uruguay;dInstituto de Pesquisa Econo ˆmica Aplicada (IPEA), Brası ´lia, Distrito Federal, Brasil;eAsesores y
consultores te ´cnicos de Centro Ame ´rica, Tegucigalpa, A.C. Honduras;fNational Health Accounts, World Health Organization,
Geneva, Switzerland;gIndependent Consultant on Health & HIV/AIDS Policy and Management, Me ´xico, DF Me ´xico;hGSD
Consultores Asociados (GSD), Guatemala, Guatemala
Correspondence to: Jose ´ A. Izazola-Licea, Fundacio ´n Mexicana para la Salud, AC, Av. Periferico Sur #4809, Me ´xico, DF
14610, Me ´xico. Tel: (52-55) 5655-9011; fax (52-55) 5655-8211.
& & 2002 Lippincott Williams & Wilkins
A common response to HIV/AIDS is the struggle to
obtain economic resources that are sufficient, oppor-
tune, properly directed to the most efficacious inter-
ventions, and provide an adequate response to peoples’
needs for prevention and care. The Joint United
Nations Programme on AIDS, following its particular
mandate to increase the flow of resources from interna-
tional donors and domestic expenditures in the fight
against AIDS, devoted efforts to describe the flow of
resources between international donors to recipient
countries . Other studies have also attempted to
estimate the economic impact of AIDS from a social
perspective, by measuring the losses of lifetime earnings
among the population mostly affected by AIDS [2,3];
and a few others have focused on evaluations of the
cost-effectiveness and benefits of specific programme
A recently published modelling exercise estimated the
basic amount of resources needed to confront the
HIV/AIDS epidemic efficiently in the developing
world. The authors projected US$9250 million an-
nually as the minimum global fund needed immediately
to curb the HIV/AIDS epidemic by 2005; for Latin
American and Caribbean (LAC) countries US$590 mil-
lion was needed annually for prevention and US$550
million for care .
As part of several technical assistance projects con-
ducted worldwide, developing countries have been
urged to draft strategic plans to confront HIV/AIDS.
These plans naturally require an initial assessment of the
current response to the epidemic and a continuous
evaluation of the implementation process, results and
achievements . On the other hand, and mainly as a
result of initiatives for reforming healthcare systems
worldwide, several tools for helping decision-makers
have been developed, for example, the National Health
Accounts (NHA), which systematically measure the
resources and flows from the original sources of funds,
the economic agents and providers of services, for care
and prevention [13–17]. The use of this tool can be
adapted to the in-depth study of expenditure on
specific diseases , such as HIV/AIDS and other
catastrophic diseases because of their economic impact
on the households of the non-insured ill.
Therefore, the purpose of the National HIV/AIDS
Accounts is to quantify all expenditures on HIV/AIDS
using NHA methodology. This accounting system
provides information aimed at decision-making and
more efficient resource allocation; it quantifies the
sector economic impact of the HIV/AIDS epidemic on
the health system and constitutes an adequate assess-
ment of countries’ financial responses to confront
HIV/AIDS . The objective of this report is to
describe the levels and flows of expenditures on HIV/
AIDS in five Latin American countries in 1997–1999.
The National HIV/AIDS Accounts consist of a sys-
tematic, periodic and exhaustive accounting of health
expenditures and financing flows related to activities on
HIV/AIDS, using the already validated NHA method-
ology [20–24]. The sources of financing are the
economic institutions or agents that provide the re-
sources to prevent HIV infection and care for those
affected by HIV or AIDS. The expenditure level is
measured on annual bases and is grouped in three
levels: from resources of financing, through the funds
or financing agents, to the institutions managing the
The level of expenditures is estimated utilizing second-
ary sources of information, official reports, and by
usually allows an estimate to be made of public
expenditure, whereas private expenditure estimation is
usually complemented with surveys. Private expendi-
ture includes institutional contributions and household
expenditures on drugs, the expenditure of additional
companies on social security in health services for its
employees, private health insurance and out-of-pocket
There are five matrices to concentrate the estimation of
financial flows and expenditures. The same approach is
accounts: (i) from the source of financing to the funds;
(ii) from the funds or financing agents to the provider
institutions; (iii) from the provider institutions by the
type of service; (iv) by the category of expenditure and
provider institution; and (v) by the category of expen-
diture and type of service.
The source of financing (government, social security,
households, external cooperation) refers to those enti-
ties that contribute with financial resources to the
prevention, diagnosis, and management of HIV/AIDS.
The financing agents (government, households) are
entities that concentrate and arrange the resources, and
the provider institutions use the resources in generating
services to the population. The funds from external
sources are classified according to their status as
reimbursable or non-reimbursable resources, i.e. reim-
bursable funds, such as loans, are considered domestic
resources, whereas donations or contracts are consid-
ered external resources.
