Health data for climate research: opportunities from longitudinal population studies
Global climate change is intensifying and already affecting human health. Extreme weather events are breaking meteorological records; a record number of people were exposed to heatwaves in 2020, with some regions experiencing deadly temperatures.
A deeper understanding of the interaction between climate and human health is critical to preventing some of the harmful impacts of climate change on people’s lives. But data that accurately captures these impacts isn’t always available to researchers.
Longitudinal population studies (LPS) offer some potential. LPS track the health of a large group of people over time and usually include a set of individuals that share some common characteristics, such as the period they were born or the area they live. Over time, data is collected about these individuals’ health alongside details about other areas of their life too. This makes it possible to study how biological, social and environmental factors affect health outcomes like the percentage of adults with heart disease.
Combining existing LPS with climate data offers some interesting opportunities. There are, of course, some challenges too. So, Wellcome commissioned Dr. Hannah Nissan and Professor Peter Diggle to investigate the feasibility and value of using these studies for climate-health research.
How can existing longitudinal population studies be valuable for climate-health research?
Firstly, these studies can be used to identify some population-level impacts of climate change. These impacts are determined by several factors, beyond the local weather and climate itself. For example, although rainfall is the main cause of flooding, whether an area is urban or not can determine how quickly water moves through the landscape — and the severity of the flooding. Similarly, a population with an action plan to deal with heatwaves is less vulnerable to extreme heat than a population without one.
Secondly, LPS can also be used to study how climate change interacts with existing health conditions or other factors which have an impact on people’s health, like their living conditions. For example, The 100 Million Brazilians cohort, which mostly includes people from low-income families, is useful for examining complex climate-health relationships and underlying population vulnerability. The fact that data is collected repeatedly over time is very important. Even if a person’s socioeconomic conditions improve (which could include changes like newly established access to piped water), they remain in the cohort, making it possible to see how improved conditions affect climate-health outcomes.
Finally, LPS often cover multiple climate zones so can also be used to explore how varying climate conditions affect people with the same health condition. For example, the impact of temperature on people with cardiovascular diseases. Understanding the relationship between climate and these underlying factors is especially important for targeting interventions in areas with limited resources. It is also useful for identifying populations that would benefit the most.
Integrating climate data into LPS is tricky
A major challenge in combining LPS and climate datasets are the differences in what the data reflects. LPS collect detailed data from individuals. In contrast, climate datasets are often based on satellite data or station data estimated over large geographical areas. This results in a mismatch which needs to be addressed (i.e. how do we match individuals to geographical areas). Also, the impacts of some climate events (like heatwaves), vary over areas with low and high elevation — and will affect populations living at different altitudes in different ways. Coarse-level climate data often cannot capture this variation. There are ways to address this issue though and adapt existing LPS infrastructure, so it is better suited for climate health research. This will require the resources for more frequent LPS follow-ups and close collaboration with climate scientists to ensure more effective data collection.
There is one piece of infrastructure we’ve already started to support that will improve our understanding of the geography of existing studies. Currently, we don’t know the exact area an LPS covers, the population density of the region and the characteristics of the landscape. We need this information to be able to match the individuals in an LPS to the climate hazards that might affect them. We’ve just commissioned a team to create a tool that will allow LPS to generate something called a ‘shapefile’, which is a way of communicating the geographical area an LPS covers in a machine-readable format. Similar work to create shapefiles for a series of large repeated cross-sectional surveys across Africa and Asia called Health and Demographic Surveillance Sites (HDSS) can be found here. Watch this space for more details — we hope this project will make all of the HDSS shapefiles open access, as well as create a tool which will allow LPS managers to generate their own shapefiles.
Finally, here are some other key findings and recommendations from the report that I found particularly interesting:
- Adapt existing LPS: supporting ongoing LPS to incorporate climate data can help answer some urgent climate-health questions in the here and now.
- Consider climate data from the outset: there is an opportunity to design LPS with the climate-health relationship in mind early on, to best meet the needs of climate and health research agendas. This will avoid downstream issues with matching data.
- Identify priority populations: LPS in those low-and-middle-income countries where populations are most affected by climate change must be the priority.
The future of LPS-climate research
In summary, LPS offer lots of interesting opportunities to use data collected over a long period to understand the complex impacts of climate on health. Issues with matching up health and climate datasets can be overcome by making some changes to how health data is collected and increasing engagement between the climate and health sectors.
You can read the report here to see the full review and list of recommendations. Building on what we’ve learned from this research, Wellcome will be considering what role we can play in bringing together the climate and health research communities to use these valuable resources in new ways.
