Disease attack rates^* amongst household contacts by their own characteristics and index case characteristics.

There was not sufficient power to fit regression model when using the definition of disease of symptomatic and RT-qPCR positive, as only 8 contacts met this definition. However, the simple frequency tabulations with this outcome were consistent with the disease (regardless of test result) outcome analysis. Attack rates were higher from children who were younger (24% (6/25) from children <18 months compared to 2% (2/106) from children ≥18 months), had multiple episodes of vomiting (15% (8/82) compared to 0% from children with 1 or no episodes of vomiting (0/79)) and amongst contacts aged less than 10 years (27% (7/26) compared to contacts aged 10 years or older (0%; 0/91)).

Odds ratios, 95% confidence intervals and Wald-test p values are given for (a) univariate models, multivariate models of (b) other case or contact characteristics and (c) other case and contact characteristics.