Impact analysis of flood-induced changes in geographical accessibility and coverage to healthcare in both public and private sector, Kenya

Impact analysis of flood-induced changes in geographical accessibility and coverage to healthcare in both public and private sector, Kenya | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Impact analysis of flood-induced changes in geographical accessibility and coverage to healthcare in both public and private sector, Kenya Bibian N. Robert, Samuel K. Muchiri, Emma W. Kahoro, Boneya H. Hindada, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7724672/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 18 Mar, 2026 Read the published version in International Journal of Health Geographics → Version 1 posted 11 You are reading this latest preprint version Abstract Background Climate change is causing more frequent and severe extreme weather events, threatening health system resilience worldwide. In April and May 2024, Kenya experienced unprecedented extensive floods with devastating outcomes. However, the quantitative impact of flooding on geographical accessibility to healthcare remains unclear. This study evaluates post-disaster accessibility to health facilities and quantifies geographical coverage losses resulting from flooding compounded by a doctors’ strike in Kenya. Methods We assembled geospatial datasets including health facility locations (public, private not-for-profit (PNfP), and private for-profit (PfP)), road networks, land use/land cover, topography, population density, and flooding extents. We defined a pre-flood baseline and three post-flood scenarios using satellite-derived flooding extents (Sentinel 1 SAR and NOAA-VIIRS satellites) and their combined maximal extents. Travel time (TT) to the nearest health facility by type was estimated using a least-cost path algorithm, accounting for ± 20% variations in travel speed and flood extent for sensitivity analysis. Population coverage was extracted within five 30-minute TT bands for each scenario, nationally and by county. Results We assembled 10,995 health facilities (public = 5,586; PNfP = 855; PfP = 4,554). Pre-floods, average TT to the nearest facility was 19.6 min (16.4–24.4), with public facilities at 20.7 min (17.3–25.7), PfP at 37.8 min (31.6–47.1), and PNfP at 49.2 min (41.1–61.4). Post-floods average TT increased across all sectors, longest across PNfP at 113.5 min (94.6–191.5 min) and shortest for public facilities at 48.5 min (40.5–74.5 min). Pre-floods, 94.0% (52.5 million) of the population had access within 30-min and 20 out of 47 counties with an average TT of < 2-hours. Under the maximal flood extents, coverage dropped to 73% (40.9 million) and only 5 counties retained < 2 hours TT. County-level 30-min coverage losses ranged from 1.0% (Nairobi) to 51.0% (Narok). In several arid counties, populations facing 2 + hours TT rose to 15–31%, up from 4–12% pre-floods. Conclusion Kenya’s health system is highly vulnerable to floods, causing unequal disruptions in geographical access across subnational region. Incorporating disaster preparedness into county health care planning to strengthen health system resilience nationwide is needed. Geographical access Travel time Flooding Healthcare system Population affected Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Ensuring equitable healthcare access is crucial for improving population health outcomes ( 1 ). Healthcare access, which involves obtaining appropriate services when needed, includes physical proximity to facilities, affecting service use ( 2 ). Physical accessibility is assessed by realised accessibility, considering actual travel time or distance, and potential accessibility, estimating travel time against barriers ( 3 , 4 ). Providing efficient access to health services is challenging for policymakers, especially during crises like natural disasters, when demand increases and the environment becomes complex. In such times, assessments based on hazards, exposure, and vulnerabilities guide targeted measures to boost population resilience and protect critical services and infrastructure ( 5 – 7 ). The increasing frequency and severity of extreme weather events, a direct consequence of climate change, pose significant threats to the resilience of health systems worldwide ( 8 ). In Africa, where health infrastructure is often strained, climate change is exacerbating vulnerabilities by increasing the frequency of disasters such as floods, droughts, and heat waves ( 9 ), and by diminishing the capacity to recover from these events ( 10 ). Floods stand out as the most prevalent natural disasters in sub-Saharan Africa (SSA) accounting for 60% of the reported natural disaster between 1970–2019 ( 11 ). By 2023, floods were responsible for nearly 75% of the 6.3 million disaster-induced displacements in the region ( 12 ). In 2024 alone, extreme rainfall affected 27 countries, resulting in widespread flooding that impacted over 11 million people, caused about 2,500 fatalities, and displaced 4 million individuals ( 13 ). These events have been reported to disrupt healthcare services and damage critical infrastructure ( 14 ) hindering care delivery. This underscores the urgent need to bolster health system resilience ( 15 ) to anticipate, prepare for, and respond to climate-related disruptions. In April and May 2024, Kenya experienced unprecedented flooding across several counties, marking the most extensive flood event in recent history. The floods resulted in 267 to 315 reported fatalities and left approximately 400,000 individuals in urgent need of humanitarian aid ( 16 , 17 ). The risk of cholera outbreaks and an increase in malaria cases was heightened due to stagnant waters, limited access to safe water and sanitation, and the overcrowding of under-resourced, haphazardly delineated temporary shelters ( 18 ). The floods caused substantial damage to numerous health facilities, schools and critical infrastructure ( 16 , 19 , 20 ). Various sources have also estimated considerable general losses to infrastructure and personal property ( 17 , 21 , 22 ), but the exact quantitative impact of the flooding on geographical accessibility to healthcare remains unclear. Modelling geographic accessibility and population coverage based on travel time to health facilities offers insights for disaster preparedness and response. Recent maps of access in Kenya highlight marginalised groups ( 3 ), and advances in remote sensing, geospatial tech, and post-disaster data collection allow accurate impact assessments, considering infrastructure damage, loss of road access, and movement barriers ( 5 , 6 , 17 ). However, disaster management frameworks often lack guidance on integrating spatial barriers into accessibility assessments. Despite guidelines for post-disaster needs ( 23 ), some static measures don't address dynamic accessibility changes caused by barriers like floods. Geographical models can identify gaps, support decision-making, and improve disaster response efficiency ( 5 ). Here, we used geospatial modelling approaches ( 5 ) to assess post-disaster accessibility to health facilities and analyse accessibility coverage losses resulting from extreme flooding in Kenya. The approach enables the precise quantification of disaster impacts on geographical healthcare accessibility, thereby providing invaluable insights for post-disaster interventions and resilience-building efforts. Specifically, we model travel time before (business-as-usual) and after flooding and quantify the proportion of the population that lost access to healthcare. We define flooding zones based on two independent satellites, disaggregate facilities by four ownership levels and implement sensitivity analyses based on travel speeds. Methods Kenya-Country context Kenya’s population was 47.6 million in 2019 based on the most recent census (24) and was predicted to be 54 million in 2022 spread heterogeneously across the country (25). Population density varies considerably across its counties, ranging from under 20 people in arid regions to over 500 people per square kilometre in urban centres. This uneven distribution significantly impacts healthcare. Since 2013, Kenya has embraced devolution, a system where decision-making and resource allocation for sub-national healthcare services have been transferred to 47 county governments (Supplementary File, Figure S1 ). This decentralisation aims to address local needs better. However, the national government retains control over major hospitals, health regulations, and overall policy direction. Kenya’s healthcare system comprises six service delivery levels, ranging from community services (level 1) to national referral hospitals (level 6). It includes primary care at dispensaries and clinics (level 2), health centres (level 3), secondary and tertiary care at sub-county hospitals and medium-sized private hospitals (level 4), primary referral hospitals (level 5), and national referral hospitals (level 6). Facilities are operated by the government, NGOs, FBOs, and private sector. Service complexity varies from basic medication and rapid diagnostic tests (RDTs) with lower-skilled staff to advanced laboratory procedures conducted by higher-skilled workers, depending on facility ownership and level. Kenya has experienced significant disruptions in healthcare services due to health worker strikes. Notably, the prolonged strikes in 2017 markedly impacted service delivery. Another nationwide health workers’ strike occurred from December 2020 to February 2021, driven by demands for better working conditions, including sufficient personal protective equipment, higher risk allowances, and comprehensive health insurance coverage (26,27). The duration of the strike varied across different health facilities and categories of workers. Although there is no comprehensive national database tracking these strikes, it is presumed that most health facilities were impacted throughout this period. The 2023/2024 period also witnessed strikes by healthcare workers, further straining the healthcare system (28) and significant delays in recruiting interns, a critical cadre in the provision of healthcare in Kenya (29). Kenya is highly vulnerable to flooding due to several factors. Situated in the drought-prone greater Horn of Africa region, the country is susceptible to extreme weather events (30). Heavy rainfall and El Niño cycles often lead to overflowing rivers and flash floods, exacerbated by widespread poverty and inadequate infrastructure, especially in urban cities and in arid and semi-arid areas (31). In April and May 2024, heavy rains that reached 111% to over 200% of the long-term average across the country (32), caused devastating floods across Kenya, that displaced hundreds of thousands of people, damaged infrastructure, and caused significant loss of life (33). This recent event underscores the urgent need for improved flood preparedness and mitigation strategies in Kenya. Methods overview We applied a four-step approach. The first step entailed data assembly including health facilities (public, Private not-for-profit (PfNP), and private for-profit (PfP)), road networks, land use/cover, elevation population distribution and the flooding extents. Second, we defined different travel scenarios to account for how flooding affected physical accessibility in Kenya. Third, we estimated travel time (TT) from every location to the nearest facility for i) all facilities, ii) public, iii) PfP and iv) PNfP using the least-cost path algorithm while also conducting a sensitivity analysis on the travel speeds and flood extents. Finally, we extracted the proportion of the population within 30 min, 30-60, 60-90, 90-120 and beyond 120 minutes of the nearest facility for each scenario. All estimates were aggregated at subnational units of decision-making (county and sub-county). Data a. Healthcare facilities The Kenya health facility data used in the study were accessed from the Kenya Ministry of Health’s 2023 health facility census that was carried out between the 14 th and 29 th of August 2023 (34). The survey aimed to map all health facilities in Kenya including public, private and PNfP (FBOs/NGOs). Some of the information collected during the census and relevant to the current analysis included facility name and code, county, sub-county, ward, latitude, longitude, Kenya Essential Package for Health (KEPH) level and ownership. We focused on all facilities offering general medical services to the public. From the mapped facilities, we excluded facilities offering services to a subset of the population, i.e. company and secondary school clinics, military, police and prison facilities (188). In addition, we excluded specialised facilities including dental, eye, cancer, tuberculosis, HIV voluntary counselling and testing (VCT) centres, maternal and nursing homes, drop-in centres, mental health centres, hospices, funeral homes, rehabilitation, family planning clinics, radiology/X-ray, blood transfusion, gynaecology, nephrology, kidney dialysis, diagnostic and pharmacies (1,192). Our final list of facilities contained 10,995 facilities ( Figure 1 ); 8,112 dispensaries (level 2), 1,995 health centres (level 3) and 888 hospitals (level 4 – 6). b. Factors affecting travel to healthcare Travel time to healthcare facilities was modelled using multiple geospatial factors including road network, land use/land cover (LULC), digital elevation model, water bodies and national parks and population distribution. The road network was classified into primary, secondary, county, and other roads based on the Kenya’s proposed 2023 road register (Supplementary File, Figure S2 ) (35). In areas without road network, LULC data derived from 2023 Sentinel-2 imagery at 10m resolution was used to estimate travel conditions (Supplementary File, Figure S3 ) (36). The terrain slope, derived from the Shuttle Radar Topographic Mission (SRTM) at 30-meter resolution digital elevation model (DEM) (37) (Supplementary File, Figure S4) , was incorporated to adjust walking and cycling speeds. Additionally, water bodies and national parks (38) were treated as impassable barriers (Supplementary File, Figure S1 & S3 ). Fine spatial resolution (100 x 100 m) population distribution for the whole country was obtained from WorldPop open spatial demographic data portal (39,40). Here, we used the constrained (population modelled on in areas containing built settlements) gridded population raster for 2023 (Supplementary File Figure S5 ). Additional details of these datasets are provided in Supplementary File. c. Flooding extent The flood extents were assessed using high-resolution imagery from the National Oceanic and Atmospheric Administration – Visible Infrared Imaging Radiometer Suite (NOAA-VIIRS) and Sentinel 1 Synthetic Aperture Radar (SAR) sensors, created specifically for flood response efforts in Kenya. The data was sourced from the United Nations Institute for Training and Research (UNITAR) and the International Center for Humanitarian Affairs (ICHA). The ICHA’s Sentinel 1 SAR flood extents data covered flood extents for the entire country from March 31 to May 9, 2024, at a spatial resolution of 30m, and was created using a change detection model to aggregate all predictions of flood extents since the start of flooding in the country (41). This was supplemented by flood extents for Tana-Galana River (April 29) based on 0.5m spatial resolution Pleidas satellite imagery and the Nyando River (May 5) based on Sentinel 1 SAR. Moreover, UNITAR provided two NOAA-VIIRS flood extents at the national level, covering April 24-28 and May 12-16, at a resolution of 375m (42), which were combined to represent the overall flood extent during that period. In addition to considering the Sentinel 1 SAR and the NOAA-VIIRS flood extents independently, we integrated all datasets to produce a maximal flooding extent covering the entire country from 31 st March to 9 th May 2024. Data that were in raster format were first converted to vector format, while those available as images were georeferenced and digitised to extract flood extents. The merged flood extents were then checked for topological errors and polygons that overlapped or shared common boundaries were dissolved into single polygons. All pre-processing steps described above were conducted in ArcGIS Pro version 3.0.3 (ESRI Inc., Redlands, CA, USA). Finally, to capture the sections of the roads closed due to floods, various newspaper reports and articles were sought out within the flooding period (April-May 2024). Using the Editor tool in ArcMap version 10.8.2 (ESRI Inc., Redlands, CA, USA), we split the sections of the flooded roads according to the reports and added an attribute of whether it was flooded or not in the attribute table. Travel scenarios We identified two main travel scenarios: i) prior to the flooding (Business-as-usual-BAU) and ii) during the flooding, applicable across the entire country. For each scenario, we outlined the various modes of transport and their corresponding speeds across different road networks and land covers. In the pre-flooding scenario, we utilised a travel framework developed from our earlier national spatial accessibility modelling in Kenya (5; Table 1 ). Briefly, this scenario encompasses a hybrid mode of transport involving either walking (areas with no roads), bicycling (lower class roads), motorized (motorbike and vehicles, both public and private in motorable roads) or combined forms of transport based on availability of motorable roads and class of roads. On the other hand, during flooding, all the roads within the flooding zone were considered impassable (barriers) while the facilities in the flooding zone were deemed non-functional. In addition, parts of the road network outside the flooding zone that the Kenyan government had closed owing