[Skip to Navigation]
Sign In
Invited Commentary
Health Policy
August 17, 2021

Was In-Person Attendance at Major College and Professional Football Games in 2020 the Right Call?

Author Affiliations
  • 1Salt Lake Informatics, Decision-Enhancement, and Analytic Sciences (IDEAS) Center, VA Salt Lake City Health Care System, Salt Lake City, Utah
  • 2Department of Internal Medicine, University of Utah School of Medicine, Salt Lake City
JAMA Netw Open. 2021;4(8):e2119580. doi:10.1001/jamanetworkopen.2021.19580

Among the countless aspects of society impacted by the global emergence of COVID-19, amateur and professional sporting events may have been one of the most publicly visible. Due to their cultural prominence in societies around the globe, and particularly in the United States, sports leagues command high levels of attention; are major drivers of local, regional, and national economies; and are a catalyst for public and private gatherings of all sizes. Indoor and outdoor stadiums typically host events with high numbers and concentrations of spectators, with some of the largest venues holding more than 100 000 people. Events are often accompanied by additional mass gatherings before and after, from tailgate parties in parking lots around or adjacent to the host stadium to packed bars and restaurants in the host (and, typically, many other) cities. It follows, then, that the magnitude of the economic, social, and cultural impacts of these events would be matched by the conspicuousness of their absence from the public spotlight.

Given the potential for these mass gatherings to efficiently spread the SARS-CoV-2 virus, the cause of COVID-19,1 it is no surprise that they were some of the first to be scrutinized during the pandemic. In North America, the onset of the pandemic occurred in the middle of 2 high-profile sports seasons—those of the National Basketball Association (NBA) and the National Hockey League (NHL)—and also shortly before the start of Major League Baseball’s spring training and the National Collegiate Athletic Association (NCAA) Division I men’s college basketball tournament, also known as March Madness. From March 11 to 16, 2020, all 3 professional leagues were suspended, and the NCAA men’s basketball tournament was cancelled.2-4 The NBA and NHL both resumed play in modified form in the fall using isolation zones or so-called bubbles in hub cities, with no spectators. Both leagues completed their seasons and were able to crown champions, with no cases of COVID-19 in either league during the time spent in their respective bubbles.

At the same time, 2 major American football organizations—the National Football League (NFL) and NCAA Division I Football Bowl Subdivision (FBS)—approached their upcoming seasons with a different strategy. The NFL, consisting of 32 member clubs based in 24 states, developed an extensive list of COVID-19 mitigation protocols for all clubs that included wearable proximity devices, restricted facility access, and frequent testing of players and staff with mandatory quarantine of high-risk contacts.5,6 More than 300 laboratory-confirmed COVID-19 cases were identified during the fall, some of which resulted from within-club transmission, requiring closure of that club’s facilities and, in a few cases, postponement and rescheduling of games. Unlike the NFL, which ultimately completed a full season and postseason, the NCAA Division I FBS season was heavily impacted by the pandemic. Although the NCAA produced a series of general recommendations regarding COVID-19 testing and mitigation, each of the 11 participating conferences—constituting 130 teams across 41 states—created protocols of their own guided by state and local regulations, with numerous differences around testing frequency, isolation and quarantine, mitigation, travel, and closing of facilities and suspension of play. Conference and team seasons were shortened, delayed, postponed, or cancelled outright, and dozens of games were postponed or cancelled due to COVID-19 exposures or outbreaks. Nevertheless, the season was ultimately completed with a reduced number of postseason games, culminating in a national championship game in January 2021.

Equally challenging during this time was the question of allowing spectators at these events. Teams, schools, and their host cities and states across the NFL and NCAA struggled to balance the concurrent pressures of increasing demand for in-person attendance and public safety concerns from steadily rising community transmission of COVID-19 across the United States during the second half of 2020. Although most teams in the NFL and NCAA began their fall seasons with stadiums empty of fans, the combined political, economic, and societal pressures eventually led some locales to relax attendance regulations around mass gatherings such as large sporting events, opening the doors (literally and figuratively) to the public. However, the rules and regulations governing in-person attendance were dictated largely by state and local authorities rather than by league officials, resulting in a heterogeneous mix of protocols governing details such as face covering, physical distancing, testing of spectators and, perhaps most notably, maximum attendance. A number of stadiums eventually opened to a small percentage of normal capacity, but those percentages varied over time and across teams. But with the potential for these gatherings to be super-spreader events, there was plausible concern about how they might escalate the already increasing community transmission of COVID-19 in the United States at that time.

In response, Toumi and colleagues7 used publicly available historical data on 2020 NFL and NCAA football games, county-level COVID-19 case counts and population data, and statewide COVID-19 nonpharmacological policies and interventions to estimate the association of limited in-person attendance at NFL and NCAA football games with COVID-19 cases in the counties in which they were held. To do this they performed a time series cross-sectional study using county matching and a difference-in-difference design to estimate the average treatment effect on the treated (ATT), in which the treatment was a county having an NFL and/or NCAA game with in-person attendance on the specified date. For each treated county, they constructed a set of matched counties based on having the identical treatment history during the specified pretreatment period and no treatments in the specified posttreatment period. They then refined the matched sets to account for control variables, such as population size and the presence of nonpharmaceutical interventions, and finally estimated the ATT using the outcome of daily COVID-19 cases per 100 000 residents at the county level.

