Close

Effects of flooding on emergency obstetric and newborn care in urban sub-Saharan Africa and AI-based forecasting to support facility preparedness (Joint PhD 2027/28)

Supervisory team

LSHTM  

Nagasaki University

The project

This PhD will be embedded within the Effects of Extreme Weather Events on Obstetric and Newborn Outcomes in African Cities (EWE-ON-AFRICITIES) project, a multi-method study of the health and economic impacts of extreme rainfall and flooding on maternal and perinatal health in Lagos, Nigeria, and Harare, Zimbabwe. The data required for this PhD will be collected through EWE-ON-AFRICITIES, giving the candidate access to a rich, multi-site dataset and the opportunity to work within an established interdisciplinary team. 

At a time when global progress in reducing maternal and newborn deaths has already stalled, climate change presents a new and escalating threat to these hard-won gains.1,2 One critical pathway is access to care. Timely access to comprehensive emergency obstetric and newborn care (EmONC), is key for maternal and newborn survival,3 yet heavy rains and floods can cut women off from it or extend journeys to care beyond safe limits.4,5  

The stakes are particularly high in sub-Saharan Africa (SSA), which accounts for about 70% of global maternal and over 40% of neonatal deaths,1,2 and where heavy rainfall and flooding are projected to intensify in cities.3 In complex urban SSA settings, insufficient transport infrastructure and traffic congestion already make journeys to care long, and flooding further disrupt travel.6 Yet how flood events affect demand and patterns of EmONC care-seeking, and whether these changes lead to worse outcomes, remain understudied. 

Unlike many barriers to care, rain and flood is often forecastable. Advances in artificial intelligence (AI), particularly geospatial AI, offer new opportunities to link climate forecasts with health data at scale, raising the opportunity to anticipate flood-related changes in care-seeking and improve facility preparedness in advance. 

The aim of this PhD project is to examine how flooding affects access to comprehensive EmONC in urban SSA, and to develop and evaluate AI-driven forecasting tools that enable health facilities to prepare for flood-related changes in care-seeking and care utilisation. This project is expected to address the following objectives: 

  1. Evidence synthesis. Review existing peer and grey literature on how AI, particularly GeoAI, has been used to predict and characterise the impact of climate hazards on the provision of and access to health care. 
  2. Impact of flooding on care-seeking and outcomes. Assess whether and how flooding and heavy rainfall affect patterns of facility arrival, including the volume and timing of admissions (day versus night) and the distribution of patients across facilities, stratified by individual underlying obstetric risk, and examine whether any such changes are associated with adverse maternal and neonatal outcomes. 
  3. Predictive modelling. Develop and validate models that use rainfall forecasts to predict changes in facility-level admissions, including at night and at facilities outside flood-affected areas, in the days before and after heavy rainfall, giving facilities time to prepare. 
  4. Feasibility and acceptability. Use implementation research frameworks to explore the feasibility and acceptability of AI-assisted decision support for EmONC among health workers, managers and policymakers in urban SSA settings. 

References 

  1. WHO, UNICEF, UNFPA, World Bank Group, UNDESA/Population Division. Trends in Maternal Mortality: 2000 to 2023. Geneva: WHO; 2025. 
  2. UN Inter-agency Group for Child Mortality Estimation. Levels and Trends in Child Mortality: Report 2024. New York: UNICEF; 2025. 
  3. Widyantari KY, Sulistyaningsih I. The Effectiveness of CEmONC on Maternal Mortality: Systematic Literature Review. InProceeding International Conference 2019 Nov 19 (pp. 169-182). 
  4. Trisos CH, Adelekan IO, Totin E, et al. Africa. In: Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the IPCC. Cambridge: Cambridge University Press; 2022. p. 1285–1455. 
  5. Pappas A, Kovats S, Ranganathan M. Extreme weather events and maternal health in low-income and middle-income countries: a scoping review. BMJ open. 2024 Jun;14(6):e079361. 
  6. Banke-Thomas A, Wong KLM, Olubodun T, et al. Geographical accessibility to functional emergency obstetric care facilities in urban Nigeria using closer-to-reality travel time estimates: a population-based spatial analysis. Lancet Glob Health. 2024 May;12(5):e848-e858.

The role of LSHTM and NU in this collaborative project

This interdisciplinary project addresses three key challenges in global health: climate change, AI, and equitable access to health of populations in highly vulnerable settings. It draws on the complementary missions and expertise of the two institutions. The project aligns with the Maternal, Adolescent, Reproductive and Child Health Centre (MARCH), the Centre on Climate Change and Planetary Health (CCCPH) and the Department of Infectious Disease Epidemiology and International Health (IDEIH). At NU, it aligns with the Interfaculty Initiative in Planetary Health, the Research Center for Global Risk (CGR) and the School of Tropical Medicine and Global Health (TMGH) at NU. 

This collaborative project will foster bi-directional learning between both institutions. LSHTM brings expertise in maternal and newborn health, geospatial access to care and implementation research, together with long-standing partnerships with researchers, policymakers and health services in Lagos and Harare. NU brings expertise in AI and data science for health, including deep learning for climate-informed disease forecasting at NEKKEN, and their experience in weather-related disaster preparedness and early warning for health services. Through the project, LSHTM will strengthen its capacity in AI-driven forecasting, and NU will extend its work to urban maternal and newborn health in sub-Saharan Africa and to knowledge translation. 

If successful, we will conduct bi-monthly online and face-to-face meetings, sharing responsibilities to ensure the student develops a robust study protocol. The student will spend time in the Japan and in the UK, as well as in Nigeria and Zimbabwe and the relevant stages of planning, conducting the study, and writing up results.  

KW will supervise the geospatial components, including urban accessibility to health care, and flood-related disruption to EmONC. LM brings expertise in environmental epidemiology and climate–health time-series methods, developed through the Multi-Country Multi-City Network, and will co-supervise the flooding analysis. ABT brings expertise in maternal, newborn and child health systems, implementation research and qualitative methods, and will lead supervision of the qualitative component. As PI of EWE-IN-AFRICITIES and the Google-funded OnTIME programme, he will secure data access and big data resources from Google. PMT brings AI and data science expertise in climate-related disease modelling and will co-supervise the forecasting models.

Particular prior educational requirements for a student undertaking this project

  • A master’s degree (or equivalent) in epidemiology, public health, global health, data science, geography, environmental science or a related quantitative discipline; or if a more quantitative academic qualification (e.g. applied mathematics, engineering, physics), then some experience/demonstrated interest in epidemiological modelling or public health.  
  • Strong quantitative and data science skills, including statistical modelling and programming in R, Python or another language 
  • A demonstrable interest in maternal and newborn health, climate change and health, or health systems in low- and middle-income countries.

Skills we expect a student to develop/acquire whilst pursuing this project

  • Literature review and evidence synthesis 
  • Data science, machine learning and AI; managing, cleaning and linking large, complex datasets from health facilities, climate sources and big data platforms, developing, validating and interpreting forecasting models, including explainable AI and model evaluation. 
  • Advanced geospatial analysis and Environmental and climate epidemiology 
  • Qualitative and mixed methods research, thematic analysis, and integrating qualitative and quantitative findings. 
  • Implementation research; assessing the feasibility, acceptability and real-world use of AI-based tools in health systems. 
  • Research ethics and data governance, including responsible use of health data and AI. 
  • Knowledge translation and policy engagement, including communicating findings to policymakers, health managers and communities.