Read on Twitter
I have a new paper in @JAMANetworkOpen today
Punchline: Many college football players underestimate their risk of injury & concussion
Context: We are all poor estimators of risk, but we don't all have football-player level exposure
Full text: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2774487
Thread
1/
Punchline: Many college football players underestimate their risk of injury & concussion
Context: We are all poor estimators of risk, but we don't all have football-player level exposure
Full text: https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2774487
Thread
1/
Debates about the risks of football often hinge around 2 positions: 1. People should be allowed to make informed decisions re: participation in risky activities; & 2. When the risks are large or the population vulnerable, risk reduction is the responsibility of policy makers.
2/
2/
Many people are poor at estimating their personal risks, but this had not been empirically assessed in football players. We used a 2017 survey of 4 NCAA Power 5 football teams to evaluate whether college football players accurately estimate their risk of injury or concussion.
3/
3/
First, we compared prior season concussion and injury to athletes' perceptions of these outcomes next season. *Many* more athletes sustained a concussion (34%) or injury (68%) last season than thought it likely they would sustain a concussion (9%) or injury (20%) next season.
4/
4/
Next, we modeled athletes' individual single-season probability of each outcome and compared model results with athlete perceptions.
Since, as they say, all models are wrong but some are useful, we looked at this comparison several different ways (detailed methods in paper).
5/
Since, as they say, all models are wrong but some are useful, we looked at this comparison several different ways (detailed methods in paper).
5/
Across a range of approaches to compare model results and athlete perceptions, we found that: between 42% and 63% of athletes underestimate their risk of concussion and between 43% and 91% underestimate their risk of injury.
6/
6/
Why might this matter?
If we take the position that people should be allowed to make informed decisions about participation in risky activities, then half +/- of football players underestimating risks raises questions about what it means to be informed in this context.
7/
If we take the position that people should be allowed to make informed decisions about participation in risky activities, then half +/- of football players underestimating risks raises questions about what it means to be informed in this context.
7/
Furthermore, although many people are poor estimators of risk, not so many people are regularly exposed to football-player levels of risk.
And while nearly all activities carry some risk, the risks of football (indeed, many sports) are uniquely socially condoned.
8/
And while nearly all activities carry some risk, the risks of football (indeed, many sports) are uniquely socially condoned.
8/
Power 5 football has continued during a pandemic, partially under the premise that athletes can make informed decisions about participation. Our results & the tendency to underestimate risks generally suggest that athletes may also underestimate the health risks of COVID-19.
9/
9/
Also noteworthy:
34% of athletes in our study sustained at least 1 suspected concussion during the previous football season, or a mean of 43 suspected concussions per team. This is quite high compared w/ previously published rates (~5-6 dx'ed concussions per team season).
10/
34% of athletes in our study sustained at least 1 suspected concussion during the previous football season, or a mean of 43 suspected concussions per team. This is quite high compared w/ previously published rates (~5-6 dx'ed concussions per team season).
10/
Our study doesn't answer important empirical questions:
Do youth or high school athletes also underestimate their risks?
Would athletes make different decisions (sports choice, retirement, injury reporting) if they had a better understanding of their risks?
11/
Do youth or high school athletes also underestimate their risks?
Would athletes make different decisions (sports choice, retirement, injury reporting) if they had a better understanding of their risks?
11/
It also doesn't tackle the ethical questions:
If many people underestimate risks does it (uniquely?) matter that college football players underestimate theirs?
What would it mean to be sufficiently informed in this context?
What obligations do universities have to inform?
12/
If many people underestimate risks does it (uniquely?) matter that college football players underestimate theirs?
What would it mean to be sufficiently informed in this context?
What obligations do universities have to inform?
12/
We do not touch the distributional justice issues:
In the case of Power 5 football, athletes (many from minoritized groups) shoulder the vast majority of risk while universities take the lion's share of the benefits.
13/
In the case of Power 5 football, athletes (many from minoritized groups) shoulder the vast majority of risk while universities take the lion's share of the benefits.
13/
