Background: Flooding is expected to increase due to climate change, population growth and urban development. The longer-term mental health impacts of flooding are not well understood. In 2015, the English National Study of Flooding and Health was established to improve understanding of the impact of flooding on health and inform future public health action.

Methods: We used 3 years of data from the English National Study of Flooding and Health. Participants who had consented to follow up were sent a questionnaire. Participants were classified into either "unaffected", "disrupted" or "flooded" according to their exposure. Logistic regression models were used to calculate adjusted odds ratios for probable depression, anxiety and post-traumatic stress disorder (PTSD) in each exposure group. The Wald test was used to assess the difference in probable mental health outcomes for those who did and did not experience "persistent damage" to their home. Conditional logistic regression was conducted to assess change in prevalence over the 3 years and to identify possible determinants of recovery.

Results: Eight hundred nineteen individuals were included in the final analysis - 119 were classified as unaffected, 421 disrupted and 279 flooded. Overall, 5.7% had probable depression, 8.1% had probable anxiety and 11.8% had probable PTSD, with higher prevalence in the flooded group compared with the unaffected group. After adjustment for potential confounders, probable mental health outcomes were higher in the flooded group compared to the unaffected group, significantly for probable depression (aOR 8.48, 95% CI 1.04-68.97) and PTSD (aOR 7.74, 95% CI 2.24-26.79). Seventy-seven (9.4%) participants reported experiencing persistent damage to their home, most commonly damp (n = 40) and visible mould (n = 26) in liveable rooms. Of the 569 participants who responded at all 3 years, a significant reduction in prevalence for all probable mental health outcomes was observed in the flooded group.

Conclusions: Flooding can have severe long-lasting consequences on mental health in affected populations. If these problems are not identified and treated early, they may persist for years. Further research is necessary to develop and evaluate interventions to increase resilience in at risk populations and to ensure timely access to support services following flooding.

The authors declare that they have no competing interests.

• Health Protection Research Unit in Evaluation of Interventions/National Institute for Health Research
• Health Protection Research Unit in Emergency Prepardness and Response/National Institute for Health Research
• Health Protection Research Unit in Environmental Change/National Institute for Health Research