providing the health interventions, including the direct costs of medicines and
health services, capital investments, complementary management and institu-
tional support, and investment in training new personnel’. This would require
by 2015 an additional $13 per person per year (2002 US$ prices) in developing
countries, an increase of more than 50 per cent, which would be sufficient to
ensure a ‘decisive drop in avoidable deaths’ (CMH, 2001:56). The total was
very small compared with spending per person in high-income countries,
which in 2001 was more than $2,000 per person per year. By the early 2000s
almost all middle-income countries were already spending sufficient resources
to ensure universal access to essential health services, but this aim was under-
mined by large inequalities within society and a lack of access by the poor to
public health services. The CMH also argued that at least $3 billion a year
should be allocated towards R&D on health priorities of the world’s poor, of
which $1.5 billion should be for R&D for new drugs, vaccines, diagnostics,
and intervention strategies against HIV/AIDs, tuberculosis, reproductive
health and other priority health conditions of the poor.
Assuming that some revenue could be raised domestically, the spending gap
was estimated at $38 billion a year by 2015. The progress made in combating
malaria (see Chapter 5) reveals this approach can sometimes work. The malaria
initiative was based on a top-down drive to promote and disseminate appropri-
ate interventions such as bed-nets and data collection. The more general prob-
lem with this big-push donor plan is that it ignores almost everything discussed
in this chapter, especially the lack of any clear link between expenditure and
health outcomes. The CMH never did make clear how they would engage with
the incentives, motivation and poor information that characterize failures in
Education and Health 143Box 6.4 Community health care insurance in SenegalCommunity-based health insurance schemes have proved successful in Senegal
when linked to hospital-based medical care. This example is of a poor rural-agri-
cultural region (Thies) with a high incidence of poverty, malnutrition and bad
health conditions. For a sign-on fee and a monthly payment (equal on average to
2 per cent of household income) the scheme pays for the cost of hospitalization
(while individuals still have to contribute to the costs of surgery). Membership
has a strong positive effect on the probability of going to hospital and members
pay hospital expenses averaging less than half the amount paid by non-
members. Without insurance a single stay in hospital can lead to expenditure
equalling more than 25 per cent of the household’s annual budget. This study
shows that community financing through pre-payment and risk-sharing reduces
financial barriers to health care. Despite the insurance scheme, access to health
care even among members of the scheme was still strongly and positively related
to household income. This scheme was also contingent on the existence of a
viable health care provider (the hospital of St Jean de Dieu) that was able and
willing to offer support for the schemes (Jutting, 2004).