to the floods were deemed inaccessible. Outside the flooding zones, we considered reduced speeds (by 50%) in areas across all roads (7; Table 1 ). Table 1: Travel modes and that were used to compute travel time to the nearest health facility for each road type and land cover category for two scenarios (business as usual and during flooding) Mode of transport Travel Scenarios (km/h) ID Category Business-as-usual During flooding 1 Flooding extents Various modes and corresponding speeds depending on road type and land cover Travel barrier Mode of transport Travel Scenarios (km/h) ID Category Business-as-usual During flooding 2 Trees Walking 2.5 1.25 3 Range land Walking 4.5 2.25 4 Crops Walking 4 2 5 Bare or built-up Walking 5 2.5 Mode of transport Travel Scenarios (km/h) ID Category Business-as-usual During flooding 6 Flooded vegetation Barrier 0.1 0.05 7 Water Barrier 0 0 8 Primary road Motorized 50 25 9 Secondary road Motorized 30 15 10 County roads Cycling 10 5 (walk) 11 Other minor roads Walking 5 2.5 Estimating travel time To estimate travel time for the different scenarios, we used AccessMod version 5.7.17, an open-source tool supported by the World Health Organization (WHO) to analyse geographic accessibility via a least-cost path algorithm (43). For the BAU scenario, we first merged the land cover, road network, water bodies and protected areas (usual barriers) via the “merge land cover” module in AccessMod Toolbox to create a merged gridded surface. Based on the resultant single raster, speeds from the BAU model ( Table 1 ) were applied to compute travel time to the nearest facility for i) all facilities, ii) public facilities, iii) PfP facilities and iv) PNfP facilities (FBOs/NGOs). This was necessary because, during the flooding period, doctors from the public health facilities were on industrial strike (44,45). Therefore, clients could only attend either the PfP or PNfP facilities. Conversely, during flooding , we created a unified gridded surface by utilising land cover, the road network outside the flooding area, and additional barriers (i.e., standard barriers), the maximum flooding extent, and the closed roads beyond the flooding zone. We then assigned speeds to this merged surface to calculate travel time to facilities, categorised into four groups (as previously done for BAU). Further, the travel speeds outside the flooding zones were reduced speeds by 50% ( Table 1; 7). For each scenario, the cumulative travel time from every populated location in Kenya based on WorldPop’s population distribution maps was computed towards (anisotropic) the closest facility via the least cost path (cost measured as time) at 100 x 100m spatial resolution. Sensitivity analyses We conducted two types of sensitivity analyses. First, in addition to the combined flooding extents, we computed spatial accessibility metrics after flooding based on the individual flooding extents from two sensors (Sentinel 1 SAR and NOAA-VIIRS). Second, we varied the travel speeds in Table 1 by ±20% (6,46) and repeated the least-cost path model to define an upper and lower bound of travel times for all travel scenarios (BAU and during flooding), facility ownership (all facilities, public, PfP and PNfP and sensor type (Sentinel 1 SAR, NOAA-VIIRS and combined extents). Population affected The geographic coverage estimates (the proportion of the total population within 30 min, between 30-60 min, 60-90 min, 90-120 min and beyond 120 min) of the nearest health facility disaggregated by travel scenario, sensor type and level of speed was extracted at national, county and sub-county levels. This was achieved by using Zonal statistics ArcGIS Pro Version 3.3. Results Health facilities assembled Of the 10,995 facilities assembled for this study over 97% were outside all three flooding zones (99.2%, 97.9% and 97.4% for the NOAA-VIIRS, Sentinel 1 SAR and combined flooding zones respectively). A total of 86, 232 and 289 facilities were identified to be within the NOAA-VIIRS, Sentinel 1 SAR and the combined flooding zones respectively ( Table 2) . Majority (68.9%) of the facilities within the three flooding zones were public facilities (45, 172 and 199 facilities for NOAA-VIIRS, Sentinel 2 and combined zones respectively) with the least (4.5%) being PNfP facilities (one, five, and 13 facilities for NOAA-VIIRS, Sentinel 1 SAR and combined zones respectively) Table 2 . Table 2: Summary of health facilities by type within each travel time scenario Type All facilities Within flooding zones Outside the flood zone NOAA-VIIRS Sentinel 1 SAR Both NOAA-VIIRS Sentinel 1 SAR Both Public 5,586 45 172 199 5,541 5,414 5,387 Private not-for-profit 855 1 5 13 854 850 842 Private for-profit 4,554 36 51 77 4,518 4,503 4,477 Total 10,995 86 232 289 10,909 10,763 10,706 Extent of flooding The total flood-affected area was 43,422.9 km², representing 7.3% of Kenya’s land area. This was mainly due to Sentinel 1 SAR, which covered 39,115.3 km² (6.6%), compared to NOAA-VIIRS, which covered 7,345.2 km² (1.2%). The overlap in sensor detection was 3,037.6 km², or 0.5% of the total area, and 7.0% of the total flood extent area ( Figure 2A ). Nyamira was the only county with no flood extents recorded from either sensor. Nine counties—Trans Nzoia, Nyeri, Bungoma, Kakamega, Nandi, Vihiga, Kericho, Bomet, and Kisii—were not included in NOAA-VIIRS flood extents but appeared in Sentinel 1 SAR data. Isiolo showed the highest proportion of flooded area at 17.0% based on maximum extents and also had the highest non-overlapping flood extents at 15.3%. Travel time to health facilities pre- and post-flooding Pre flooding Figure 3 illustrates the average travel times to the nearest facility across per subcounty, categorised into 30-minute intervals and disaggregated by facility type and scenario. Under the BAU scenario, the national average travel time to the closest facility was 19.6 minutes (range: 16.4 - 24.4). When disaggregated by facility type, the average travel times were 20.7 minutes (17.3–25.7) for public facilities, 37.8 minutes (31.6–47.1) for PfP facilities, and 49.2 minutes (41.1–61.4) for PNfP facilities. Nairobi County had the shortest mean travel times: 2.5 minutes (2.2 - 3.1) overall, 3.8 minutes (3.3 - 4.7) for public facilities, 2.9 minutes (2.5 - 3.5) for private facilities, and 4.1 minutes (3.5 - 4.9) for PNfP facilities ( Figure 3 ). In contrast, Marsabit County recorded the longest average travel times across all types of health facilities: 76.3 minutes (63.7 - 95.3) for all facilities, 77.1 minutes (64.3 - 96.2) for public facilities, and 185.4 minutes (154.6 - 231.6) for private facilities (Supplementary File, Figures S6-S8 ). Mandera county, however, showed the highest average travel time for PNfP facilities at 363.0 minutes (302.6 - 453.7) (Supplementary File, Figure S9 ). Thirteen counties (Garissa, Isiolo, Kajiado, Kitui, Lamu, Mandera, Marsabit, Narok, Samburu, Tana River, Turkana, Wajir and West Pokot) had travel times longer than the national average travel times for all, public, and private facilities (Supplementary File, Figures S6-S8 ), while twelve counties surpassed the national average travel time for PNfP facilities (Supplementary File, Figure S9 ). Post flooding In all worst-case scenarios (NOAA-VIIRS, Sentinel 1 SAR, and the combined data), the national average travel time to health facilities exceeded that of the BAU scenario ( Figure 3 ). Specifically, the average travel time to the nearest facility (covering all facilities) was 55.6 minutes (range: 46.5 - 69.6) for Sentinel 1 SAR-derived extents, 45.8 minutes (38.3 - 57.1) for NOAA-VIIRS, and 56.3 minutes (47.0 – 70.3) for the combined flooded extents. Similar patterns were observed across the different facility ownership types and flooding scenarios, with the longest travel times for PNfPs facilities was 113.5 minutes (range: 94.6 - 191.5) and the shortest for public facilities was 48.5 minutes (range: 40.5 - 74.5), regardless of flooding extent. The combined flooding scenario resulted in the greatest reduction in travel times ( Figure 3 ; Supplementary Figures S6-S9 ). Geographic coverage pre- and post-floods Pre flooding Figure 4 shows the proportion of Kenya’s population within 30-minute travel time bands to all facilities across different flooding scenarios. Before the floods (BAU), 94.0% of the population had access to health facilities within 30 minutes ( Figure 4, panel 1 ) and 0.7% lived outside 2 hrs travel time ( Figure 4, panels 2-5 ). County level population access within 30-min access varied between 66.4% (Garissa) to 100% (Kisii). Overall, 30 of 47 counties had over 94% of their population within the 30-minute access range, matching the national average. Only eight counties had their entire population within 60 minutes, 16 counties within 90 minutes and 20 counties within 2 hrs travel time. Post flooding During flooding events (under combined scenarios), the proportion of the population with timely access within 30 minutes decreased from 94.0% to 73.3%. Specifically, between sensors, only 19 counties (Sentinel 1 SAR scenario), 20 counties (NOAA-VIIRS), and 17 counties (combined) maintained at least 75% of their population coverage within 30 minutes, as shown in Figure 4, panel 1 . The reduction in population coverage within this 30-minute threshold at the county level ranged from as low as 1.0% in Nairobi to as high as 51.0% in Narok. The most significant declines (41.0-51.0%) were seen in Narok, Turkana, Makueni, Tana River, Kitui, Laikipia, Lamu, Wajir, Nyandarua, and West Pokot. Conversely, counties such as Nairobi, Vihiga, Kiambu, Kisii, Mombasa, and Nyamira experienced relatively small coverage losses (1-10% ; Figure 4 ). Only five counties- Kakamega, Kisii, Nairobi, Nyamira, and Vihiga- retained full population access within a 2-hour travel time. Counties including Wajir, Garissa, Turkana, Marsabit, Samburu, Isiolo, and Tana River saw 15-31% of their populations falling beyond 2 hours of access after the floods, an increase from 4-12% before the floods ( Figure 4, panel 5 ). Similar reductions in access to health services across public health facilities were observed, with the most severe impacts in counties in northern and northeastern Kenya, which have fewer private and PNfP facilities (Supplementary Files, Figures S10-S12 ). Discussion The 2024 Kenya floods recovery needs assessment and other humanitarian impact reports highlighted extensive disruptions to health services following the March–May 2024 floods ( 17 ). While these reports documented cross-sectoral impacts and proposed broad mitigation measures, they did not quantify health service coverage losses. Our study addresses this critical gap by assessing the impact of floods on geographical access to health care facilities by quantifying travel time during pre-floods (BAU) and post-flood. The post-floods scenarios were analysed using Sentinel 1 SAR and NOAA-VIIRS satellite data and a combined scenario integrating the two datasets. Our results reveal a substantial reduction in timely (within 30-min) access to health care facilities, with population coverage dropping from 94% (52,488,181 million) pre-flood to 73% (40,916,717 million) post-floods. The increased travel time to health facilities was also observed across all health facility types. These findings underscore the sensitivity of Kenya’s health system to extreme weather events and highlight the urgent need for frameworks that aid in disaster preparedness to mitigate the risk of loss of access to healthcare during crises ( 47 ). Flooded extents varied substantially across counties (Supplementary Figure S13) reflecting a complex interplay of topographical and socio-environmental factors beyond rainfall intensity ( 48 ). For instance, the largest flood extents were observed in diverse ecological and geographic settings: semi-arid counties such as Isiolo, Laikipia, and Turkana, which typically experience low rainfall and limited drainage infrastructure; Mombasa, a coastal urban county prone to tidal influences and poor urban drainage; and Nakuru, a highland region with urban zones where topography and land use may contribute to localized flooding. These spatial differences were further influenced by the characteristics of specific satellite sensors. For example, Sentinel-1 SAR, with its finer spatial resolution (5 × 20 m), was able to detect small-scale flooded areas across all counties that may have been missed by NOAA-VIIRS, likely due to its coarser resolution (375 m). Notably, in Mombasa County NOAA-VIIRS detected wider flooding extents than Sentinel 1 SAR, while in counties like Narok, Nakuru and Laikipia, the opposite was true. This sensitivity of results and flood extent estimates to the satellite sensor selected highlights the crucial role of high-resolution, real-time geospatial data in disaster response efforts ( 49 ). In Nyamira County, neither sensor detected flood extents despite reports of infrastructural damages caused by the heavy rainfall ( 50 , 51 ). Such limitations in satellite-based flood monitoring emphasise the need for standardized, multi-source flood assessment methods to enhance detection accuracy, especially in low-resource settings ( 7 , 52 ). Under the BAU scenario, over 90% of the population resided within half an hour of a health facility. However, this national coverage masks out some of the huge inequalities particularly in arid and semi-arid counties such as Garissa, Isiolo, Marsabit and Wajir where travel times exceeded 1 hr, reflecting persistent access challenges in marginalised regions ( 3 , 46 ). The floods further exacerbated these disparities reducing the coverage of people within 30 minutes nationally to 73% (combined scenario), with northeastern counties experiencing longer travel times above national average. These counties are characterised by limited road infrastructure, and they heavily rely on public facilities. For instance, Mandera had an average travel time of 37 minutes to a public facility under the BAU scenario as compared to over 3.5 hrs average travel time to PfP and PNfP facilities. After floods travel time to PNfP facilities alone in this county doubled (6hrs). In addition, PfP and PNfP facilities that serve as critical alternatives during public sector strikes ( 28 , 44 ), were concentrated in urban centres, leaving rural populations with even fewer options post-disaster. Floods have disproportionately disrupted timely access to health facilities, exposing critical geographic, hydrological, and infrastructural vulnerabilities across counties ( 48 , 53 ). Counties with the greatest loss in 30-minute coverage—such as Narok, Tana River, Lamu, Kitui, and Makueni—are located in major flood-prone areas identified in flood advisories issued months before the floods ( 54 , 55 ). Notably, Kitui experienced the highest coverage loss at 51% and also sustained the most extensive damage to health facilities at 47%, including a referral hospital ( 17 ). Turkana, Tana River, Wajir, and Lamu had the highest proportion of facilities within flooded areas, with at least nine facilities in Tana River reported damaged ( 20 ). Laikipia and Nyandarua were among those with the largest flooded extents. The correlation between health facility exposure and coverage decline was also evident in longer travel times (60 to 120 minutes) across counties like Marsabit, Turkana, Tana River, Wajir, Isiolo, and Garissa, where limited road infrastructure further hindered access. Our findings reveal that even minor disruptions in coverage (1–10%) in densely populated counties such as Vihiga, Kiambu, Kisii, Mombasa, and Nyamira can lead to significant increases in the number of people facing longer travel times, despite generally good infrastructure. These delays have varied impacts on health service utilization. Notably, Kiambu experienced some of the steepest declines in hospital admissions, outpatient visits, and immunization services ( 17 ). This underscores the need to interpret coverage losses not just in percentage terms but through the lens of absolute population affected and spatial context. The fact that only five counties had the entire population within 2 hours travel time post-floods down from 20 pre-floods, further underscores the urgency of integrating disaster preparedness policies to ensure responsive, equitable healthcare access during extreme weather events ( 5 ). Policy implications Kenya continues to experience recurrent and severe flood events resulting in fatalities, destruction and disruption of essential infrastructural services, including health care ( 56 ). Our study highlights the sensitivity of Kenya’s health system to flood-related shocks, offering insights that complement existing recovery strategies and help mitigate long-term impacts on the population. Policymakers should prioritise resource allocation to high-risk counties, not only those vulnerable to flooding but also those susceptible to post-disaster health risks. For instance, Tana River, Lamu and Siaya counties experienced cholera outbreaks following the March-May floods. Mapping access to care losses including at the sub-county level (Fig. 3 ) offers actionable insights for targeted emergency response. Further, leveraging multi-source flood assessment approaches, including geospatial and drone technologies ( 57 ) and including participatory approaches to gather historical flood impacts from communities, can enhance informed decision-making and develop contingency strategies, such as mobile clinics and emergency transport corridors, to buffer access disruptions during disasters. Enhancing health system resilience amid escalating climate risks will require stronger intersectoral coordination among health, infrastructure, humanitarian, and meteorological agencies. It is also critical that both national and county governments treat flood advisories and weather