Parachute use to prevent death and major trauma when jumping from aircraft: randomized controlled trial
A satirical paper that is also a real randomised trial, with ethics approval, block randomisation and an intention-to-treat analysis. In the PARACHUTE trial, Yeh and colleagues screened 92 aircraft passengers; the 23 who agreed to be enrolled were randomised to jump from a biplane or helicopter wearing a parachute or an empty backpack. Death or major traumatic injury was 0% in both arms, within five minutes of impact and at 30 days. The catch is in the comparison with the 69 screened but not enrolled: participants were on aircraft at a mean altitude of 0.6 metres travelling at 0 km/h, against 9,146 metres and 800 km/h, and most were study investigators. The trial was in part a response to Smith and Pell's call to (broken) arms.
The serious point is about who enters trials. When beliefs about an intervention's effectiveness exist in the community, Yeh and colleagues argue, trials might selectively enrol people with a lower perceived likelihood of benefit, diminishing the applicability of the results to clinical practice. They hold that randomised trials remain the gold standard for most new treatments, and suggest running definitive trials before a treatment becomes routine and reporting how participants differ from non-participants, while allowing that clinical beliefs may sometimes justifiably prevent a true randomised evaluation. The trial illustrates the selection problem rather than measuring how common it is. The bridge to psychological safety is ours: Tom draws from it that controlled studies of psychological safety necessarily come from lower altitudes, where failure is not catastrophic, and that the meaningful data come from retrospective investigation of disasters where voice failed.
Save this paper to a bibliography, keep notes on it, and export citations. Create a free account →