Part 3 How decisions are currently made

Process versus outcome

A good outcome does not prove a good process.

This chapter argues that process and outcome have to be judged separately, not read off each other. The authors build a simple grid of good and bad process against good and bad outcome, then work four emergency cases through it: septic shock handled well, a missed bleeding ulcer, a risky pre-intubation transfer that happens to succeed and a low-risk chest pain death that followed a textbook pathway. The good outcome from a bad process is named the most dangerous square, because the reward locks in an unsafe habit. They propose a feedback-focused logbook, protected time for mutual coaching and performance dashboards, and they fault mortality meetings for only ever examining bad outcomes.

Three takeaways

  1. A good outcome can mask a dangerous process that then repeats
  2. Logbooks skew to the wins, losing half the useful feedback
  3. Mortality meetings review bad outcomes and ignore lucky escapes