forecasts ( 54 , 58 ) as actionable early warnings, prompting proactive disaster prevention and preparedness rather than reactive responses. Strengths This study offers several strengths. Unlike previous similar studies ( 59 , 60 ) that rely solely on assumptions such as reduced travel speeds during flood events, we incorporated actual flooded extents to comprehensively capture access disruptions caused by flooding. To overcome the limitations of any single sensor, we employed multi-sensor data. Further, the Sentinel 1-derived flood extents was an aggregation of satellite data since the flooding started and hence not from a single point in time. In addition, Sentinel 1 SAR sensor is not affected by cloud cover and by lack of daylight which ensured uninterrupted monitoring of flooding events. We also integrated official records on road closures provided by Kenyan road authorities ( 19 , 61 ). Additionally, since travel speeds used in the analysis are estimates rather than directly observed, we conducted a sensitivity analysis increasing/decreasing speed by 20% to adjust for uncertainties. We also used the most recent and comprehensive database of health facilities, compiled through a nationwide survey conducted by the Ministry of Health. Finally, we disaggregated our analysis by health sector owing to doctor strike and household wealth differentials in terms of money that influence the ability to access care. Limitations Notwithstanding the valuable insights on understanding the impacts of flooding on access to health care in Kenya, there are several limitations. The change detection model that produced the Sentinel 1 SAR-defined flood extents may have resulted to some false positives within the semi-arid areas for instance in Wajir. NOAA-VIIRS is an optical sensor, and its flood detection capabilities is constrained by cloud cover and lack of daylight, further contributing to potential underestimation of inundation. This study incorporated flooded road segments in addition to the flood extents, but only segments reported to have flooded were accounted for possibly missing other affected roads not documented. Furthermore, we did not know the extent of damaged roads, limiting our mapping of infrastructure disruption. We assumed that population within flood zones had no alternative means of accessing health facilities. However, in some affected areas boats were reportedly used. We also lacked data on conditions outside the mapped flood zones, so we couldn’t account for disruptions caused by unusually heavy rainfall in those areas. We assumed individuals travel to the nearest facility, which may not reflect actual healthcare-seeking behaviour or preferences. All flooded roads were assumed impassable although some vehicles (e.g. trucks) may have been able to navigate certain segments. Lastly, this study only focused on quantifying the impact of floods on geographical access to health facilities, highlighting spatial disparities in health care access during floods. However, further research is needed to assess county-level adaptive capacity which is essential for understanding overall health system vulnerabilities to extreme weather events. Conclusions One of the core responsibilities of the health system is to prevent, prepare for, detect and respond to public health threats ( 62 ) including floods, the second most prevalent natural disaster in Kenya. Our findings show that flood events significantly reduce geographical access to healthcare services, with pronounced subnational disparities across flood-prone counties, semi-arid regions with persistent infrastructural deficits, and densely populated regions. The loss of access to health care is largely driven by impacts on flood-exposed health facilities, spatial extent of flooding, and damaged transport system inhibiting access. However, the impact on health services utilisation varies sub-nationally. Considering compounding health risks associated with floods such physical injuries, the spread of communicable diseases (e.g., measles) due to overcrowding in displacement settlements, disease outbreaks (malaria, cholera, etc), it is critical to ensure that health care system can sustain service delivery amid recurring climate shocks. Abbreviations BAU Business-as-usual DEM Digital elevation model ESRI Environmental Systems Research Institute FBOs Faith-based Organisations ICHA International Center for Humanitarian Affairs KEPH Kenya Essential Package for Health LULC Land use/cover MoH Ministry of Health NGOs Non-government organisations NOAA-VIIRS National Oceanic and Atmospheric Administration – Visible Infrared Imaging Radiometer Suite PfP Private-for-profit PNfP Private-not-for-profit RDTs Rapid diagnostic tests SRTM Shuttle Radar Topographic Mission SSA sub-Saharan Africa TT Travel time UNITAR United Nations Institute for Training and Research VCT Voluntary counselling and testing WHO World Health Organisation Declarations Availability of data and material: The health facility data is available from the Ministry of Health (https://www.health.go.ke/contact-us).Sentinel-2 flooding extent from the International Center for Humanitarian Affairs (ICHA; https://redcrosske.maps.arcgis.com/home/item.html?id=bce5234218c547439ce2a71f7cbeb4e2) and NOAA-VIIRS flooding extent from the United Nations Institute for Training and Research (https://unosat.org/products/). The roads data from the Kenya Roads Board portal (https://maps.krb.go.ke/kenya-roads-board12769/maps). The Digital elevation model can be downloaded from the Regional Centre for Mapping of Resources for Development (RCMRD) geoportal (https://gmesgeoportal.rcmrd.org/datasets/rcmrd::kenya-srtm-dem-30meters/about).The land use land cover from the Environmental Systems Research Institute (ESRI’s) ArcGIS platform (https://www.arcgis.com/home/item.html?id=6df9ed7a1eda4ed58023456b7c5484fd), protected areas for the Global database of protected areas (https://www.protectedplanet.net/en/about) and population data from the WorldPop open spatial demographic data portal (https://www.worldpop.org/) . Acknowledgements: We are grateful to the International Centre for Humanitarian Affairs (Kenya Red Cross Society) for providing Sentinel 1 SAR flood extent data and the Ministry of Health (MoH) for providing health facilities data Competing interests Not Applicable Funding: EAO is supported by the Wellcome Trust Senior Fellowship (#224272) which supported BNR. SKM is supported by the Wellcome Trust Principal Fellowship (#212176). PMM is supported by the Fonds voor Wetenschappelijk Onderzoek – Research Foundation Flanders Senior Postdoctoral Fellowship (#1201925N). BNR, SKM, PMM and EAO are grateful for the support of the Wellcome Trust to the Kenya Major Overseas Programme (#203077). The views expressed in this publication are those of the authors and not necessarily those of Wellcome Trust. The funders had no role in study design, data collection, data analysis, data interpretation, or writing of the report. Ethics approval and consent to participate. No individual patient-level data was used in this publication. Consent for publication Not applicable - The manuscript does not contain data from any individual person. Authors' contributions BNR: Data curation, Formal Analysis, Conceptualisation, Investigation, Methodology, Software, Validation, Visualisation, Writing – original draft, Writing – review & editing; SKM: Data curation, Formal Analysis, Conceptualisation, Investigation, Methodology, Software, Validation, Visualisation, Writing – original draft, Writing – review & editing; EWK : Data curation , Writing – review final draft; BHH: Data curation , Writing – review final draft; HK : Data curation, Writing – review final draft; EAO: Conceptualisation, Investigation, Methodology, Funding acquisition, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing; PMM: Data curation, Formal Analysis, Conceptualisation, Investigation, Methodology, Validation, Visualisation, Supervision, Writing – original draft , Writing – review & editing; All authors contributed to the final manuscript. References Penchansky R, Thomas JW. The concept of access: definition and relationship to consumer satisfaction. Med Care. 1981 Feb;19(2):127–40. Levesque JF, Harris MF, Russell G. Patient-centred access to health care: conceptualising access at the interface of health systems and populations. Int J Equity Health. 2013 Mar 11;12:18. Moturi AK, Suiyanka L, Mumo E, Snow RW, Okiro EA, Macharia PM. Geographic accessibility to public and private health facilities in Kenya in 2021: An updated geocoded inventory and spatial analysis. Front Public Health. 2022;10:1002975. Ouma PO, Maina J, Thuranira PN, Macharia PM, Alegana VA, English M, et al. Access to emergency hospital care provided by the public sector in sub-Saharan Africa in 2015: a geocoded inventory and spatial analysis. Lancet Glob Health. 2018 Mar;6(3):e342–50. Hierink F, Rodrigues N, Muñiz M, Panciera R, Ray N. Modelling geographical accessibility to support disaster response and rehabilitation of a healthcare system: an impact analysis of Cyclones Idai and Kenneth in Mozambique. BMJ Open. 2020 Nov 3;10(11):e039138. Tariverdi M, Nunez-Del-Prado M, Leonova N, Rentschler J. Measuring accessibility to public services and infrastructure criticality for disasters risk management. Sci Rep. 2023 Jan 28;13(1):1569. Patrascu FI, Mostafavi A, Vedlitz A. Relationship between Access to Critical Facilities During Normal Times and Disrupted Access During Extreme Weather Events, and Underlying Disparities [Internet]. Rochester, NY: Social Science Research Network; 2022 [cited 2025 Aug 23]. Available from: https://papers.ssrn.com/abstract=4108983 Wade R. Climate Change and Healthcare: Creating a Sustainable and Climate-Resilient Health Delivery System. J Healthc Manag Am Coll Healthc Exec. 2023 Aug 1;68(4):227–38. Wright CY, Kapwata T, du Preez DJ, Wernecke B, Garland RM, Nkosi V, et al. Major climate change-induced risks to human health in South Africa. Environ Res. 2021 May;196:110973. Africa Risk Capacity. The State of Natural Disasters in Africa [Internet]. 2024 [cited 2025 Aug 23]. Available from: https://www.arc.int/sites/default/files/2024-07/ARC_white_paper_2024.pdf World Meteorological Organisation (WMO). State of the Climate in Africa 2024 [Internet]. [cited 2025 Jun 9]. Available from: https://library.wmo.int Internal Displacement Monitoring Center (IDMC). Internal Displacement in Africa (2024) [Internet]. Internal Displacement Monitoring Center (IDMC); 2024 Nov [cited 2025 Jun 9]. Available from: https://www.internal-displacement.org/publications/internal-displacement-in-africa/ Studies the AC for S. Record Levels of Flooding in Africa [Internet]. Africa Center. [cited 2025 Aug 23]. Available from: https://africacenter.org/spotlight/record-levels-of-flooding-in-africa-compounds-stress-on-fragile-countries/ Petricola S, Reinmuth M, Lautenbach S, Hatfield C, Zipf A. Assessing road criticality and loss of healthcare accessibility during floods: the case of Cyclone Idai, Mozambique 2019. Int J Health Geogr. 2022 Oct 12;21(1):14. Health systems resilience toolkit: a WHO global public health good to support building and strengthening of sustainable health systems resilience in countries with various contexts [Internet]. [cited 2025 Aug 23]. Available from: https://www.who.int/publications/i/item/9789240048751 OCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #5 (10 May 2024) | OCHA [Internet]. 2024 [cited 2025 Aug 2]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-5-10-may-2024 Kenya Floods Recovery Needs Assessment 2024 [Internet]. [cited 2025 Aug 22]. Available from: https://www.undp.org/sites/g/files/zskgke326/files/2025-05/kenya_floods_recovery_needs_assessment_2024.pdf WHO. Greater Horn of Africa - WHO 2024 Health Emergency Appeal [Internet]. 2024 [cited 2025 Aug 7]. Available from: https://cdn.who.int/media/docs/default-source/documents/emergencies/2024-appeals/greater-horn-of-africa---who-2024-health-emergency-appeal.pdf OCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #1 (11 April 2024) | OCHA [Internet]. 2024 [cited 2025 Aug 22]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-1-9-april-2024 OCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #6 (17 May 2024) | OCHA [Internet]. 2024 [cited 2025 Aug 22]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-6-17-may-2024 Kenya Red Cross. Floods operations, 2024 [Internet]. 2024 [cited 2025 Aug 7]. Available from: https://www.redcross.or.ke/wp-content/uploads/2024/06/MAM-SitRep-18th-June-2024.pdf OCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #7 (19 June 2024) | OCHA [Internet]. 2024 Jun [cited 2025 Aug 7]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-7-19-june-2024 Post-Disaster Needs Assessment Guidelines [Internet]. [cited 2025 Aug 23]. Available from: https://www.who.int/publications/i/item/post-disaster-needs-assessment-guidelines Kenya National Bureau of Statistics, editor. 2019 Kenya population and housing census. Nairobi: Kenya National Bureau of Statistics; 2019. KNBS. Kenya National Bureau of Statistics [Internet]. 2021 [cited 2025 Aug 23]. Available from: https://www.knbs.or.ke/ Wambua S, Malla L, Mbevi G, Nwosu AP, Tuti T, Paton C, et al. The indirect impact of COVID-19 pandemic on inpatient admissions in 204 Kenyan hospitals: An interrupted time series analysis. PLOS Glob Public Health. 2021 Nov 17;1(11):e0000029. Irimu G, Ogero M, Mbevi G, Kariuki C, Gathara D, Akech S, et al. Tackling health professionals’ strikes: an essential part of health system strengthening in Kenya. BMJ Glob Health [Internet]. 2018 Nov 28 [cited 2025 Aug 23];3(6). Available from: https://gh.bmj.com/content/3/6/e001136 Nakweya G. Kenya: Patients are turned away from hospitals as doctors’ strike enters fourth week. BMJ. 2024 Apr 11;385:q845. Kenya Medical Association. Kenya Medical Association Statement on delayed posting of medical interns 2024 [Internet]. [cited 2025 Aug 7]. Available from: https://kma.co.ke/images/KMA_Statement_on_Delayed_Posting_of_Medical_Interns_20240222.pdf Osima S, Indasi VS, Zaroug M, Endris HS, Gudoshava M, Misiani HO, et al. Projected climate over the Greater Horn of Africa under 1.5 °C and 2 °C global warming. Environ Res Lett. 2018 May;13(6):065004. Extreme Rainfall and Flooding over Central Kenya Including Nairobi City during the Long-Rains Season 2018: Causes, Predictability, and Potential for Early Warning and Actions [Internet]. [cited 2025 Aug 23]. Available from: https://www.mdpi.com/2073-4433/9/12/472 NDMA - KnowledgeWeb [Internet]. [cited 2025 Jun 10]. Available from: https://knowledgeweb.ndma.go.ke/Public/Resources/ResourceDetails.aspx?doc=28861990-6d58-4404-ba1a-bf7d0a940ea6 Kenya: Floods - Apr 2024 | ReliefWeb [Internet]. 2024 [cited 2025 Jun 10]. Available from: https://reliefweb.int/disaster/fl-2024-000045-ken Ministry of Health. Kenya Health Facility Census Report September 2023 [Internet]. 2024. Available from: https://www.health.go.ke/sites/default/files/2024-01/Kenya%20Health%20Facility%20Census%20Report%20September%202023.pdf KRB. KENYA ROADS BOARD (KRB) MAP PORTAL [Internet]. [cited 2024 Mar 1]. Available from: https://maps.krb.go.ke/kenya-roads-board12769/maps Karra K, Kontgis C, Statman-Weil Z, Mazzariello JC, Mathis M, Brumby SP. Global land use / land cover with Sentinel 2 and deep learning. In: 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS [Internet]. 2021 [cited 2025 Aug 23]. p. 4704–7. Available from: https://ieeexplore.ieee.org/document/9553499 RCMRD. Kenya SRTM DEM 30meters [Internet]. [cited 2024 Mar 1]. Available from: https://gmesgeoportal.rcmrd.org/datasets/rcmrd::kenya-srtm-dem-30meters/about UNEP-World Conservation Monitoring Centre (WCMC),International Union for Conservation of Nature (IUCN). The world database on protected areas. 2023 [cited 2023 Dec 9]. About. Available from: https://www.protectedplanet.net/en/about WorldPop. WorldPop, open data for spatial demography | Scientific Data [Internet]. [cited 2025 Aug 23]. Available from: https://www.nature.com/articles/sdata20174 Tatem AJ. WorldPop, open data for spatial demography. Sci Data. 2017 Jan 31;4:170004. International Centre for Humanitarian Affairs. Kenya Flood Extent Layer [Internet]. 2025. Available from: https://redcrosske.maps.arcgis.com/home/item.html?id=bce5234218c547439ce2a71f7cbeb4e2 United Nations Institute for Training and Research. UNITAR/UNOSAT [Internet]. 2024. Available from: https://unosat.org/products/ Ray N, Ebener S. AccessMod 3.0: computing geographic coverage and accessibility to health care services using anisotropic movement of patients. Int J Health Geogr. 2008 Dec 16;7(1):63. ON-GOING DOCTORS STRIKE [Internet]. [cited 2025 Aug 23]. Available from: https://www.knchr.org/Articles/ArtMID/2432/ArticleID/1192/ON-GOING-DOCTORS-STRIKE Nakweya G. Doctors in Kenya end strike after threat of court action. BMJ. 2024 May 14;385:q1088. Macharia PM, Mumo E, Okiro EA. Modelling geographical accessibility to urban centres in Kenya in 2019. PLoS One. 2021;16(5):e0251624. Nyandiko NO. Devolution and disaster risk reduction in Kenya: Progress, challenges and opportunities. Int J Disaster Risk Reduct. 2020 Dec 1;51:101832. Opere A. Chapter 21 - Floods in Kenya. In: Paron P, Olago DO, Omuto CT, editors. Developments in Earth Surface Processes [Internet]. Elsevier; 2013 [cited 2025 Aug 1]. p. 315–30. (Kenya: A Natural Outlook; vol. 16). Available from: https://www.sciencedirect.com/science/article/pii/B9780444595591000219 Vanderhoof MK, Alexander L, Christensen J, Solvik K, Nieuwlandt P, Sagehorn M. High-frequency time series comparison of Sentinel-1 and Sentinel-2 satellites for mapping open and vegetated water across the United States (2017–2021). Remote Sens Environ. 