Abstract:
Background
Malaria remains a major public health problem in the world. In Ghana, the entire
population of24.2 million is at risk of malaria infection. Malaria is end emic and perennial
in all parts of the country, with seasonal variations that are more pronounced in the north.
From 2010 to 2015, Ghana has reduced by 45% malaria deaths. Progress in reduction of
malaria prevalence has been recorded in the routine surveillance system through Health
Management Information System but unfortunately that data suffers from reliability from
presumed malaria.
Therefore, Ghana established the sentinel surveillance system in 2013, with the aim to
monitor the prevalence of malaria in the country and minimizing the proportion of cases of
presumed malaria. Since this establishment the sentinel surveillance system, the indicators
generated by the system show a clear progression in the control of malaria. Nevertheless,
there is no evidence about how effective the system is performing.
This study seeks to evaluate the performance of the malaria sentinel surveillance system,
by assessing its attributes in the Greater Accra region; and also, determine the prevalence
of malaria during that same period.
Methods: This study was cross-sectional and used secondary quantitative data and
methods to derive malaria positivity rate in sentinel sites in the Greater Accra region. Data
on malaria indicators were extracted from District Health Management Team e-database
for January 2014 to December 2016. The data were analyzed to show both slide and RDT
malaria positivity rate, proportion of suspected malaria case, and testing rate. Based upon
CDC the Centre for Disease Control, Atlanta updated guideline for evaluating Public health
surveillance system, keys system attributes were assessed and described. Epi info was used
to generate frequencies, proportions, and chi square test at 5% confidence level.
Results: In general, the rate of malaria positivity and the proportion of suspected cases of
malaria prescribed with ACTs have decreased overtime. From 2014 to 2016, this decrease
ranged from 25% to 12.2% for malaria positivity and from 61.4% to 29.6% for proportion
of suspected malaria cases. There was also an increase of testing rate from 81.7 to 98 %
over the study period. Data quality is particularly poor in the Qbom health center sentinel
surveillance site. However, overall, the internal completeness of the surveillance system
was satisfactory. Data from Sentinel sites was getting more and more accurate over time,
when comparing with Noguchi Memorial Institute for Medical Research data. Positive
predictive value ranged from 12.0010 to 20.4 % in 2014 to 2016, declining over time while
Sensitivity increased leading to the increase of number of suspected cases since 2014.
Conclusion: The testing rate in the malaria surveillance sites during the study period
increased, resulting in a decrease in the use of ACTs. Majority of suspected cases were
tested and classified according to outcomes. Malaria positivity rate also decreased
significantly in the course of these three years. The data are generally of good quality,
representing very well the community in terms of place and people. The application ofT3
(Test, Treat and Track) and case definition by the system, increased the systems sensitivity
to the detriment of the positive predictive value.