ABSTRACT
Household secondary attack rate (HSAR) by risk factor might have a higher transmission rate between spouses. We investigated risk factors for the HSAR among non-spousal household contacts of patients with coronavirus disease 2019 (COVID-19). We studied household contacts of index cases of COVID-19 in Tsuchiura, Japan, from August 2020 through February 2021. The HSARs of the whole household contacts and non-spousal household contacts were calculated and compared across risk factors. We used a generalized linear mixed regression model for multivariate analysis. We enrolled 496 household contacts of 236 index COVID-19 cases. The HSAR was higher for spousal household contacts (37.8%) than for other contacts (21.2%). The HSAR was lower for non-spousal household contacts with a household size (number of household members) of two (18.2%), compared to the HSAR for contacts with a household size ≥4. The HSAR was higher for non-spousal household contacts of index patients with ≥3 days of diagnostic delay (period between onset and diagnosis) (26.0%) compared to those with ≤2 days' delay (12.5%) (p = 0.033). Among non-spousal household contacts, the HSAR was low for those with a household size of two and was high for contacts of index patients with a long diagnostic delay.