Table 2—

Agreement between the in-laboratory flow respiratory disturbance index (RDI) and apnoea/hypopnoea index (AHI)

Correlation between AHI and flow-RDI r0.950.930.94 (0.91–0.96)
Difference between AHI and flow-RDI-8.2±9.1-10.8±12.0-9.5±10.7
Diagnostic ability of flow-RDI
 AHI >5 events·h−1
  Sensitivity0.960.960.96 (0.91–1.00)
  Specificity0.830.800.82 (0.59–1.00)
  P-LR5.74.85.3 (2.4–11.7)
  N-LR0.050.060.05 (0.03–0.09)
  AUC0.940.950.95 (0.90–0.99)
 AHI >15 events·h−1
  Sensitivity0.900.910.91 (0.84–0.98)
  Specificity0.900.730.82 (0.70–0.95)
  P-LR8.63.45.2 (3.9–6.8)
  N-LR0.110.120.11 (0.08–0.15)
  AUC0.970.940.96 (0.92–0.99)
 AHI >30 events·h−1
  Sensitivity0.830.930.89 (0.80–0.97)
  Specificity0.921.000.96 (0.90–1.00)
  P-LR10.8NA21.2 (8.3–54.6)
  N-LR0.180.070.12 (0.09–0.16)
  AUC0.960.990.98 (0.94–1.00)
  • Data are presented as n or mean±sd, unless otherwise stated. 95% confidence intervals are given in parentheses. FNH: Fukuoka National Hosiptal, Fukuoka, Japan; TCH: Tenri City Hospital, Tenri, Japan; r: Pearson's correlation coefficient; P-LR: positive likelihood ratio; N-LR: negative likelihood ratio; AUC: area under the receiver-operating characteristic curve; NA: not available.