Neurology · Lancet Neurology · 2026
Mice with spinal-leak patients’ gene changes leaked more often. Paper: not chance. Its counts: maybe chance.
How often chance alone would give this gap if the gene changes did nothing
The paper says
Mice given any of three gene changes from spinal-leak patients leaked more often than normal mice when fluid was pumped around the spinal cord: 79%, 76%, 73% versus 57%. Chance alone would produce gaps this big under 1 time in 1,000, the paper says.
Its own data show
Its own counts: 15 of 19, 22 of 29 and 8 of 11 mice leaked, versus 20 of 35 normal mice. The test the paper names gives about 1 time in 9, 1 in 9 and 1 in 3. No standard test gets any of the three groups past the paper’s 1-in-20 bar.
“1 time in N” is how often chance alone would produce a gap this big if the changes did nothing; the paper’s bar for calling a gap significant is 1 in 20. With 11 to 35 mice per group, the counts fit anything from no difference to roughly double the risk: this test doesn’t show the changes make leaks likelier.