Design: International stepped-wedge cluster-randomised trial, masked endpoints
Population: 67,624 emergency presentations with suspected MI at 19 hospitals
Primary outcome: 30-day death or type 1 MI; emergency department length of stay
Effect: 1.1% vs 1.2%, adjusted OR 0.93 (0.77 to 1.13); median stay 309 vs 309 min

PRESC1SE-MI was a stepped-wedge cluster-randomised trial at 19 hospitals in 10 countries. Hospitals already using a 0/3 h high-sensitivity troponin pathway switched to the guideline-recommended 0/1 h pathway at a randomised time. It covered 67,624 presentations with suspected MI.

Death or new type 1 MI at 30 days occurred in 1.2% on 0/3 h and 1.1% on 0/1 h (adjusted OR 0.93, 0.77 to 1.13), meeting non-inferiority. Median emergency department stay was 309 minutes on both pathways.

The 0/1 h pathway is safe to adopt. But the hoped-for gain in throughput did not appear, because the bottleneck is usually what happens after the second troponin: waiting for reports, reviews and beds.

The 0/1 h pathway is safe to adopt with a validated high-sensitivity assay
Use the assay-specific cut-offs, not generic ones
Measure what delays discharge after the result before expecting faster flow
Keep clinical judgement for early presenters and ongoing pain

Why it matters

Faster blood tests alone did not move patients through the department any faster.

The statistics, in plain English

Non-inferiority means the trial showed the new pathway was not worse than the old by more than a set margin, here an odds ratio of 1.3. The upper bound of 1.13 sits inside that margin. The length-of-stay ratio of 1.00 (0.97 to 1.02) is a precise null: a difference of more than a few per cent is unlikely.