Flow from sources to financing agents
This matrix registers the origin and amount of the
National HIV/AIDS Accounts in LAC Izazola-Licea et al.S67
contribution of each source of financing as well as the
destination of these resources towards the financing
agents. The sources of financing of the sector identified
domestic and international non-governmental organiza-
tions (NGO). In order to fill in the matrix, the
expenditures are disaggregated according to the source
of financing in private (households and companies) and
public (governmental or public social security).
Resources from agents to provider institutions
This matrix registers the flow of resources of each of
the financial agents towards the institution that delivers
the service. Service providers and financing agents
constitute the dimension of this matrix, financing
agents include the social security system and public and
private agents. Each transfer should be quantified in
order to establish the profile of the public–private,
public–public or private–private mix of each institu-
Institutional expenditures according to type of
The matrix presents the breakdown of expenditures
performed by each service provider institution in each
programme. Similar to estimating NHA, the first step is
to identify all the programmes and institutions that
carry out health activities on HIV/AIDS. The dimen-
sion of the matrix is that each institution must share a
generic name: public services for insured individuals
(social security services); public services for uninsured
individuals (public services); and private healthcare
Provider institution by type of expenditure
The dimension of the matrix is formed by the same
service provider institutions described above as well as
by the classification as an objective of expenditure, such
as personal services, materials and supplies, general
Expenditure by programme according to type of
The total expenditure must be consistent throughout
all the matrices, and the distribution of total expendi-
ture will change in each matrix. Frequently, the
services that have registries lack data on expenditure;
the principal effort consists of carrying out the collec-
tion of existing data. HIV/AIDS expenditures may or
may not be classified as health expenditures according
to their nature, but are included in the total .
For comparative purposes the financial flows and the
expenditures were analysed in current US dollars, as
well as their adjusted value as a result of purchasing
power parity (PPP) according to standard methods set
by the World Bank Development Reports. The adjust-
ment caused by PPP provides monetary units of
international US dollars, i.e. adjusting for differentials
in inflation and exchange rates; in other words, adjust-
ing for the cost of living in each country. The results
for the estimations for Brazil, Guatemala and Mexico
are for 1997 and 1998; the results for Uruguay are for
1998 and those for Honduras are for 1999. Emphasis
was placed on the comparison across countries; more
in-depth results of each country project might be found
in the published final reports for each country and year
The data analysis is descriptive, allowing the identifica-
tion of flows of resources as well as of the major actors
of the national response towards HIV/AIDS, by means
of the identification of the major providers of services,
for example social security institutions. To analyse the
financial response to HIV/AIDS and evaluate the
patterns of expenditures critically, we combined the
final figures with other economic, demographic, and
epidemiological data, in order to build useful and
complex indicators for decision making.
Total national HIV/AIDS expenditures estimated for
the year 1998 were US$PPP29.5 million in Guatemala,
US$PPP32.5 million in Uruguay, US$PPP257 million
in Mexico, and US$PPP587.4 million in Brazil, and for
the year 1999 US$PPP33.9 million in Honduras. All
the results were interpreted in the context of the
demographic and epidemiological situation in the
country. Table 1 shows that the health investment
varies widely across the five countries under study, and
presents demographic, epidemiological and economic
data, per capita expenditure on HIV/AIDS crude and
adjusted by PPP, and the relative distribution of the
health expenditure allocated to HIV/AIDS [31–33].
Expenditure on care and prevention
The expenditures allocated to personal healthcare ser-
vices represented the majority of the expenditures in
each country, as opposed to prevention; averaging 66%
in the estimations of the five countries across the period
of study (Table 2). Conversely, expenditures on pre-
ventative services averaged less than 40% (Table 3).
Other expenditures were devoted, for example, to
institutional development, advocacy or mitigation. Ad-
ministration costs were divided into their share attribu-
table to institutional expenditures, or were included
within the prevention and care estimates.
Uruguay had the highest unadjusted per capita expen-
diture, but after adjusting for PPP it ranked second to
Honduras. The largest share of Uruguay’s expenditures
AIDS 2002, Vol 16 (suppl 3) S68
(62%) was devoted to care, which is not surprising as
a result of their policy of universal access to antire-
troviral drugs (ARV). Uruguay’s expenditure was
executed mainly by private health providers (56%) and
by direct government (43%), which is consistent with
the operations mechanism of its health system and the
overall investment in health. The change in ranking
after adjusting by PPP to the second position is
consistent with the high prices Uruguay pays for
technological inputs and in general with a higher cost
Honduras had the highest per capita expenditure after
adjusting by PPP, partly explained by a larger number
of AIDS cases and significant international financial
cooperation. In contrast to Uruguay and Brazil, where
AIDS patients have universal access to ARV, in
Honduras the access to ARV was minimal. Most of the
provision of care was palliative, including hospitaliza-
tion and ambulatory care mainly by private health
providers (63%), and to a minor extent by public
providers (27%) and NGO (10%). In Honduras, ARV
were not provided using public resources; the provision
of ARV through private sector agents represented 7%
of the out-of-pocket expenditures used obtaining ser-
vices from private for-profit providers; the provision of
ARV accounted for the totality of NGO expenditures
Table 1. Background demographic, economic and epidemiological information and national expenditure on HIV/AIDS in Brazil, Guatemala,
Honduras, Mexico and Uruguay.
