Climate disaster effects on acute health care: a case study and model of the 2021 heatwave in British Columbia, Canada

Background

Acute health-care systems are a final layer of protection against growing climate impacts on population health. Climate disasters over the past decade have resulted in surges of patients seeking emergency care when preventive measures fall short. We aimed to understand how acute health-care delivery and access is vulnerable to climate disasters.

Methods

We built a discrete event simulation model to replicate acute health-care system dynamics during Canada's deadliest climate disaster—the 2021 heatwave. We used public data and government reports to estimate resource capacity per capita and interconnected trajectories to define the movement of patients between resources. In an intervention scenario, we evaluated the efficacy of a package of three interventions in the emergency department and prehospital settings (upstaffing before the disaster, mass casualty procedures, and outpatient cooling beds). Across a 29-day period, we measured six key performance indicators (KPIs) to compare statistical changes in waiting times between baseline and intervention models (physician initial assessment waiting time; waiting time for emergency department bed among most acute patients; waiting time for emergency department bed among least acute patients; ambulance response time; boarding time; and total time in the emergency department). Using Monte Carlo methods, we ran both baseline and intervention models 100 times.

Findings

We validated baseline model outputs against real-world data, with no statistically significant differences in all KPI medians. The baseline model showed significant negative effects on five of the six KPIs during the heatwave compared with the preheatwave period. Under the intervention model, four KPIs had significant improvements during the heatwave compared with the preheatwave period while the other two KPIs did not significantly change. Notably, emergency department waiting times decreased by over 35% with the interventions.

Interpretation

The model replicated real-world patterns and was a valid representation of system dynamics. Our findings showed that even a small surge in patients can be detrimental to health-care access and delivery. The model also suggests that health-care delays during climate disasters can be avoidable with proactive planning.

Funding

The Government of British Columbia.