2023 Apr 1;288:113498. Nyamira: Locals warned of potential flooding amid heavy rains [Internet]. [cited 2025 Aug 2]. Available from: https://citizen.digital/wananchi-reporting/nyamira-locals-warned-of-potential-flooding-amid-heavy-rains-n362416 Storey Building Collapses in Gesima, Nyamira County as Floods Persist - Kenyans.co.ke [Internet]. 2024 [cited 2025 Jul 28]. Available from: https://www.kenyans.co.ke/news/100001-5-storey-building-collapses-gesima-nyamira-county-floods-persist Johary R, Révillion C, Catry T, Alexandre C, Mouquet P, Rakotoniaina S, et al. Detection of Large-Scale Floods Using Google Earth Engine and Google Colab. Remote Sens. 2023 Jan;15(22):5368. Juma B, Olang LO, Hassan MA, Chasia S, Mulligan J, Shiundu PM. Flooding in the urban fringes: Analysis of flood inundation and hazard levels within the informal settlement of Kibera in Nairobi, Kenya. Phys Chem Earth Parts ABC. 2023 Dec 1;132:103499. Water Resources Authority (WRA). Flood Advisory – Water Resources Authority [Internet]. [cited 2025 Aug 2]. Available from: https://wra.go.ke/flood-advisory-2/ Response-To-Floods-in-Kenya [Internet]. [cited 2025 Aug 2]. Available from: https://www.oagkenya.go.ke/wp-content/uploads/2023/08/Response-To-Floods-in-Kenya.pdf International Federation of Red Cross and Red Crescent Societies. Kenya Flood and Cholera Outbreak 2025 - DREF Operation (Appeal: MDRKE066) - Kenya | ReliefWeb [Internet]. 2025 May [cited 2025 Aug 25]. Available from: https://reliefweb.int/report/kenya/kenya-flood-and-cholera-outbreak-2025-dref-operation-appeal-mdrke066 ICHA and Kenya Red Cross. Leveraging Technology in the Face of Disaster - Kenya | ReliefWeb [Internet]. 2024 [cited 2025 Aug 23]. Available from: https://reliefweb.int/report/kenya/leveraging-technology-face-disaster Kenya Meteorological Department (KMD). THE CLIMATE OUTLOOK FOR APRIL 2024 AND REVIEW FOR MARCH 2024 [Internet]. [cited 2025 Aug 25]. Available from: https://meteo.go.ke/documents/17/April_2024_Forecast_Rev_D.pdf Makanga PT, Schuurman N, Sacoor C, Boene HE, Vilanculo F, Vidler M, et al. Seasonal variation in geographical access to maternal health services in regions of southern Mozambique. Int J Health Geogr. 2017 Jan 13;16(1):1. Espinet Alegre X, Stanton-Geddes Z, Aliyev S. Analyzing Flooding Impacts on Rural Access to Hospitals and Other Critical Services in Rural Cambodia Using Geo-Spatial Information and Network Analysis [Internet]. Rochester, NY: Social Science Research Network; 2020 [cited 2025 Aug 25]. Available from: https://papers.ssrn.com/abstract=3614143 Daily Nation. Daily Nation. 2024 [cited 2025 Aug 22]. Kura announces road closures in Nairobi due to flooding. Available from: https://nation.africa/kenya/counties/nairobi/kura-announces-road-closures-in-nairobi-due-to-flooding-4601706 Ministry of Health (MoH). Kenya Public Health Emergency Operations Center (KPHEOC) Handbook [Internet]. [cited 2025 Aug 2]. Available from: https://www.nphi.go.ke/sites/default/files/KPHEOC%20Handbook.pdf Additional Declarations No competing interests reported. Supplementary Files AccesstoFloodsSI260925.docx Cite Share Download PDF Status: Published Journal Publication published 18 Mar, 2026 Read the published version in International Journal of Health Geographics → Version 1 posted Editorial decision: Revision requested 23 Jan, 2026 Reviews received at journal 15 Jan, 2026 Reviewers agreed at journal 22 Dec, 2025 Reviewers agreed at journal 06 Nov, 2025 Reviewers agreed at journal 06 Nov, 2025 Reviews received at journal 15 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers invited by journal 30 Sep, 2025 Editor assigned by journal 29 Sep, 2025 Submission checks completed at journal 29 Sep, 2025 First submitted to journal 26 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7724672","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":527768715,"identity":"6e3f26f6-5039-48e1-beec-dd8c0f9b997e","order_by":0,"name":"Bibian N. Robert","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCElEQVRIiWNgGAWjYBACAwYGNhAtJwEiE4CYj504LQbGcC1szERqSZwBEyGoxVz68LEHH/f8SZ/Z3p324QHDYTk2ZuaHj25UMMiZ9y/AqsWyLy3dcMYzg9zZPGc3z0hgOGzMxsxmbJxzhsFY5sYD7A47w2MmzXPAIHeeRO5mhsR/txPbmHnYpHPbGBJnSBzAreXPAYN0Ofm3m4Hev10P0fKPgBaGAwYJ0hK8YC0JbGAtDUAt/A3Y/dLDlm7Yc8DYcGZPLkjLf8M2sF+OSRhLSOAIMR7mYw9+HJCTlzh+djPjD4Y0eX725oePc2ps5CT4sTsMJwBaIZFAmhYgINWWUTAKRsEoGK4AANSCUy6Wo/94AAAAAElFTkSuQmCC","orcid":"","institution":"KEMRI Wellcome Trust Research Programme","correspondingAuthor":true,"prefix":"","firstName":"Bibian","middleName":"N.","lastName":"Robert","suffix":""},{"id":527768716,"identity":"f775abeb-8da8-4b25-8058-6a2595c8f503","order_by":1,"name":"Samuel K. Muchiri","email":"","orcid":"","institution":"KEMRI Wellcome Trust Research Programme","correspondingAuthor":false,"prefix":"","firstName":"Samuel","middleName":"K.","lastName":"Muchiri","suffix":""},{"id":527768717,"identity":"dcffd434-b5b6-463e-9df2-6baa8ace9eff","order_by":2,"name":"Emma W. Kahoro","email":"","orcid":"","institution":"KEMRI Wellcome Trust Research Programme","correspondingAuthor":false,"prefix":"","firstName":"Emma","middleName":"W.","lastName":"Kahoro","suffix":""},{"id":527768718,"identity":"1bb9c44f-1693-47c7-9e10-7c770dbfcded","order_by":3,"name":"Boneya H. Hindada","email":"","orcid":"","institution":"Kenya Red Cross Society","correspondingAuthor":false,"prefix":"","firstName":"Boneya","middleName":"H.","lastName":"Hindada","suffix":""},{"id":527768719,"identity":"25cecf5d-4c6d-4295-9c09-6e9e6b02564e","order_by":4,"name":"Helen Kiarie","email":"","orcid":"","institution":"Ministry of Health","correspondingAuthor":false,"prefix":"","firstName":"Helen","middleName":"","lastName":"Kiarie","suffix":""},{"id":527768722,"identity":"19782c83-e5b2-4f2f-9690-ff9289a7a134","order_by":5,"name":"Emelda A. Okiro","email":"","orcid":"","institution":"KEMRI Wellcome Trust Research Programme","correspondingAuthor":false,"prefix":"","firstName":"Emelda","middleName":"A.","lastName":"Okiro","suffix":""},{"id":527768726,"identity":"1289993a-3f0e-4743-bf2a-1be3c3eab65c","order_by":6,"name":"Peter M. Macharia","email":"","orcid":"","institution":"KEMRI Wellcome Trust Research Programme","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"M.","lastName":"Macharia","suffix":""}],"badges":[],"createdAt":"2025-09-26 20:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7724672/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7724672/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12942-026-00461-x","type":"published","date":"2026-03-18T15:59:46+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93471019,"identity":"5e8f0d6d-0236-4e15-acde-7262b7a98cdd","added_by":"auto","created_at":"2025-10-14 08:19:19","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6101119,"visible":true,"origin":"","legend":"","description":"","filename":"AccesstoFloodsManuscript260925.docx","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/099be92151c62781dc9971de.docx"},{"id":93470752,"identity":"4a482abc-57a7-40f1-b417-9609f8bcccf6","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11008,"visible":true,"origin":"","legend":"","description":"","filename":"d140688205fa41d398adf7e7490836e8.json","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/86fec65924de00e599d3401b.json"},{"id":93470770,"identity":"a72ece2b-1e58-4f13-a4f5-3c15da63dec9","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20257064,"visible":true,"origin":"","legend":"","description":"","filename":"AccesstoFloodsSI260925.docx","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/0b59a4ca5bf415aff3c3451c.docx"},{"id":93470760,"identity":"5de723f3-48e6-43b7-942c-6f201f21c84b","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":137186,"visible":true,"origin":"","legend":"","description":"","filename":"d140688205fa41d398adf7e7490836e81enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/861fa81ca758f3895c49c8dd.xml"},{"id":93470762,"identity":"56526a2b-d920-46f4-b7e2-94197c03c61d","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3593079,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/291bdb6f36f6b9c1c56fd12c.jpeg"},{"id":93471016,"identity":"2fde59db-5a7f-40f6-bd4b-c84e1855cce0","added_by":"auto","created_at":"2025-10-14 08:19:19","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1081700,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/cc90440f9f65387373c9f339.jpeg"},{"id":93472486,"identity":"62a6522e-3cf5-480c-b2ca-6312a405a658","added_by":"auto","created_at":"2025-10-14 08:27:19","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":406976,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/bd362684cae512af4aebe6d9.jpeg"},{"id":93470764,"identity":"c8475f11-4957-46ce-abb0-1e89c15f1c45","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":838096,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/e26b0542fbcdafd705bcc805.jpeg"},{"id":93471015,"identity":"2757c338-8d49-48c8-bccf-b485623f159c","added_by":"auto","created_at":"2025-10-14 08:19:19","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1386062,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/d00950bfc5631c1158a8bde9.png"},{"id":93471020,"identity":"4cda6fd9-c2d1-4bea-bb31-3484ae8091e3","added_by":"auto","created_at":"2025-10-14 08:19:19","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":148427,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/82ae6fd598a5c37479fe6ecb.png"},{"id":93470758,"identity":"729977a2-584d-42c4-8392-2d57cd854879","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":210551,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/3187032cde99620b6129720d.png"},{"id":93471017,"identity":"b9b29891-0859-446c-a231-14e6e60b778e","added_by":"auto","created_at":"2025-10-14 08:19:19","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79473,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/c6f5b88e2a9c8081b51cd574.png"},{"id":93470767,"identity":"fb3f4971-e724-4814-ab1e-70eb190346ea","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":137888,"visible":true,"origin":"","legend":"","description":"","filename":"d140688205fa41d398adf7e7490836e81structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/c34e89795ec684bebce41c51.xml"},{"id":93470769,"identity":"1541a3fe-c508-44aa-9427-8ef674c4466c","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":151025,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/524497f692c6030959ffaf54.html"},{"id":93470754,"identity":"9fa1af84-4bc6-4b65-ac43-1fbf086e0b5a","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":546730,"visible":true,"origin":"","legend":"\u003cp\u003eThe spatial distribution of health facilities in Kenya by sector in 2023: Public (n = 5,586), Private not-for-profit (n=855), private for-profit (n = 4,554) and total (both public and private) health facilities (n = 10,995).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/7f2c5fa55db54bb272753538.png"},{"id":93470755,"identity":"05276161-aa30-4733-bab9-f0f4938ced8a","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":518972,"visible":true,"origin":"","legend":"\u003cp\u003e(A) The spatial extent of flooding by satellite sensors: NOAA-VIIRS and Sentinel 1 SAR; (B) the percentage of flooded regions that are covered by both Sentinel 1 SAR and NOAA-VIIRS that do not overlap and those that overlap across counties.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/daec3bec6c63235ffdcd967a.png"},{"id":93470756,"identity":"2b78a97a-0d48-45d0-8f37-e2ac46466e41","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":3189865,"visible":true,"origin":"","legend":"\u003cp\u003eAverage travel time to the nearest health facility at sub-county level by facility type (all, public, PfP and PNfP) for the BAU and worst-case scenarios (Sentinel 1 SAR, NOAA-VIIRS and the combined flooded extents).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/8c4f3a5f2ff0e43a585c603e.png"},{"id":93470753,"identity":"e07a11bd-c8b7-46df-b6c6-d2c4e9d3f444","added_by":"auto","created_at":"2025-10-14 08:11:19","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":194688,"visible":true,"origin":"","legend":"\u003cp\u003eProportion of population within 30-minute travel time bands for all facilities per scenario (BAU, Sentinel 1 SAR, NOAA-VIIRS and combined flooded extents).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/3599cb3045718fdc8a4f1d3b.png"},{"id":105223744,"identity":"be4f58ed-d3a9-43d3-a262-1cde218b8364","added_by":"auto","created_at":"2026-03-23 16:09:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5063945,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/a9fc2bcc-3e39-4ec4-b0a3-532119c7f3af.pdf"},{"id":93470771,"identity":"d73869e8-f2c0-4d5c-9d18-cc52c03fdafe","added_by":"auto","created_at":"2025-10-14 08:11:20","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":20257064,"visible":true,"origin":"","legend":"","description":"","filename":"AccesstoFloodsSI260925.docx","url":"https://assets-eu.researchsquare.com/files/rs-7724672/v1/af5201f3324321e191e812da.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Impact analysis of flood-induced changes in geographical accessibility and coverage to healthcare in both public and private sector, Kenya","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEnsuring equitable healthcare access is crucial for improving population health outcomes (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). Healthcare access, which involves obtaining appropriate services when needed, includes physical proximity to facilities, affecting service use (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). Physical accessibility is assessed by realised accessibility, considering actual travel time or distance, and potential accessibility, estimating travel time against barriers (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Providing efficient access to health services is challenging for policymakers, especially during crises like natural disasters, when demand increases and the environment becomes complex. In such times, assessments based on hazards, exposure, and vulnerabilities guide targeted measures to boost population resilience and protect critical services and infrastructure (\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe increasing frequency and severity of extreme weather events, a direct consequence of climate change, pose significant threats to the resilience of health systems worldwide (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). In Africa, where health infrastructure is often strained, climate change is exacerbating vulnerabilities by increasing the frequency of disasters such as floods, droughts, and heat waves (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e), and by diminishing the capacity to recover from these events (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Floods stand out as the most prevalent natural disasters in sub-Saharan Africa (SSA) accounting for 60% of the reported natural disaster between 1970\u0026ndash;2019 (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). By 2023, floods were responsible for nearly 75% of the 6.3\u0026nbsp;million disaster-induced displacements in the region (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). In 2024 alone, extreme rainfall affected 27 countries, resulting in widespread flooding that impacted over 11\u0026nbsp;million people, caused about 2,500 fatalities, and displaced 4\u0026nbsp;million individuals (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). These events have been reported to disrupt healthcare services and damage critical infrastructure (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e) hindering care delivery. This underscores the urgent need to bolster health system resilience (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) to anticipate, prepare for, and respond to climate-related disruptions.\u003c/p\u003e\u003cp\u003eIn April and May 2024, Kenya experienced unprecedented flooding across several counties, marking the most extensive flood event in recent history. The floods resulted in 267 to 315 reported fatalities and left approximately 400,000 individuals in urgent need of humanitarian aid (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). The risk of cholera outbreaks and an increase in malaria cases was heightened due to stagnant waters, limited access to safe water and sanitation, and the overcrowding of under-resourced, haphazardly delineated temporary shelters (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). The floods caused substantial damage to numerous health facilities, schools and critical infrastructure (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Various sources have also estimated considerable general losses to infrastructure and personal property (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e), but the exact quantitative impact of the flooding on geographical accessibility to healthcare remains unclear.\u003c/p\u003e\u003cp\u003eModelling geographic accessibility and population coverage based on travel time to health facilities offers insights for disaster preparedness and response. Recent maps of access in Kenya highlight marginalised groups (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e), and advances in remote sensing, geospatial tech, and post-disaster data collection allow accurate impact assessments, considering infrastructure damage, loss of road access, and movement barriers (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). However, disaster management frameworks often lack guidance on integrating spatial barriers into accessibility assessments. Despite guidelines for post-disaster needs (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e), some static measures don't address dynamic accessibility changes caused by barriers like floods. Geographical models can identify gaps, support decision-making, and improve disaster response efficiency (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eHere, we used geospatial modelling approaches (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) to assess post-disaster accessibility to health facilities and analyse accessibility coverage losses resulting from extreme flooding in Kenya. The approach enables the precise quantification of disaster impacts on geographical healthcare accessibility, thereby providing invaluable insights for post-disaster interventions and resilience-building efforts. Specifically, we model travel time before (business-as-usual) and after flooding and quantify the proportion of the population that lost access to healthcare. We define flooding zones based on two independent satellites, disaggregate facilities by four ownership levels and implement sensitivity analyses based on travel speeds.