# Cross-Sectional Study of Flood Damage Assumptions in Medical Facilities Using Geographic Information Systems

## TL;DR

This study uses geographic data to estimate flood damage to medical facilities in Japan and finds that floods could significantly impact healthcare access.

## Contribution

The study provides the first assessment of flood impact on medical facilities in Japan using nationwide geographic data and facility characteristics.

## Key findings

- Planned-scale flooding could damage 8.0% of general clinics and 10.8% of their beds in Japan.
- Maximum-scale flooding could damage 23.6% of general clinics and 23.9% of their beds.
- Non-disaster base hospitals had lower bed damage rates at planned-scale flooding compared to disaster base hospitals.

## Abstract

Introduction

Floods not only directly damage medical facilities but also hinder access to medical facilities, potentially disrupting local medical services. The scale of damage that medical facilities suffer from floods in Japan is unknown. In this study, we assessed the potential impact of floods on Japanese healthcare facilities by facility characteristics.

Methods

We conducted a cross-sectional study involving medical facilities registered in the Japan Medical Association Regional Medical Information System. Geographic data for the inundation area was obtained from open data of the Japanese government. Facilities that overlap with flooded areas were designated as affected facilities. The primary outcomes were the percentage of damaged facilities and beds. We calculated odds ratios (OR) and 95% confidence intervals (95%CI) using the Wald method to assess the impact of disaster base hospital designation on damage extent.

Results

We included 140,826 general clinics and 8,126 hospitals, which had 137,731 and 1,483,347 beds, respectively. The planned scale of flooding is estimated to affect 8.0% of general clinics and 10.8% of their beds. For hospitals, these figures were 8.8% and 7.8%, respectively. The maximum potential scale of flooding is estimated to affect 23.6% of general clinics and 23.9% of their beds. For hospitals, these figures were 22.5% and 20.6%, respectively. At the planned scale of flooding, there was no difference found in the rate of damaged facilities between disaster base hospitals and non-disaster base hospitals, and the rate of damaged beds was lower at non-disaster base hospitals (OR = 0.92, 95%CI = 0.71-1.18 for damaged facilities and OR = 0.79, 95%CI = 0.78-0.80 for damaged beds). At the maximum potential scale of flooding, there was no difference found in the expected damage between disaster base hospitals and non-disaster base hospitals (OR = 1.14, 95%CI = 0.95-1.38 for damaged facilities and OR = 0.99, 95%CI = 0.98-1.00 for damaged beds).

Conclusion

In Japan, floods can hinder nationwide medical functions, particularly in certain regions. Healthcare professionals should assess potential flood damage in advance and ensure that their workplace's business continuity plan includes appropriate countermeasures.