The global health community has long grappled with how to improve health outcomes in low- and middle-income countries (LMICs). Effective methods to manage many acute and chronic health conditions have been developed, but LMICs continue to experience poor health outcomes relative to high-income countries (HICs). While we know what care most patients require, we must learn how to deliver that care in LMICs. This problem of how to implement effective care practices is what could be called the “navigation problem” in global health.

Healthcare leaders in LMICs encounter the navigation problem every day. They must make decisions about how their organizations care for patients despite the many barriers to high-quality care that they face. Some of these obstacles include inadequate resources, limited specialty training for providers, insufficient healthcare infrastructure, and lack of knowledge about the state of operations of clinics and hospitals. Healthcare leaders must navigate these barriers and many more as they lead their organizations to improve care.  

Data science provides valuable tools to help leaders navigate challenging obstacles. Analytical methods such as statistical modeling, machine learning, and deep learning provide healthcare leaders with the information they need to make decisions. Technologies, such as electronic medical records, mobile devices, databases, and analytical software are important components of the infrastructure that supports data analyses. Like compasses, GPS, or other navigation tools, data science methods and technologies can help leaders know where they are, where to go next, and how to overcome obstacles they encounter along the way.

One good example of data science technologies allowing leaders to make better decisions comes from the Middle East and Northern Africa (MENA) Coalition for Human Papillomavirus (HPV) Elimination (1). This group used open-access data to create a dashboard of information about HPV vaccination rates, cancer incidence rates, and death rates in the MENA region. Before this project, this information was not readily available to decision-makers because there was no standard way to combine the data sources. The team integrated the data to create a comprehensive view of the region’s disease burden, allowing leaders to make informed decisions about how to improve vaccination rates and reduce cervical cancer incidence.

Data science tools can also assist with diagnosis when specialized resources are limited. A team of researchers developed a mobile device-based diagnostic system for malaria in LMICs (2). The system uses image processing and artificial intelligence algorithms to identify Plasmodium falciparum species in peripheral blood samples. The researchers demonstrated that the system was 91% accurate at diagnosing malaria. This system could provide health centers in remote communities the ability to diagnose malaria without needing extensive pathology expertise or expensive equipment.

Data science methods can also help providers identify patients at high risk for poor health outcomes. A group of researchers in Tanzania developed a clinical prediction model to identify women at risk of perinatal death (3). The model showed good discrimination and calibration using predictors obtained from the history and physical examination at the time of admission to the labor ward. By using the model to focus improvement efforts on the highest-risk patients, healthcare leaders can efficiently use their organization’s time and resources to care for those who would most benefit.  

Data science offers powerful tools to help leaders in LMIC solve the navigation problem. With these tools, healthcare leaders can better understand their health systems, identify areas for improvement, and develop strategies to overcome challenges. The global health landscape is filled with complex terrain to traverse. Data science can help healthcare leaders navigate this landscape and make more informed decisions that lead to better outcomes for patients.

We believe in changing healthcare one connection at a time. If you are interested in the opinions in this piece, in connecting with the author, or the opportunity to submit an article, let us know. We love to help bring people together! [email protected] 

References

  1. Somai M, Levy S, Mhirsi Z. Data Science in Global Health—Highlighting the Burdens of Human Papillomavirus and Cervical Cancer in the MENA Region Using Open Source Data and Spatial Analysis. In: Celi LA, Majumder MS, Ordóñez P, Osorio JS, Paik KE, Somai M, editors. Leveraging Data Science for Global Health [Internet]. Cham: Springer International Publishing; 2020 [cited 2022 Nov 28]. p. 373–82. Available from: https://doi.org/10.1007/978-3-030-47994-7_23
  2. Oliveira AD, Prats C, Espasa M, Serrat FZ, Sales CM, Silgado A, et al. The Malaria System MicroApp: A New, Mobile Device-Based Tool for Malaria Diagnosis. JMIR Res Protoc. 2017 Apr 25;6(4):e6758.
  3. Housseine N, Rijken MJ, Weller K, Nassor NH, Gbenga K, Dodd C, et al. Development of a clinical prediction model for perinatal deaths in low resource settings. eClinicalMedicine [Internet]. 2022 Feb 1 [cited 2022 Dec 9];44. Available from: https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(22)00018-9/fulltext