Direct comparison of the ‘Reduced’ HUNT model against NLST criteria, for the same number of individuals as selected by the NLST (n=918), on the control group of DLCST (n=1999). The HUNT model shows increased predictive performance in all metrics (sensitivity, specificity, PPV, NPV)
| LC (N) | Without LC (N) | Total (N) | Predictive value | |
| Population | 52 | 1 947 | 1999 | |
| ‘Reduced’ HUNT | ||||
| Criteria positive | 40 TP (4.36%) | 878 FP (95.64%) | 918† | PPV 4.36% |
| Criteria negative | 12 FN (1.11%) | 1069 TN (98.89%) | 1081 | NPV 98.89% |
| Sensitivity | 76.92% | |||
| Specificity | 54.90% | |||
| NLST‡ | ||||
| Criteria positive | 35 TP (3.81%) | 883 FP (96.19%) | 918* | PPV 3.81% |
| Criteria negative | 17 FN 1.57% | 1064 TN 98.43% | 1081 | NPV 98.43% |
| Sensitivity | 67.31% | |||
| Specificity | 54.65% |
‘Reduced’ HUNT model.
*‘Reduced’ HUNT model criteria.
†The number of those picked by the NLST, but with the top risk score by the ‘Reduced’ HUNT model.
‡NLST criteria (: >30 pack years of smoking, <15 years since quitting and ages between 55 and 74 years).
FN, false negative; FP, false positive; NPV, negative predictive value; PPV, positive predictive value; TN, true negative; TP, true positive.