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eKenya-Country context\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eKenya\u0026rsquo;s population was 47.6 million in 2019 based on the most recent census (24) and was predicted to be 54 million in 2022 spread heterogeneously across the country (25). Population density varies considerably across its counties, ranging from under 20 people in arid regions to over 500 people per square kilometre in urban centres. This uneven distribution significantly impacts healthcare. Since 2013, Kenya has embraced devolution, a system where decision-making and resource allocation for sub-national healthcare services have been transferred to 47 county governments (Supplementary File, \u003cstrong\u003eFigure S1\u003c/strong\u003e). This decentralisation aims to address local needs better. However, the national government retains control over major hospitals, health regulations, and overall policy direction.\u003c/p\u003e\n\u003cp\u003eKenya\u0026rsquo;s healthcare system comprises six service delivery levels, ranging from community services (level 1) to national referral hospitals (level 6). It includes primary care at dispensaries and clinics (level 2), health centres (level 3), secondary and tertiary care at sub-county hospitals and medium-sized private hospitals (level 4), primary referral hospitals (level 5), and national referral hospitals (level 6). Facilities are operated by the government, NGOs, FBOs, and private sector. Service complexity varies from basic medication and rapid diagnostic tests (RDTs) with lower-skilled staff to advanced laboratory procedures conducted by higher-skilled workers, depending on facility ownership and level.\u003c/p\u003e\n\u003cp\u003eKenya has experienced significant disruptions in healthcare services due to health worker strikes. Notably, the prolonged strikes in 2017 markedly impacted service delivery. Another nationwide health workers\u0026rsquo; strike occurred from December 2020 to February 2021, driven by demands for better working conditions, including sufficient personal protective equipment, higher risk allowances, and comprehensive health insurance coverage (26,27). The duration of the strike varied across different health facilities and categories of workers. Although there is no comprehensive national database tracking these strikes, it is presumed that most health facilities were impacted throughout this period. The 2023/2024 period also witnessed strikes by healthcare workers, further straining the healthcare system (28) and significant delays in recruiting interns, a critical cadre in the provision of healthcare in Kenya (29).\u003c/p\u003e\n\u003cp\u003eKenya is highly vulnerable to flooding due to several factors. Situated in the drought-prone greater Horn of Africa region, the country is susceptible to extreme weather events (30). Heavy rainfall and El Ni\u0026ntilde;o cycles often lead to overflowing rivers and flash floods, exacerbated by widespread poverty and inadequate infrastructure, especially in urban cities and in arid and semi-arid areas (31). In April and May 2024, heavy rains that reached 111% to over 200% of\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003ethe long-term average across the country (32), caused devastating floods across Kenya, that displaced hundreds of thousands of people, damaged infrastructure, and caused significant loss of life (33). This recent event underscores the urgent need for improved flood preparedness and mitigation strategies in Kenya.\u003c/p\u003e\n\u003ch2\u003eMethods overview\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eWe applied a four-step approach. The first step entailed data assembly including health facilities (public, Private not-for-profit (PfNP), and private for-profit (PfP)), road networks, land use/cover, elevation population distribution and the flooding extents. \u0026nbsp;Second, we defined different travel scenarios to account for how flooding affected physical accessibility in Kenya. Third, we estimated travel time (TT) from every location to the nearest facility for i) all facilities, ii) public, iii) PfP and iv) PNfP using the least-cost path algorithm while also conducting a sensitivity analysis on the travel speeds and flood extents. Finally, we extracted the proportion of the population within 30 min, 30-60, 60-90, 90-120 and beyond 120 minutes of the nearest facility for each scenario. All estimates were aggregated at subnational units of decision-making (county and sub-county).\u003c/p\u003e\n\u003ch2\u003eData\u0026nbsp;\u003c/h2\u003e\n\u003ch3\u003ea. Healthcare facilities\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe Kenya health facility data used in the study were accessed from the Kenya Ministry of Health\u0026rsquo;s 2023 health facility census that was carried out between the 14\u003csup\u003eth\u003c/sup\u003e and 29\u003csup\u003eth\u003c/sup\u003e of August 2023 (34). The survey aimed to map all health facilities in Kenya including public, private and PNfP (FBOs/NGOs). Some of the information collected during the census and relevant to the current analysis included facility name and code, county, sub-county, ward, latitude, longitude, Kenya Essential Package for Health (KEPH) level and ownership.\u003c/p\u003e\n\u003cp\u003eWe focused on all facilities offering general medical services to the public. From the mapped facilities, we excluded facilities offering services to a subset of the population, i.e. company and secondary school clinics, military, police and prison facilities (188). In addition, we excluded specialised facilities including dental, eye, cancer, tuberculosis, HIV voluntary counselling and testing (VCT) centres, maternal and nursing homes, drop-in centres, mental health centres, hospices, funeral homes, rehabilitation, family planning clinics, radiology/X-ray, blood transfusion, gynaecology, nephrology, kidney dialysis, diagnostic and pharmacies (1,192). Our final list of facilities contained 10,995 facilities (\u003cstrong\u003e\u003cu\u003eFigure 1\u003c/u\u003e\u003c/strong\u003e); 8,112 dispensaries (level 2), 1,995 health centres (level 3) and 888 hospitals (level 4 \u0026ndash; 6).\u003c/p\u003e\n\u003ch3\u003eb. Factors affecting travel to healthcare\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eTravel time to healthcare facilities was modelled using multiple geospatial factors including road network, land use/land cover (LULC), digital elevation model, water bodies and national parks and population distribution. The road network was classified into primary, secondary, county, and other roads based on the Kenya\u0026rsquo;s proposed 2023 road register (Supplementary File, \u003cstrong\u003eFigure S2\u003c/strong\u003e) (35). In areas without road network, LULC data derived from 2023 Sentinel-2 imagery at 10m resolution was used to estimate travel conditions (Supplementary File, \u003cstrong\u003eFigure S3\u003c/strong\u003e) (36). The terrain slope, derived from the Shuttle Radar Topographic Mission (SRTM) at 30-meter resolution digital elevation model (DEM) (37) (Supplementary File, \u003cstrong\u003eFigure S4)\u003c/strong\u003e, was incorporated to adjust walking and cycling speeds. Additionally, water bodies and national parks (38) were treated as impassable barriers (Supplementary File, \u003cstrong\u003eFigure S1 \u0026amp; S3\u003c/strong\u003e). Fine spatial resolution (100 x 100 m) population distribution for the whole country was obtained from WorldPop open spatial demographic data portal (39,40). Here, we used the constrained (population modelled on in areas containing built settlements) gridded population raster for 2023 (Supplementary File \u003cstrong\u003eFigure S5\u003c/strong\u003e). Additional details of these datasets are provided in \u003cstrong\u003eSupplementary File.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ch3\u003ec. \u0026nbsp;Flooding extent\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe flood extents were assessed using high-resolution imagery from the National Oceanic and Atmospheric Administration \u0026ndash; Visible Infrared Imaging Radiometer Suite (NOAA-VIIRS) and Sentinel 1 Synthetic Aperture Radar (SAR) sensors, created specifically for flood response efforts in Kenya. The data was sourced from the United Nations Institute for Training and Research (UNITAR) and the International Center for Humanitarian Affairs (ICHA). The ICHA\u0026rsquo;s Sentinel 1 SAR flood extents data covered flood extents for the entire country from March 31 to May 9, 2024, at a spatial resolution of 30m, and was created using a change detection model to aggregate all predictions of flood extents since the start of flooding in the country (41). This was supplemented by flood extents for Tana-Galana River (April 29) based on 0.5m spatial resolution Pleidas satellite imagery and the Nyando River (May 5) based on Sentinel 1 SAR. Moreover, UNITAR provided two NOAA-VIIRS flood extents at the national level, covering April 24-28 and May 12-16, at a resolution of 375m (42), which were combined to represent the overall flood extent during that period.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn addition to considering the Sentinel 1 SAR and the NOAA-VIIRS flood extents independently, we integrated all datasets to produce a maximal flooding extent covering the entire country from 31\u003csup\u003est\u003c/sup\u003e March to 9\u003csup\u003eth\u003c/sup\u003e May 2024. Data that were in raster format were first converted to vector format, while those available as images were georeferenced and digitised to extract flood extents. The merged flood extents were then checked for topological errors and polygons that overlapped or shared common boundaries were dissolved into single polygons. All pre-processing steps described above were conducted in ArcGIS Pro version 3.0.3 (ESRI Inc., Redlands, CA, USA).\u003c/p\u003e\n\u003cp\u003eFinally, to capture the sections of the roads closed due to floods, various newspaper reports and articles were sought out within the flooding period (April-May 2024). Using the \u003cem\u003eEditor\u003c/em\u003e tool in ArcMap version 10.8.2 (ESRI Inc., Redlands, CA, USA), we split the sections of the flooded roads according to the reports and added an attribute of whether it was flooded or not in the attribute table.\u003c/p\u003e\n\u003ch2\u003eTravel scenarios\u003c/h2\u003e\n\u003cp\u003eWe identified two main travel scenarios: i) prior to the flooding (Business-as-usual-BAU) and ii) during the flooding, applicable across the entire country. For each scenario, we outlined the various modes of transport and their corresponding speeds across different road networks and land covers. In the pre-flooding scenario, we utilised a travel framework developed from our earlier national spatial accessibility modelling in Kenya (5; \u003cstrong\u003e\u003cu\u003eTable 1\u003c/u\u003e\u003c/strong\u003e). Briefly, this scenario encompasses a hybrid mode of transport involving either walking (areas with no roads), bicycling (lower class roads), motorized (motorbike and vehicles, both public and private in motorable roads) or combined forms of transport based on availability of motorable roads and class of roads.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOn the other hand, during flooding, all the roads within the flooding zone were considered impassable (barriers) while the facilities in the flooding zone were deemed non-functional. In addition, parts of the road network outside the flooding zone that the Kenyan government had closed owing to the floods were deemed inaccessible. Outside the flooding zones, we considered reduced speeds (by 50%) in areas across all roads (7; \u003cstrong\u003e\u003cu\u003eTable 1\u003c/u\u003e\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1:\u0026nbsp;\u003c/strong\u003eTravel modes and that were used to compute travel time to the nearest health facility for each road type and land cover category for two scenarios (business as usual and during flooding)\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"595\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMode of transport \u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 264px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTravel Scenarios (km/h)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eID\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCategory\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBusiness-as-usual\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDuring flooding\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eFlooding extents\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 281px;\"\u003e\n \u003cp\u003eVarious modes and corresponding speeds depending on road type and land cover\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 131px;\"\u003e\n \u003cp\u003eTravel barrier\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMode of transport \u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 264px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTravel Scenarios (km/h)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eID\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCategory\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBusiness-as-usual\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDuring flooding\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eTrees\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eWalking\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e2.5\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e1.25\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eRange land\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eWalking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e4.5\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e2.25\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eCrops\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eWalking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e4\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eBare or built-up\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eWalking\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e5\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e2.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMode of transport \u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 264px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTravel Scenarios (km/h)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eID\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCategory\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBusiness-as-usual\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDuring flooding\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eFlooded vegetation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eBarrier \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eWater\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eBarrier \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003ePrimary road\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eMotorized \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e50\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eSecondary road\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eMotorized \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e30\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eCounty roads\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eCycling\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e10\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e5 (walk)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 36px;\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 147px;\"\u003e\n \u003cp\u003eOther minor roads\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 148px;\"\u003e\n \u003cp\u003eWalking\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 133px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 131px;\"\u003e\n \u003cp\u003e2.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch2\u003eEstimating travel time\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eTo estimate travel time for the different scenarios, we used AccessMod version 5.7.17, an open-source tool supported by the World Health Organization (WHO) to analyse geographic accessibility via a least-cost path algorithm (43). \u0026nbsp;For the \u003cem\u003eBAU\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/em\u003escenario, we first merged the land cover, road network, water bodies and protected areas (usual barriers) via the \u0026ldquo;merge land cover\u0026rdquo; module in AccessMod Toolbox to create a merged gridded surface. Based on the resultant single raster, speeds from the BAU model (\u003cstrong\u003eTable 1\u003c/strong\u003e) were applied to compute travel time to the nearest facility for i) all facilities, ii) public facilities, iii) PfP facilities and iv) PNfP facilities (FBOs/NGOs). This was necessary because, during the flooding period, doctors from the public health facilities were on industrial strike (44,45). Therefore, clients could only attend either the PfP or PNfP facilities.