In a recent study published in BMJ Medicine, researchers explored the long-term effects of flooding on mortality.

Study: Floods and cause-specific mortality in the UK: a nested case-control study. Image Credit: Valery Zotev/Shutterstock.com

Background

Floods are a major global issue that inflicts fatalities, property damage, and disruptions to critical services such as water, communications, energy, and public transportation. They can transmit infections, cause accidents, and expose individuals to hazardous chemicals.

Floods can increase noncommunicable illnesses such as cardiovascular disease, cancer, chronic respiratory disease, and diabetes. Stressors may cause mental health disorders, which can have long-term health consequences.

However, there is limited epidemiological data on the long-term mortality effects and studies on the implications of flooding exposure.

About the study

In the present study, researchers explored the influence of flood exposures on mortality outcomes.

The researchers retrieved flood data from the 2000–2020 Dartmouth Flood Observatory (DFO) linked with 499,487 United Kingdom Biobank participant data.

Individuals aged 37 to 73 years were enrolled in the UK Biobank between 2006 and 2010 from 21 centers across Scotland, Wales, and England. They excluded individuals lacking latitude and longitude data of residence and those lacking information on ethnicity, age, gender, sex, and deceased within the recruitment year.

The team followed the study participants until the end of the study period (December 31, 2020) or death, whichever occurred first. They determined the yearly cumulative flood exposure by multiplying flooding duration and severity and summing each year’s values.

They evaluated flood-associated any-cause and specific-cause mortality risks using the International Classification of Diseases, tenth edition (ICD-10) codes and explored lag patterns in mortality risks.

The researchers matched every case with eight controls and modeled flooding exposure using non-linear and lag-type conditional logistic regressions to determine the odds ratios (OR) for analysis.

Study covariates included educational attainment, body mass index (BMI), healthy diet score, physical activity, overall health rating, smoking status, alcohol intake, annual income, and Townsend deprivation index (TDI).

They used the European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA-5) dataset to obtain hourly humidity and temperature data and mapped meteorological data to the geocoded residential addresses of the participants.

Results

The mean participant age was 61 years; 59% (n=170,549) were males, and 281,175 (97%) were white. In total, the researchers noted 33,021 death cases.

Deceased individuals showed a higher likelihood of smoking, having a high BMI, having a lower household income, and being university graduates. They consumed less vegetables and fruits, more processed and red meat, and rated their health as fair to poor.

Any-cause mortality risk increased by 6.7% (OR, 1.1) per unit rise in flooding index after confounder control. In the crude model, per unit rise in flooding index increased any-cause mortality risk by 9.2% (OR: 1.1), with similar results after socioeconomic adjustment (OR, 1.1) but attenuated association after adjusting for lifestyle habits.

The team noted similar effects for specific-cause mortality in the fully adjusted model, wherein a higher flooding index was related to a higher death risk from neurodegenerative disorders (OR: 1.1), neoplasms (OR: 1.1), respiratory disorders (OR: 1.1), suicides (OR: 1.1), cardiovascular disorders (OR: 1.1), mental disorders (OR: 1.1), and digestive disorders (OR: 1.0).

For any-cause deaths, the strength of relationships rose from current times (OR, 1.0) to the third lag year (OR, 1.0) and reduced to nil by the fifth lag year. For death due to neurodegenerative and mental disorders, mortality risks were negligible during the current period but highest in the third and fourth lag years.

Contrastingly, mortality risks from suicides were the highest in current times (OR, 1.0) and decreased by the fifth lag year.

Individuals with higher educational attainment and annual income showed higher death risks for most reasons, and suicide-related mortality risks were higher among obese, non-moderate alcohol consumers residing in more deprived regions with lower income and lower physical activity.

Females aged less than 64 years shower higher death risks from any cause, including respiratory disease and neoplasms, but lower death risks from mental and digestive disorders.

In the sensitivity analysis, utilizing several imputed data points did not alter the findings, and the coronavirus disease 2019 (COVID-19) did not affect the results.

Odds ratios for any-cause deaths and tumors increased somewhat, whereas other causes remained stable. The odds ratios for any-cause deaths rose as the monthly flooding index increased.

Conclusion

The study findings showed that long-term flood exposure raises mortality risks, with varied lag times across different death causes. Flood exposures have long-term implications for neurological and mental illnesses, as well as an acute influence on suicide.

The vulnerable groups for flood-associated deaths varied by death cause. Over six years, every unit rise in the flood index increased the chance of all-cause death by 6.7%.

Flood exposure can have a variety of health repercussions over time. The findings could enhance public health practices and lessen the burden of illness caused by floods.