Brazil GuatemalaMexicoUruguay Honduras
199819971998 1997 199819981999
GNP (billion US$)
GNP (billion US$PPP)c
GNP per capita (US$)
GNP per capita (US$PPP)
Cumulative AIDS cases
by end of yeard
Total health expenditure per
Total health expenditure per
Total health expenditure as
percentage of GDP (%)
Public health expenditure as
percentage of GDP (%)
Total expenditure on HIV/
AIDS (million US$)
Total expenditure on HIV/
AIDS (million US$PPP)
HIV/AIDS expenditure per
HIV/AIDS expenditure per
Percentage of HIV/AIDS
expenditures in care
Percentage of HIV/AIDS
expenditures in prevention
from external sources
Percentage of HIV/AIDS funds
from external sources
– 309– 78–202 62174
– 453– 155–371 823 210
435.8 5.911.989.7 124.621.2 22.1
587.415.1 29.5 197.9 257.032.5 33.9
–0.87%–1.38%–0.62%1.06% 4.86 %
73%81%63%70%55% 66% 51%69%
13%10%25% 15%38% 29% 36% 29%
GDP, Gross domestic profit; GNP, gross national profit.
aEstimated expenditures for Brazil are only of direct government expenditures.
bInstituto de Pesquisa Economica (IPEA) .
cPurchasing power parity (PPP) based on the World Bank development reports for each year.
dBoletı ´n de Vigilancia del SIDA en las Ame ´ricas .
eMinisterio da Saude .
fHealth expenditures are for 1998 or the closest estimate. World development indicators; Health expenditure and use; Health nutrition and
population. http://www.worldbank.org/data/wdi2001/pdfs/tab2 15.pdf. Accessed 29 November 2001.
gThe Brazilian team did not estimate the external non-reimbursable funds for Brazil in 1997 and 1998. Another study by J Rojas, unpublished,
estimated an average of US$24 million non-reimbursable flow from bilateral and multilateral sources per year, or approximately 7 and 5.5% per
year in 1997 and 1998, respectively. The World Bank loan was considered as a domestic fund because it was reimbursable.
National HIV/AIDS Accounts in LAC Izazola-Licea et al.S69
As might be expected, the largest bulk of the expendi-
tures by far was for ARV. The category labelled ‘other’
included research, infrastructure and managerial expen-
In the three countries with estimates for two consecu-
tive years, their expenditure was increased. The largest
increase in the total HIV/AIDS expenditure was ob-
US$5.9 million in 1997 to US$11.9 million in 1998.
This increment in expenditures is mainly explained by
an increase in hospitalizations and antiretroviral therapy
by the social security system (from US$1.7 million
in 1997 to US$5.7 million in 1998). A much-limited
increase was observed in preventative expenditures. A
similar but more moderate increase in expenditures for
HIV/AIDS was observed in Mexico from the first to
the second year of the estimation, also attributed
mainly to a mild increase in ARV coverage.
In Brazil, the increased expenditures in 1998 compared
with 1997 were attributable to condoms, which was an
item missing in the 1997 estimation. Condom expendi-
tures reached US$100 million in 1998 as out-of-pocket
expenditure through for-profit private providers (e.g.
pharmacies). The rest of the figures remained relatively
stable in Brazil, but the expenditure on ARV increased
fromUS$167 millionin1997toUS$260 millionin1998,
showing only a small decrease in expenditures caused by
Prevention activities rely mainly on services or products
provided by the private sector in Brazil, Guatemala and
Uruguay, such as the purchase of condoms in pharma-
cies, which are paid as out-of-pocket expenditures. In
Mexico, these activities are still a significant portion of
the expenditures on prevention, but they are surpassed
by blood bank screening in public institutions, mainly
by the social security institutions. In Honduras, an
Table 2. HIV/AIDS care estimated expenditures by type of service and provider institution in Brazil, Guatemala, Honduras, Mexico and Uruguay
Provided by public sector Provided by private sectora
Curative servicesSocial securityDirect government For profit NGO
PPP, Purchasing power parity; n/a, not available.
aThese expenditures are paid for by out-of-pocket expenditures, pre-paid health plans or private insurance when using for-profit private providers
(i.e. physicians office, hospitals, etc.), and when using non-governmental organization (NGO) services they usually are paid as out-of-pocket
expenditures or subsidized by the NGO.
