Four Square Step Test Performance in Hip Fracture Patients.

Small and selective: 40 people were enrolled and only 27 produced test data. The 13 who could not do the test were older and scored worse on nearly every cognitive and physical measure, so the results describe the better-performing part of an already restricted group. About 60% of the observations with an impaired visual organisation score were lost because those people did not perform the test, which is exactly the group the cognition question needed. Everyone scoring 73 or below on the cognitive screen was excluded, so the cognitive measure could not show what happens at more severe levels of impairment. The parent trial excluded people who could already walk 300 metres in six minutes, so everyone here was a slow walker to begin with, and entry was restricted in other ways because it was an efficacy trial. These findings do not describe the wider hip fracture population. Cross-sectional: everyone was measured once, before treatment, and nobody was followed to see who actually fell, so the 15-second cut-off is not validated for predicting falls after hip fracture. The authors did not record leg dominance, fear of falling or pre-fracture physical ability, so the fracture-side findings cannot be fully explained. The models were also chosen partly on how easy the measures would be to collect in practice, not on statistics alone.

The article text supplied contains no funding or conflict-of-interest statement, so this paper does not say who paid for the work or what the authors declared. The publication record lists NIH extramural research support.

Do not use the 15-second cut-off as a falls-risk threshold in hip fracture patients. It comes from community-dwelling older adults without a fracture, and this study measured people once, before treatment, and never recorded who went on to fall. Treat a slow time here as a sign of poor balance to work on, not as a validated prediction that the patient will fall.