|
About ADF 2000
Partners
ADF '99
|
|
ADF
2000 Daily Updates
Home > Documents
> Costs of Scaling HIV
Programme Activities to a National Level in Sub-Saharan Africa: Methods and
Estimates
PREVIOUS PAGE < CONTENTS > NEXT PAGE
Annex 2
Calculation
of unit cost data
Financial costs represent
actual expenditure on goods and services purchased. Costs are thus described in terms of
how much money has been paid for the resources used in the project or service. Economic
costs include the estimated value of goods or services for which there were no financial
transactions or when the price of the good did not reflect the cost of using it
productively elsewhere (including donated goods and services and other inputs whose prices
may be incorrect). A full cost analysis estimates the costs of all resources that are
being employed in running a project or programme, including basic infrastructure. An
incremental analysis looks at the cost of adding or implementing the additional project or
programme to existing services.
For the calculation of the unit
cost data, all prevention and basic care costs are from the perspective of the provider.
Where possible, economic and full costs were obtained. Unless stated all unit costs
include recurrent and annualized costs for capital inputs. A detailed review regarding the
costs of prevention is found in Kumaranayake and Watts (2000b). Countries which were
classified as high income for the purposes of the analysis were Botswana, Djibouti, Gabon,
Mauritius, Namibia, South Africa, and Swaziland.
| |
|
TABLE A2: Description of Source of Unit Cost
Estimates |
Youth
Interventions
· Cost per teacher trained primary school
education
· Cost per teacher trained secondary school
education
· Cost per youth targeted/peer-education for
out-of-school youth |
Note: 1. For
all cost scenarios, activities are primarily coordinated through the ministry of
education. 2. No available information regarding the costs of primary school education,
other than approximations by Boerma and Bennett (1997) for Tanzania. Low-cost: $75; simple programme with teacher
training and provision of basic material
Medium: $200; includes the development
of training materials and establishment of school curricula
Low: $121; simple programme ; assumed
to be financial and incremental cost; (Boerma and Bennett, 1997); Tanzania
Medium: $241; more extensive
programme; assumed to be financial and incremental cost; Tanzania.
Assumed a peer-education programme in place
for out-of-school youth. Given the lack of data, it was assumed that the costs would be
higher than a workplace intervention, but lower than a CSW intervention because the
clients are easier to reach than with the latter. Thus the cost calculations are a
straight average of the relevant scenarios for the CSW and the workplace peer-education
programmes.
Low: $8.00
Medium: $10.81 |
Sex
Worker Interventions
· Cost per CSW targeted |
Peer-education
project in Cameroon, educators not salaried and condoms not distributed freely. Economic
and full costing. (Kumaranayake et al, 1998). Low: $15.83
Medium: $21.12 |
| · Cost per male
condom distributed, urban |
Costs from a CSW
programme in Zimbabwe. Assumed to be economic costing (Soderlund et al 1993). This
provided a figure for the medium scenario, extrapolated for low-cost. Low: $0.10
Medium: $0.14 |
| · Cost per female
condom distributed |
Commodity and
marketing/distribution costs from existing Population Services International (PSI) CSM
programmes in Zambia and Zimbabwe. (Source: personal communication, Guy
Stallworthy). Financial and incremental costs. Low: $1.00
Medium: $2.00
These figures include estimates of market and
distribution costs associated with the female condom. The negotiated wholesale price for
the female condom is about $0.64 in year 2000 prices. Marketing is much more intensive for
the female condom, relative to the male condom. |
| Strengthening
public sector condom distribution ·
Cost per male condom distributed in the public sector
· Cost per male condom for strengthening
condom logistics |
Taken from Zambia,
where there was free distribution of condoms through public channels. Economic and full
costing (Goodman and Watts, 1995). Both urban and rural costs are assumed to be the same. Low: $0.10
Medium: $0.34
Low: $0.045
Medium: $0.07 |
Condom
social marketing
· Cost per male condom distributed urban
· Cost per male condom distributed rural
· Cost per female condom distributed |
Figures were taken
from Stallworthy and Meekers (1998), which presented range of costs for PSI's CSM
programmes by low, medium and high-cost programmes in 1996 dollars. Assumed to be
financial and full cost. Low:
$0.12
Medium: $0.29
Low: $0.25
Medium: $0.45
Same as for cost per female condom
distributed in CSW interventions. |
Strengthening
STD services
· Cost per STD case treated/visited
(syndromic treatment) |
Low: $12.65;
intensified intervention through existing health services in Tanzania with syndromic
management; economic and incremental (Gilson et al, 1998) Medium: $15; integrated STD/HIV control
programme in Mozambique; included costs for pre-consultation, partner notification,
syndromic management, but excluded planning and management; assumed to be financial and
incremental (Bastos et al, 1992) |
| |
|
|
|
|
Table A2 (continued): Description of Source of Unit
Cost Estimates |
| · Cost per women
screened for syphillis in reproductive health services · Cost per STD case treated, ANC service |
Low: $0.91;
cost per woman screened in Tanzania, assumed to be financial (Kigadye et al 1993} Medium: $2.00 from Mozambique programme
described above (Bastos et al, 1992)
Same as for cost per STD case
treated/visited. |
VCT
· Cost per person counseled and tested |
Low: $3.80;
cost of adding VCT to a rural South African hospital, excluding all overhead costs, but
including all commercial costs of test kits, laboratory staff and equipment used; rapid
Capillus test, economic and incremental. Medium: $13.82, estimated costs of running VCT in
free-standing clinic in Uganda (Alwano-Edyegu and Marum, 1999). Assumed to be economic and
full. |
| Workplace
interventions · Cost per
person in employment reached (peer-education)
· Cost per STD case treated/visited
· Cost per male condom distributed |
The unit cost data
for peer-education are taken from Soderlund et al (1993) and comes from a workplace
intervention in Uganda. Figures are taken for low-cost and are extrapolated upwards to
obtain a medium cost. Assumed to be economic and full costing. Low: $0.26
Medium: $0.50
As per strengthening STD treatment.