AIDS 2002, Vol 16 (suppl 3)S70
important proportion of the expenditures on preven-
tion are services provided by the public sector and are
allocated for information, education and communica-
tion (IEC) activities and the purchase of condoms,
followed closely by condoms paid out of pocket.
The subsidy for the health sector (public expenditure
divided by the total expenditure on health) varies
widely between countries, and ranges from a low 21%
subsidy in Uruguay (1.9% of the gross domestic product
from public sources) to a high 60% in Mexico (2.8% of
the gross domestic produce from public sources, Table
1). The subsidy for HIV/AIDS (or the public share of
the total expenditure) was 68% for Brazil in 1998; 60%
in 1997 and 72% in 1998 for Guatemala; 77% in 1997
and 84% in 1998 for Mexico; 34% for Honduras in
1999; and 30% for Uruguay in 1998. The high public
expenditure shares are associated with the provision of
ARV, except in Honduras where no ARV are pro-
vided, and in Uruguay where there is a high expendi-
ture on condoms from out-of-pocket sources.
Most of the public HIV/AIDS expenditure remains
centralized and is managed at the national or federal
level. In Brazil, only 31% in 1997 and 29% in 1998 of
the total HIV/AIDS public expenditure was exercised
at the states and municipalities. The decentralized
expenditures in Brazil included drugs for treating
Table 3. HIV/AIDS prevention expenditures by type of service and provider institution in Brazil, Guatemala, Honduras, Mexico and Uruguay (in
Provided by public sectorProvided by private sectora
PreventionSocial security Direct governmentFor profitNGO
PPP, Purchasing power parity.
aPreventative activities are paid for as out-of-pocket expenditures when using for-profit providers (i.e. pharmacies or private laboratories for HIV
testing and counselling), whereas using non-governmental organization (NGO) services, they could be subsidized by the NGO or paid
completely or partly from out-of-pocket expenditures.
bThe Brazil study did not estimate the private expenditures of information, education and communication (IEC) for 1997.
cThe Brazil study did not estimate the private expenditures of condoms for 1997.
National HIV/AIDS Accounts in LAC Izazola-Licea et al. S71
opportunistic infections, HIV screening in blood banks,
testing and counselling, IEC activities, and to a small
extent the purchasing of ARV as a counterpart of the
federal provision. Federal expenditure was executed
mainly through the Brazilian unified health system
(Sistema Unico de Saude, SUS).
In Mexico, only 19% in 1997 and 14% in 1998 of the
total HIV/AIDS public expenditure was decentralized,
covering mainly the costs of care through social
security institutions, HIV screening in blood banks and
a limited number of IEC activities. In Guatemala,
Honduras and Uruguay there is no information avail-
able for the years under study.
The relative distribution of HIV/AIDS expenditures
might reflect the pattern of a national response in a
given country (see Table 2 and Table 3). Brazil in 1998
allocated the largest amount of expenditures paid from
public sources to ARV (56%), followed by private
expenditures on condoms (22%) and blood banks,
which accounted for almost US$23 m or 4.9% of the
In Mexico, during 1997 most of the expenditures were
allocated to HIV/testing (30.5%) and blood bank
testing (30.4%), followed by expenditures on ARV
(19.7%), the treatment of opportunistic infections
(14.4%), and private expenditures on condoms (11.2%).
In 1998, the largest item was ARV (43.1% from
public sources and 3.4% out of pocket), blood banks
accounted for US$21.5 million (21.2%) and private
expenditures on condoms accounted for US$8 million
(7.9% of the total expenditures).
In 1997, Guatemala allocated half of the HIV/AIDS
expenditure (50.5%) to hospital admissions (without
ARV) paid from public sources, 23% was spent on
ARV from private sources (or US$1.05 million) and
US$693100 was spent on condoms from out-of-pocket
sources. In 1998, 37% was spent on hospitalization
using public sources; expenditures on ARV accounted
for 32.1% from public sources (US$3.16 million) and
13.6% from private sources (US$1.3 million). Only
16% of the public and private expenditures were spent
on IEC and condoms (or US$1.6 million).
In 1998, Uruguay spent 59% out of the total on care
(e.g. hospitalization, ARV, treatment of opportunistic
infections and ambulatory care); most of these expendi-
tures were for services provided by private agents (33%
of the total) compared with the proportion provided
by the public sector (26%). Almost US$7.3 million was
paid out of pocket in condoms (37% of the total), and
only US$600000 of condoms were purchased through
In 1999, Honduras spent 72% on care; 24% of these
expenditures on hospital admissions to public services
(US$2.9 million) and 28% on hospital services through
private providers (US$3.5 million). Twenty-three per
cent of the total was spent out of pocket on ARV
(US$2.8 million). Blood bank testing through public
services represented 2.3% of the total. Expenditures on
condoms totalled US$485100 through public services
and US$1.5 million from private sources.
