Severe Acute Respiratory Syndrome Coronavirus 2 and Respiratory Virus Sentinel Surveillance, California, USA, May 10, 2020–June 12, 2021

Gail L. Sondermeyer Cooksey; Christina Morales; Lauren Linde; Samuel Schildhauer; Hugo Guevara; Elena Chan; Kathryn Gibb; Jessie Wong; Wen Lin; Brandon J. Bonin; Olivia Arizmendi; Tracy Lam-Hine; Ori Tzvieli; Ann McDowell; Kirstie M. Kampen; Denise L. Lopez; Josh Ennis; Linda S. Lewis; Eyal Oren; April Hatada; Blanca Molinar; Matt Frederick; George S. Han; Martha Sanchez; Michael A. Garcia; Alana McGrath; Nga Q. Le; Eric Boyd; Regina M. Bertolucci; Jeremy Corrigan; Stephanie Brodine; Michael Austin; William R. K. Roach; Robert M. Levin; Brian M. Tyson; Jake M. Pry; Kristin J. Cummings; Debra A. Wadford; Seema Jain


Emerging Infectious Diseases. 2022;28(1):9-19. 

In This Article


During April–October 2020, the CDPH recruited local county health departments (LHDs) and public health laboratories from 10 counties (Santa Clara, San Luis Obispo, Marin, Imperial, Contra Costa, Tulare, San Diego, Humboldt, Butte, and Ventura Counties) representing the geographic, demographic, and socioeconomic diversity of California. LHDs partnered with ≥1 outpatient clinical sites (e.g., urgent care, primary care or pediatric clinic, university clinic, drive-through or pop-up SARS-CoV-2 testing site) in their jurisdiction. Several LHDs used CalSRVSS as an opportunity to offer SARS-CoV-2 testing and conduct enhanced surveillance in settings that had lower access to testing (i.e., testing deserts) or in clinics serving populations that had a potentially higher risk for infection (e.g., serving particular demographic groups, university setting).

Partner clinical sites collected respiratory specimens and demographic, clinical, and epidemiologic data from a convenience sample of ≤50 persons/week/jurisdiction (i.e., participating county). Clinical sites sampled from adult and pediatric populations who were asymptomatic and seeking SARS-CoV-2 screening, which included contacts of confirmed COVID-19 cases; or populations that had mild symptoms (≥1 of the following new or worsening symptoms: fever [measured or subjective], cough, shortness of breath or difficulty breathing, chills or rigors, myalgia, headache, sore throat, new olfactory or taste disorder, congestion or runny nose, nausea or vomiting, diarrhea, or fatigue) where SARS-CoV-2 testing was recommended as part of clinical care. Site case definitions for enrollment, data elements, and launch dates (date of first specimen collection by participating county) varied. Launch dates by county were as follows: Santa Clara, May 10, 2020; San Luis Obispo, June 1, 2020; Marin, July 6, 2020; Imperial, July 12, 2020; Contra Costa, July 27, 2020; Tulare, August 25, 2020; San Diego, August 28, 2020; Humboldt, September 21, 2020; Butte, October 6, 2020; and Ventura, January 28, 2021. Persons could be tested multiple times to represent each testing incident; however, subsequent positive results were excluded.

Respiratory specimens were tested for SARS-CoV-2 by using US Food and Drug Administration–authorized PCRs. The CDPH Viral and Rickettsial Disease Laboratory tested specimens from all counties, except San Diego and San Luis Obispo Counties, for 20 pathogens (influenza A H1 and H3 and B [Yamagata and Victoria] viruses; parainfluenza types 1–4 viruses; human coronaviruses NL63, 229E, OC43, and HKU1; RSV; adenovirus; human metapneumovirus; rhinovirus; enterovirus; and Mycoplasma pneumoniae) by using a multiplex respiratory panel assay, and positive results were confirmed by singleplex PCR.[7,8] The San Luis Obispo County Public Health Laboratory tested specimens from San Luis Obispo County for a number of pathogens (influenza A and B viruses; parainfluenza types 1–4 viruses; human coronaviruses NL63, 229E, and HKU1; RSV; adenovirus; human metapneumovirus; rhinovirus/enterovirus; M. pneumoniae, Bordetella pertussis, and Chlamydia pneumoniae) by using the BioFire Respiratory 2.0 or 2.1 panels (

Respiratory panel testing was not conducted for specimens from San Diego County. In this analysis, we combined rhinovirus and enterovirus results as rhinovirus/enterovirus because both viruses belong to the genus Enterovirus and the assays used for this project did not distinguish between the different species. Co-infections were characterized as detection of SARS-CoV-2 and ≥1 respiratory pathogens in the same sample.

We described trends and participant characteristics by laboratory result; a respiratory panel positive result was defined as detection of ≥1 pathogens included in the respiratory panels. We also analyzed the 5-year American Community Survey 2018 data ( by demographic group (sex, age, race/ethnicity) for California to show how these data compared with demographic characteristics of the participant population. If participant ethnicity was reported as Latino/Hispanic, then race/ethnicity was listed as Latino; otherwise, race/ethnicity was listed as the reported race. When analyzing data on reported concurrent conditions, we restricted this analysis to ≥18 years of age for smoking, diabetes, and hypertension and to ≥5 years of age for obesity and asthma. For persons ≥16 years of age who reported being employed, free-text data on industry and occupation were coded according to the National Institute for Occupational Safety and Health Industry and Occupation Computerized Coding System and supplemented by manual review.[9] Occupation categories reported by <50 patients were excluded from analyses because of small numbers and unreliable estimates.

We calculated percentage positivity and corresponding 95% CIs by using the Wilson method, including Yates' continuity correction for small cell sizes (<5), for SARS-CoV-2 and respiratory panel (positive for ≥1 pathogen in the respiratory panels) by demographic group. We limited analysis to participants with a test result for both SARS-CoV-2 and the respiratory panel. We compared clinical manifestations and reported symptoms of participants who had positive results for SARS-CoV-2, rhinovirus/enterovirus, or non–SARS-CoV-2 human coronavirus and participants who had negative results for all pathogens. We also calculated the sensitivity and specificity of clinical case definitions.

We defined ILI in accordance with the California Influenza Surveillance Program[4] as any illness with fever and a cough or sore throat. We defined COVID-19–like illness (CLI) in accordance with the National Syndromic Surveillance Program as any illness with fever and cough or shortness of breath or difficulty breathing and symptomatic (reported ≥1 listed symptom) for SARS-CoV-2, rhinovirus/enterovirus, and non–COVID-19 coronavirus.

To assess associations between demographic, clinical, and epidemiologic characteristics and SARS-CoV-2 PCR results, we used mixed effects Poisson regression to calculate relative risks (RRs) and 95% CIs, both unadjusted and adjusted for sex, age (categorical), race/ethnicity, and county, allowing for random effects at the county level. We performed all analyses by using R software version 4.0.0 ( and created the map by using the R software Maps version 3.3.0 package.