Table 3

Performance of the NeoHoP score ≥2 for the diagnosis of any HAI in the NeoHoP validation cohort, the NeoOBS-SA cohort and the NeoOBS-SA-VLBW cohort

External validation cohort*External validation cohort: VLBW only†Internal validation cohort‡
95% CI95% CI95% CI
n573373552
Sensitivity (%)65.060.5–69.466.160.2–71.354.247.3–60.9
Specificity (%)91.684.6–96.189.780.8–95.596.493.9–98.2
PLR7.74.1–14.56.43.3–12.515.28.6–26.9
NLR0.380.33–0.440.380.32–0.450.480.41–0.55
PPV (%)97.194.7–98.496.192.6–97.990.784.7–94.5
NPV (%)37.634.4–40.841.236.8–45.276.773.9–79.2
Diagnostic accuracy (%)70.066.1–73.771.266.0–75.479.976.3–83.2
Youden index0.570.560.51
  • *External validation cohort includes 346 infants with 573 episodes of suspected HAI selected from the NeoOBS-SA study.17

  • †External validation cohort: VLBW only is a subset of 222 VLBW infants within the NeoOBS-SA cohort (NeoOBS-SA-VLBW).17

  • ‡Internal validation cohort: the NeoHoP study included 406 VLBW infants with 552 episodes of suspected HAI that were used for internal validation of the NeoHoP score.16

  • HAI, healthcare-associated infection; NeoOBS, Neonatal Sepsis Observational study; NLR, negative likelihood ratio; NPV, negative predictive value; PLR, positive likelihood ratio; PPV, positive predictive value; VLBW, very low birth weight.