The results of these case studies document the levels
and flow of resources for HIV/AIDS in five Latin
American countries. Very few attempts have been
focused on the level of monetary resources being
mobilized and used within countries in the fight against
AIDS. An early World Bank and European Commis-
sion publication focused on the public policies to cope
with HIV/AIDS in developing countries. That report
analysed the levels and determinants of the expenditure
on HIV/AIDS in Tanzania, Ivory Coast, Thailand,
Brazil and Mexico [34,35].
Information on the economic funding of countries’
responses to the epidemic is sparse or incomplete at
best. Several approaches have been taken to estimate
the resources needed to confront the epidemic, and
have been focused on costing specific preventative
interventions and more frequently estimating the cost
of providing care in developed and developing coun-
The estimated funds needed for LAC were US$1,140
million annually to curb the HIV/AIDS epidemic by
2005 (US$590 million prevention,
care); our report on five countries shows expenditures
of US$755.8 million in 2000 (US$350 million preven-
tion, US$406 million care). Although insufficient, these
levels of resources clearly represent the national coun-
tries responses towards HIV/AIDS, and thus the
willingness of governments and the ability of house-
holds to subsidize and pay for prevention and care. The
estimates for 2005 might show a decrease in care costs,
mainly caused by the price of ARV or an under-
estimation of the resources needed.
The lowest share of public expenditures in the five
countries was in prevention, ranging from 20% in
Brazil to 49% in Uruguay. The steady domestic in-
crease in private expenditures on prevention (e.g.
condoms) more likely constitutes an important achieve-
ment of the National AIDS Programmes (NAP),
whereas the large portion of care expenditures from
private sources demonstrates governments’ inability to
provide services for this catastrophic disease that has
devastating effects on households.
AIDS 2002, Vol 16 (suppl 3)S72
Most of the actions to prevent HIV infections in these
countries are specifically executed by the NAP,
whereas most care actions occur within the existing
healthcare system. It is of vital importance to quantify
the public and private funds these programmes manage
in order to optimize their allocation and to estimate the
balance of the public–private mix of funding and the
provision of these activities. Unfortunately, only a
tenth of the countries affiliated to the World Health
Organization (WHO) have a recurrent flow of financial
data for health, as was reported for the NHA system for
191 countries published by WHO.
Our observations are descriptive and are not intended
for prescriptive purposes; in fact, the data say nothing
about the right level of expenditures. However, they
allow international comparisons and are useful as a
reference, because they are based on a standard
methodology. The economic development of these
countries is contrasting, and ranges from a per capita
gross national product (GNP) of US$760 in Honduras
to US$9480 in Uruguay. The participating countries
have different health system organizations and responses
to the epidemic: such as Brazil whose governmental
programme is considered to be very successful; Guate-
mala whose response to HIV/AIDS is weak with
varying degrees of leadership; or Honduras with a
highly dependent programme on external cooperation.
To analyse the financial response to HIV/AIDS, the
expenditures were combined with other economic,
demographic and epidemiological data to build useful
indicators to identify policy decisions. We found that
the relative distribution of the health expenditure
allocated to HIV/AIDS ranged from 0.6% in Mexico
to 4.9% in Honduras. Despite the formidable amount
of resources spent in Honduras, the access to ARV is
minimal and most of the provision of care is palliative,
paid for by individuals. Interestingly, the per capita
expenditure on HIV/AIDS is apparently unrelated
to the contextual indicators of GNP. The unadjusted
per capita expenditure on HIV/AIDS ranged from
US$0.53 (1997) in Guatemala to US$6.63 in Uruguay.
The adjustment as a result of PPP indicates that
countries with relatively similar economic levels de-
crease their differences when measured in per capita
US$PPP. In 1998, Brazil, Guatemala and Mexico spent
US$PPP3.62, 2.68 and 2.57, respectively; but the
expenditures of Uruguay and Honduras, the richest and
the poorest countries, appear to be much further apart
from the other countries than was observed on the
unadjusted estimates (US$PPP10.17 and 11.29, respec-
Many developing countries with very limited domestic
resources for health finance their activities against
HIV/AIDS with external sources; World Bank loans
have become a major source of funds for middle-
income countries in the region. External cooperation is
another important source of funding, such as in
Guatemala, Honduras, and many others in the region.
The Joint United Nations Programme on HIV/AIDS
(UNAIDS) reported that the HIV/AIDS funding that
official organization agencies channelled through multi-
lateral organizations dropped from over 70% in 1987 to
22% in 1997 . The estimated flow from international
sources per capita ranged from US$0.01 in Mexico to
US$1.04 in Honduras in 1998. The World Bank
(IBRD) loan to Brazil provided US$44.6 million and
US$41.6 million for 1997 and 1998, but such funds are
considered domestic because they are reimbursable.
