With approximately 10% of adults aged 60 years and older across SSA having an estimated travel time to the nearest hospital of greater than six hours, physical access to healthcare will likely play a major role in whether older adults in this world region will be able to seek care for COVID-19. By precisely identifying where older adults are residing who have a particularly high estimated travel time to the nearest hospital, our high-resolution maps can inform policy makers as to where interventions to increase physical access to hospital care are needed most urgently. Such interventions could include transport programs to existing hospitals and the establishment of make-shift hospitals. In addition, our maps of estimated travel time to the nearest healthcare facility of any type could help guide policy makers as to which populations are least likely to present to the healthcare system when they suffer from COVID-19 symptoms due to a lack of physical access to healthcare. This information in turn could be helpful in interpreting monitoring data on new cases of COVID-19 from different areas within countries, and in targeting testing efforts to those populations that have the greatest need for such tests.

The usefulness and policy relevance of this analysis goes beyond informing countries’ response to the SARS-CoV-2 pandemic. Physical access – that is, the time required to travel to a healthcare facility, available transport options, and costs for transport – is thought to be one of the main barriers to accessing healthcare in SSA.^5-10 Yet, there is currently very little detailed evidence on how physical access to healthcare varies across SSA, particularly within countries. Such evidence, however, is crucial to guide policy makers in identifying those areas that have the greatest need for community outreach programs, the establishment of new healthcare facilities, and improved transport infrastructure. Our study helps fill this important evidence gap for older adults in the region and is, thus, of high relevance for informing countries’ efforts to improve care for conditions that affect older adults, particularly chronic non-communicable diseases. Specifically, this study builds on the, to our knowledge, four existing studies that have mapped physical access to healthcare in SSA at a subnational level within countries, as well as studies that have examined physical access to man-made resources more generally.²⁸ Ouma et al. have examined access to emergency hospital care in SSA.²⁹ This study differs from ours in that it focussed on women of child-bearing age (15 to 49 years) rather than older adults, did not include primary healthcare facilities nor any private-sector healthcare facilities, did not use OSM data, employed a cut-off of travel time less or greater than two hours (based on a target set by the Lancet Commissions for Global Surgery 2030³⁰) rather than analysing the whole distribution, analysed data from 2015, and did not provide detailed country-by-country maps. Other relevant studies have focussed on the effect of physical access to a healthcare facility on the probability of seeking care for a febrile episode in children,³¹ estimated travel time to healthcare facilities among populations at risk of viral haemorrhagic fevers,³² and physical access to major district and regional hospitals.³³

Another key contribution of our study is the collation of a new dataset on geo-tagged healthcare facilities in SSA. By making this dataset available in the public domain and including the location of other age groups (not merely adults aged 60 years and older), we allow researchers and policy makers to run their own analyses for a variety of demographic groups and add to (or alter) the list of geo-tagged healthcare facilities in a country. There is currently no authoritative source of the location of all healthcare facilities in SSA. We have combined data from the only two existing sources of data for the geolocation of healthcare facilities in the region. We chose this approach because it is highly likely that neither dataset is complete, as is evidenced by the fact that in some countries the MFL dataset listed a higher number of healthcare facilities than the OSM data while the opposite was the case in other countries. Because the OSM project relies on volunteers to map and tag healthcare facilities, the OSM data by itself is particularly likely to underestimate the density of healthcare facilities in an area. In addition, because the categorization of a healthcare facility as a primary care facility or a hospital relies on the judgement or knowledge of the person tagging the facility, the OSM data likely has inaccuracies in the categorization of healthcare facilities into different types. For instance, OSM listed far more hospitals than primary care facilities for the Central Africa Republic, which appears unlikely to be correct. The fact that OSM contained a higher number of healthcare facilities in many countries than did the MFL dataset, especially for hospitals, is, however, encouraging in that OSM appears to be a useful source of information for the geolocation of healthcare facilities. Importantly, OSM data are likely to improve over time as coverage of smartphones increases in SSA and more volunteers map out their local areas. We will, therefore, update our dataset on a regular basis. Moving forward, it will be important to continuously monitor the validity of the data entered into the OSM and MFL dataset; a task that would ideally be accomplished by Ministries of Health in SSA.

This study has several limitations. First and foremost, while we (by combining MFL with OSM data) likely provide the most comprehensive source of data for the geolocation of healthcare facilities that is currently available, it is still likely to miss a substantial proportion of healthcare facilities. The degree to which this is the case probably varies between countries as both the participation in the OSM project and the degree to which documentation used for the MFL dataset were available and complete differ across countries.³⁴ Second, we do not have any data on the readiness of healthcare facilities to provide care, nor the quality of care provided at healthcare facilities. Similarly, we did not have information on the functioning of referral systems from primary to secondary and tertiary care, which impacts access to effective healthcare for COVID-19 and other conditions requiring more specialized care. These factors are likely to vary across and within countries.³⁵ Third, a limitation of our analysis for the COVID-19 response is that governments may decide that not all hospitals in a country should be providing care for COVID-19. Fourth, our analysis does not take into account that vulnerability to COVID-19 is likely affected by factors beyond age that vary across and within countries, including HIV, tuberculosis, and malnutrition. We decided against including these factors in our analysis because which conditions increase the risk of experiencing a severe disease course, and to what degree, is still largely unknown, especially in SSA. Fifth, we did not examine to what degree the MFL and OSM dataset contain the same healthcare facilities. Our findings are, thus, estimates for travel time to the nearest healthcare facility regardless of whether the facility is contained in the MFL or the OSM dataset. This strategy does not introduce any bias as long as the same healthcare facility has the same or very similar geographic coordinates in both datasets. It is, however, possible that the geographic coordinates for the same healthcare facility differed between the two datasets, in which case our analysis would consider these to be two different healthcare facilities and, thus, underestimate the true travel time. Lastly, our travel time numbers are approximations that, for example, do not take into account the frequency of transport services and assign an estimated (rather than measured) travel speed to different types of roads. Similarly, we assumed that individuals were able to cross national borders and incurred no additional time cost from doing so. In border regions where these assumptions do not hold true, our estimated travel times would, thus, underestimate the true travel time. Lastly, our analysis focusses on only one aspect of access to healthcare and thus does not, for instance, take into account financial barriers to accessing care.

Most countries in SSA contain areas in which older adults have little to no physical access to a hospital and (albeit to a lesser extent) healthcare facilities of any type. If COVID-19 becomes a generalized epidemic that infects large swaths of countries’ populations in the region, then it will be older adults living in these areas who are in a particularly high need for either improved transport options to existing hospitals or the provision of make-shift hospital care. Beyond their usefulness for the COVID-19 response, our maps could inform health system planning for conditions that commonly affect older adults, such as expansions of care for myocardial infarctions and strokes.