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConversely, \u003cem\u003eduring flooding\u003c/em\u003e, we created a unified gridded surface by utilising land cover, the road network outside the flooding area, and additional barriers (i.e., standard barriers), the maximum flooding extent, and the closed roads beyond the flooding zone. We then assigned speeds to this merged surface to calculate travel time to facilities, categorised into four groups (as previously done for BAU). Further, the travel speeds outside the flooding zones were reduced speeds by 50% (\u003cstrong\u003eTable 1;\u003c/strong\u003e 7).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor each scenario, the cumulative travel time from every populated location in Kenya based on WorldPop\u0026rsquo;s population distribution maps was computed towards (anisotropic) the closest facility via the least cost path (cost measured as time) at 100 x 100m spatial resolution.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eSensitivity analyses\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eWe conducted two types of sensitivity analyses. First, in addition to the combined flooding extents, we computed spatial accessibility metrics after flooding based on the individual flooding extents from two sensors (Sentinel 1 SAR and NOAA-VIIRS). Second, we varied the travel speeds in Table 1 by \u0026plusmn;20% (6,46) and repeated the least-cost path model to define an upper and lower bound of travel times for all travel scenarios (BAU and during flooding), facility ownership (all facilities, public, PfP and PNfP and sensor type (Sentinel 1 SAR, NOAA-VIIRS and combined extents).\u003c/p\u003e\n\u003ch2\u003ePopulation affected\u003c/h2\u003e\n\u003cp\u003eThe geographic coverage estimates (the proportion of the total population within 30 min, between 30-60 min, 60-90 min, 90-120 min and beyond 120 min) of the nearest health facility disaggregated by travel scenario, sensor type and level of speed was extracted at national, county and sub-county levels. This was achieved by using Zonal statistics ArcGIS Pro Version 3.3.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eHealth facilities assembled\u003c/h2\u003e\n\u003cp\u003eOf the 10,995 facilities assembled for this study over 97% were outside all three flooding zones (99.2%, 97.9% and 97.4% for the NOAA-VIIRS, Sentinel 1 SAR and combined flooding zones respectively). A total of 86, 232 and 289 facilities were identified to be within the NOAA-VIIRS, Sentinel 1 SAR and the combined flooding zones respectively (\u003cstrong\u003e\u003cu\u003eTable\u0026nbsp;2)\u003c/u\u003e\u003c/strong\u003e. Majority (68.9%) of the facilities within the three flooding zones were public facilities (45, 172 and 199 facilities for NOAA-VIIRS, Sentinel 2 and combined zones respectively) with the least (4.5%) being PNfP facilities (one, five, and 13 facilities for NOAA-VIIRS, Sentinel 1 SAR and combined zones respectively) \u003cstrong\u003e\u003cu\u003eTable 2\u003c/u\u003e\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2:\u0026nbsp;\u003c/strong\u003eSummary of health facilities by type within each travel time scenario\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"619\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003eType\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 34px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAll facilities\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 205px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWithin flooding zones\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 237px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOutside the flood zone\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003eNOAA-VIIRS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003eSentinel 1 SAR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eBoth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 87px;\"\u003e\n \u003cp\u003eNOAA-VIIRS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003eSentinel 1 SAR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003eBoth\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003ePublic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 34px;\"\u003e\n \u003cp\u003e5,586\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e172\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e199\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 87px;\"\u003e\n \u003cp\u003e5,541\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e5,414\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003e5,387\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003ePrivate not-for-profit\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 34px;\"\u003e\n \u003cp\u003e855\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 87px;\"\u003e\n \u003cp\u003e854\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e850\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003e842\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003ePrivate for-profit\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 34px;\"\u003e\n \u003cp\u003e4,554\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 87px;\"\u003e\n \u003cp\u003e4,518\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e4,503\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003e4,477\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003eTotal\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 34px;\"\u003e\n \u003cp\u003e10,995\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 70px;\"\u003e\n \u003cp\u003e86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 79px;\"\u003e\n \u003cp\u003e232\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e289\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 87px;\"\u003e\n \u003cp\u003e10,909\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e10,763\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 69px;\"\u003e\n \u003cp\u003e10,706\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003ch2\u003eExtent of flooding\u003c/h2\u003e\n\u003cp\u003eThe total flood-affected area was 43,422.9 km\u0026sup2;, representing 7.3% of Kenya\u0026rsquo;s land area. This was mainly due to Sentinel 1 SAR, which covered 39,115.3 km\u0026sup2; (6.6%), compared to NOAA-VIIRS, which covered 7,345.2 km\u0026sup2; (1.2%). The overlap in sensor detection was 3,037.6 km\u0026sup2;, or 0.5% of the total area, and 7.0% of the total flood extent area (\u003cstrong\u003e\u003cu\u003eFigure 2A\u003c/u\u003e\u003c/strong\u003e). Nyamira was the only county with no flood extents recorded from either sensor. Nine counties\u0026mdash;Trans Nzoia, Nyeri, Bungoma, Kakamega, Nandi, Vihiga, Kericho, Bomet, and Kisii\u0026mdash;were not included in NOAA-VIIRS flood extents but appeared in Sentinel 1 SAR data. Isiolo showed the highest proportion of flooded area at 17.0% based on maximum extents and also had the highest non-overlapping flood extents at 15.3%.\u003c/p\u003e\n\u003ch2\u003e\u003cem\u003eTravel time to health facilities pre- and post-flooding\u0026nbsp;\u003c/em\u003e\u003c/h2\u003e\n\u003ch3\u003ePre flooding\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 3\u003c/strong\u003e illustrates the average travel times to the nearest facility across per subcounty, categorised into 30-minute intervals and disaggregated by facility type and scenario. Under the BAU scenario, the national average travel time to the closest facility was 19.6 minutes (range: 16.4 - 24.4). When disaggregated by facility type, the average travel times were 20.7 minutes (17.3\u0026ndash;25.7) for public facilities, 37.8 minutes (31.6\u0026ndash;47.1) for PfP facilities, and 49.2 minutes (41.1\u0026ndash;61.4) for PNfP facilities. Nairobi County had the shortest mean travel times: 2.5 minutes (2.2 - 3.1) overall, 3.8 minutes (3.3 - 4.7) for public facilities, 2.9 minutes (2.5 - 3.5) for private facilities, and 4.1 minutes (3.5 - 4.9) for PNfP facilities (\u003cstrong\u003eFigure 3\u003c/strong\u003e). In contrast, Marsabit County recorded the longest average travel times across all types of health facilities: 76.3 minutes (63.7 - 95.3) for all facilities, 77.1 minutes (64.3 - 96.2) for public facilities, and 185.4 minutes (154.6 - 231.6) for private facilities (Supplementary File, \u003cstrong\u003eFigures S6-S8\u003c/strong\u003e). Mandera county, however, showed the highest average travel time for PNfP facilities at 363.0 minutes (302.6 - 453.7) (Supplementary File, \u003cstrong\u003eFigure S9\u003c/strong\u003e). Thirteen counties (Garissa, Isiolo, Kajiado, Kitui, Lamu, Mandera, Marsabit, Narok, Samburu, Tana River, Turkana, Wajir and West Pokot) had travel times longer than the national average travel times for all, public, and private facilities (Supplementary File, \u003cstrong\u003eFigures S6-S8\u003c/strong\u003e), while twelve counties surpassed the national average travel time for PNfP facilities (Supplementary File, \u003cstrong\u003eFigure S9\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003ePost flooding\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eIn all worst-case scenarios (NOAA-VIIRS, Sentinel 1 SAR, and the combined data), the national average travel time to health facilities exceeded that of the BAU scenario (\u003cstrong\u003eFigure 3\u003c/strong\u003e). Specifically, the average travel time to the nearest facility (covering all facilities) was 55.6 minutes (range: 46.5 - 69.6) for Sentinel 1 SAR-derived extents, 45.8 minutes (38.3 - 57.1) for NOAA-VIIRS, and 56.3 minutes (47.0 \u0026ndash; 70.3) for the combined flooded extents. Similar patterns were observed across the different facility ownership types and flooding scenarios, with the longest travel times for PNfPs facilities was 113.5 minutes (range: 94.6 - 191.5) and the shortest for public facilities was 48.5 minutes (range: 40.5 - 74.5), regardless of flooding extent. The combined flooding scenario resulted in the greatest reduction in travel times (\u003cstrong\u003eFigure 3\u003c/strong\u003e; Supplementary \u003cstrong\u003eFigures S6-S9\u003c/strong\u003e).\u003c/p\u003e\n\u003ch2\u003eGeographic coverage pre- and post-floods\u003c/h2\u003e\n\u003ch3\u003ePre flooding\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 4\u003c/strong\u003e shows the proportion of Kenya\u0026rsquo;s population within 30-minute travel time bands to all facilities across different flooding scenarios. Before the floods (BAU), 94.0% of the population had access to health facilities within 30 minutes (\u003cstrong\u003eFigure 4, panel 1\u003c/strong\u003e) and 0.7% lived outside 2 hrs travel time (\u003cstrong\u003eFigure 4, panels 2-5\u003c/strong\u003e). County level population access within 30-min access varied between 66.4% (Garissa) to 100% (Kisii). Overall, 30 of 47 counties had over 94% of their population within the 30-minute access range, matching the national average. Only eight counties had their entire population within 60 minutes, 16 counties within 90 minutes and 20 counties within 2 hrs travel time.\u003c/p\u003e\n\u003ch3\u003ePost flooding\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eDuring flooding events (under combined scenarios), the proportion of the population with timely access within 30 minutes decreased from 94.0% to 73.3%. Specifically, between sensors, only 19 counties (Sentinel 1 SAR scenario), 20 counties (NOAA-VIIRS), and 17 counties (combined) maintained at least 75% of their population coverage within 30 minutes, as shown in \u003cstrong\u003eFigure 4, panel 1\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe reduction in population coverage within this 30-minute threshold at the county level ranged from as low as 1.0% in Nairobi to as high as 51.0% in Narok. The most significant declines (41.0-51.0%) were seen in Narok, Turkana, Makueni, Tana River, Kitui, Laikipia, Lamu, Wajir, Nyandarua, and West Pokot. Conversely, counties such as Nairobi, Vihiga, Kiambu, Kisii, Mombasa, and Nyamira experienced relatively small coverage losses (1-10%\u003cstrong\u003e; Figure 4\u003c/strong\u003e). Only five counties- Kakamega, Kisii, Nairobi, Nyamira, and Vihiga- retained full population access within a 2-hour travel time. Counties including Wajir, Garissa, Turkana, Marsabit, Samburu, Isiolo, and Tana River saw 15-31% of their populations falling beyond 2 hours of access after the floods, an increase from 4-12% before the floods (\u003cstrong\u003eFigure 4, panel 5\u003c/strong\u003e). Similar reductions in access to health services across public health facilities were observed, with the most severe impacts in counties in northern and northeastern Kenya, which have fewer private and PNfP facilities (Supplementary Files, \u003cstrong\u003eFigures S10-S12\u003c/strong\u003e).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe 2024 Kenya floods recovery needs assessment and other humanitarian impact reports highlighted extensive disruptions to health services following the March\u0026ndash;May 2024 floods (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). While these reports documented cross-sectoral impacts and proposed broad mitigation measures, they did not quantify health service coverage losses. Our study addresses this critical gap by assessing the impact of floods on geographical access to health care facilities by quantifying travel time during pre-floods (BAU) and post-flood. The post-floods scenarios were analysed using Sentinel 1 SAR and NOAA-VIIRS satellite data and a combined scenario integrating the two datasets. Our results reveal a substantial reduction in timely (within 30-min) access to health care facilities, with population coverage dropping from 94% (52,488,181\u0026nbsp;million) pre-flood to 73% (40,916,717\u0026nbsp;million) post-floods. The increased travel time to health facilities was also observed across all health facility types. These findings underscore the sensitivity of Kenya\u0026rsquo;s health system to extreme weather events and highlight the urgent need for frameworks that aid in disaster preparedness to mitigate the risk of loss of access to healthcare during crises (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFlooded extents varied substantially across counties (Supplementary \u003cb\u003eFigure S13)\u003c/b\u003e reflecting a complex interplay of topographical and socio-environmental factors beyond rainfall intensity (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e). For instance, the largest flood extents were observed in diverse ecological and geographic settings: semi-arid counties such as Isiolo, Laikipia, and Turkana, which typically experience low rainfall and limited drainage infrastructure; Mombasa, a coastal urban county prone to tidal influences and poor urban drainage; and Nakuru, a highland region with urban zones where topography and land use may contribute to localized flooding. These spatial differences were further influenced by the characteristics of specific satellite sensors. For example, Sentinel-1 SAR, with its finer spatial resolution (5 \u0026times; 20 m), was able to detect small-scale flooded areas across all counties that may have been missed by NOAA-VIIRS, likely due to its coarser resolution (375 m). Notably, in Mombasa County NOAA-VIIRS detected wider flooding extents than Sentinel 1 SAR, while in counties like Narok, Nakuru and Laikipia, the opposite was true. This sensitivity of results and flood extent estimates to the satellite sensor selected highlights the crucial role of high-resolution, real-time geospatial data in disaster response efforts (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e). In Nyamira County, neither sensor detected flood extents despite reports of infrastructural damages caused by the heavy rainfall (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e). Such limitations in satellite-based flood monitoring emphasise the need for standardized, multi-source flood assessment methods to enhance detection accuracy, especially in low-resource settings (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eUnder the BAU scenario, over 90% of the population resided within half an hour of a health facility. However, this national coverage masks out some of the huge inequalities particularly in arid and semi-arid counties such as Garissa, Isiolo, Marsabit and Wajir where travel times exceeded 1 hr, reflecting persistent access challenges in marginalised regions (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). The floods further exacerbated these disparities reducing the coverage of people within 30 minutes nationally to 73% (combined scenario), with northeastern counties experiencing longer travel times above national average. These counties are characterised by limited road infrastructure, and they heavily rely on public facilities. For instance, Mandera had an average travel time of 37 minutes to a public facility under the BAU scenario as compared to over 3.5 hrs average travel time to PfP and PNfP facilities. After floods travel time to PNfP facilities alone in this county doubled (6hrs). In addition, PfP and PNfP facilities that serve as critical alternatives during public sector strikes (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e), were concentrated in urban centres, leaving rural populations with even fewer options post-disaster.