Flood stressors and mental distress among community-dwelling adults in Ghana: a mediation model of flood-risk perceptions

ABSTRACT Extant research has linked flooding with mental distress (MD) among flood-prone populations, but the mechanisms underlying this association remain largely unknown. This paper examines the association of flood stressors with MD among households in flood-prone communities in Ghana and estimates whether flood risk perceptions mediated the association. The study involved 767 household heads aged ≥20 years [mean = 47.3 ± 13.7); males = 61.4%]. Flood stressors were assessed using a 15-item Flood Stress-related Scale, while the Kessler Psychological Distress Scale (K10) was used to assess MD. Linear regression-based mediation analysis using Hayes’ PROCESS macro was performed to assess the indirect effect of flood stressors with MD through flood risk perceptions. After adjusting for potential confounders, flood stressors were significantly linked with higher risks of MD (β = .060-.080, p < .001). The mediation and bootstrapping analyses suggested that flood health risk perception partially mediated and accounted for 26.7% of the relationship between flood stressors and MD (direct effect: β = .060, Boots 95%CI: .041-.079; indirect effect: β = .022, Boots 95%CI: .015-.031; total effect: β = .082, Boots 95%CI: .063-.101). The link between flood stressors and MD risk is explained partially by flood health risk perception. Knowledge of households' flood risk perception is therefore critical for effectively managing the effects of flood stressors on MD. Policy Highlights The mechanisms underlying flood stressors-mental distress (MD) association remain largely unknown. Flood stressors were significantly associated with higher risks of MD. Flood stressors and MD risk linkage was partially explained by the health-related risk perception. Timely psychological interventional services to flood-prone households via improvement in their health-risk perceptions is desirable.