Reliability and validity of a smartphone heart rate application for the assessment of resting and elevated heart rate

Purpose/Hypothesis: The purpose of this study was to investigate the test-retest reliability and concurrent validity of a heart rate (HR) application on both Android and iPhone smartphones as compared to Polar monitoring.

Number of Subjects: 111 healthy volunteers ranging from 21 to 61 years.

Materials/Methods: Methodological study assessing the test-retest and concurrent validity of the Azumio's Instant Heart Rate (version 3.0.1) application using both an iPhone and Android device. Both devices utilize the smartphone's camera via photoplethysmography to measure HR. Polar FT7 HR monitors were used as the gold standard comparison. Procedure: Participants were randomly assigned a device/side to be tested. Participants donned a Polar FT7 chest strap that was then assessed for proper reading. With the participants placed in a seated position, the testers simultaneously recorded initial resting HR from the participant's index fingers with the Android and iPhone devices. Outcomes from the smart phones were blinded to the testers. A third tester recorded information from the Polar device. Following 5 minutes of rest, the same procedure was repeated for a second reading. Participants then completed a one-minute step test before returning to the chair for a final recording.

Results: The Android, iPhone, and Polar devices all showed high test-retest reliability, with ICC values of .824, .758, and .843, respectively. At rest, both the Android and iPhone demonstrated a high degree of concurrent validity with the Polar device (Android ICC=.947, iPhone ICC=.919), however this was not the case after the one-minute step test, with the Android's ICC=.169 and iPhone ICC=.212. The standard error of measurement (SEM) at rest for the Android, iPhone and Polar were 4.09, 5.79 and 4.52 beats per minute respectively.

Conclusions: The poor agreement between the smart phone apps and the Polar post-exercise warrants further investigation. This may be due to the high degree of test variability captured by the Polar device following exercise. Additionally, further research is needed to compare the various devices following more vigorous exercise, in patients with pre-existing conditions, and using a metronome for step pace to reduce individual variability. Limitations that may have contributed to discrepancies in the results included hardware/ software inconsistencies such as measurement lag and non-reads.

Clinical Relevance: The Azumio Instant Heart Rate application is a free and popular application, that when used by either platform appears to be a reliable and valid tool to assess HR in healthy individuals in a clinical setting or by an individual as part of a home exercise program at rest. However, this application may not provide valid readings immediately following exercise.