The world's scientific and social network for malaria professionals
Subscribe to free Newsletter | 10206 malaria professionals are enjoying the free benefits of MalariaWorld today

Measuring the wrong parameter for evaluating impact of malaria control

March 14, 2015 - 14:51 -- William Jobin

In the latest Annual Report of PMI (April 2014), they cite the reduction in mortality rates of children under five as proof of the beneficial impact of the anti-malarial work of PMI. Figure 1 of the report cites the following figures for the 15 PMI focus countries which have the best data. However, as in the past, they have not done a comparison with other countries in Africa. The same is true in general of the reports from the Roll Back Malaria program. For some strange reason they don't want to measure changes in malaria prevalence. It isn't that hard, even I know how to do it!

PMI country Reduction in mortality rate from 2000 to 2010
Rwanda 50%
Mali 49%
Senegal 46%
Benin 44%
Mozambique 37%
Kenya 36%

( UNICEF 34% mean for all countries in sub-Saharan Africa )

Uganda 34%
Zambia 29%
Ethiopia 28%
Tanzania 28%
Ghana 26%
Angola 23%
Madagascar 23%
Liberia 18%
Malawi 16%

At first glance, this looks impressive, especially for Rwanda and Mali where the rate is now half of what it was before. However, a comparison is needed, and the UNICEF report makes that possible.

UNICEF recently published their 2014 Report on ‘Levels and Trends in Child Mortality’ which showed that the over-all reduction for all countries in sub-Saharan Africa for this same time period was 34%. This included all countries in Africa, approximately 30 which were not receiving malaria suppression help from the US PMI.

Thus only 6 PMI countries did better than the mean, while 9 PMI countries had a lower reduction than the mean; not a very good record. What went wrong with PMI operations in Malawi, Liberia, Madagascar, Angola, etc?

I think the problem is that PMI measured the wrong parameter in the wrong age-group. All-cause mortality in infants probably reflects primarily the improvements in maternal health, made by other programs, not PMI or Roll Back Malaria. Such improvements as pre-natal care, provision of mid-wives, improved nutrition and sanitation for mothers, and other public health programs aimed at mothers and infants were probably the primary factors in reducing the all-cause mortality.

To measure the impact of malaria, they should measure malaria prevalence. Isn't that obvious?

If we had better data on the impact of malaria suppression measures, we could do a better job.

Bill, still puzzled


Submitted by David Warhurst on

It may need clarification. When you say malaria prevalence do you mean prevalence of malaria parasites symptomatic or asymptomatic in that age group, as normally understood.
Or do you mean prevalence of parasitemia with symptoms (what I would call malaria).
Mortality while infected with malaria is easier to measure, though clearly not a perfect parameter because it may not be the cause of death.

Submitted by jobin (not verified) on

Thank you David,

When I suggest simply measuring malaria prevalence, I refer to the normal procedure of picking a sentinel population and following them every year, making blood slides and checking them microscopically for parasites. The sentinel population should be in the same place every year - such as the schools of a randomly selected village.


Submitted by David Warhurst on

No cost for this? which is the ideal method, you would have to do it in several places. Compared with this, death rate in the age group is a quick and dirty method.


Submitted by William Jobin (not verified) on

The US Presidential Malaria Initiative spends over half a billion dollars a year in 19 African countries, using their narrowly conceived and ephemeral strategy, which requires - absolutely - repeated application year after year.

Shouldn't they spend some money on evaluating the impact?

Bill, worried that they will fall into the Immunity Trap as well