Health care

For a magnitude 7.0 earthquake under Greater Vancouver on a weekday afternoon, the province’s earthquake immediate response plan forecasts 1,000 people critically injured, 6,500 needing hospital care that is not critical, and 21,000 needing paramedics or first aid, from damage to buildings alone. [2] The province excludes injuries from fires, landslides, vehicle collisions and falls, so those counts do not cover everything that would send a person to a hospital. [2] The same plan says first responders and medical facilities would be overwhelmed. [2]

Hospitals are hindered as well by “thousands of uninjured, distraught people … looking for loved ones and seeking shelter”. [3]

A building can come through the shaking and stop being a hospital. The frame holds and the things fixed to it do not: the ceilings, the ductwork, the pipes overhead, the fittings that hold equipment in place. Damage to any of those, which engineers call non-structural damage, closes a department in a building that did not fall down.

A case study for the province sorted all 127 buildings owned by Vancouver Coastal Health by the era they were built in. [1]

That 65 per cent is one health authority’s buildings, not the region’s. It counts buildings expected to be completely damaged at a single design ground motion, which is a statement about a portfolio and not a forecast for any particular earthquake. The method takes the year a building went up as a stand-in for how it would perform, and its authors say reliable results need a detailed engineering assessment of each building. [1] It says nothing about whether any particular hospital would collapse.

When the power goes, the generator carries the building, and the generator burns fuel that arrives by truck. Once the water held in the building is gone, more of it has to be delivered the same way.

The Vancouver Building By-law sets emergency power run times for getting people out of a building safely: not less than two hours for elevators in tall buildings, for equipment supplying water to fight a fire and for smoke-control fans, and between thirty minutes and two hours for lighting. [4] For hospitals the by-law points elsewhere, to CSA Z32, the standard for essential electrical systems in health care facilities, which CSA sells. It does not publish the full text online. [4]

The project agreements for the new Surrey hospital and the new St. Paul’s each require, in identical terms, 72 hours of self-sufficiency for fuel, water and sanitary holding, which is wastewater kept on site. [5] [6] Neither hospital has opened.

Three engineers at UBC published a study in April 2026 modelling how a magnitude 9.0 Cascadia earthquake would affect people’s ability to reach the 16 Metro Vancouver hospitals that have emergency departments, counting damage to the hospitals and damage to the roads together. [7]

In the City of Vancouver the study’s measure of how well people can reach emergency care falls from about 0.9 before the earthquake to about 0.3 after it, mostly because of damage to the hospitals. Road damage sits outside it. [7] Across the region the model leaves 54,339 people unable to reach any hospital on the roads it considers, because the bridges they would cross are out, and retrofitting those bridges inside the model cuts that to under 27,000. [7] The bridges themselves are on the roads.

The model excludes non-structural damage, utility failures and staff who cannot get to work, and its authors say this leaves it “likely overestimating post-earthquake hospital functionality and underestimating accessibility loss”. [7] It also assumes the major river crossings stay open and that everyone drives on free-flowing roads. [7]

The Mass Casualty Incident Framework for British Columbia Hospitals, published by the Provincial Health Services Authority in January 2026, is the official document for this question and it does not answer it. [8] It standardises how hospitals respond to a mass casualty incident, and the words earthquake, seismic, Cascadia and post-disaster do not appear in it. Its one quantified surge figure, that 10 to 20 per cent of a hospital’s beds could be freed up quickly by moving patients, is drawn from published literature. It was not measured in British Columbia hospitals. [8]

The capacity figures that are public are academic. The 2026 study carries a table of acute care beds for its 16 hospitals and labels them “Assumed”: the authors compiled them from public health-authority pages. They count acute care beds; intensive care and emergency department beds are not separated. The hospitals are anonymised by city. [7] The only per-hospital intensive care counts in public come from a 2020 medical journal case series covering the six hospitals designated as COVID-19 centres, which is not an inventory of the region. [9]

That describes the St. Paul’s on Burrard Street. Its replacement on Station Street is built to the Vancouver Building By-law’s post-disaster requirement. [6] How many patients the other three could take is not in the plan.

Three major projects are specified to post-disaster importance category, the code’s term for a building expected to keep working after an earthquake: the new Surrey hospital and BC Cancer Centre [5], the new St. Paul’s [6] and phase two of Royal Columbian Hospital. [10] At Royal Columbian the acute care tower is post-disaster and the support building beside it is specified as normal importance category. [10] The standard is applied where care is delivered. A hospital is a group of buildings, and they are not all rated the same.

The new St. Paul’s also carries the only published number for how long a Lower Mainland hospital would be out of service, a median repair time of 30 days or less at the 1-in-2,475-year earthquake, which is the shaking the current building code designs against. [6] It is a target in a contract for a building that has not opened.

Richmond Hospital [11] and Lions Gate Hospital [12] predate current seismic standards. Richmond’s redevelopment is funded and under way in phases. [13]

The seismic ratings for the region’s hospital buildings do exist, held in a British Columbia Health Seismic Database maintained by the engineering firm Bush, Bohlman & Partners and used by the 2026 study under data-sharing arrangements with two health authorities. [7] It is not published.