Why the rate—not the raw count—is the best evidence
The project’s stated aim is to
reduce postpartum hemorrhage (PPH) among women giving birth in the hospital. That is an
outcome aim, so the strongest evidence must be an
outcome measure expressed as a rate per 100 births. A rate adjusts for how many women actually delivered during each period, which lets the team compare performance fairly even if delivery volume changed.
Option 1 shows a fall from 3.0 to 1.8 hemorrhages per 100 births while the number of births remained similar, which is a true reduction in the frequency of PPH. The denominator is stable, so the change in the numerator reflects a genuine improvement in outcomes rather than a change in patient volume.
Why raw counts can mislead
Option 2 appears to show fewer hemorrhage cases—
30 down to
24—but the number of births also fell sharply, from
1,000 to
600. Converting to rates reveals the problem:
| Quarter | Hemorrhage cases | Births | Rate per 100 births |
|---|
| Before | 30 | 1,000 | 3.0 |
| After | 24 | 600 | 4.0 |
The hemorrhage rate actually rose from 3.0 to 4.0 per 100 births, so the outcome worsened despite the smaller raw count. Watch out! In quality-improvement data, a falling numerator can hide a rising rate when the denominator shrinks. Always calculate the rate before declaring success.
Process and structure measures support—but do not prove—the outcome
Option 3 reports that timely oxytocin administration rose from
70% to
98% of births. This is a
process measure: it shows the team is delivering a recommended intervention more consistently. Option 4 reports that all delivery room staff completed a refresher course on the third stage of labor. This is a
structure measure: it shows the environment or workforce capacity changed.
Both are valuable in a quality-improvement project. Improved uterotonic availability and staff education can plausibly contribute to fewer hemorrhages, as quality-improvement evaluations of evidence-based PPH protocols and uterotonic access strategies suggest . However,
a process or structure measure alone does not confirm that the clinical outcome—the actual PPH rate—improved. The team could administer oxytocin perfectly and still see no change in hemorrhage rates if other factors are at play.
Linking measures to the aim
A well-designed improvement project tracks all three types of measures, but the
outcome measure is the one that directly answers whether the aim was achieved. In this scenario, the aim is explicitly about reducing PPH, so the rate of hemorrhage per 100 births is the most direct evidence. A clinical pathway implementation study similarly evaluated PPH outcomes before and after the intervention by comparing actual hemorrhage-related results, not just whether staff followed the pathway .
Key point! When a project aims to reduce a clinical event, the best evidence of success is a change in the
outcome rate, not a change in raw counts, process compliance, or training completion.