Draw weight, posture and injury risk: Ji 2025 study

Thirteen students from a US university club shot three recurve bows of 20, 25 and 32 pounds at 10 and 15 yards while wearing a suit of inertial sensors. Ergonomics software rebuilt their posture at the moment of release and estimated the loads on the lumbar spine. Compression between the fourth and fifth lumbar vertebrae averaged between 469 and 773 newtons, depending on sex and on the combination of bow and distance, against the 3,400 newtons usually taken as a safety limit; shear forces ranged from 36 to 64 newtons front to back and from 18 to 55 sideways, against a limit of 700. On this measure none of the six combinations came near values regarded as risky. Compression was higher in the men (581 to 773 newtons) than in the women (469 to 555), which the authors link to the women leaning back further.

The differences showed up in the joint angles, and almost only with the heaviest bow. With the 32-pound bow men and women differed in flexion of the drawing shoulder (p = 0.05 at 10 yards, 0.02 at 15) and in trunk flexion and rotation (p between 0.003 and 0.04), by 9 to 15 degrees; with the 25-pound bow they differed in abduction of the bow shoulder. Only in the women did the bow-arm elbow change from one bow to another: with the 32-pound bow it was hyperextended by about 18 degrees on average (−17.7 and −17.9), while in the men it stayed between −2.4 and −7.5. The women also leaned back and rotated the trunk more, with more scattered angles. The correlations between the trunk angles and those of the shoulders and elbow grew with draw weight: between trunk rotation and flexion of the drawing shoulder, for instance, they went from 0.29 with the 20-pound bow to 0.76 with the 32.

The participants were 7 men and 6 women, split into 4 experts, 4 intermediates and 5 novices; only one was left-handed, and their data were mirrored to match the others. They shot three of the club's bare recurve bows, marked 20, 25 and 32 pounds at a 28-inch draw, with an arm guard and a tab, holding the string with three fingers under the nock. Each archer shot 5 arrows per combination, always in the same order: lightest bow at 10 yards, then at 15, then the 25-pound bow and finally the 32, for 30 arrows in all, with the target centre 140 centimetres off the ground. Before starting, a certified coach took everyone through a shot sequence pared down to the basics.

Movement was recorded with the Xsens MVN Awinda system, 17 wireless inertial sensors at 60 Hz, and fed into a digital human model in Siemens' JACK ergonomics software. The posture analysed was the one at release; from it the Force Solver module estimated the forces on the L4/L5 disc, using the bow weights and the draw force measured with a digital force gauge. Men and women were compared with t-tests, one combination at a time, and the relationships between angles with Pearson correlation. The same group has published two other papers on the same thirteen students, with partly different tasks: in the registry they count as a single sample.

The spinal forces are not measured: they come from a body model that starts from the joint angles and two forces applied at the hands, and this is the part most sensitive to the software's assumptions. The angles come from inertial sensors at 60 Hz, which capture overall posture well but the finer detail of the shoulder less so, and they are read at a single instant of the shot. The comparisons between men and women rest on 7 people against 6, with dozens of t-tests and no correction for multiple comparisons, so p values close to 0.05 call for caution. Sex, strength and skill overlap, and the paper does not say how experts and novices were split between the men and the women. The fixed order, from lightest to heaviest bow, also mixes fatigue in with draw weight.

The study measures neither injury nor pain and follows nobody over time: the "risk" in the title is inferred by comparison with thresholds developed for occupational settings, such as 90 degrees of shoulder elevation. The distances, 10 and 15 yards, are also far shorter than competition distances. There is at least one slip in the text, which has the bow hand anchoring at the corner of the mouth where the drawing hand is meant.

For anyone shooting bows between 20 and 32 pounds, the study indicates that the lumbar spine, at release, works under estimated loads well away from the limits used in ergonomics: even with the heaviest bow, compression stayed below a quarter of the threshold. It says nothing about hours of practice, the hundreds of arrows in a session or the higher draw weights used in competition.

The finding of most interest to a coach concerns the bow arm. With the 32-pound bow the women in the sample held the elbow hyperextended and compensated by leaning back and rotating the trunk; the authors put this down to insufficient strength, which they did not measure, and read it as a sign of a bow too heavy for the archer. It is a plausible reading, but the study does not test it, since it records neither strength nor pain nor score. What remains is the observation that changing bow changes posture, and that in this sample the change appeared first in the women, with the heaviest bow.