Transmission of SARS-COV-2 Infections in Households — Tennessee and Wisconsin, A...

Transmission of SARS-COV-2 Infections in Households — Tennessee and Wisconsin, April–September 2020

Improved understanding of transmission of SARS-CoV-2, the virus that causes coronavirus disease 2019 (COVID-19), within households could aid control measures. However, few studies have systematically characterized the transmission of SARS-CoV-2 in U.S. households (1). Previously reported transmission rates vary widely, and data on transmission rates from children are limited. To assess household transmission, a case-ascertained study was conducted in Nashville, Tennessee, and Marshfield, Wisconsin, commencing in April 2020. In this study, index patients were defined as the first household members with COVID-19–compatible symptoms who received a positive SARS-CoV-2 reverse transcription–polymerase chain reaction (RT-PCR) test result, and who lived with at least one other household member. After enrollment, index patients and household members were trained remotely by study staff members to complete symptom diaries and obtain self-collected specimens, nasal swabs only or nasal swabs and saliva samples, daily for 14 days. For this analysis, specimens from the first 7 days were tested for SARS-CoV-2 using CDC RT-PCR protocols.† A total of 191 enrolled household contacts of 101 index patients reported having no symptoms on the day of the associated index patient’s illness onset, and among these 191 contacts, 102 had SARS-CoV-2 detected in either nasal or saliva specimens during follow-up, for a secondary infection rate of 53% (95% confidence interval [CI] = 46%–60%). Among fourteen households in which the index patient was aged <18 years, the secondary infection rate from index patients aged <12 years was 53% (95% CI = 31%–74%) and from index patients aged 12–17 years was 38% (95% CI = 23%–56%). Approximately 75% of secondary infections were identified within 5 days of the index patient’s illness onset, and substantial transmission occurred whether the index patient was an adult or a child. Because household transmission of SARS-CoV-2 is common and can occur rapidly after the index patient’s illness onset, persons should self-isolate immediately at the onset of COVID-like symptoms, at the time of testing as a result of a high risk exposure, or at the time of a positive test result, whichever comes first. Concurrent to isolation, all members of the household should wear a mask when in shared spaces in the household.§

The data presented in this report are from an ongoing, CDC-supported study of household transmission of SARS-CoV-2 in Nashville, Tennessee and Marshfield, Wisconsin, and include households enrolled during April–September 2020.¶ Households were eligible if the index patient had symptom onset <7 days before household enrollment and the household included at least one other person who was not symptomatic at the time of the index patient’s illness onset and was thus deemed to be at risk. Characteristics of the index patients, household members, and their interactions were ascertained using Research Electronic Data Capture (REDCap),** an online application for data collection, or through paper-based surveys. The 7-day secondary infection rate was calculated by dividing the number of laboratory-confirmed SARS-CoV-2 infections identified during the 7-day follow-up period by the number of household members at risk per 100 population.†† Because saliva samples are considered an emerging diagnostic approach but are not yet standard for SARS-CoV-2 detection (2), secondary infection rates were also calculated using positive test results from nasal swab specimens only. To account for household members possibly having been infected when the index case became ill, secondary infections rates were also conservatively calculated excluding household members who had positive test results at enrollment. The study protocol was reviewed and approved by the Vanderbilt University Medical Center’s and Marshfield Clinic Research Institute’s Institutional Review Board, and was conducted consistent with applicable federal law and CDC policy.§§

For this analysis, 101 households (including 101 index patients and 191 household members) were enrolled and completed ≥7 days of follow-up. The median index patient age was 32 years (range = 4–76 years; interquartile range [IQR] = 24–48 years); 14 (14%) index patients were aged <18 years, including five aged <12 years and nine aged 12–17 years. Among index patients, 75 (74%) were non-Hispanic White, eight (8%) were non-Hispanic persons of other races, and 18 (18%) were Hispanic or Latino (Table 1). Index patients received testing for SARS-CoV-2 a median of 1 day (IQR = 1–2) after illness onset and were enrolled in the study a median of 4 days (IQR = 2–4) after illness onset.

The median number of household members per bedroom was one (IQR = 0.8–1.3). Seventy (69%) index patients reported spending >4 hours in the same room with one or more household members the day before and 40 (40%) the day after illness onset. Similarly, 40 (40%) of index patients reported sleeping in the same room with one or more household members before illness onset and 30 (30%) after illness onset.

Among all household members, 102 had nasal swabs or saliva specimens in which SARS-CoV-2 was detected by RT-PCR during the first 7 days of follow-up, for a secondary infection rate of 53% (95% CI = 46%–60%) (Table 2). Secondary infection rates based only on nasal swab specimens yielded similar results (47%, 95% CI = 40%–54%). Excluding 54 household members who had SARS-CoV-2 detected in specimens taken at enrollment, the secondary infection rate was 35% (95% CI = 28%–43%).

Forty percent (41 of 102) of infected household members reported symptoms at the time SARS-CoV-2 was first detected by RT-PCR. During 7 days of follow-up, 67% (68 of 102) of infected household members reported symptoms, which began a median of 4 days (IQR = 3–5) after the index patient’s illness onset. The rates of symptomatic and asymptomatic laboratory-confirmed SARS-CoV-2 infection among household members was 36% (95% CI = 29%–43%) and 18% (95% CI = 13%–24%), respectively.