A survey of 3,437 households inside Quebec’s flooded postal codes. The worst-hit town reported a similar symptom rate to everywhere else, and what tracked with symptoms was a house still not fully back in use.
The basement took water in the spring. You pulled the carpet, ran fans for a week, and by June it looked fine. This study asks what your nose and throat were doing in October.
Elsa Landaverde and Mélissa Généreux of the Université de Sherbrooke, with Danielle Maltais of UQAC and Philippe Gachon of UQAM, surveyed households across the six Quebec regions hit hardest by the 2019 spring floods, eight to ten months after the water went down. The paper was published on 17 September 2022 in the International Journal of Environmental Research and Public Health.
What they did
A polling firm dialed at random inside every postal code where at least one residence or workplace had flooded. 3,437 households completed the questionnaire, 3,138 by telephone and 299 online, for an overall response rate of 15.3%.
Everyone in the sample lived in a flooded neighborhood, which is what makes the comparison worth reading. What separated the groups was what happened inside the individual house. Someone was counted as flooded if at least one liveable room took water. Someone was disrupted if no liveable room flooded but they were evacuated, lost home services, had trouble reaching community services, or had water in a non-liveable area such as a basement or garage. Unaffected meant neither.
Symptoms were counted only if they were frequent and persistent: at least two symptoms, reported at a frequency of at least 2-3 days per week over the previous six months, that improved or stayed the same when the person was away from their home. 11 ENT symptoms were on the list, including stuffy nose, runny nose, sneezing, sore or dry throat, itchy or watery eyes and ear pressure, along with five respiratory symptoms: cough, sputum, wheezing, shortness of breath and chest tightness. Episodes of flu, colds and seasonal allergies were excluded. That scale is not improvised. It is the one Quebec public health authorities use clinically to identify probable cases of mold-related respiratory illness linked to building sanitation.
What they found
Eight to ten months after the water receded, frequent symptoms sorted cleanly by exposure.
34.7% of flooded households reported frequent ENT symptoms, against 21.7% of the disrupted and 14.6% of the unaffected. 19.8% of flooded households reported frequent respiratory symptoms, against 8.8% and 6.9%.
Adjusted for personal characteristics and compared with the unaffected group, flooded households had an adjusted odds ratio of 3.18 (95% CI 2.45-4.14) for ENT symptoms and 3.41 (95% CI 2.45-4.75) for respiratory symptoms. The disrupted group, whose liveable rooms never took water, still came in at 1.76 (95% CI 1.45-2.14) for ENT symptoms and 1.45 (95% CI 1.10-1.91) for respiratory symptoms.
The part that surprises people
Severity did not sort the outcomes the way you would expect. Sainte-Marthe-sur-le-Lac lost a dike in 2019, flooding 700 to 800 residences and forcing a third of the town out. Residents there reported far worse conditions than the other regions: 24.5% had water at 30 cm or more on the first floor, against 3.4% elsewhere, along with larger material losses. Their ENT and respiratory symptom rates came out similar to flooded households everywhere else.
The paper is careful here, and so are we. Every primary stressor the team measured, water height in the home included, was significantly associated with both ENT and respiratory symptoms, and greater exposure did carry higher odds ratios. What failed was the gradient: the confidence intervals for the depth categories (under 30 cm, 30 to 100 cm, over 100 cm) overlapped, so the authors declined to claim a dose-response relationship. Depth mattered. It did not rank the households.
What did track with symptoms were the conditions that stretch a repair out. Among people who had been flooded or disrupted, for frequent ENT symptoms:
2.77 (95% CI 1.96-3.91) - unable to use all rooms normally a year after the flood. 2.54 (95% CI 1.66-3.89) - took a bank loan to meet expenses. 2.44 (95% CI 1.51-3.93) - received a sum covering less than half the damage costs. 1.87 (95% CI 1.11-3.13) - received less concrete or moral help than expected.
Most of those held for respiratory symptoms as well. Lacking flood insurance, on its own, did not significantly raise the odds of reporting symptoms. The authors offer the connection as a hypothesis rather than a finding: these are all markers of cleaning and restoration that ran slow, ran cheap, or got done without protective equipment.
Why the clock matters
Mold and bacteria multiply in environments that hold sufficient moisture for a certain period of time, and the risk to residents comes from the aerosolization of spores, toxins and other byproducts. The paper cites the standard guidance that cleaning and drying should be performed within 48 hours of water exposure to prevent mold growth. Duration is a variable the household controls in a way that water depth is not.
Read the secondary stressors against that clock and they stop looking like economics. A household waiting on an insurance decision, or rebuilding in stages as money allows, or still unable to use one room a year later, is a household where some assembly stayed wet well past 48 hours.
The limits, stated plainly
This study is cross-sectional and self-reported on both the exposure side and the symptom side, so it cannot establish cause. The 15.3% response rate leaves real room for who chose to answer to shape the result, and the authors raise that possibility themselves, including the chance that the most frustrated residents were likelier to participate. Nobody knows what these people’s symptoms were like before 2019, and many of the same households flooded in 2017, which the authors flag directly. Cellphone-only households could not be reached from postal codes, which under-represents younger adults.
What survives all of it is narrow and still useful: eight to ten months out, people whose rooms took water were about three times as likely to report frequent, building-linked symptoms as their neighbors on the same street, independent of personal characteristics.
What gets lost in the flood-cleanup marketing
Fans and dehumidifiers move air. They do not tell you whether the cavity behind the baseboard is dry. Drying equipment running in a room is evidence of effort, not evidence of a dry wall.
Fogging and ozone treatment work on surfaces and smell. The EPA says ozone generators sold as air cleaners are not safe or effective at concentrations that meet public health standards, ozone itself being a lung irritant, and no aerosol treatment removes water from an assembly. If the material is still wet, the smell comes back with the growth.
“It’s been dried out” is a claim. The measurement is a moisture reading taken inside the assembly, or air and surface sampling afterward. Ask which one you were given.
What the data supports
Treat 48 hours as the deadline, not “sometime this weekend.” That window is the difference between a wet floor and a colonized subfloor.
Open the assembly. Water travels under flooring, behind baseboard, into insulation and into drywall cavities, where surface appearance tells you nothing.
Judge dryness by measurement rather than by touch or smell. Both of those instruments are worse than people expect.
Treat a room you still can’t use as a health question, not just an unfinished project. In this data, that single condition carried the strongest association of any secondary stressor.
Translate the geography. Portland and Vancouver flood by atmospheric river, failed sump pump and sewer backup rather than by dike break. Different cause, same clock, same wet insulation.
None of this is medical advice, and none of it says that a particular flooded home is unsafe. The study establishes something narrower and still worth knowing: months after the water is gone, homes that took water were associated with substantially more frequent ENT and respiratory symptoms, and the conditions that delayed the repair tracked with symptoms without the clean gradient that water depth failed to produce. illumenair tests indoor air quality and mold. We do not perform water extraction, drying or remediation.
Wondering whether your house actually dried out?
illumenair tests indoor air quality and mold in Portland and Vancouver homes, room by room. We measure and report, and we do not remediate, so the result is not a sales pitch for the repair.
971-363-5626 · info@illumenair.com · www.illumenair.com
Sources: Landaverde E, Généreux M, Maltais D, Gachon P. Respiratory and Otolaryngology Symptoms Following the 2019 Spring Floods in Quebec. International Journal of Environmental Research and Public Health, 2022;19(18):11738, doi:10.3390/ijerph191811738.