Skip to main content

Table 6 Number and percentages of the properties identified by the algorithm in the modified descriptions per dataset

From: Contextual property detection in Dutch diagnosis descriptions for uncertainty, laterality and temporality

Property/dataset Amsterdam UMC dataset
(n = 73,280)
Multicenter dataset
(n = 175,210)
Amsterdam UMC dataset (n = 73,280) Multicenter dataset
(n = 175,210)
N Apparent prevalence % (95% CI) N Apparent prevalence % (95% CI) N Actual prevalence % (95% CI) N Actual prevalence % (95% CI)
Laterality 17,656 24.1 (23.8–24.4) 54,081 30.9 (30.7–31.1) 17,934 24.5 (24.2–24.8) 54,931 31.4 (31.1–31.6)
Uncertainty 11,834 16.1 (15.9–16.4) 16,779 9.6 (9.4–9.7) 12,235 16.7 (16.4–17.0) 17,347 9.9 (9.5–10.3)
Temporality 10,758 14.7 (14.4–14.9) 16,582 9.5 (9.3–9.6) 11,130 15.2 (14.9–15.4) 17,156 9.8 (9.7–9.9)
Removal of uncertainty 677 0.9 (0.9–1.0) 1,991 1.1 (1.1–1.2) 751 1.0 (1.0–1.1) 2,208 1.3 (1.2–1.3)
  1. Ordered by descending percentages of contextual properties