Olympic sports injuries and archery: Gimigliano 2021

This is a scoping review of musculoskeletal injuries in adult athletes across the 33 sports present at Tokyo 2020, not a study on archery. On PubMed, 1,022 titles were found and 162 read in full. For ten sports out of thirty-three no data emerged at all, shooting among them.

For archery the review reports four figures, all taken from a single included study: among overuse injuries 86.2% involve the upper limb and 11.8% the trunk; among acute ones 83.4% involve the upper limb; overuse injuries are 67.9% to tendons, ligaments and joints; acute ones are 83.3% fractures. The incidence rate is 0.00536 injuries per thousand hours of exposure.

From this the authors draw two statements that have to be read together. The first: archery is the sport with the highest risk of upper-limb injury, mainly to the shoulder of the dominant side. The second: among the twelve disciplines reporting a rate per thousand hours, archery has the lowest, 0.00536 against the marathon's 65.0. The two are not in conflict, because the first is a share of that sport's own injuries and the second is a frequency: archery is injured rarely and, when it is, almost always in the upper limb. By comparison, in the same table baseball has 51.4% of its injuries in the upper limb.

The authors also recall that at the 2016 Rio Games archery was among the sports with the lowest proportion of injured athletes, between zero and three per cent, against eight per cent overall and thirty-eight per cent for BMX. And they note that the ranking between sports changes with the metric: the three measures used in the literature cover twelve, thirteen and four disciplines respectively, and archery comes out lowest only in the one per thousand hours.

The review follows the PRISMA-ScR model for scoping reviews. The search was run on PubMed on 25 June 2021, with a string combining the terms for epidemiology and incidence with those for the musculoskeletal system and athletic injuries, crossed with the names of the 33 sports that were to make up the Tokyo 2020 programme.

Included were works published between 1 January 2001 and 31 May 2021 that had: a population of adult athletes over 18, professional or not, non-Paralympic, in one of the 33 sports; the incidence rate reported in one of the three formats used in the literature, that is injuries per thousand hours of exposure, per thousand athlete-exposures or per thousand player-hours; any study design; and the text in English.

The definition of injury is the one from the International Olympic Committee's manual: new or recurring musculoskeletal complaints incurred during competition or training that require medical attention, regardless of any absence from competition or training. The three ways of calculating incidence are explained by the authors because they are not interchangeable: athlete-exposures count participation in a session or a competition without regard to actual time, while hours of exposure and player-hours do take it into account.

The results are gathered in a table reporting sites, types and rates for each sport, with the sources indicated sport by sport. For archery, the table's row carries a single reference.

For archery this review is not an independent confirmation of anything: every figure in its row comes from a single included study. Anyone citing it is citing that study a second time, not two concordant studies.

The rate reported does not survive an arithmetic check. An incidence of 0.00536 injuries per thousand hours of exposure means one injury about every 186,000 hours. An archer training four hours a week for fifty weeks accumulates two hundred hours a year: at that rate they would have to shoot for more than nine hundred years before being hurt once. In the same table the marathon sits at 65.0 per thousand hours, that is twelve thousand times as much. The number as printed is not plausible as an injury rate, and it is inherited from the primary study; the review reports it without comment and uses it to place archery last among twelve disciplines.

The sentence that circulates most often, that archery is the sport at highest risk for the upper limb, has to be read for what it is. The 86.2% is the share of this sport's own overuse injuries that fall in the upper limb, not a frequency comparable across sports. In the same table baseball has 51.4%, so archery does sit higher on this share, but the share says where those who get hurt get hurt. Set beside the lowest rate of the twelve disciplines, the correct reading is that in archery injuries are few and concentrated in one place.

«The shoulder of the dominant side» is a phrase that does not resolve itself in archery, and the review does not resolve it. An archer's two shoulders do different jobs, one pushes the bow and the other draws the string, and knowing which of the two is the prevalent site is exactly the information needed: it is not here.

The limitations the authors declare remain, and they are substantial. The search was run on PubMed alone, where much of the sport science literature is not indexed, and for archery this weighs more than for other sports. The ways of reporting incidence are too many and not comparable, and the authors write explicitly that it cannot be said which sports have the absolutely highest or lowest rates. The table's data, except for the marathon, are general injuries rather than specifically musculoskeletal ones, even though this is a review of musculoskeletal injuries. And for ten sports out of thirty-three no data were found.

Finally the population: adults over 18 only, and no Paralympic athletes. On young archers, where an important part of the archery injury literature sits, this review says nothing.

The useful way to read this review is to hold its two figures about archery side by side. Archery has the lowest injury rate among the twelve disciplines using the same metric, and at the same time the highest share of upper-limb injuries, 86.2% of overuse cases. This is not a contradiction: it is the profile of a sport in which few people get hurt, and those who do almost always get hurt in the same area.

For an archer, the part that matters is not the frequency but the nature. Among overuse injuries, 67.9% involve tendons, ligaments and joints: that is the picture of repeated load on structures that do not adapt quickly, not of trauma. Among acute injuries, 83.3% are fractures, which in archery come from falls or accidents rather than from the shot. These are two different stories, and only the first has anything to do with how many arrows are shot.

Two things not worth leaning on. The rate of 0.00536 per thousand hours, which corresponds to one injury every one hundred and eighty-six thousand hours of shooting, is not a usable number: at that pace a club archer would have to shoot for nine centuries. It comes from a single study and the review passes it on without checking it. And the sentence about the highest risk to the upper limb, which is often quoted, does not mean archery is a risky sport: it says where the risk there is concentrates.

One last note for anyone trying to work out which shoulder to watch: the review speaks of «the shoulder of the dominant side» without saying whether it means the bow arm or the string arm. In archery that distinction is the point, and to have it you have to go to the primary study.