Predictors of SARS-CoV-2 Infection in Youth at a Large, Urban Healthcare Center in California, March-September 2020

Abstract

Objective: To understand which social, epidemiologic, and clinical risk factors are associated with SARS-CoV-2 infection in youth accessing care in a large, urban academic institution. Methods: We conducted a prospective cohort study with case-control analyses in youth who received testing for SARS-CoV-2 at our academic institution in Los Angeles during the first wave of the COVID-19 pandemic (March-September 2020). Results: A total of 27,976 SARS-CoV-2 assays among 11,922 youth aged 0-24 years were performed, including 475 youth with positive SARS-CoV-2 results. Positivity rate was higher among older, African American, and Hispanic/Latinx youth. Cases were more likely to be from non-English-speaking households and have safety-net insurance. Zip codes with higher proportion of Hispanic/Latinx and residents living under the poverty line were associated with increased SARS-CoV-2 cases. Youth were more likely to have positive results if tested for exposure (OR 21.5, 95% CI 14.6-32.1) or recent travel (OR 1.5, 95% CI 1.0-2.3). Students were less likely to have positive results than essential worker youth (OR 0.5, 95% CI 0.3-0.8). Patterns of symptom presentation varied significantly by age group; number of symptoms correlated significantly with age in SARS-CoV-2 cases (r = 0.030, p < 0.001). SARS-CoV-2 viral load did not vary by symptom severity, but asymptomatic youth had lower median viral load than those with symptoms (21.5 vs. 26.7, p = 0.009). Conclusions: Socioeconomic factors are important drivers of SARS-CoV-2 infection in youth. Presence of symptoms, exposure, and travel can be used to drive testing in older youth. Policies for school reopening and infection prevention should be tailored differently for elementary schools and universities.

Keywords: COVID-19; LA County; SARS-CoV-2; testing; viral load; youth.

Figure 1

(A) SARS-CoV-2 test positivity by age. (B) SARS-COV-2 test positivity by race/ethnicity. In each panel, bars represent the mean proportion of positives +/−95% confidence interval. ^*p < 0.05, ^**p < 0.01, ^***p < 0.001, ^****p < 0.0001.

Figure 2

(A) Zip codes in Southern California with SARS-CoV-2 tests at UCLA. (B) Zip codes within Los Angeles County with positive SARS-CoV-2 tests at UCLA: Percentage of positive tests. (C) Zip codes within Los Angeles County with positive SARS-CoV-2 tests at UCLA: Absolute number of positive tests. (A) Zip codes in green reflect residents who were tested at UCLA and were negative. Zip codes in red reflect residents who were tested at UCLA and were positive. (A–C) Zip codes of youth tested for SARS-CoV-2 at UCLA included zip codes in Los Angeles, Santa Barbara, Ventura, San Bernardino, Riverside, San Diego, and Orange Counties. The number of cases of SARS-CoV-2 positive youth under 25 per zip code was significantly associated with the proportion of residents who identified as Hispanic/Latinx.

Figure 3

Number of reported symptoms at time of SARS-CoV-2 testing by age and test result. Linear regression and correlation coefficients estimate the change in number of symptoms by age. Youth with positive SARS-CoV-2 results had a significant correlation between age and number of symptoms (r = 0.30, p < 0.001); youth with negative results had no significant correlation between age and number of symptoms (r = 0.06, p = 0.25). Presence of symptoms increased significantly with age (p < 0.001). Youth with positive SARS-CoV-2 results are plotted in blue and youth with negative SARS-CoV-2 results are in red.

Figure 4

SARS-CoV-2 PCR cycle threshold (Ct) for available nasopharyngeal specimens (NP) from 475 youth with COVID-19. (A) Days of symptoms vs. cycle threshold (Ct) according to COVID-19 severity. (B) Median Ct values of positive PCR results in symptomatic vs. asymptomatic youth. (A) shows SARS-CoV-2 Ct values according to days since symptom onset for each positive SARS-CoV-2 PCR test. Asymptomatic/pre-symptomatic cases were distributed along the entire height of the y-axis from Ct values of 10 (representing the highest viral load) to Ct values of 40 (negative cutoff for a positive RT-PCR result). Mild/moderate cases represented the majority of data points (n = 213) and demonstrated a wide range of Ct values (10–40). Severe cases were fewer in number (n = 59) and generally had lower Ct values. (B) Median PCR Ct values of asymptomatic youth were significantly lower than that of symptomatic youth (21.5 vs. 26.7, p = 0.009).

Figure 5

Youth SARS-CoV-2 testing results by epidemic week. The figure reflects the evolution of the pandemic in Los Angeles County, with very few positive tests in the first week (two positive SARS-CoV-2 PCR tests) and in the last week of the study period (five positive SARS-CoV-2 PCRs). The number of positive SARS-CoV-2 PCR tests peaked in week 18 (July 6, 2020 to July 12, 2020), corresponding roughly with the summer peak of COVID-19 in Los Angeles County (23, 24). During week 18, the number of both positive and negative SARS-CoV-2 PCR tests reached the highest number during our observation period, with 60 positive SARS-CoV-2 PCRs and 704 negative SARS-CoV-2 PCRs, resulting in a positivity rate of 7.8%. Week 1 also showed a high positivity rate (6.7%), but a low number of tests were performed (n = 30) with the lowest number of positive tests (n = 2) observed that week. After the Week 18 peak, the number of positive tests remained elevated (22–37 positive tests/week) until week 25 (August 24, 2020 to August 30, 2020). At that point, the number of positive results began to decline.