The accuracy of portable cholesterol analyzers in public screening programs.

To determine the accuracy of portable cholesterol analyzers in public settings, four screening organizations were accompanied to cholesterol screenings where consenting participants completed the finger-stick procedure and provided a blood sample by venipuncture. The finger-stick values were compared later with the participants' blood cholesterol values obtained in a reference laboratory. The results indicated that only one of the organizations produced cholesterol measurements entirely within the acceptable range (+/- 14.2%), while the accuracy of the other three organizations ranged from 76.5% to 96.4%. Those finger-stick values that did not fall within the acceptable range tended to underestimate the laboratory cholesterol values. Additionally, classification of the persons screened based on the National Cholesterol Education Program risk categories indicated that the finger-stick values primarily tended to produce false-negative results. The variability of the results across organizations was caused partially by insufficient operator training. However, inadequate quality-control procedures for field settings and dilution of capillary blood by tissue fluid also may have contributed to the inaccurate finger-stick results.