Design New Analytics Approaches for Malaria Elimination (Round 17)

The Opportunity

The Bill & Melinda Gates Foundation sees surveillance as the backbone of effective malaria elimination efforts, but we believe that malaria surveillance needs a fundamental “rethink” to achieve the goal of a malaria-free world within a generation.

For malaria elimination, surveillance is defined as the information flow that people use to make decisions about how to find malaria parasites, how to eliminate them from human populations, and how to prevent parasites from coming back. And advances in digital technology and the increased reach of cell phone networks put the goal of timely, affordable, geo-located reporting within reach. There is a real opportunity to transform malaria surveillance in a way that can accelerate national and regional elimination efforts.

Malaria transmission events take place in a dynamic, interconnected, and complex system. But malaria elimination efforts can thrive if they have access to novel solutions that enable rapid data exchange via new, user-friendly platforms. The purpose of this call is to engage a broad spectrum of innovators to identify how tools and strategies that have been developed for other sectors can be applied to the fight against malaria.

The Challenge

Despite the acknowledgement of the role of surveillance in malaria elimination, the full potential of available information communication technology, data systems, and data harmonization has not yet been effectively applied to the challenge of malaria elimination.

There are multiple breakdowns in the data life cycle for malaria elimination. 

Data availability and interoperability

Data that is collected is stored across organizations and different levels of the health system in multiple locations and in multiple formats, resulting in long and inefficient processes to harmonize and compile information. Some analyses require data from multiple stakeholders where sharing between inter- and intra-national organizations, ministry departments, public and private, is complex and sometimes prohibited.

Availability of novel and robust data analysis

There is no formal market or market leader for specialized malaria surveillance analysis tools, nor have any existing systems reached global scale. Further complicating innovation is that solutions require national-level buy-in to be funded and implemented at scale.

What we are looking for:

We are looking for proposals of innovative solutions for improving data availability and use in decision-making for malaria elimination that focus on ONE of the following areas:

• Innovation in interoperability: solutions that will reduce the time for data harmonization by automation or skill set simplification. This can include the use of machine learning or heuristics, and setting standards. Proposals should increase the availability of interoperable tools and/or facilitate the adoption of interoperable systems in country.
   
• Innovation in analytics: solutions for how to analyze and present the most pertinent, actionable information, personalized by audience. This can include recommendation engines. Solutions should be designed with the end goal of feeding into a tool, but they should not include the development of visualization. Proposals should improve the quantity and timeliness of data analysis at all levels of National Malaria Control Programs (NMCPs), and/or increase the availability of easy to use data analysis tools.

Because many countries are currently undergoing the process of reorienting their programs and systems to an elimination context, we do not yet have broad-based examples of places where all the data sources exist. The community is working on data collection standards and tools in another area, and we expect these changes to happen in elimination settings over the next several years. We do not want respondents to develop new data collection tools. Solutions must, at a minimum, incorporate into analysis and identify the following sources of data (locations mentioned below are examples and not prescriptive).

• Geolocation to the household level, travel history, treatment history, demographics and speciation of diagnosed cases
• Population denominator (for example, from WorldPop)
• Human movement patterns
• Vector species identification abundance, and behavior (for example, from VectorBase or the Malaria Atlas Project)
• Stock levels for diagnostic tests, treatment, and bednets

Because analysis based on frequently-updated geo-located information that is often proprietary to the countries from which it originates, access to country-level incidence data is not a requirement for applicants to propose a demonstration project.

We are especially interested in applicants and solutions that come from sectors outside of malaria, or even outside health. Solutions do not have to be brand new; they can be an existing method or tool used in another context repurposed and applied. However, in this case we will require proposals to include a demonstration project.

Winning proposals should:

• Focus on simplicity of methods using routinely available data
• Explain rationale for targeting specific countries or regions
• Describe approach for interoperability with country data systems including DHIS2
• Provide evidence that the activity is aligned to relevant national strategies on eHealth
• State the expected improvements over existing solutions, including streamlining of communications and reporting burden
• Address cost of the solution and a description of how it can be brought to scale and be sustainable in the malaria-endemic developing world context, including any computing resources
• Explain online and offline capabilities and limitations
• Explain how ease of use will be measured – i.e. back end analytics that show improvement in timeliness of data collection to analysis, and from analysis to decision-making
• Include user performance metrics (i.e. completeness and timeliness of analysis conducted with the information available)
• Explain how the solution will be strengthening existing health systems instead of building a parallel system
• If the solution involves building on or integrating a reporting system for additional disease areas, explain how it will strengthen malaria elimination decision-making
• Explain how the solution will improve key malaria elimination indicators
• If a proposal recommends new data be collected, it must provide details on what information will become routine surveillance data and what is expected to be uniquely collected for the analysis

A few examples of work that would be considered for funding:

• Developing a novel algorithm or API to pull together data for models or data interchange
• Advanced identification and tagging methods (of people, habitats, structures, etc) utilizing remote sensing platforms
• Real-time recommendation engines for precision interventions
• Automated monitoring for early warning or other alert systems
• Heuristic or machine learning algorithms for data validation
• Ontology, translation and other data integration services
• A data integration method (i.e. ELT-related tooling) that is optimized and scalable for malaria elimination settings
• Interoperable software modules for popular malaria elimination or other global health platforms
• Harmonization data sources into a tool that people in country malaria control program offices can update and interact with
• A solution that pulls together data on malaria cases and health facility stock. Or case data and vector control coverage. Or population and travel history. Or incorporating new types of data, such as population estimates through household maps.

We will not consider funding for:

• Proposals that do not focus on surveillance in countries with malaria elimination goals
• Development of new primary data collection apps
• Proposals that do not address the issue of interoperability
• New modeling approaches for risk mapping and malaria transmission; although incorporating these elements into the analysis is highly desirable
• Proposals that only focus on standalone research or survey contexts