Brazil has an average non-reimbursable flow of US$5
million from external bilateral and multilateral sources
The quality of the financial information on HIV/AIDS
services depends on the availability of information and
the quality of the sources of information or estimations.
In practically all cases we use different formats of
presentation. Until now the estimation of expenditures
for a specific programme was relatively easy when the
value of an average service is estimated or known and
is then multiplied by the total volume.
The household registry of expenditures, frequently used
for private expenditure estimation, has the problem of
lacking specificity, e.g. income and expenditure surveys
do not have information on every disease, just for
health expenditures. In this case it is advisable to seek
specific indicators to disaggregate the expenditure. Such
indicators can be obtained from ad hoc studies on the
market prices of products and services that households
consume when they have patients with a specific
disease, according to their socioeconomic character-
The estimates presented here are based on the approach
developed by the Harvard School of Public Health for
NHA. However, other recently developed method-
ologies might facilitate data abstraction and the inter-
pretation of the results by NAP managers and other
HIV/AIDS stakeholders. For example the Organization
for Economic Cooperation and Development (OECD)
approach provides information about health expendi-
tures in a comparable way across countries. At the
beginning of 2000, the OECD published a version of
the NHA system that seeks to generate a set of
comparable categories, therefore providing an inte-
grated, consistent and flexible system of accounts. It
establishes a conceptual basis of statistical reporting
rules, and proposes an innovative international classifi-
cation of health accounts covering three dimensions:
healthcare by health functions, healthcare providers,
and sources of financing. Our results are estimates of
the first five countries of a large-scale ongoing project
that includes 21 LAC countries, which in a second
National HIV/AIDS Accounts in LAC Izazola-Licea et al. S73
phase are incorporating the methods proposed by the
OECD, thus offering a more comparable presentation
of country results with a fixed and exhaustive classifica-
tion system. This method also allows the accounting of
non-health expenditures, such as advocacy and lobby-
ing, educational programmes in schools, food supply,
hospices and shelters, etc.
An additional advantage of this accounting method of
expenditures is that the reporting categories are better
suited for policy analysis. The Latin American countries
are characterized by having social security systems that
cover an important proportion of salaried workers and
civil servants. Despite a coverage lower than 50% of
the population in most countries, these systems provide
antiretroviral therapy for a substantial number of people
with AIDS in the region. In fact, the major component
of public expenditure in the five countries was allo-
cated to care services, ranging from 51% in Mexico to
80% in Brazil. Antiretroviral therapy, far from univer-
sal, accounted for 63% of the expenditures on curative
services, and its coverage for AIDS patients ranged
from 10% in Guatemala to universal in Brazil. The
major sources of these funds were households and
enterprises, as they supply the resources for the social
security system (flows from sources to agents or
providers). However, the decision to provide ARV
resided within the social security system as the health
provider, without participation or knowledge of the
sources of the funds, such as households’ and compa-
nies’ fees directed to these institutions.
One of the strengths of the data is the application of
the same methodology in the different countries, thus
allowing international comparisons. However, we have
to acknowledge that our estimation method is not free
of limitations. Information available in most countries
usually allows a more accurate estimation of public
expenditure, whereas private expenditure estimation is
usually complemented with surveys, for the most part
outside the health sector. The major area of uncertainty
about the precision of our estimates comes from private
expenditure, mainly in estimating household and out-
In conclusion, our results are relevant for the planning
process in the health sector, both at the institutional
and national level and document governmental deci-
sions. Also, identifying the level of expenditures will
establish the need to generate alternative sources of
financing. These results also provide a series of financial
indicators describing the country’s response to HIV/
AIDS, including governmental, non-governmental and
international cooperation activities. As the data allow
international comparison, the National HIV/AIDS
Accounts accurately portray the current situation of the
national response towards HIV/AIDS, and can there-
fore easily be adopted as a measure for benchmarking
and as a tool for implementing national AIDS strategic
1.UNAIDS/WHO. Level and flow of national and international
resources for the response to HIV/AIDS, 1996–1997. Geneva:
Joint United Nations Programme on HIV/AIDS (UNAIDS)/World
Health Organisation (WHO); 1999.
Camara B, Nicholls S, Malean R, et al. Creacio ´n de modelos y
proyecciones de la epidemia del VIH/SIDA y su impacto
econo ´mico en el Caribe. La experiencia de Trinidad y Tobago y
Jamaica. [Models and projections of the HIV/AIDS epidemic
and its economic impact in the Caribbean. The experience in
Trinidad and Tobago and Jamaica.] In: Situacio ´n epidemiolo ´gica
y econo ´mica del SIDA en Ame ´rica Latina y el Caribe. [Epidemio-
logical and economical situation of AIDS in Latin America and
the Caribbean.] Izazola-Licea JA (editor). Mexico City: Fundacio ´n
Mexicana para la Salud; 1998. pp. 169–244.