As per CSW intervention. |
| Strengthening
blood transfusion system ·
Cost per safe unit collected |
The
unit cost data is derived from Soderlund et al (1993). These figures are based on national
and centralized blood transfusion systems in Zimbabwe and Uganda. Assumed to be full and
economic costing. Low:
$5.34
Medium: $18.22 |
| |
|
|
Table A2 (continued): Description of Source of Unit
Cost Estimates |
| MTCT · Cost per woman screened
· Cost per woman testing HIV-positive
and receiving regimen |
As per VCT
intervention All these costs
are just the drug cost.
Low: $5; HIVNET 012 Nevirapine
regimen; cost per course based on dose given to women at labor and then dose for infant
after birth, based on Marseille et al (1999). Assumed to be financial and incremental.
Medium: $50; CDC Thai - ZDV only pre
and intrapartum. Cost per course if using Thai generically manufactured drug (UNAIDS,
1999); no inclusion of freight costs just the drug prices. Assumed to be financial and
incremental. |
| · Cost
per woman testing HIV-positive of strengthening delivery services in facilities to
undertake regimen · Cost per
woman of six months formula milk |
We extrapolate the
cost of providing training and additional staff to meet these needs. These costs are
derived from Wilkinson et al (1998), and are based on the costs of additional nurse
training and midwives in rural South Africa. Low: $13.70 - 40% of Wilkinson et al (1998) costs
Medium: $24.00 - 70% of Wilkinson et
al (1998) costs.
Low-cost data from Pazvakavambwa (1999).
Assumed to be financial and incremental. Does not include freight and transport costs. The
medium cost is extrapolated from the low price. Costs of ensuring access to safe water
have not been included.
Low: $50
Medium: $55 |
| |
|
|
|
Table A2 (continued): Description of Source of Unit
Cost Estimates |
| Mass
media · Cost per campaign |
The cost data for
mass media campaigns come from Soderlund (1993) and Kumaranayake et al (1998). Both are
economic and full costings. The low-cost scenario corresponds to a programme in Gabon
where the campaign was contracted to a private firm and includes salary and overheads. The
medium cost scenario relates to a mass media programme run for three months in Cameroon,
but paying commercial rates for broadcast time. This excludes overhead costs and salaries
of people involved. Low:
$489,565
Medium: $516,817 |
Palliative
care
· Cost per patient year |
These costs are
based on the estimated frequency of these symptoms for PLHA in Sub-Saharan Africa and the
drug costs of treating common symptoms (such as fever, cough, diarrhea, skin rashes,
headaches, nausea), which are used to derive a cost per patient-year. Assumed to be
economic and incremental. (World Bank (1997a). Low: $21.50
Medium: $25.80 |
| Clinical
management of OI · Cost
per adult person year of treatment |
The costs of
clinical management include the costs of treating common OI in Sub-Saharan Africa (such as
tuberculosis, oral thrush, and pneumonia/septicemia). The cost is based on the estimated
frequency of OI, drug costs and costs of inpatient and outpatient care in Sub-Saharan
Africa. The cost of the drugs is taken from World Bank 1997. Costs of inpatient and
outpatient facilities were estimated separately for low-income countries (based on World
Bank 1997 and Hansen et al 2000). These costs included both direct patient-related costs
(such as drugs and laboratory services) and indirect costs (staff and facility costs). In
order to obtain figures for high-income countries, we calculated multipliers for direct
and indirect costs based on the relative costs of ambulatory and inpatient TB treatment in
South Africa (Floyd et al, 1997) and Malawi, Mozambique and Tanzania (De Jonghe et al
1994). Assumed to be economic and full. Low-income countries:
Low: $247
Medium: $359
High-income countries:
Low: $471
Medium: $698 |
Table A2 (continued): Description of Source of Unit
Cost Estimates |
Prevention
of opportunistic illnesses
· Cost per adult person year of prophylaxis
for OI (INH and Cotrimoxazole) |
Given the
duplication of activities for provision of preventive TB therapy and cotrimoxazale (in
terms of monitoring, costs of outpatient visits, etc), the joint costs of administering a
cost of INH and Cotrimoxazole were estimated. For low-income countries the cost data are
adapted from Aisu et al (1995) in Uganda for six months INH preventive therapy, and
include the costs of the initial skin test, screening (chest x-ray and sputum smears) for
active TB, personnel and administration costs of the programme. High-income country costs
were adapted from Masobe et al (1995) in South Africa, and include the costs of testing,
administration of drugs (INH 5 times a week for six months), monitoring, personnel costs
