Background: An unplanned admission to the intensive care unit within 24 h of a procedure (UIA) is a recommended clinical indicator in surgical patients. Often regarded as a surrogate marker of adverse events, it has potential as a direct measure of patient safety. Its true validity for such use is currently unknown.

Methods: The authors validated UIA as an indicator of safety in surgical patients in a prospective cohort study of 44,130 patients admitted to their hospital. They assessed the association of UIA with intraoperative incidents and near misses, increased hospital length of stay, and 30-day mortality as three constructs of patient safety.

Results: The authors identified 201 patients with a UIA; 104 (52.2%) had at least one incident or near miss. After adjusting for confounders, these incidents were significantly associated with UIA in all categories of surgical procedures analyzed; odds ratios were 12.21 (95% confidence interval [CI], 6.33-23.58), 4.06 (95% CI, 2.74-6.03), and 2.13 (95% CI, 1.02-4.42), respectively. The 30-day mortality for patients with UIA was 10.9%, compared with 1.1% in non-UIA patients. After risk adjustment, UIA was associated with excess mortality in several types of surgical procedures (odds ratio, 3.89; 95% CI, 2.14-7.04). The median length of stay was increased if UIA occurred: 16 days (interquartile range, 10-31) versus 2 days (interquartile range, 0.5-9) (P < 0.001). For patients with a UIA, the likelihood of discharge from hospital was significantly decreased in most surgical categories analyzed, with adjusted hazard ratios of 0.41 (95% CI, 0.23-0.77) to 0.58 (95% CI, 0.37-0.93).

Conclusions: These findings provide strong support for the construct validity of UIA as a measure of patient safety.