Validation of a new algorithm for the BPM-100 electronic oscillometric office blood pressure monitor

Background To test the accuracy of a new algorithm for the BPM-100, an auto mated oscillometric blood pressure (BP) monitor, using stored data from an independently conducted validation trial comparing the BPM-100(Beta) with a mercury sphygmomanometer. Design Raw pulse wave and cuff pressure data were stored electronically usi ng embedded software in the BPM-100(Beta) during the validation trial. The 391 sets of measurements were separated objectively into two subsets. A sub set of 136 measurements was used to develop a new algorithm to enhance the accuracy of the device when reading higher systolic pressures. The larger s ubset of 255 measurements (three readings for 85 subjects) was used as test data to validate the accuracy of the new algorithm. Methods Differences between the new algorithm BPM-100 and the reference (me an of two observers) were determined and expressed as the mean difference S D, plus the percentage of measurements within 5, 10, and 15 mmHg, Results The mean difference between the BPM-100 and reference systolic BP w as -0.16 +/- 5.13 mmHg, with 73.7% less than or equal to 5 mmHg, 94.9% less than or equal to 10 mmHg and 98.8% :! 15 mmHg. The mean difference between the BPM-100 and reference diastolic BP was -1.41 +/- 4.67 mmHg, with 78.4% less than or equal to 5 mmHg, 92.5% less than or equal to 10 mmHg, and 99. 2% less than or equal to 15 mmHg. These data improve upon that of the BPM-1 00(Beta) and pass the AAMI standard, and 'A' grade BHS protocol. Conclusion This study illustrates a new method for developing and testing a change in an algorithm for an oscillometric BP monitor utilizing collected and stored electronic data and demonstrates that the new algorithm meets t he AAMI standard and BHS protocol. (C) 2001 Lippincott Williams & Wilkins.