Their results indicated that limited in-person attendance at NFL and NCAA football games did not significantly increase the community spread of COVID-19 cases in the counties in which they were held. Sensitivity analysis on the duration of the pretreatment and posttreatment periods demonstrated the robustness of their results, even when the posttreatment period was extended to 21 days. While the message is encouraging, the interpretation of the results requires some caution. Because the authors did not have access to exact attendance figures for the NCAA games, the representation of in-person attendance as a binary value prevented a more precise understanding of how the number (or concentration) of spectators may have influenced the outcome, which could be useful for state or local decision-makers. Additionally, other unmeasured variables not included in the model, such as other types of mass indoor or outdoor gatherings that may have occurred in treatment or control counties, could have biased the observed results toward the null. Furthermore, many NFL and NCAA football stadiums are outdoor, open-air facilities, where aerosols are rapidly dispersed, limiting the relevance of these results to leagues such as the NBA and NHL that largely hold indoor events.

Other groups have attempted to answer the same question with mixed results, although there remains a paucity of high-quality, published, peer-reviewed studies. Several independent, unpublished online reports have supported the idea that in-person attendance at NFL games was not associated with local clusters of COVID-19 cases, such as a study done in response to the introduction of measures by team, state, and local officials allowing limited in-person attendance at a pair of NFL games in Buffalo, New York, in January 2021.8 Others have found the opposite; an oft-cited but not peer-reviewed and still-unpublished preprint study from earlier this year using a similar, county-level analysis did find significant surges in infection rates in the second and third weeks following NFL games that were played with more than 5000 individuals in attendance.9

While the use of isolation zones or bubbles has proven to be effective at preventing outbreaks of COVID-19 among players, staff, and referees, the evidence to support different approaches to protect the safety of in-person spectators is scant. As vaccination against SARS-CoV-2 increases in the United States and around the world and restrictions on the size of gatherings and the use of facial coverings in public relaxes in different communities, the landscape around attendance at major sporting events is in a period of constant evolution. Studies such as that by Toumi et al,7 which add to our understanding of the virus and its transmission, are essential for providing decision-makers with the information needed to design and enact policies that align with evidence and protect people in the community. More like it are needed as the United States and other countries confront the mounting political and societal pressure of regaining a sense of normalcy that comes with returning our large and beloved sports arenas to full—and full-throated—capacity.

Back to top
Article Information

Published: August 17, 2021. doi:10.1001/jamanetworkopen.2021.19580

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Rubin MA. JAMA Network Open.

Corresponding Author: Michael A. Rubin, MD, PhD, MS, Salt Lake Informatics, Decision-Enhancement, and Analytic Sciences (IDEAS) Center, VA Salt Lake City Health Care System, 500 Foothill Dr, M/C 182, Salt Lake City, UT 84148 (michael.rubin@hsc.utah.edu).

Conflict of Interest Disclosures: Dr Rubin reported receiving support from the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research and Development Service.

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government.

McCloskey  B, Zumla  A, Ippolito  G,  et al; WHO Novel Coronavirus-19 Mass Gatherings Expert Group.  Mass gathering events and reducing further global spread of COVID-19: a political and public health dilemma.   Lancet. 2020;395(10230):1096-1099. doi:10.1016/S0140-6736(20)30681-4PubMedGoogle ScholarCrossref
Botkin  B, Kaskey-Blomain  M. NBA suspends season due to coronavirus outbreak; owners preparing for no games until June, per report. CBS Sports. March 11, 2020. Accessed May 5, 2020. https://www.cbssports.com/nba/news/nba-suspends-season-due-to-coronavirus-outbreak-owners-preparing-for-no-games-until-june-per-report/
NHL statement on coronavirus. NHL.com. March 12, 2020. Accessed May 5, 2020. https://www.nhl.com/news/nhl-coronavirus-status/c-316155530
NCAA cancels remaining winter and spring championships. NCAA.org. March 12, 2020. Accessed May 5, 2020. https://www.ncaa.org/about/resources/media-center/news/ncaa-cancels-remaining-winter-and-spring-championships
DiFiori  JP, Green  G, Meeuwisse  W, Putukian  M, Solomon  GS, Sills  A.  Return to sport for North American professional sport leagues in the context of COVID-19.   Br J Sports Med. 2021;55(8):417-421. doi:10.1136/bjsports-2020-103227PubMedGoogle ScholarCrossref
Mack  CD, Wasserman  EB, Perrine  CG,  et al; NFL COVID-19 Advisory and Operational Team.  Implementation and evolution of mitigation measures, testing, and contact tracing in the National Football League, August 9-November 21, 2020.   MMWR Morb Mortal Wkly Rep. 2021;70(4):130-135. doi:10.15585/mmwr.mm7004e2PubMedGoogle ScholarCrossref
Toumi  A, Zhao  H, Chhatwal  J, Linas  BP, Ayer  T.  Association of limited in-person attendance in US National Football League and National Collegiate Athletic Association games with county-level COVID-19 cases.   JAMA Netw Open. 2021;4(8):e2119621. doi:10.1001/jamanetworkopen.2021.19621Google Scholar
Reisman  N. Analysis: Bills games did not lead to rise in COVID cases. Spectrum News 1. February 9, 2021. Accessed May 5, 2021. https://spectrumlocalnews.com/nys/central-ny/ny-state-of-politics/2021/02/09/analysis--bills-games-did-not-lead-to-rise-in-covid-cases
Kurland  J, Piquero  A, Leal  W, Sorrell  E, Piquero  N.  COVID-19 incidence following fan attendance: a case study of the National Football League 2020-2021 season.  Preprints with the Lancet. Preprint published March 31, 2021. doi:10.2139/ssrn.3805754.
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words