Why good data is crucial to achieving universal healthcare

Designing strategies to reach this admirable goal, monitoring progress towards it and measuring its achievement requires comprehensive and high-quality data on not only healthcare services, but the populations of countries themselves.

In many regions data is lacking

Across the low- and middle-income regions of the world, data on health systems, services offered, and costs are often lacking, particularly at subnational scales. Improvement of these data to fill substantial gaps in our abilities to assess progress towards universal healthcare is vital and has rightly received significant attention recently.

However, accurate, recent and reliable data on population numbers and distributions remain equally as important. The goal of assuring access to healthcare for everyone, necessitates knowing where everyone is to ensure it is achieved.

Knowing where to expand or upgrade health facilities, target interventions and reach those left behind is challenging without reliable data on populations and their characteristics.

Processes improve, but gaps remain

In high-income countries with strong national statistical systems, it is often taken for granted that accurate population and housing censuses can be undertaken regularly, and the data produced from these kept up to date in the inter-censal period through comprehensive civil and vital registration systems. Such systems are so strong in some countries that it has been decided that there is no need for expensive and logistically challenging national censuses.

The picture across many low- and middle-income countries can be different, however. While census processes continue to be improved and registry systems established and strengthened, significant gaps remain.

With the 2020 round of censuses yet to take place in many countries, there can still be a strong reliance on projections from previous enumerations, often eight to ten years ago.

Projections can be highly uncertain

In settings where migration rates, urbanisation and fertility rates are high and heterogeneous across space and time, these projections are highly uncertain, especially at small areas.

In the most resource-poor settings, the last census can be more than 15+ years ago, with registry systems almost non-existent. Examples include Afghanistan, the Democratic Republic of Congo and Iraq where the last censuses were 1979, 1984 and 1997, respectively. Here, health systems and data on them are weak, and the lack of recent and reliable data on population distributions compounds this problem, making even simple assessments of healthcare coverage challenging.

UHC is a geographical issue

Ensuring access to healthcare for all is a fundamentally geographical issue – if the people in one remote village do not have access to healthcare, then universal healthcare cannot be claimed. But without basic data on where all the villagers across the world are, we cannot even get to this stage of assessment.

The importance of population data to not only healthcare access, but all government operations and beyond is prompting innovation in obtaining recent and reliable small area population data, even in those settings where a new census seems a long way off due to conflict, logistical and financial limitations.

Improvements in the availability of detailed satellite imagery across large areas, and computing power and tools to process it to extract data on even the most remote residences, is however facilitating new cost-effective approaches to mapping populations.

WorldPop datasets

The development of powerful statistical methods to integrate satellite-derived data on buildings and settlements with other geospatial layers has enabled the disaggregation of population estimates at coarse scales to 100 x 100m grid square maps of population counts and characteristics globally.

Such datasets have been built and made freely available by WorldPop at the University of Southampton, through www.worldpop.org (see image).

Moreover, in those settings where census data are outdated, incomplete or inaccurate, population estimation approaches are being tested and adopted, whereby rapid and cheap small area ”micro-census” surveys are used to train and validate predictive models of population counts and demographics at small areas using satellite and other geospatial datasets.

GRID3 programme

These approaches have been applied in Afghanistan and Nigeria (see image) and are being scaled up to many more countries in collaborations with governments through the GRID3 programme at grid3.org.

Capacity to utilise the datasets produced is being strengthened, and use cases across governments are being developed, with healthcare access a key focus.

The simultaneous collection of data on the locations, services and capacities of health facilities across countries is enabling the analysis and measurement of geographical access to healthcare for even the most remote rural populations that have been put on the map and counted for the first time.

New technology is driving forward approaches to obtaining data

Every person should have the right to access healthcare without financial hardship, but many across the world currently do not, especially in low- and middle-income settings. Universal healthcare is something the international community should strive for; but knowing when and how it can be achieved is impossible without even the most basic data on population distributions and infrastructure.

New technologies, computing power and forward-thinking governments are driving forward approaches to obtain these data however, where financial and logistical barriers once existed.

Putting people and healthcare infrastructure on the map ensures that no one is left behind, pinpointing those populations without access and facilitating targeted and efficient strategies to reaching universal coverage.