Abstract
Introduction:
Mobile health (‘mHealth’) technologies include unregulated smartphone apps. Instant Blood Pressure (IBP; AuraLife, Newport Beach, CA) estimates blood pressure (BP) using the integrated sensors in the phone, without requiring a cuff. For 392 days in 2014-15, it was a top 50 best-selling iPhone app in the US (requiring ≥950 copies sold on those days). App store reviews document use of the app for treatment of hypertension. IBP has not been independently validated.
Hypothesis:
The absolute mean difference (MD) between IBP and a standard device will be ≤5.0 mmHg for systolic and diastolic BP measurements.
Methods:
Between Jul. and Sep. 2015, we enrolled adults age ≥18 yrs old, at Johns Hopkins clinics. Exclusion criteria were an active arrhythmia, internal device, or contraindication for arm BP measurement. Two order-randomized measurements were taken on two devices: IBP and a validated standard device (Omron-907). Each IBP measurement was compared to the average of the standard measurements. SBP and DBP MD were log transformed and T-tests were used to compare MD and the cutoff of ≤5.0 mmHg.
Results:
Of 85 participants, 52% (44 of 85) were women, mean ± SD age was 57 ± 16 years, and mean BMI was 28 ± 6 kg/m2; 53% (45 of 85) had hypertension. The mean ± SD absolute BP differences (IBP minus standard) were 12.4 ± 10.5 mmHg for systolic (P<0.001) and 10.1 ± 8.1 mmHg for diastolic BP (P<0.001). The app overestimated low BP and underestimated high BP (Figure). Sensitivity and specificity for hypertensive measurements (systolic ≥140 or diastolic ≥90 mmHg) were 0.2 and 0.9.
Conclusions:
BP measurements from an unregulated mHealth app with >148,000 units sold were highly inaccurate. The low sensitivity for hypertensive readings means that four-fifths of individuals with hypertensive BP levels will be falsely reassured that their BP is in the non-hypertensive range.