Comparison of SARS-CoV-2 Test Positivity in NCAA Division I Student Athletes vs Nonathletes at 12 Institutions

This cross-sectional study compares SARS-CoV-2 test positivity in college student athletes vs nonathlete students at 12 NCAA institutions.


Introduction
Collegiate and professional athletics were shut down temporarily in spring 2020 owing to the COVID-19 pandemic. As various collegiate and professional organizations debated the resumption of athletic activities, 2 primary concerns came to the forefront: adverse cardiac sequelae postinfection and potentially increased viral transmission within the athletic footprint. While initial studies raised concerns of widespread cardiac involvement postinfection, larger follow-up studies found low rates of myocarditis and other cardiac abnormalities in young athletes. [1][2][3] Strategies to mitigate disease transmission ranged from a complete bubble (in the National Basketball Association and Women's National Basketball Association) to a hybrid bubble (in Major League Baseball) to local implementation of strict distancing, face covering, and testing protocols (in the National Football League and National Collegiate Athletic Association [NCAA]). However, the specific risk of transmission within a collegiate athletic team setting including meals, practice, travel, competition, and communal housing with these various protocols is unknown. Although there have been anecdotal reports of outbreaks of SARS-CoV-2 infection within athletic teams, 4 these outbreaks have also been seen in other communal living settings. It is not known if collegiate student athlete infection rates are significantly higher than those of the general student or community population. [5][6][7]  In the NCAA's Resocialization of Collegiate Sport document, specific guidelines on polymerase chain reaction (PCR) testing, training, physical distancing, and face coverings were implemented to mitigate the risk of transmission within athletics. 8 Several studies have found these measures to be largely effective. [9][10][11] Although these minimum NCAA guidelines were implemented across all institutions, there was still variation in the frequency of student athlete testing owing to additional individual university or county protocols on testing. 12 Conversely, while most universities implemented a surveillance testing cadence for members of the university community (ie, students, faculty and staff, etc) using antigen or PCR tests, there was no minimum standard requirement to do so. To our knowledge, no study to date has looked at the association of participation in intercollegiate athletics with SARS-CoV-2 test positivity compared with those of the general university student population. This study examines test positivity of student athletes and nonathlete students at various universities during the 2020 to 2021 academic year to investigate if intercollegiate sport participation was associated with an increased risk of SARS-CoV-2 infection.

Methods
This cross-sectional study used publicly available deidentified data and was approved by the Stanford

Statistical Analysis
Data for each group were obtained through official press releases or public data dashboards, and time frames were matched when possible. These raw numbers were used to calculate test positivity for each group within a university, dividing the number of positive SARS-CoV-2 tests by the total number of tests administered during the study period. The test proportions of positives were then used to calculate the relative risk of a student athlete positive test compared with a nonathlete at a given institution. (0.01%-0.79%) for student athletes and 1.04% (0.40%-6.58%) for nonathlete students (Figure 1).
The relative risk for student athletes at these schools vs nonstudent athletes ranged from 0.002 (95% CI, 0.0005-0.01) for the University of Arkansas to 0.61 (95% CI, 0.54-0.70) for the University of Virginia (Figure 2). Of the remaining 3 schools, there was no statistically significant difference in test positivity at 2 of them, and 1 institution had a statistically significantly increased test positivity

Discussion
In this cross-sectional study, SARS-CoV-2 test positivity among student athletes did not vary by institution. This could be associated with implementation of surveillance and containment strategies across the NCAA. The 3 schools that did not have statistically significantly decreased student athlete test positivity compared with nonathlete student positivity were all located in California. Stringent public health guidelines at the local level could have been associated with this finding. Furthermore, all 3 had required weekly or twice weekly testing for the entire academic year. They had similar student athlete test positivity as the rest of the schools analyzed, but some of the lowest nonathlete student test positivity, possibly associated with these public health restrictions, as well as the frequency of regular surveillance testing for nonathlete students.
The specific mitigation protocols implemented and the frequency of surveillance testing varied widely between student athlete and nonstudent athlete populations and among universities. By NCAA standards, all in-season student athletes participating in high contact risk sports were tested a minimum of once per week, but out-of-season student athletes or those in low or medium contact Athlete vs nonathlete SARS-CoV-2 test positivity, relative risk (95% CI) risk sports were required to test once per month. Several institutions exceeded these standards and tested all student athletes and nonathlete students at least weekly, while others did not have any required surveillance testing for nonathlete students. Increased rates of surveillance testing in individuals without symptoms could be associated with decreased test positivity if these individuals would not otherwise have been testing; thus, among institutions that tested student athletes more frequently than nonathlete students, one could expect a decreased proportion testing positive.
There also appeared to be an inverse association in testing frequency with positivity in nonathlete students. Institutions that implemented regular surveillance testing for nonathlete students had decreased test positivity compared with institutions that did only on-arrival, random, or symptomatic testing, which is consistent with repetitive testing in many other settings. In fact, the 4 universities that required weekly or twice weekly surveillance testing for nonathlete students throughout the academic year had the lowest nonathlete student test positivity of the cohort. Conversely, there are also several factors which could be associated with increased test positivity among student athletes.
First, student athletes were traveling for competition regularly and may have been in close contact with a larger number of individuals outside their athletic teams and universities compared with nonathlete students. Moreover, student athletes were also frequently in close contact with their teammates during practice and competitions, particularly in sports designated as high risk for transmission; in these settings, social distancing and face coverings were not mandated by the NCAA or universities. These factors could be associated with increased risk of SARS-CoV-2 infection in student athletes.

Limitations
Despite being one of the first studies, to our knowledge, comparing test positivity in collegiate student athletes with nonathlete students, there are several limitations to this study. Most significantly, only 12 of 65 Power 5 institutions had publicly available testing data for analysis. It is unknown if these 12 schools are representative of the overall collegiate student athlete and nonathlete student population, particularly given that there has been significant geographic variability in SARS-CoV-2 infection rates and public health measures across the country. There may be a selection bias present in those universities that opted to make their data publicly available. As we noted previously, surveillance test frequency varied significantly among institutions and student populations, which was likely associated with test positivity. It is also possible that infection rates may have varied among sports; however, sport-specific data were not available for analysis. Given that the definition of student athlete was left up to the reporting institution, it is unknown if all student athletes were actively participating in sport during the period analyzed. Another important consideration is the association of vaccinations with positivity rates. Because the study analyzed tests from the 2020 to 2021 academic year, vaccinations became more widely available during the latter part of the study period, and the association of vaccination with positivity rates in team and university settings in the future is unclear. Moreover, the study period took place prior to the rise of the SARS-CoV-2 Delta and Omicron variants, and the association of this and future variants with test positivity in collegiate athletic settings is still to be determined.

Conclusions
This study found that varsity collegiate student athletes did not have increased risk of SARS-CoV-2 infection compared with nonathlete students, and at many institutions, they had a decreased risk.
The COVID-19 mitigation strategies implemented by the NCAA and individual universities may have been associated with these results. However, the association of future SARS-CoV-2 variants and more widespread vaccination with positivity outcomes is unknown.