and transport costs. Data related to the cost of Cotrimoxazole was taken from Guinness
(2000). No additional costs for HIV-testing were included as they were assumed to be
included in the costs related to VCT. Costs are assumed to be financial and incremental. Low-income countries:
Low: $30
Medium: $36
High-income countries:
Low: $64
Medium: $79 |
Home-based
care
· Cost per person with AIDS supported |
The cost estimates
come from projects in Zimbabwe and Zambia (Chela et al (1994), Gilks et al (1998). They
are from interventions that tend to have a very low coverage among its target population
(less than 10% of the eligible population in Zambia and Zimbabwe). Assumed to be financial
and full costing. Low:
$63
Medium: $197 |
| Support
for orphans Cost per child
in an orphanage |
Living expenses
include food, clothing and basic commodities. The school expenses include subsidies for
school fees and uniforms. Based on estimates from Boerma and Bennett (1997) for orphanage
care in Tanzania. Assumed to be financial and full cost. Low: $120
Medium: $180 |
| |
|
|
|
Table A2 (continued): Description of Source of Unit
Cost Estimates |
| · Cost
per child for community assistance with living expenses · Cost per child for school expenses |
Taken
from Drew et al (1998) for programmes in Zimbabwe, implemented by community-based
organizations, using volunteers who visit families with orphans. Cost assumed to be
financial and full. Low:
$9
Medium: $35
Derived from Boerma and Bennett (1997), in the context of district-based programmes for
communities in a high-prevalence setting. This includes support to new orphans and
community-feeding posts. Cost assumed to be financial and full.
Based on estimates from Boerma and Bennett
(1997) for Tanzania. Weighted average of primary and secondary school costs. Assumed to be
financial and full cost.
Low: $25
Medium: $33 |
| Psychosocial
support and counselling
|
Costs
are extrapolated from VCT cost. There are no published estimates of costs. Low: $3
Medium: $6 |
| Treatment
(HAART)
|
The
structure for the calculation of the costs of HAART follows the UNAIDS Care Model (UNAIDS
Secretariat 2000). The cost per person includes the costs of drugs, monitoring, staff
training, transport, strengthening of facilities for administration of ARV treatment and
appropriate clinical management, and provision of drugs to deal with side-effects and
adverse reactions and basic pain relief, as well as prophylaxis for OI. The largest
component of the unit cost is the cost of the drugs. These prices have been changing
rapidly and also reflect the different combinations of drugs that may be prescribed. For
the low-cost scenario, we used an average drug price of $US1,400 per person annually. This
was based on the recent negotiations between Senegal and the pharmaceutical companies
related to the bulk-buying of ARV (Wall Street Journal, October 2000). The drug price for
the medium cost scenario was $US2,635 based on the 1999 Brazilian experience with ARV
purchases (Panos, 2000). In comparison the negotiated price for Uganda's purchases in May
2000 was $US4,201 and the price in the US for drugs was $US9,905. |
| |
|
Table A2 (continued): Description of Source of Unit Cost
Estimates |
| |
The costs of
monitoring reflect the costs of viral load monitoring, CD-4 cell counts, blood chemistry,
transport and the cost of outpatient visits drawn from a range of sources (Guinness 2000,
Panos 2000, World Bank 1997). The staff training courses were based on Masobe et al (1995)
for the costs of training for implementation of preventive therapy, and were adjusted
downwards for low-income countries based on the ambulatory multiplier for direct costs
described in the calculation of the costs of clinical management above. The costs for
appropriate clinical management, and provision of drugs to deal with side-effects and
adverse reactions and basic pain relief, as well as prophylaxis for OI were drawn from the
same sources as described above. The costs are assumed to financial and incremental. Low-income countries: Low: $1,993
Medium: $3,468
High-income countries: Low: $2,393
|
| Institutional
strengthening · Cost per
capita for very low programme strength countries
· Cost per capita for low programme strength
countries
· Cost per capita for medium programme
strength countries
· Cost per capita for strong programme
strength countries |
These are related
to existing programme strength (very low, low, medium and strong). Costs are based on
estimates for the implementation of the MTP-III in Tanzania. Low: $0.021
Medium: $0.026
Low: $0.015
Medium: $0.019
Low: $0.010
Medium: $0.013
Low: $0.006
Medium: $0.008 |
[Go to Top]
PREVIOUS
PAGE < CONTENTS > NEXT PAGE
|
