Does soccer headgear reduce the incidence of sport-related concussion? A cluster, randomised controlled trial of adolescent athletes

Key points

1. 01Headgear made no measurable difference: concussion rate was 4.4% with headgear versus 4.1% without (HR 0.98, p=0.935).
2. 02Headgear also did not shorten recovery, with about 13 days lost from soccer in both groups (p=0.583).
3. 03Female players were concussed at roughly twice the rate of males (5.3% vs 2.2%).
4. 04Most concussions came from player-to-player contact (55.4%), not heading the ball, so padding the head helped little.
5. 05A possible benefit in females (as-treated HR 0.70, p=0.154) was not statistically significant and the study was underpowered to confirm it.

How it was conducted

Design

Cluster randomised controlled trial, with schools randomised to headgear or no headgear over the 2016/2017 and 2017/2018 seasons

Participants

2766 US high school soccer players (about 67% female, mean age 15.6), across 3050 player-seasons and 151,157 soccer exposures

Groups

Headgear group wore an ASTM-approved headgear model of their choice; control group wore none

Primary outcome

Time to first sport-related concussion, analysed with Cox proportional hazards models adjusted for sex, age, cohort year, prior concussion and baseline symptom severity

Secondary outcome

Severity measured as days lost from soccer, compared with a Wilcoxon rank-sum test

What they found

• A total of 130 concussions occurred, with females affected more (n=108, 5.3%, 1.10 per 1000 exposures) than males (n=22, 2.2%, 0.42 per 1000 exposures).
• Overall intent-to-treat analysis showed no difference in concussion rate between headgear (n=68, 4.4%) and no headgear (n=62, 4.1%): HR 0.98 (0.64 to 1.51), p=0.935.
• Days lost from soccer did not differ: 13.5 (11.0 to 18.8) days with headgear versus 13.0 (9.0 to 18.8) days without, p=0.583.
• Days of reported symptoms (p=0.831) and days in return-to-play protocol (p=0.075) also did not differ between groups.
• Non-concussion acute contact injuries did not differ either (RR 0.91, 0.64 to 1.29, p=0.584), arguing against the idea that headgear makes players take more risks.
• An as-treated female subgroup suggested a possible reduction (HR 0.70, 0.43 to 1.15, p=0.154), but it was not significant and the authors caution it was underpowered.
• Concussion rates varied by headgear model from 2.5% (Storelli ExoShield, HR 0.54, 0.20 to 1.43, p=0.213) to 5.4% (Forcefield Sweatband, HR 1.09, 0.62 to 1.91, p=0.765), though the study was not powered to compare models.
• Most concussions resulted from contact with another player (55.4%) rather than contact with the ball (35.4%).

Limitations

• Players in the headgear group chose among several different models rather than using one standard headgear, which mixes products with different lab performance.
• Players were not fitted with accelerometers and injuries were not filmed, so the study cannot say whether headgear reduced actual impact forces.
• Subgroup analyses, including the female trend toward benefit, were underpowered and exploratory.
• There is potential selection bias in which schools and players agreed to take part, though prior concussion history was similar between groups.

Why it matters

For patients

Buying soccer headgear is unlikely to protect you or your child from concussion, since most concussions come from colliding with other players rather than the ball.

For clinicians

Current soccer headgear should not be recommended as a concussion-prevention measure, consistent with consensus statements citing limited supporting evidence.

For readers

A well-run, large trial found no protective effect from headgear, reinforcing that concussion prevention in soccer needs strategies beyond head padding.