Story: The 0.05 Line
Scene 1 — Friday Formulary Meeting
Riverside Wound Center debates HydroX. Vendor slide: “Time‑to‑closure improved by 0.4 days (95% CI 0.1–0.7), p=0.04.” Cost: +$58/patient. The committee approves based on p<0.05. Six months later: outcomes unchanged; spend +$74,000; nurses note no real‑world benefit.
Takeaway: statistically significant but trivial effect → costly decision.
Scene 2 — Same Team, Different Decision
OxyPatch pilot: closure at day 28 is 56% vs 40% (risk diff +16%, 95% CI −1% to +33%), p=0.07, +$8/patient. Rather than dismissing it, the team runs a pragmatic ward trial. Result: +12% closures (95% CI +3% to +21%), p=0.02, fewer painful changes, and net savings. OxyPatch is added with clear use criteria.
Takeaway: p≈0.07 ≠ “no effect.” Weigh effect size, CI, patient impact, and costs; gather local data when promising.
Cartoon placeholder
Drop your image here (replace the src of the <img> below) — or keep the SVG sketch.
