Non-adherence to COVID-19 containment behaviours: results from an all-Ireland telephone survey

BACKGROUND: COVID-19 public health measures like handwashing and social distancing can help stem the spread of the virus. Adherence to guidelines varies between individuals. This study aims to identify predictors of non-adherence to social distancing and handwashing guidelines. METHODS: A cross-sectional weekly telephone survey was conducted over eight weeks (11/06/2020-05/08/2020). The sample included adults resident on the island of Ireland (75:25 split between ROI and NI). Data were collected on demographics, threat perceptions, fear of COVID-19, response efficacy and self-efficacy, response cost and social norms, COVID-19 behaviours, mood, loneliness, and self-reported health. RESULTS: 3011 participants were surveyed. Handwashing non-adherers were more likely to be male (OR: 5.2, 95% CI: 2.4 - 11.3), to have higher levels of loneliness (OR: 1.86, 95% CI: 1.1 - 3.1), and higher perceptions of handwashing costs (OR: 3.4, 95% CI: 2.2 - 5.2). Those reporting rarely engaging in social distancing were more likely to be members of lower socioeconomic groups, to be younger (OR: 0.97, 95% CI: 0.96 - 0.98), male (OR: 1.67, 95% CI: 1.1 - 2.5), healthcare workers (OR: 1.98, 95% CI: 1.1 - 3.4), to report lower mood (OR: 1.72, 95% CI: 1.3 - 2.2), were less likely to live in households with people aged under-18 (OR: 0.75, 95% CI: 0.6 - 0.9), and to have lower fear of COVID-19 (OR: 0.79, 95% CI: 0.6 - 0.9). CONCLUSIONS: Non-adherers to handwashing differ to social distancing non-adherers. Public health messages should target specific demographic groups and different messages are necessary to improve adherence to each behaviour. KEY MESSAGES: • non-adherers have different demographic/psychological covariates public health messages should be tailored to the specific behaviour/populations of interest, rather than generic messaging. • both behaviours (hand washing & social distancing) share some predictive variables, including male gender, perceptions of response cost, social norms and self-efficacy.