Izazola-Licea JA. Una aproximacio ´n al ana ´lisis costo-beneficio
de la prevencio ´n de la transmisio ´n vertical del VIH en el Cono
Sur. [A preliminary approach to a cost-benefit analysis for the
prevention of vertical HIV transmission in LAC Southern Cone.]
In: Prevencio ´n de la transmisio ´n vertical de VIH/SIDA. Taller
Subregional. [Prevention of mother to child HIV transmission.
Subregional Workshop.] Buenos Aires, Argentina: Ministerio de
Salud y Accio ´n Social, Repu ´blica Argentina/UNAIDS/SIDALAC;
Foster S, Buve A. Benefits of HIV screening of blood transfusions
in Zambia. Lancet 1995, 346:225–227.
Gilson L, Mkanje R, Grosskurth H, et al. Cost-effectiveness
analysis of improved treatment services for sexually transmitted
diseases in preventing HIV-1 infection in Kwanzaa Region,
Tanzania. Lancet 1997, 350:1805–1810.
Holtgrave DR, Qualls NL, Graham JD. Economic evaluation of
HIV prevention programs. Annu Rev Public Health 1996,
Kahn J. The cost-effectiveness of HIV prevention targeting: how
much bang for the buck? Am J Public Health 1996, 86:1709–
Kumaranayake L, Pepperall J, Goodman H, Mills A, Walker D.
Costing guidelines for HIV/AIDS prevention strategies. Economy
and Financing Program of Health. London: School of Hygiene
and Tropical Medicine; 1998.
Simpson KN. Design and assessment of cost-effectiveness studies
in AIDS populations. J Acquir Immune Defic Syndr Hum Retro-
virol 1995, 10 (suppl 4):528–532.
Soderlund N, Lavis J, Broomberg J, Mills A. The cost of HIV
prevention strategies in developing countries. Bull WHO 1993,
Schwartlander B, Stover J, Walker N, et al. Resource needs for
HIV/AIDS. Science 2001, 292:2434–2436.
Bronfman M, Herrera C. Planificacio ´n estrate ´gica e investigacio ´n
en Cuentas Nacionales en VIH/SIDA: avances en la region
latinoamericana y del Caribe. [Strategic planning and research
on National HIV/AIDS Accounts: advancements in Latin Amer-
ica and the Caribbean.] Mexico City: Fundacio ´n Mexicana para
la Salud/ONUSIDA; 2001. pp. 325–342.
Zschock DK. Health care financing in developing countries.
Monograph series No. 1. Washington DC: American Public
Health Association; 1979.
Mach EP, Abel-Smith B. Planning the finances of the health
sector: a manual for developing countries. Geneva: World Health
WHO. Evaluation of recent changes in the financing of health
services: report of a WHO study group. WHO technical report
series 829. Geneva: World Health Organization; 1993.
Abel-Smith B. Paying for health services: a study of the costs and
sources of finance in six countries. Public health papers No. 17.
Geneva: World Health Organization; 1993.
Waldo DR. Creating health accounts for developed and develop-
ing countries (draft). Human Development Department. Washing-
ton, DC, USA World Bank; 1996.
Shepard DS. Levels and determinants of HIV/AIDS expenditures
AIDS 2002, Vol 16 (suppl 3)S74
in five developing countries. In: Confronting AIDS: evidence
from the developing world. Ainsworth M, Fransen L, Over M
(editors). Selected background papers for the World Bank Policy
Research Report. Confronting AIDS: public priorities in a global
epidemic. Luxembourg: The European Commission; 1998. pp.
Avila-Figueroa C. The AIDS epidemic and the health sector
reform. In: AIDS in Latin America and the Caribbean: a multi-
disciplinary view. Izazola-Licea JA (editor). Mexico City, Mexico
FUNSALUD/SIDALAC, UNAIDS; 1999.
Berman PA, Cooper DM. National Health Accounts handbook.
Data for decision making project. Boston, MA: Harvard School of
Public Health; 1996.
OECD. A system of health accounts for international data
collection, Version 1.0. Beta version for publication and pilot
implementations (draft) 28.1.2000. Geneva: OECD Health Policy
Rannan Eliya RP, Berman PA. National Health Accounts in
developing countries: improving the foundation. DDM publica-
tion No. 2. Boston MA: Harvard School of Public Health; 1993.
Berman PA. National Health Accounts in developing countries:
appropriate methods and recent applications. Data for decision
making project and partnership for health reform. Boston, MA:
Harvard School of Public Health; 1996.
