OBJECTIVE: This study explored the cost-effectiveness of the use of PA
PNET testing-supplemental, neural network-based testing used to increa
se the accuracy of cervical smear screening practice. STUDY DESIGN: Us
ing a model adapted from one developed under contract from the United
States Congress Office of Technology Assessment to evaluate cost-effec
tiveness of cervical cancer screening in elderly women, the study eval
uated U.S. women aged 20-64. The study reviewed various screening inte
rvals and examined the course of follow-up (i.e., repeat abnormal smea
r vs. immediate colposcopy) to offer a range into which most routine m
edical practice will fall. RESULTS: Assuming a false negative rate of
25% for low grade squamous intraepithelial lesion (SIL) and 15% for hi
gh grade SIL and applying increased sensitivity of 30% from PAPNET, we
can expect a cost of $48,474 per life-year saved among biennially scr
eened women. For women screened triennially, Its recommended by most g
roups, the cost per life-year saved is $25,185. CONCLUSION: The PAPNET
rescreening program is an economical approach to decreasing mortality
and morbidity from cervical cancer and compares favorably to other co
mmonly used interventions and diagnostic procedures.