When the Plan-EO team describes their platform, one idea comes up repeatedly: filling in the gaps. In global health, particularly in low- and middle-income countries, data gaps on enteric pathogens and diarrhoeal disease are big. Epidemiologists may spend years conducting a single study that documents a few hundred cases. Compare that with climatologists, who can tell you, down to the day, what the weather was like in a specific location years ago, with some certainty.
Plan-EO, an interdisciplinary, translational research initiative founded by epidemiologist Josh Colston, is built on bridging these two worlds. And now, through a new partnership with the EU-funded SPRINGS project, Plan-EO’s modelling approach is being extended into the future: from predicting current disease patterns to projecting how climate change and public-health interventions may reshape disease risks over coming decades.
Plan-EO originated nearly a decade ago, seeded by a collaboration between Colston and a group of climatologists during his PhD and postdoctoral work. The concept was ambitious: take climate data: high-resolution, historically consistent, and globally available, and use it to build disease models that can “fill in” the epidemiological blind spots across regions where data is sparse or absent.
Most existing diarrhoeal disease studies rely on highly localised sampling. A study in one district may reveal the prevalence of Campylobacter, Shigella, Cryptosporidium or other enterics pathogen among children visiting a particular health facility. But these snapshots are not enough to offer a broader picture.
Plan-EO uses this climatological precision to build models linking weather variables to pathogen transmission patterns. For some pathogens, prevalence increases with temperature; for others, it declines. By teasing apart these relationships, Plan-EO generates predictions of disease prevalence across wide geographic areas in low and middle income countries.
This is the backbone of what Colston describes as a “living evidence system”: a continuously updated platform that integrates newly published findings, new datasets shared by partners, and new climatic inputs to refine its outputs over time.
At the centre of Plan-EO is a growing network of collaborators across roughly 30 studies that have shared individual-level data. Additional results are incorporated as aggregate values from published literature. These inputs underpin the pathogen models, such as the existing system for Shigella, which visualises how predicted prevalence “undulates” throughout a typical year.
On the user-facing side, Plan-EO offers a dashboard highlighting:
%202.gif)
Colston gives a practical example: a funder considering where to trial a new Campylobacter vaccine. By exploring observed or predicted prevalence across candidate locations, decision-makers can identify where transmission is sufficiently high to detect an effect. Even in regions with little or no direct evidence, Plan-EO’s prediction models can provide a starting point.
Plan-EO is also expanding its suite of risk-factor layers. Following published work on environmental and infrastructural indicators, the next additions will likely cover sanitation and hygiene access.
“It’s a big scaling-up phase,” Colston notes. “The aim is that, by around 2026, users will be able to explore modelled maps for all the SPRINGS pathogens” (Rotavirus, Campylobacter, Cryptosporidium and Giardia).
This is where the SPRINGS partnership becomes central. While Plan-EO currently focuses on predicting present-day prevalence, SPRINGS introduces another dimension: future climate scenarios.
SPRINGS researchers, through the climate modelling working group, are producing downscaled climate projections for the project’s four case study regions (Ghana, Tanzania, Italy and Romania) spanning multiple future scenarios. These projections will eventually feed into Plan-EO’s modelling framework, allowing the platform not only to predict current prevalence but to project how patterns may evolve by 2050 and beyond.
“SPRINGS is providing the projections part,” Colston says. “Given what climate experts think conditions will be like in 2050, we’ll be able to model how patterns of Campylobacter or other pathogens might change.”
Beyond climate, the collaboration will also incorporate intervention modelling, drawing on SPRINGS’ stakeholder consultations led by the Policy Translation working group. These teams develop priority intervention scenarios, such as vaccines, sanitation improvements and coverage targets, that Plan-EO can integrate into disease models.
The relationship is mutually reinforcing: Plan-EO provides the platform on which SPRINGS’ more complex projection and intervention layers can be built, while SPRINGS ensures that Plan-EO’s future-focused capabilities remain aligned with real-world policy priorities.
This alignment is particularly evident in SPRINGS’ case study in Ghana, which Colston describes as especially exciting. Unlike many previous studies, SPRINGS integrates environmental data collection, such as weather characterisation, hydrology and epidemiology, from the outset. It represents one of the first attempts to embed planetary-health thinking directly into study design.
In the coming year, Plan-EO will continue adding data retrieved through systematic reviews, integrate new risk-factor layers, and make its interface more navigable and user-friendly. A demo is also expected.
By the end of the SPRINGS project (2028), users will likely have the ability to toggle between current predictions and future projections, select specific climate scenarios, and explore model outputs for all SPRINGS pathogens.
The partnership stands at the intersection of climate science, epidemiology, and public-health action, bringing together data that rarely coexist in a single system.
As Colston puts it, “The core of Plan-EO is already there. Now it’s about building out the layers: different pathogens, risk factors, and visualisation options, and integrating the future scenarios from SPRINGS to answer the questions that matter for public health.