\u003c/p\u003e\u003cp\u003eFloods have disproportionately disrupted timely access to health facilities, exposing critical geographic, hydrological, and infrastructural vulnerabilities across counties (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e, \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). Counties with the greatest loss in 30-minute coverage\u0026mdash;such as Narok, Tana River, Lamu, Kitui, and Makueni\u0026mdash;are located in major flood-prone areas identified in flood advisories issued months before the floods (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). Notably, Kitui experienced the highest coverage loss at 51% and also sustained the most extensive damage to health facilities at 47%, including a referral hospital (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Turkana, Tana River, Wajir, and Lamu had the highest proportion of facilities within flooded areas, with at least nine facilities in Tana River reported damaged (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Laikipia and Nyandarua were among those with the largest flooded extents. The correlation between health facility exposure and coverage decline was also evident in longer travel times (60 to 120 minutes) across counties like Marsabit, Turkana, Tana River, Wajir, Isiolo, and Garissa, where limited road infrastructure further hindered access.\u003c/p\u003e\u003cp\u003eOur findings reveal that even minor disruptions in coverage (1\u0026ndash;10%) in densely populated counties such as Vihiga, Kiambu, Kisii, Mombasa, and Nyamira can lead to significant increases in the number of people facing longer travel times, despite generally good infrastructure. These delays have varied impacts on health service utilization. Notably, Kiambu experienced some of the steepest declines in hospital admissions, outpatient visits, and immunization services (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). This underscores the need to interpret coverage losses not just in percentage terms but through the lens of absolute population affected and spatial context. The fact that only five counties had the entire population within 2 hours travel time post-floods down from 20 pre-floods, further underscores the urgency of integrating disaster preparedness policies to ensure responsive, equitable healthcare access during extreme weather events (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003ePolicy implications\u003c/h2\u003e\u003cp\u003eKenya continues to experience recurrent and severe flood events resulting in fatalities, destruction and disruption of essential infrastructural services, including health care (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e). Our study highlights the sensitivity of Kenya\u0026rsquo;s health system to flood-related shocks, offering insights that complement existing recovery strategies and help mitigate long-term impacts on the population. Policymakers should prioritise resource allocation to high-risk counties, not only those vulnerable to flooding but also those susceptible to post-disaster health risks. For instance, Tana River, Lamu and Siaya counties experienced cholera outbreaks following the March-May floods. Mapping access to care losses including at the sub-county level (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) offers actionable insights for targeted emergency response. Further, leveraging multi-source flood assessment approaches, including geospatial and drone technologies (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e) and including participatory approaches to gather historical flood impacts from communities, can enhance informed decision-making and develop contingency strategies, such as mobile clinics and emergency transport corridors, to buffer access disruptions during disasters. Enhancing health system resilience amid escalating climate risks will require stronger intersectoral coordination among health, infrastructure, humanitarian, and meteorological agencies. It is also critical that both national and county governments treat flood advisories and weather forecasts (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e) as actionable early warnings, prompting proactive disaster prevention and preparedness rather than reactive responses.\u003c/p\u003e\u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\u003ch2\u003eStrengths\u003c/h2\u003e\u003cp\u003eThis study offers several strengths. Unlike previous similar studies (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e) that rely solely on assumptions such as reduced travel speeds during flood events, we incorporated actual flooded extents to comprehensively capture access disruptions caused by flooding. To overcome the limitations of any single sensor, we employed multi-sensor data. Further, the Sentinel 1-derived flood extents was an aggregation of satellite data since the flooding started and hence not from a single point in time. In addition, Sentinel 1 SAR sensor is not affected by cloud cover and by lack of daylight which ensured uninterrupted monitoring of flooding events. We also integrated official records on road closures provided by Kenyan road authorities (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e). Additionally, since travel speeds used in the analysis are estimates rather than directly observed, we conducted a sensitivity analysis increasing/decreasing speed by 20% to adjust for uncertainties. We also used the most recent and comprehensive database of health facilities, compiled through a nationwide survey conducted by the Ministry of Health. Finally, we disaggregated our analysis by health sector owing to doctor strike and household wealth differentials in terms of money that influence the ability to access care.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\u003ch2\u003eLimitations\u003c/h2\u003e\u003cp\u003eNotwithstanding the valuable insights on understanding the impacts of flooding on access to health care in Kenya, there are several limitations. The change detection model that produced the Sentinel 1 SAR-defined flood extents may have resulted to some false positives within the semi-arid areas for instance in Wajir. NOAA-VIIRS is an optical sensor, and its flood detection capabilities is constrained by cloud cover and lack of daylight, further contributing to potential underestimation of inundation. This study incorporated flooded road segments in addition to the flood extents, but only segments reported to have flooded were accounted for possibly missing other affected roads not documented. Furthermore, we did not know the extent of damaged roads, limiting our mapping of infrastructure disruption. We assumed that population within flood zones had no alternative means of accessing health facilities. However, in some affected areas boats were reportedly used. We also lacked data on conditions outside the mapped flood zones, so we couldn\u0026rsquo;t account for disruptions caused by unusually heavy rainfall in those areas. We assumed individuals travel to the nearest facility, which may not reflect actual healthcare-seeking behaviour or preferences. All flooded roads were assumed impassable although some vehicles (e.g. trucks) may have been able to navigate certain segments. Lastly, this study only focused on quantifying the impact of floods on geographical access to health facilities, highlighting spatial disparities in health care access during floods. However, further research is needed to assess county-level adaptive capacity which is essential for understanding overall health system vulnerabilities to extreme weather events.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eOne of the core responsibilities of the health system is to prevent, prepare for, detect and respond to public health threats (\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e) including floods, the second most prevalent natural disaster in Kenya. Our findings show that flood events significantly reduce geographical access to healthcare services, with pronounced subnational disparities across flood-prone counties, semi-arid regions with persistent infrastructural deficits, and densely populated regions. The loss of access to health care is largely driven by impacts on flood-exposed health facilities, spatial extent of flooding, and damaged transport system inhibiting access. However, the impact on health services utilisation varies sub-nationally. Considering compounding health risks associated with floods such physical injuries, the spread of communicable diseases (e.g., measles) due to overcrowding in displacement settlements, disease outbreaks (malaria, cholera, etc), it is critical to ensure that health care system can sustain service delivery amid recurring climate shocks.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"669\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBAU\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eBusiness-as-usual\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDEM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eDigital elevation model\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eESRI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eEnvironmental Systems Research Institute\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFBOs\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eFaith-based Organisations\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eICHA\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eInternational Center for Humanitarian Affairs\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eKEPH\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eKenya Essential Package for Health\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLULC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eLand use/cover\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMoH\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eMinistry of Health\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNGOs\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eNon-government organisations\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNOAA-VIIRS\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eNational Oceanic and Atmospheric Administration \u0026ndash; Visible Infrared Imaging Radiometer Suite\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePfP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003ePrivate-for-profit\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePNfP\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003ePrivate-not-for-profit\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRDTs\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eRapid diagnostic tests\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSRTM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eShuttle Radar Topographic Mission\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSSA\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003esub-Saharan Africa\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTT\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eTravel time\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUNITAR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eUnited Nations Institute for Training and Research\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVCT\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eVoluntary counselling and testing\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 111px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWHO\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 558px;\"\u003e\n \u003cp\u003eWorld Health Organisation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAvailability of data and material:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe health facility data is available from the Ministry of Health (https://www.health.go.ke/contact-us).Sentinel-2 flooding extent from the International Center for Humanitarian Affairs (ICHA; https://redcrosske.maps.arcgis.com/home/item.html?id=bce5234218c547439ce2a71f7cbeb4e2) and NOAA-VIIRS flooding extent from the United Nations Institute for Training and Research (https://unosat.org/products/). The roads data from the Kenya Roads Board portal (https://maps.krb.go.ke/kenya-roads-board12769/maps). The Digital elevation model can be downloaded from the Regional Centre for Mapping of Resources for Development (RCMRD) geoportal\u003c/p\u003e\n\u003cp\u003e(https://gmesgeoportal.rcmrd.org/datasets/rcmrd::kenya-srtm-dem-30meters/about).The land use land cover from the Environmental Systems Research Institute (ESRI\u0026rsquo;s) ArcGIS platform (https://www.arcgis.com/home/item.html?id=6df9ed7a1eda4ed58023456b7c5484fd), protected areas for the Global database of protected areas (https://www.protectedplanet.net/en/about) and population data from the WorldPop open spatial demographic data portal (https://www.worldpop.org/)\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to the International Centre for Humanitarian Affairs (Kenya Red Cross Society) for providing Sentinel 1 SAR flood extent data and the Ministry of Health (MoH) for providing health facilities data\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot Applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEAO is supported by the Wellcome Trust Senior Fellowship (#224272) which supported BNR. SKM is supported by the Wellcome Trust Principal Fellowship (#212176). PMM is supported by the Fonds voor Wetenschappelijk Onderzoek \u0026ndash; Research Foundation Flanders Senior Postdoctoral Fellowship (#1201925N). BNR, SKM, PMM and EAO are grateful for the support of the Wellcome Trust to the Kenya Major Overseas Programme (#203077). The views expressed in this publication are those of the authors and not necessarily those of Wellcome Trust. The funders had no role in study design, data collection, data analysis, data interpretation, or writing of the report.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo individual patient-level data was used in this publication.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable - The manuscript does not contain data from any individual person.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBNR:\u003c/strong\u003e Data curation, Formal Analysis, Conceptualisation, Investigation, Methodology, Software, Validation, Visualisation, Writing \u0026ndash; original draft, Writing \u0026ndash; review \u0026amp; editing; \u003cstrong\u003eSKM:\u003c/strong\u003e Data curation, Formal Analysis, Conceptualisation, Investigation, Methodology, Software, Validation, Visualisation, Writing \u0026ndash; original draft, Writing \u0026ndash; review \u0026amp; editing; \u003cstrong\u003eEWK\u003c/strong\u003e: Data curation , Writing \u0026ndash; review final draft; \u003cstrong\u003eBHH:\u003c/strong\u003e Data curation , Writing \u0026ndash; review final draft; \u003cstrong\u003eHK\u003c/strong\u003e: Data curation, Writing \u0026ndash; review final draft; \u003cstrong\u003eEAO:\u003c/strong\u003e Conceptualisation, Investigation, Methodology, Funding acquisition, Resources, Supervision, Validation, Writing \u0026ndash; original draft, Writing \u0026ndash; review \u0026amp; editing; \u003cstrong\u003ePMM:\u003c/strong\u003e Data curation, Formal Analysis, Conceptualisation, Investigation, Methodology, Validation, Visualisation, Supervision, Writing \u0026ndash; original draft , Writing \u0026ndash; review \u0026amp; editing; All authors contributed to the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003ePenchansky R, Thomas JW. The concept of access: definition and relationship to consumer satisfaction. Med Care. 1981 Feb;19(2):127\u0026ndash;40. \u003c/li\u003e\n\u003cli\u003eLevesque JF, Harris MF, Russell G. Patient-centred access to health care: conceptualising access at the interface of health systems and populations. Int J Equity Health. 2013 Mar 11;12:18. \u003c/li\u003e\n\u003cli\u003eMoturi AK, Suiyanka L, Mumo E, Snow RW, Okiro EA, Macharia PM. Geographic accessibility to public and private health facilities in Kenya in 2021: An updated geocoded inventory and spatial analysis. Front Public Health. 2022;10:1002975. \u003c/li\u003e\n\u003cli\u003eOuma PO, Maina J, Thuranira PN, Macharia PM, Alegana VA, English M, et al. Access to emergency hospital care provided by the public sector in sub-Saharan Africa in 2015: a geocoded inventory and spatial analysis. Lancet Glob Health. 2018 Mar;6(3):e342\u0026ndash;50. \u003c/li\u003e\n\u003cli\u003eHierink F, Rodrigues N, Mu\u0026ntilde;iz M, Panciera R, Ray N. Modelling geographical accessibility to support disaster response and rehabilitation of a healthcare system: an impact analysis of Cyclones Idai and Kenneth in Mozambique. BMJ Open. 2020 Nov 3;10(11):e039138. \u003c/li\u003e\n\u003cli\u003eTariverdi M, Nunez-Del-Prado M, Leonova N, Rentschler J. Measuring accessibility to public services and infrastructure criticality for disasters risk management. Sci Rep. 2023 Jan 28;13(1):1569. \u003c/li\u003e\n\u003cli\u003ePatrascu FI, Mostafavi A, Vedlitz A. Relationship between Access to Critical Facilities During Normal Times and Disrupted Access During Extreme Weather Events, and Underlying Disparities [Internet]. Rochester, NY: Social Science Research Network; 2022 [cited 2025 Aug 23]. Available from: https://papers.ssrn.com/abstract=4108983\u003c/li\u003e\n\u003cli\u003eWade R. Climate Change and Healthcare: Creating a Sustainable and Climate-Resilient Health Delivery System. J Healthc Manag Am Coll Healthc Exec. 2023 Aug 1;68(4):227\u0026ndash;38. \u003c/li\u003e\n\u003cli\u003eWright CY, Kapwata T, du Preez DJ, Wernecke B, Garland RM, Nkosi V, et al. Major climate change-induced risks to human health in South Africa. Environ Res. 2021 May;196:110973. \u003c/li\u003e\n\u003cli\u003eAfrica Risk Capacity. The State of Natural Disasters in Africa [Internet]. 2024 [cited 2025 Aug 23]. Available from: https://www.arc.int/sites/default/files/2024-07/ARC_white_paper_2024.pdf\u003c/li\u003e\n\u003cli\u003eWorld Meteorological Organisation (WMO). State of the Climate in Africa 2024 [Internet]. [cited 2025 Jun 9]. Available from: https://library.wmo.int\u003c/li\u003e\n\u003cli\u003eInternal Displacement Monitoring Center (IDMC). Internal Displacement in Africa (2024) [Internet]. Internal Displacement Monitoring Center (IDMC); 2024 Nov [cited 2025 Jun 9]. Available from: https://www.internal-displacement.org/publications/internal-displacement-in-africa/\u003c/li\u003e\n\u003cli\u003eStudies the AC for S. Record Levels of Flooding in Africa [Internet]. Africa Center. [cited 2025 Aug 23]. Available from: https://africacenter.org/spotlight/record-levels-of-flooding-in-africa-compounds-stress-on-fragile-countries/\u003c/li\u003e\n\u003cli\u003ePetricola S, Reinmuth M, Lautenbach S, Hatfield C, Zipf A. Assessing road criticality and loss of healthcare accessibility during floods: the case of Cyclone Idai, Mozambique 2019. Int J Health Geogr. 