How accurate are home blood pressure monitors used by patients ?

Problem

An accurate BP monitor is fundamental to the diagnosis and management of hypertension. Home blood pressure (BP) monitoring is recommended in hypertension guidelines and the ownership and use of home BP monitors is increasing. However the reliability and accuracy of these devices is currently unknown. We therefore tested the accuracy of a sample of home blood pressure monitors, and also gathered data on the approximate amount of usage, length of time participants had owned their monitors, and characteristics such as make, model, validation status, and cost.

Approach

Accuracy was evaluated by comparison to an Omron PA350 digital BP monitor tester, at 50mmHg intervals across the range 0-300mmHg following a standard process as recommended by each monitor manufacturer and the British Hypertension Society. Following manufacturer protocol, a difference from the reference monitor of +/-3mmHg was considered a failure. We also tested deflation, air leakage, and all cuffs in use. We assessed the relationship of monitor accuracy to length of time in use and number of recorded uses.A sample size of 385 testable monitors was conservatively estimated to be required, based on an assumed (because unknown) failure rate of 50%.

Findings

We have tested 281 monitors to date (and will have full results by the time of the conference). 62 further monitors proved untestable. Of the 281 devices tested 211 (75%) passed on all tests and 70 (25%) failed. The accuracy test was failed by 40 (14%) monitors, 17 (6%) of them at the 150 mmHg level closest to the threshold used for diagnosis and treatment. The largest difference from the reference monitor was 11.4mmHg; 8 monitors (2.8%) failed by more than 6mmHg. Of the 40 monitors that failed, 28 overestimated the pressure, 12 underestimated it. The other devices that failed did so because of cuff leakage. A total of 40 cuffs have failed to date.Analysis of precise test-reference monitor differences by usage rate and service duration will be presented. We will also present on the characteristics of home blood pressure devices that fail (notably length of time in service and amount of usage) so that appropriate advice can be given to patients.

Consequences

It is a matter of concern that one-quarter of the devices fail overall, though the monitor failure rate of 14% is similar to that previously identified in devices used in general practices. Approximately half of failures to date have been the result of cuff air leakage. This suggests storage and maintenance of cuffs may be a particular problem in home BP monitors.