By Doris Kwesiga | In mid-2017, I attended a meeting preparing for introduction of maternal immunization in Low and Middle Income Countries (LMICs). A large part of the discussion was around the need to be able to track adverse outcomes in the community, but options on an available cohort to do so were few. The debate about the best platform was endless. Additionally, students, researchers and policy makers frequently cite Demographic and Health Surveys (DHS) as the “best” data source for background statistics.
However, I have learned through my work that DHS data, while providing a good idea of country level statistics, is not as current as it appears. For instance, the 2016 Ugandan DHS, whose report was released in 2018, has data going as far back as 5 years before the study. No way is this current data!
With the global push to reduce maternal mortality to below 70 deaths per 100,000 live births and newborn deaths to 12 per 1,000 live births in the Sustainable Development Goals (SDGs), good and reliable data is required for tracking progress to achieving these. What are the alternative data sources, given the challenges with the DHS? Are these a good alternative anyway?
The HDSS sites are geographical areas in countries that have been specially mapped out, right down to the household level, including the use of geographical information systems and unique identification numbers. They regularly track specific indicators of population health, for instance age, socio-economic status, pregnancies and their outcomes (births, newborn deaths, maternal deaths), fertility rates, deaths and their causes, among others. The frequency of this longitudinal data collection varies depending on resources, and can be as regularly as every two months or twice a year. In between these rounds, community informants provide information on key events, which are then registered.
What are the strengths of HDSS sites for MNH data?
- They collect real-time data prospectively. They do not use modelling, and so they provide a more realistic picture of current MNH status.
- HDSS sites collect data on other determinants of health that could be linked to MNH mortality, for example antenatal care attendance.
- They could contribute to monitoring different indicators in the SDG agenda or other key research priorities like maternal immunization. Where they are not currently collecting this data, it is possible to add it to their routine tools, or do standalone rounds when resources allow. More so, this data can be linked to health facility utilization, as some HDSS sites are already doing.
- It is possible for HDSS sites to pool or merge data and do comparisons within and across countries, thus providing national and regional MNH information
Importantly, HDSS sites have shown eagerness to collaborate across regions and are already doing so. They want to increase visibility of the work they do, but most importantly to have this impact policy and improve the lives of their communities. Additionally, they have a wide base of technical experts and scientists.
Current challenges in the HDSS sites
Nevertheless, they also have barriers that must be honestly acknowledged.
- Definitions of key outcomes are not harmonized across the different sites, often making it cumbersome to pool data. Data collection tools and information collected also differs.
- Many still use paper data collection, a very slow and laborious process that affects accuracy and timeliness of their results.
- The high cost of running an HDSS site, especially when combined with the need to give back to the communities which provide the data. Some HDSS sites like the Ethiopian ones are more fortunate to have Government funding, but most remain privately backed. Indeed, a few have started to shut down their activities.
- At a higher level, HDSS sites so far have barely been able to demonstrate how useful they are and can be. People are unaware of their immense capabilities they often have no business model, and their priority to the global community is not yet well established.
What can be done to improve on mining of the HDSS gold?
Through multiple discussions with HDSS teams, researchers and other stakeholders, one key issue is the importance of strengthening HDSS data collection processes. This includes standardization and harmonization of definitions and tools and having an indication of the minimum data that all sites should collect on MNH.
Linking HDSS sites to the policy makers and MNH development agenda in their countries and beyond: Advocacy should be done to show the importance of their results, so they can complement the DHS, WHO data, health facility and local data. Their data can reflect national data and should be translated to inform governments of varied characteristics across the country.
HDSS sites should identify high impact, answerable questions and have a clear research agenda. These could be descriptive epidemiology questions, discovery related questions, or delivery questions#
The author is a Research Associate with the Makerere University Centre of Excellence for Maternal Newborn and Child Health.