Scoping Review: Analytical Approaches to Estimating Malaria Intervention Effectiveness and Impact

Malaria is an acute febrile illness caused by a parasitic

infection transmitted by Anopheles mosquitoes. Human malaria is caused by five

different Plasmodium parasites, with P. falciparum being the predominant

species in sub-Saharan Africa (SSA) [1]. In the past 15–20 years, the combined

efforts of Ministries of Health (MOHs) and National Malaria Control Programs

(NMCPs), and their partners, including PMI, have made tremendous progress

against malaria. This progress resulted from the massive scale-up of various

malaria prevention and control interventions, including facility and

community-based confirmatory testing and treatment of malaria cases,

intermittent preventive treatment in pregnancy (IPTp), and seasonal malaria

chemoprevention (SMC), along with indoor residual spraying (IRS) and

insecticide-treated nets (ITNs).

MOH / NMCPs use programmatic intervention coverage and

effectiveness data to regularly monitor impact of interventions; modify

intervention implementation approaches (e.g., if coverage estimates are

sub-par) or switch interventions altogether (e.g., if effectiveness is observed

to be lower than expected). Intervention coverage and effectiveness has

traditionally been assessed by post-intervention campaign surveys or periodic

nationally-representative surveys (e.g., Demographic and Health Surveys [DHS],

Malaria Indicator Surveys [MIS], Multiple Indicator Cluster Surveys [MICS]), or

estimated by complex mathematical modelling. The limitations of surveys are

that—while generally robust—they only occur every 2–5 years; take time; require

significant human, logistical and financial resources and capabilities; and may

not be powered sufficiently enough to provide sub-national intervention

estimates. Similarly, mathematical modelling may be limited by the available

data and the significant technical expertise needed to develop and run the

models, let alone run them continuously. Additionally, neither surveys or

modeling may avail necessary estimates at key strategic moments in the malaria

programming planning, implementation and monitoring cycle, such as the

development of national strategies or design of necessary donor documents

(e.g., Global Fund Concept Notes or PMI Malaria Operational Plans).

Countries health

management information systems (HMIS) have been dramatically strengthened in

the past few years, with countries being able to consistently and fully report

on outpatient, inpatient and other programmatic data—much of this progress has

been made by adopting, piloting and rolling out the district health management

information system 2 (DHIS2), an open-source data-system software specifically

developed to capture health data in lower-and-middle income countries. Because

of their sheer volume across space and time, data collected and reported through

HMIS like DHIS2 complement and even offer an alternative to nationally

representative and other ad hoc surveys to assess health intervention coverage

and effectiveness, and ultimately impact on health outcomes. [2–4]

The objective of this scoping review is to map and summarize different

analytical approaches for estimating the impact and effectiveness of malaria

interventions using routine surveillance and health management information

system data. To our knowledge, such review has not been conducted. A

preliminary search for existing scoping and systematic reviews on the topic was

conducted on December 15, 2023, using PubMed and no similar reviews were found.

We will follow the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews

and Meta-Analysis extension for Scoping Reviews) to conduct and report this

scoping review [5].

A systematic search of PubMed was conducted on January 2,

2024, to identify studies and analyses that had used routine surveillance and

health management information systems data to assess the effectiveness of

malaria interventions. A detailed search strategy was designed and piloted to

identify the optimal combination of keywords used.

We examined the available electronic databases using

combination searches of the following Boolean terms: “Malaria” AND

“Intervention” AND “effect*” OR “impact*” AND “System” OR “Surveillance.” Other

key terms for such as "routine” or “information systems” control were not

included in the search strategy to have a more comprehensive search and will be

used during abstract and full text screening.

All identified studies will be imported into Rayyan, a systematic review

management software, to screen (title, abstract, and full text) and manage the

results of the search. Two reviewers will independently assess the titles and

abstracts of the included articles based on the inclusion criteria. In the

event of discordance between the two reviewers, a third reviewer will review

the titles and abstracts and will come to a final decision. From the included

articles, the two reviewers will identify relevant publications by reviewing

the full text. Any discordance will again be resolved by a third reviewer. A

PRISMA flow diagram will be used to report final numbers of articles that are

included and excluded at each stage.

We limited inclusion to studies and analyses that were

conducted in the past decade (i.e., 2014), which we purposefully chose as a

time point when countries’ routine health management information systems had

begun to substantially mature, with data reported by these systems

progressively becoming more robust.

From the included articles, each reviewer will work

independently to extract data from the articles following a pre-specified

extraction sheet. The following data will be extracted from each paper into an

MS Excel spreadsheet: (1) author; (2) year of publication; (3) geography; (4)

study design; (5) study period / time period covered; (6) intervention(s) for

which effectiveness and impact was measured; (7) approach to measure

effectiveness and impact; (8) health information system platform used for

analyses; (9) indicator variables included in the analyses; (10) target

population; (11) key findings; and (12) items from the Template for

Intervention Description and Replication (TIDieR) checklist. TIDieR is a

12-item checklist that includes the brief name, why, what (materials), what

(procedure), who provided, how, where, when and how much, tailoring,

modifications, how well (planned), how well (actual) of a program.

The proposed scoping review will outline the approaches to

estimate the effectiveness of malaria interventions using routine surveillance

and health management information systems data. The scoping review does not

involve data on human subjects and ethical approval is not required.