2022 Oct 12;21(1):14. \u003c/li\u003e\n\u003cli\u003eHealth systems resilience toolkit: a WHO global public health good to support building and strengthening of sustainable health systems resilience in countries with various contexts [Internet]. [cited 2025 Aug 23]. Available from: https://www.who.int/publications/i/item/9789240048751\u003c/li\u003e\n\u003cli\u003eOCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #5 (10 May 2024) | OCHA [Internet]. 2024 [cited 2025 Aug 2]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-5-10-may-2024\u003c/li\u003e\n\u003cli\u003eKenya Floods Recovery Needs Assessment 2024 [Internet]. [cited 2025 Aug 22]. Available from: https://www.undp.org/sites/g/files/zskgke326/files/2025-05/kenya_floods_recovery_needs_assessment_2024.pdf\u003c/li\u003e\n\u003cli\u003eWHO. Greater Horn of Africa - WHO 2024 Health Emergency Appeal [Internet]. 2024 [cited 2025 Aug 7]. Available from: https://cdn.who.int/media/docs/default-source/documents/emergencies/2024-appeals/greater-horn-of-africa---who-2024-health-emergency-appeal.pdf\u003c/li\u003e\n\u003cli\u003eOCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #1 (11 April 2024) | OCHA [Internet]. 2024 [cited 2025 Aug 22]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-1-9-april-2024\u003c/li\u003e\n\u003cli\u003eOCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #6 (17 May 2024) | OCHA [Internet]. 2024 [cited 2025 Aug 22]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-6-17-may-2024\u003c/li\u003e\n\u003cli\u003eKenya Red Cross. Floods operations, 2024 [Internet]. 2024 [cited 2025 Aug 7]. Available from: https://www.redcross.or.ke/wp-content/uploads/2024/06/MAM-SitRep-18th-June-2024.pdf\u003c/li\u003e\n\u003cli\u003eOCHA K. Kenya: Heavy Rains and Flooding Update - Flash Update #7 (19 June 2024) | OCHA [Internet]. 2024 Jun [cited 2025 Aug 7]. Available from: https://www.unocha.org/publications/report/kenya/kenya-heavy-rains-and-flooding-update-flash-update-7-19-june-2024\u003c/li\u003e\n\u003cli\u003ePost-Disaster Needs Assessment Guidelines [Internet]. [cited 2025 Aug 23]. Available from: https://www.who.int/publications/i/item/post-disaster-needs-assessment-guidelines\u003c/li\u003e\n\u003cli\u003eKenya National Bureau of Statistics, editor. 2019 Kenya population and housing census. Nairobi: Kenya National Bureau of Statistics; 2019. \u003c/li\u003e\n\u003cli\u003eKNBS. Kenya National Bureau of Statistics [Internet]. 2021 [cited 2025 Aug 23]. Available from: https://www.knbs.or.ke/\u003c/li\u003e\n\u003cli\u003eWambua S, Malla L, Mbevi G, Nwosu AP, Tuti T, Paton C, et al. The indirect impact of COVID-19 pandemic on inpatient admissions in 204 Kenyan hospitals: An interrupted time series analysis. PLOS Glob Public Health. 2021 Nov 17;1(11):e0000029. \u003c/li\u003e\n\u003cli\u003eIrimu G, Ogero M, Mbevi G, Kariuki C, Gathara D, Akech S, et al. Tackling health professionals\u0026rsquo; strikes: an essential part of health system strengthening in Kenya. BMJ Glob Health [Internet]. 2018 Nov 28 [cited 2025 Aug 23];3(6). Available from: https://gh.bmj.com/content/3/6/e001136\u003c/li\u003e\n\u003cli\u003eNakweya G. Kenya: Patients are turned away from hospitals as doctors\u0026rsquo; strike enters fourth week. BMJ. 2024 Apr 11;385:q845. \u003c/li\u003e\n\u003cli\u003eKenya Medical Association. Kenya Medical Association Statement on delayed posting of medical interns 2024 [Internet]. [cited 2025 Aug 7]. Available from: https://kma.co.ke/images/KMA_Statement_on_Delayed_Posting_of_Medical_Interns_20240222.pdf\u003c/li\u003e\n\u003cli\u003eOsima S, Indasi VS, Zaroug M, Endris HS, Gudoshava M, Misiani HO, et al. Projected climate over the Greater Horn of Africa under 1.5 \u0026deg;C and 2 \u0026deg;C global warming. Environ Res Lett. 2018 May;13(6):065004. \u003c/li\u003e\n\u003cli\u003eExtreme Rainfall and Flooding over Central Kenya Including Nairobi City during the Long-Rains Season 2018: Causes, Predictability, and Potential for Early Warning and Actions [Internet]. [cited 2025 Aug 23]. Available from: https://www.mdpi.com/2073-4433/9/12/472\u003c/li\u003e\n\u003cli\u003eNDMA - KnowledgeWeb [Internet]. [cited 2025 Jun 10]. Available from: https://knowledgeweb.ndma.go.ke/Public/Resources/ResourceDetails.aspx?doc=28861990-6d58-4404-ba1a-bf7d0a940ea6\u003c/li\u003e\n\u003cli\u003eKenya: Floods - Apr 2024 | ReliefWeb [Internet]. 2024 [cited 2025 Jun 10]. Available from: https://reliefweb.int/disaster/fl-2024-000045-ken\u003c/li\u003e\n\u003cli\u003eMinistry of Health. Kenya Health Facility Census Report September 2023 [Internet]. 2024. Available from: https://www.health.go.ke/sites/default/files/2024-01/Kenya%20Health%20Facility%20Census%20Report%20September%202023.pdf\u003c/li\u003e\n\u003cli\u003eKRB. KENYA ROADS BOARD (KRB) MAP PORTAL [Internet]. [cited 2024 Mar 1]. Available from: https://maps.krb.go.ke/kenya-roads-board12769/maps\u003c/li\u003e\n\u003cli\u003eKarra K, Kontgis C, Statman-Weil Z, Mazzariello JC, Mathis M, Brumby SP. Global land use / land cover with Sentinel 2 and deep learning. In: 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS [Internet]. 2021 [cited 2025 Aug 23]. p. 4704\u0026ndash;7. Available from: https://ieeexplore.ieee.org/document/9553499\u003c/li\u003e\n\u003cli\u003eRCMRD. Kenya SRTM DEM 30meters [Internet]. [cited 2024 Mar 1]. Available from: https://gmesgeoportal.rcmrd.org/datasets/rcmrd::kenya-srtm-dem-30meters/about\u003c/li\u003e\n\u003cli\u003eUNEP-World Conservation Monitoring Centre (WCMC),International Union for Conservation of Nature (IUCN). The world database on protected areas. 2023 [cited 2023 Dec 9]. About. Available from: https://www.protectedplanet.net/en/about\u003c/li\u003e\n\u003cli\u003eWorldPop. WorldPop, open data for spatial demography | Scientific Data [Internet]. [cited 2025 Aug 23]. Available from: https://www.nature.com/articles/sdata20174\u003c/li\u003e\n\u003cli\u003eTatem AJ. WorldPop, open data for spatial demography. Sci Data. 2017 Jan 31;4:170004. \u003c/li\u003e\n\u003cli\u003eInternational Centre for Humanitarian Affairs. Kenya Flood Extent Layer [Internet]. 2025. Available from: https://redcrosske.maps.arcgis.com/home/item.html?id=bce5234218c547439ce2a71f7cbeb4e2\u003c/li\u003e\n\u003cli\u003eUnited Nations Institute for Training and Research. UNITAR/UNOSAT [Internet]. 2024. Available from: https://unosat.org/products/\u003c/li\u003e\n\u003cli\u003eRay N, Ebener S. AccessMod 3.0: computing geographic coverage and accessibility to health care services using anisotropic movement of patients. Int J Health Geogr. 2008 Dec 16;7(1):63. \u003c/li\u003e\n\u003cli\u003eON-GOING DOCTORS STRIKE [Internet]. [cited 2025 Aug 23]. Available from: https://www.knchr.org/Articles/ArtMID/2432/ArticleID/1192/ON-GOING-DOCTORS-STRIKE\u003c/li\u003e\n\u003cli\u003eNakweya G. Doctors in Kenya end strike after threat of court action. BMJ. 2024 May 14;385:q1088. \u003c/li\u003e\n\u003cli\u003eMacharia PM, Mumo E, Okiro EA. Modelling geographical accessibility to urban centres in Kenya in 2019. PLoS One. 2021;16(5):e0251624. \u003c/li\u003e\n\u003cli\u003eNyandiko NO. Devolution and disaster risk reduction in Kenya: Progress, challenges and opportunities. Int J Disaster Risk Reduct. 2020 Dec 1;51:101832. \u003c/li\u003e\n\u003cli\u003eOpere A. Chapter 21 - Floods in Kenya. In: Paron P, Olago DO, Omuto CT, editors. Developments in Earth Surface Processes [Internet]. Elsevier; 2013 [cited 2025 Aug 1]. p. 315\u0026ndash;30. (Kenya: A Natural Outlook; vol. 16). Available from: https://www.sciencedirect.com/science/article/pii/B9780444595591000219\u003c/li\u003e\n\u003cli\u003eVanderhoof MK, Alexander L, Christensen J, Solvik K, Nieuwlandt P, Sagehorn M. High-frequency time series comparison of Sentinel-1 and Sentinel-2 satellites for mapping open and vegetated water across the United States (2017\u0026ndash;2021). Remote Sens Environ. 2023 Apr 1;288:113498. \u003c/li\u003e\n\u003cli\u003eNyamira: Locals warned of potential flooding amid heavy rains [Internet]. [cited 2025 Aug 2]. Available from: https://citizen.digital/wananchi-reporting/nyamira-locals-warned-of-potential-flooding-amid-heavy-rains-n362416\u003c/li\u003e\n\u003cli\u003eStorey Building Collapses in Gesima, Nyamira County as Floods Persist - Kenyans.co.ke [Internet]. 2024 [cited 2025 Jul 28]. Available from: https://www.kenyans.co.ke/news/100001-5-storey-building-collapses-gesima-nyamira-county-floods-persist\u003c/li\u003e\n\u003cli\u003eJohary R, R\u0026eacute;villion C, Catry T, Alexandre C, Mouquet P, Rakotoniaina S, et al. Detection of Large-Scale Floods Using Google Earth Engine and Google Colab. Remote Sens. 2023 Jan;15(22):5368. \u003c/li\u003e\n\u003cli\u003eJuma B, Olang LO, Hassan MA, Chasia S, Mulligan J, Shiundu PM. Flooding in the urban fringes: Analysis of flood inundation and hazard levels within the informal settlement of Kibera in Nairobi, Kenya. Phys Chem Earth Parts ABC. 2023 Dec 1;132:103499. \u003c/li\u003e\n\u003cli\u003eWater Resources Authority (WRA). Flood Advisory \u0026ndash; Water Resources Authority [Internet]. [cited 2025 Aug 2]. Available from: https://wra.go.ke/flood-advisory-2/\u003c/li\u003e\n\u003cli\u003eResponse-To-Floods-in-Kenya [Internet]. [cited 2025 Aug 2]. Available from: https://www.oagkenya.go.ke/wp-content/uploads/2023/08/Response-To-Floods-in-Kenya.pdf\u003c/li\u003e\n\u003cli\u003eInternational Federation of Red Cross and Red Crescent Societies. Kenya Flood and Cholera Outbreak 2025 - DREF Operation (Appeal: MDRKE066) - Kenya | ReliefWeb [Internet]. 2025 May [cited 2025 Aug 25]. Available from: https://reliefweb.int/report/kenya/kenya-flood-and-cholera-outbreak-2025-dref-operation-appeal-mdrke066\u003c/li\u003e\n\u003cli\u003eICHA and Kenya Red Cross. Leveraging Technology in the Face of Disaster - Kenya | ReliefWeb [Internet]. 2024 [cited 2025 Aug 23]. Available from: https://reliefweb.int/report/kenya/leveraging-technology-face-disaster\u003c/li\u003e\n\u003cli\u003eKenya Meteorological Department (KMD). THE CLIMATE OUTLOOK FOR APRIL 2024 AND REVIEW FOR MARCH 2024 [Internet]. [cited 2025 Aug 25]. Available from: https://meteo.go.ke/documents/17/April_2024_Forecast_Rev_D.pdf\u003c/li\u003e\n\u003cli\u003eMakanga PT, Schuurman N, Sacoor C, Boene HE, Vilanculo F, Vidler M, et al. Seasonal variation in geographical access to maternal health services in regions of southern Mozambique. Int J Health Geogr. 2017 Jan 13;16(1):1. \u003c/li\u003e\n\u003cli\u003eEspinet Alegre X, Stanton-Geddes Z, Aliyev S. Analyzing Flooding Impacts on Rural Access to Hospitals and Other Critical Services in Rural Cambodia Using Geo-Spatial Information and Network Analysis [Internet]. Rochester, NY: Social Science Research Network; 2020 [cited 2025 Aug 25]. Available from: https://papers.ssrn.com/abstract=3614143\u003c/li\u003e\n\u003cli\u003eDaily Nation. Daily Nation. 2024 [cited 2025 Aug 22]. Kura announces road closures in Nairobi due to flooding. Available from: https://nation.africa/kenya/counties/nairobi/kura-announces-road-closures-in-nairobi-due-to-flooding-4601706\u003c/li\u003e\n\u003cli\u003eMinistry of Health (MoH). Kenya Public Health Emergency Operations Center (KPHEOC) Handbook [Internet]. [cited 2025 Aug 2]. Available from: https://www.nphi.go.ke/sites/default/files/KPHEOC%20Handbook.pdf\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-health-geographics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijhg","sideBox":"Learn more about [International Journal of Health Geographics](http://ij-healthgeographics.biomedcentral.com/)","snPcode":"12942","submissionUrl":"https://submission.nature.com/new-submission/12942/3","title":"International Journal of Health Geographics","twitterHandle":"@IJHGeo","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Geographical access, Travel time, Flooding, Healthcare system, Population affected","lastPublishedDoi":"10.21203/rs.3.rs-7724672/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7724672/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eClimate change is causing more frequent and severe extreme weather events, threatening health system resilience worldwide. In April and May 2024, Kenya experienced unprecedented extensive floods with devastating outcomes. However, the quantitative impact of flooding on geographical accessibility to healthcare remains unclear. This study evaluates post-disaster accessibility to health facilities and quantifies geographical coverage losses resulting from flooding compounded by a doctors\u0026rsquo; strike in Kenya.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eWe assembled geospatial datasets including health facility locations (public, private not-for-profit (PNfP), and private for-profit (PfP)), road networks, land use/land cover, topography, population density, and flooding extents. We defined a pre-flood baseline and three post-flood scenarios using satellite-derived flooding extents (Sentinel 1 SAR and NOAA-VIIRS satellites) and their combined maximal extents. Travel time (TT) to the nearest health facility by type was estimated using a least-cost path algorithm, accounting for \u0026plusmn;\u0026thinsp;20% variations in travel speed and flood extent for sensitivity analysis. Population coverage was extracted within five 30-minute TT bands for each scenario, nationally and by county.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eWe assembled 10,995 health facilities (public\u0026thinsp;=\u0026thinsp;5,586; PNfP\u0026thinsp;=\u0026thinsp;855; PfP\u0026thinsp;=\u0026thinsp;4,554). Pre-floods, average TT to the nearest facility was 19.6 min (16.4\u0026ndash;24.4), with public facilities at 20.7 min (17.3\u0026ndash;25.7), PfP at 37.8 min (31.6\u0026ndash;47.1), and PNfP at 49.2 min (41.1\u0026ndash;61.4). Post-floods average TT increased across all sectors, longest across PNfP at 113.5 min (94.6\u0026ndash;191.5 min) and shortest for public facilities at 48.5 min (40.5\u0026ndash;74.5 min). Pre-floods, 94.0% (52.5\u0026nbsp;million) of the population had access within 30-min and 20 out of 47 counties with an average TT of \u0026lt;\u0026thinsp;2-hours. Under the maximal flood extents, coverage dropped to 73% (40.9\u0026nbsp;million) and only 5 counties retained\u0026thinsp;\u0026lt;\u0026thinsp;2 hours TT. County-level 30-min coverage losses ranged from 1.0% (Nairobi) to 51.0% (Narok). In several arid counties, populations facing 2\u0026thinsp;+\u0026thinsp;hours TT rose to 15\u0026ndash;31%, up from 4\u0026ndash;12% pre-floods.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eKenya\u0026rsquo;s health system is highly vulnerable to floods, causing unequal disruptions in geographical access across subnational region. Incorporating disaster preparedness into county health care planning to strengthen health system resilience nationwide is needed.\u003c/p\u003e","manuscriptTitle":"Impact analysis of flood-induced changes in geographical accessibility and coverage to healthcare in both public and private sector, Kenya","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-14 08:11:14","doi":"10.21203/rs.3.rs-7724672/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-23T16:06:38+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-15T13:58:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"267244063722488671005755269226233254258","date":"2025-12-22T08:58:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"142892037258590134342991724571884276172","date":"2025-11-06T10:04:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"154688136378712308475663450521472083050","date":"2025-11-06T08:09:03+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-15T06:25:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"60036547594703406912926280749089555936","date":"2025-10-02T05:02:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-30T15:18:28+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-29T22:48:54+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-29T22:48:48+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal of Health Geographics","date":"2025-09-26T20:25:42+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-health-geographics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijhg","sideBox":"Learn more about [International Journal of Health Geographics](http://ij-healthgeographics.biomedcentral.com/)","snPcode":"12942","submissionUrl":"https://submission.nature.com/new-submission/12942/3","title":"International Journal of Health Geographics","twitterHandle":"@IJHGeo","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"81f02b4f-bc7a-481b-a6bf-388ae3333fed","owner":[],"postedDate":"October 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-23T16:06:31+00:00","versionOfRecord":{"articleIdentity":"rs-7724672","link":"https://doi.org/10.1186/s12942-026-00461-x","journal":{"identity":"international-journal-of-health-geographics","isVorOnly":false,"title":"International Journal of Health Geographics"},"publishedOn":"2026-03-18 15:59:46","publishedOnDateReadable":"March 18th, 2026"},"versionCreatedAt":"2025-10-14 08:11:14","video":"","vorDoi":"10.1186/s12942-026-00461-x","vorDoiUrl":"https://doi.org/10.1186/s12942-026-00461-x","workflowStages":[]},"version":"v1","identity":"rs-7724672","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7724672","identity":"rs-7724672","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}