Poullier JP, Hernandez P. Estimates of National Health Accounts
(NHA) for 1997. GPE discussion paper series No. 27. EIP/GPE/
FAR. Geneva: World Health Organization; 2000.
Foulon A. Proposal for a homogeneous treatment of health
expenditures in the National Health Accounts. Rev Income
Wealth 1982, 28:45–70.
Piola S, Nunes J, Teixeira L, Conde F. Cuentas Nacionales de
VIH/SIDA. Flujos de financiamiento y gasto en VIH/SIDA en
Brasil 1997–98. [National Health Accounts on HIV/AIDS. Flows
of financing and expenditure for Brazil 1997–1998.] Izazola-
Licea JA (editor). FUNSALUD/SIDALAC, UNAIDS; 2000. Also
available at http://www.sidalac.org.mx/
Perdomo R. Cuentas Nacionales de VIH/SIDA. Flujos de finan-
ciamiento y gasto en VIH/SIDA en Honduras 1997–98. [Na-
tional Health Accounts on HIV/AIDS. Flows of financing and
expenditure for Honduras 1999.] Final Report of research
project. Unpublished. Also available at http://www. sidalac.
Valladares R, Barrillas E. Cuentas Nacionales de VIH/SIDA.
Flujos de financiamiento y gasto en VIH/SIDA en Guatemala
1997–98. [National Health Accounts on HIV/AIDS. Flows of
financing and expenditure for Guatemala 1997–1998.] Izazola-
Licea JA (editor). FUNSALUD/SIDALAC, UNAIDS; 2000. Also
available at http://www.sidalac.org.mx/
Savedra J, Redorta E, Avila-Figueroa C. Cuentas Nacionales de
VIH/SIDA. Flujos de financiamiento y gasto en VIH/SIDA en
Me ´xico 1997–98. [National Health Accounts on HIV/AIDS.
Flows of financing and expenditure for Mexico 1997–1998.]
Izazola-Licea JA (editor). FUNSALUD/SIDALAC, UNAIDS; 2000.
Also available at http://www.sidalac.org.mx/
Ara ´n D, Berriolo R. Cuentas Nacionales de VIH/SIDA. Flujos de
financiamiento y gasto en VIH/SIDA en Uruguay 1997–98.
[National Health Accounts on HIV/AIDS. Flows of financing
and expenditure for Uruguay 1998.] Izazola-Licea JA (editor).
FUNSALUD/SIDALAC, UNAIDS; 2000. Also available at http://
31.Instituto de Pesquisa Economica (IPEA). Economic indicators.
http://www.ipeadata.gov.br. Accessed 3 December 2001.
Boletı ´n de Vigilancia del SIDA en las Ame ´ricas, Washington, DC,
USA, Organizacio ´n Panamericana de la Salud; 2001.
Ministerio da Saude. Boletim epidemiologico, ano XIV, No. 1,
January–March 2001. Brasilia, DF, Brazil, Ministerio da Saude.
World Bank. Confronting AIDS: public priorities in a global
epidemic. A World Bank Policy Research Report. Washington,
DC: Oxford University Press, Inc.; 1997.
Ainsworth M, Fransen L, Over M (editors). In: Confronting AIDS:
evidence from the developing world. Selected background papers
for the World Bank Policy Research Report. Confronting AIDS:
public priorities in a global epidemic, Part III, Chapter 12,
Sections A–F. Luxembourg: The European Commission; 1998.
Scitovsky A, Rice D. Estimates of the direct and indirect costs of
acquired immunodeficiency syndrome in the United States,
1985, 1986 and 1991. Public Health Rep 1987, 102:5–17.
Appendix of additional study sites and
The methodological approach was developed by Jose
A. Izazola-Licea MD, and Carlos Avila-Figueroa, MD,
DSc, with continuous support from Patricia Hernan-
dez, MD, PhD, currently working for WHO.
The participants in each country project are listed
according to authorships in their final project reports.
Brazil: Fabricio Conde, Luciana Da Silva Teixeira, Joel
Sadi Dutra Nunes, Sergio Francisco Piola. Acknowl-
edgements: Julio Barrios, Alexandre Grangeiro, Claudia
Cunha, Elisa Cazue Sudo, Dulce Maria Moreira,
Rosemarie Munoz, Maria do Carmen Soares Pereira,
Ferreira Pinto, Mamad De Souza, Naira De Bem
Alves, Francisco Dos Santos.
Guatemala: Edgar Barrillas, Ricardo Valladares.
Mexico: Carlos Avila, Esthela Redorta, Jorge Saavedra.
Uruguay: Daniel Aran, Ruben Berriolo.
Honduras: Rodulio Perdomo. Acknowledgements:
Ramon J. Soto, Rosa Linda Hernandez, Marco Alvar-
National HIV/AIDS Accounts in LAC Izazola-Licea et al.S75