Table 3—

Performance of the risk score in the three different cohorts

SensitivitySpecificityPV+PV−% Need subsequent testing
First half Inter99
 Cutoff
  2979.3 (71.4–86.3)68.7 (67.1–70.3)9.9 (8.2–11.6)98.7 (98.2–99.2)33.3 (31.7–34.9)
  3077.0 (69.9–83.6)70.2 (68.6–71.9)10.1 (8.3–12.1)98.6 (98.1–99.1)31.8 (30.1–33.3)
  3173.3 (66.1–80.9)74.3 (72.7–75.6)11.0 (9.1–13.2)98.5 (98.0–98.9)27.7 (26.1–29.3)
  3268.9 (61.0–77.1)78.0 (76.6–79.5)12.0 (9.8–14.5)98.3 (97.8–98.8)23.9 (22.4–25.4)
Second half Inter99 Cutoff 3166.7 (58.1–74.5)73.6 (71.9–75.2)9.7 (7.5–11.7)98.1 (97.5–98.7)28.1 (26.4–29.7)
ADDITION pilot study Cutoff 3175.9 (58.3–90.3)72.2 (69.3–75.1)7.3 (4.5–10.3)99.0 (98.3–99.6)29.2 (26.1–32.0)
  • Data are percentages and 95% CI calculated by bootstrapping (1,000). PV+, positive predictive value; PV−, negative predictive value.