7. How are new devices tested for accuracy?
All blood pressure devices have to be tested against reference manometers under laboratory conditions. In addition, clinical validation is recommended to evaluate clinical performance. Automated and non-automated devices are assessed against different International Organisation for Standardisation (ISO) standards, published in 2009 and 2007 respectively. A further standard covering clinical validation of automated devices is in preparation.
The standards specify such details as accuracy of cuff pressure measurements, and resistance to shock and vibration.
Continued accuracy in use relies on regular metrological testing. The most important thing to verify is the error of pressure measurement of the manometer, which must be tested in steps of no more than 50 mmHg over its entire range. Mercury manometers are not suited to use as reference devices for these tests as their resolution is not good enough and the mercury column is hard to read to the required precision (1.0 mmHg or less). The most accurate tests require a digital manometer which uses a piezoelectric sensor – made of material which produces a tiny electric current when it is stressed.
Clinical validation does not demand this degree of precision, and automated devices have normally been assessed against a mercury sphygmomanometer. Specialist medical organisations publish clinical validation protocols which specify how this should be done. All require the use of mercury devices as reference, although one protocol, from the CEN (European Committee for Standardisation) protocol also allows use of alternatives.
The normal protocol calls for a consecutive series of measurements over a wide range of blood pressures, comparing results with a mercury sphygmomanometer as a reference. Test devices are graded for accuracy and the results published after independent review by committees of the professional organisations concerned, such as the British Hypertension Society. Many devices have yet to be validated, but the number of validation studies has gone up from only 10 in 1990 to 104 in 2009. One of the alternative blood pressure measurement devices may prove suitable as a reference device for clinical validation in future. In the meantime, mercury sphygmomanometers will be needed for clinical validation of aneroid and automated devices, even if they pass out of routine clinical use.