Evidence Required for Adoption of New Vector Control Methods in Public Health - Biological and Environmental Control of Disease Vectors - page 191

11.2 Perspectives on the
Effectiveness Question

interventions are appropriate for a given setting.
In particular, the Global Fund to Fight AIDS,
Tuberculosis and Malaria (GFATM), which has
fi nanced most of the recent scaling-up of
malaria control, explicitly relies on WHO
guidance for such decisions (GFATM, 2011). As
noted above, WHO already has a well-known
and infl uential system for the assessment of
new pesticide products within existing
categories of vector control technology
(WHOPES), but at the time of writing, a
comparable system for the assessment of new
vector control technologies (or 'new paradigms'
as they are known in the trade) has not yet been
created. The WHO is, however, planning such a
system, and has been engaged in discussion
with stakeholders as to how such a system
might operate.
So far, these discussions have concentrated
on the evidence needed to justify the fi rst use of
a new intervention in a public health
programme. This is, of course, the point at
which the fi rst income from sales can be expected
by a product developer, but for an intervention
that will eventually become widely and generally
used, this is only the start of the process. For
example, in the case of insecticide-treated bed
nets (ITNs), the fi rst trial showing that this new
technology could produce a clear and dramatic
reduction in child mortality in Africa was
published in 1991 (Alonso et al ., 1991) and
soon afterwards, at a meeting in 1993, the WHO
drafted its fi rst recommendations for ITN use in
public health (WHO, 1995). These were rather
guarded: it was specifi cally noted that ITNs
could not yet be regarded as a substitute for IRS,
and that further evidence was needed on ITN
ef ectiveness in a wider variety of settings. At
that time, notably, WHO was recommending
that all vector control should be 'selective' - in
other words there was specifi c guidance against a
policy of universal coverage. It took another
decade, until 2006, for WHO to start to
recommend ITNs/LLINs as a primary form of
malaria vector control for 'tropical African
savannah' (although IRS remained the
intervention of choice for highland and desert
fringes, and for plains and valleys outside Africa)
(WHO, 2006). Then in 2007, the UN secretary
general recommended a policy of universal
coverage with treated nets or another form of
ef ective vector control.

The question 'Is this intervention ef ective?' can
have dif erent meanings for a product developer
and for a public health programme manager.
This is because these two professions have
dif erent short-term objectives, and work with
dif erent indicators. Product developers spend
most of their time trying to compare alternative
prototypes, and thus prefer to work with an
established series of laboratory-based tests, each
measuring a single simplifi ed aspect or function,
often at the entomological level. However, the
relevance of these entomological tests to health
outcomes in the real world is often distant and
sometimes undemonstrated. Public health
programme managers, on the other hand, are
occupied with the task of disease prevention:
they are interested only in health outcomes, and
need epidemiological data from trials carefully
designed to separate the ef ects of the
intervention from those of other confounding
factors. Moreover, among trials with health
outcomes, there are 'ei cacy' trials (designed to
show that the intervention has the desired ef ect
in carefully controlled and favourable con-
ditions) and 'ef ectiveness' trials (designed to
measure performance in real-world programme
conditions).
Programme managers focus on these
distinctions because they have a duty to ensure
that limited public health funds are deployed to
achieve the maximum possible health benefi ts.
The decisions they make are liable to be
challenged and criticized. In order to justify the
selection of a given intervention (or mix of
interventions), they require a solid body of
evidence showing that in the target setting, the
chosen strategy will be feasible, ef ective in terms
of health outcomes, reliable in a variety of
situations, and good value-for-money in
comparison with the existing alternatives. Note
that the main alternatives - IRS and LLINs - are
supported by a remarkably large and robust
body of evidence (Lengeler, 2004), so the
standards of evidence required to argue for the
introduction of a new intervention are high.
We must also consider the special role of
the World Health Organization (WHO) in this
context. Many developing countries rely on
WHO guidance for decisions about which