BackgroundThe novel coronavirus (COVID-19) has disproportionately impacted the Latinx community in the United States. Environmental risk factors, including community level pollution burden and exposure to smoking and secondhand smoke, have not been evaluated in relation to risk for infection with COVID-19.MethodsWe evaluated self-reported infection rates of COVID-19 in three, preexisting, longitudinal, Latinx family cohorts in the San Francisco Bay Area from May through September 2020 (N=383 households, 1,875 people). All households were previously recruited during pregnancy and postpartum at Zuckerberg San Francisco General Hospital (ZSFG) and UCSF Benioff before the pandemic. For the COVID-19 sub-study, participants responded to a 15-minute telephonic interview where we assessed food consumption patterns, housing and employment status, and history of COVID-19 infection based on community and hospital-based testing. We also evaluated secondhand smoke exposure based on previously collected self-reported data. Environmental pollution exposure was determined from census tract residence using California’s EnviroScreen 2.0 data. Non-parametric tests and multiple logistic regression were used to assess possible associations and independent predictors of COVID-19 infection.ResultsIn the combined Latinx, Eating and Diabetes Cohort (LEAD) and Hispanic, Eating and Nutrition (HEN) cohorts there was a 7.6% household infection rate (14/183) with a lower rate of 3.5% (7/200) in the Telomeres at Birth (TAB) cohort. Larger household size increased risk for infection (OR, 1.43 (95%CI 1.10-1.87)) in the combined LEAD/HEN cohorts and increasing number of children trended towards significance in the TAB cohort (OR 1.82, 95% CI 0.98-3.37). Any exposure to secondhand smoke in the household also trended towards increasing risk after adjusting for household size and other exposures (OR 3.20, 95%CI 0.80-12.73) and (OR 4.37, 95% CI 0.80-23.70). We did not find any associations between neighborhood pollution level based on census track location and risk of infection. Furthermore, we found weak evidence between dietary exposure and risk of COVID-19 infection after adjusting for possible confounders.ConclusionCrowding as indicated by household size increases risk for COVID-19 infection in Latinx families. Exposure to secondhand smoke may also increase risk for COVID-19 through increased coughing, respiratory impairment and increased travel of virus on smoke particles. Public policy and health interventions need to ensure that multiunit residential complexes prevent any exposure to secondhand smoke.