Table 4.
Surprise Question accuracy by profession
| Total | Died | Survived | Sensitivity | Specificity | PPV | NPV | c-statistic | |
|---|---|---|---|---|---|---|---|---|
| Doctors | ||||||||
| No | 218 | 71 (33) | 147 (67) | |||||
| Yes | 442 | 26 (6) | 416 (94) | 73 | 74 | 33 | 94 | 0.735 |
| Total | 660 | 97 (15) | 563 (85) | (63–82) | (70–78) | (26–39) | (92–96) | (0.688–0.783) |
| Nurses | ||||||||
| No | 720 | 210 (29) | 510 (71) | |||||
| Yes | 2516 | 259 (10) | 2257 (90) | 45 | 82 | 29 | 90 | 0.632 |
| Total | 3236 | 469 (14) | 2767 (86) | (40–49) | (80–83) | (26–33) | (86–91) | (0.608–0.655) |
Sensitivity (the ability to recognise those who were dying, e.g. 71/97 for doctors)
Specificity (the ability to recognise those who were not dying, e.g. 26/97 for doctors)
PPV positive predictive value, the proportion of patients who died when the clinician predicted dying, e.g. 71/218 for doctors), NPV negative predictive value, the proportion of patients who survived when the clinician predicted survival, e.g. 416/442 for doctors)