The Integrated Infectious Disease Capacity Building Evaluation (IDCAP) team evaluated
two malaria case management interventions that were implemented in 36 health facilities
in Uganda using facility-based and population-based data. The population survey was
administered in the catchment areas surrounding the IDCAP facilities to measure
utilization of care at the facilities as well as the effects of the intervention on mortality of
children under five years of age (U5). The relationship between health facility-based
data, population-based data, and the quality of malaria case management in children U5
is explored.
Methods:
Using malaria case management facility-based data, variables for health facility
performance were created. Using the population survey, healthcare utilization questions
were used to define the following three performance variables: delay in seeking care,
whether a child had a blood test taken, and whether a child received a drug. Linear and
logistic regressions were used to explore the relationship between the two sources of data
on health facility performance.
Results:
The regression coefficients for the delay in seeking care variable regressed on having a
blood test taken are as follows: best performing (reference): 0, 2nd best: 0.78 (0.43, 1.14),
3rd best: 0.19 (-0.20, 0.58), lowest performing: -0.39 (-0.76, -0.02); p <0.001. The
regression coefficients for the delay in seeking care variable regressed on being given a
drug are as follows: best performing (reference): 0, 2nd best: -0.10 (-0.47, 0.27), 3rd best:
-0.36 (-0.76, 0.04), lowest performing: 0.33 (-0.08, 0.74); p =0.01. The odds ratios for the
blood test taken variable are as follows: best performing (reference): 1, 2nd best: 0.96
(0.57, 1.61), third best performing: 0.92 (0.52, 1.62), lowest performing: 0.73 (0.43,
1.24); p =0.76. The odds ratios for the drug(s) given variable are as follows: (reference):
1, 2nd best: 1.26 (0.54, 2.97), 3rd best: 0.66 (0.30, 1.45), lowest performing: 1.21 (0.51,
2.85); p =0.14.
Discussion:
The delay in seeking care variable shows a statistically significant difference across
categories of the facility-based measures of malaria test recorded and malaria drug(s)
given. Overall, there is the need for more research, as this study is one of the first of its
kind.