Situational Awareness and Health Protective Responses to Pandemic Influenza A (H1N1) in Hong Kong: A Cross-Sectional Study

Liao, Qiuyan; Cowling, Benjamin; Lam, Wing Tak; Ng, Man Wai; Fielding, Richard

BACKGROUND: Whether information sources influence health protective behaviours during influenza pandemics or other emerging infectious disease epidemics is uncertain. METHODOLOGY: Data from cross-sectional telephone interviews of 1,001 Hong Kong adults in June, 2009 were tested against theory and data-derived hypothesized associations between trust in (formal/informal) information, understanding, self-efficacy, perceived susceptibility and worry, and hand hygiene and social distancing using Structural Equation Modelling with multigroup comparisons. PRINCIPAL FINDINGS: Trust in formal (government/media) information about influenza was associated with greater reported understanding of A/H1N1 cause (β = 0.36) and A/H1N1 prevention self-efficacy (β = 0.25), which in turn were associated with more hand hygiene (β = 0.19 and β = 0.23, respectively). Trust in informal (interpersonal) information was negatively associated with perceived personal A/H1N1 susceptibility (β = −0.21), which was negatively associated with perceived self-efficacy (β = −0.42) but positively associated with influenza worry (β = 0.44). Trust in informal information was positively associated with influenza worry (β = 0.16) which was in turn associated with greater social distancing (β = 0.36). Multigroup comparisons showed gender differences regarding paths from trust in formal information to understanding of A/H1N1 cause, trust in informal information to understanding of A/H1N1 cause, and understanding of A/H1N1 cause to perceived self-efficacy. CONCLUSIONS/SIGNIFICANCE: Trust in government/media information was more strongly associated with greater self-efficacy and handwashing, whereas trust in informal information was strongly associated with perceived health threat and avoidance behaviour. Risk communication should consider the effect of gender differences.