Jason N Doctor1, Greg Strylewicz. 1. Department of Clinical Pharmacy & Pharmaceutical Economics & Policy, School of Pharmacy, University of Southern California, 1540 East Alcazar Street, CHP-140, Lost Angeles, CA 90089-9004, United States. addresses: jdoctor@pharmacy.usc.edu
Abstract
OBJECTIVE: In an effort to address the problem of laboratory errors, we develop and evaluate a method to detect mismatched specimens from nationally collected blood laboratory data in two experiments. METHODS: In Experiments 1 and 2 using blood labs from National Health and Nutrition Examination Survey (NHANES) and values derived from the Diabetes Prevention Program (DPP) respectively, a proportion of glucose and HbA1c specimens were randomly mismatched. A Bayesian network that encoded probabilistic relationships among analytes was used to predict mismatches. In Experiment 1 the performance of the network was compared against existing error detection software. In Experiment 2 the network was compared against 11 human experts recruited from the American Academy of Clinical Chemists. Results were compared via area under the receiver-operator characteristic curves (AUCs) and with agreement statistics. RESULTS: In Experiment 1 the network was most predictive of mismatches that produced clinically significant discrepancies between true and mismatched scores ((AUC of 0.87 (±0.04) for HbA1c and 0.83 (±0.02) for glucose), performed well in identifying errors among those self-reporting diabetes (N=329) (AUC=0.79 (±0.02)) and performed significantly better than the established approach it was tested against (in all cases p<.0.05). In Experiment 2 it performed better (and in no case worse) than 7 of the 11 human experts. Average percent agreement was 0.79 and Kappa (κ) was 0.59, between experts and the Bayesian network. CONCLUSIONS: Bayesian network can accurately identify mismatched specimens. The algorithm is best at identifying mismatches that result in a clinically significant magnitude of error.
OBJECTIVE: In an effort to address the problem of laboratory errors, we develop and evaluate a method to detect mismatched specimens from nationally collected blood laboratory data in two experiments. METHODS: In Experiments 1 and 2 using blood labs from National Health and Nutrition Examination Survey (NHANES) and values derived from the Diabetes Prevention Program (DPP) respectively, a proportion of glucose and HbA1c specimens were randomly mismatched. A Bayesian network that encoded probabilistic relationships among analytes was used to predict mismatches. In Experiment 1 the performance of the network was compared against existing error detection software. In Experiment 2 the network was compared against 11 human experts recruited from the American Academy of Clinical Chemists. Results were compared via area under the receiver-operator characteristic curves (AUCs) and with agreement statistics. RESULTS: In Experiment 1 the network was most predictive of mismatches that produced clinically significant discrepancies between true and mismatched scores ((AUC of 0.87 (±0.04) for HbA1c and 0.83 (±0.02) for glucose), performed well in identifying errors among those self-reporting diabetes (N=329) (AUC=0.79 (±0.02)) and performed significantly better than the established approach it was tested against (in all cases p<.0.05). In Experiment 2 it performed better (and in no case worse) than 7 of the 11 human experts. Average percent agreement was 0.79 and Kappa (κ) was 0.59, between experts and the Bayesian network. CONCLUSIONS: Bayesian network can accurately identify mismatched specimens. The algorithm is best at identifying mismatches that result in a clinically significant magnitude of error.
Authors: Curt L Rohlfing; Hsiao-Mei Wiedmeyer; Randie R Little; Jack D England; Alethea Tennill; David E Goldstein Journal: Diabetes Care Date: 2002-02 Impact factor: 19.112
Authors: Elizabeth A Wagar; Lorraine Tamashiro; Bushra Yasin; Lee Hilborne; David A Bruckner Journal: Arch Pathol Lab Med Date: 2006-11 Impact factor: 5.534