Smartphone camera applications may have high sensitivity and specificity for diagnosing atrial fibrillation (AFib), but their modest positive predictive value suggests these devices may generate a higher number of false-positive than true-positive results, according to a study published April 3 in JAMA Network Open.

Jack W. O'Sullivan, MBBS, DPhil, et al., analyzed 10 primary diagnostic accuracy studies to determine the precision of smartphone camera applications that diagnose AFib. Sensitivity and specificity were measured with a bivariate random-effects meta-analysis. To simulate the use of these applications as a screening tool, the positive predictive value and negative predictive value for different population groups were modeled.

Out of the 10 primary diagnostic accuracy studies, 3,852 participants and four applications were included. The applications analyzed the pulse wave signal for a mean of two minutes.

Results showed that the sensitivity and specificity for all applications combined were 94.2% and 95.8%, respectively, based on the meta-analysis.

The authors found that the positive predictive value for smartphone camera applications detecting AFib in an asymptomatic population aged 65 years and older was between 19.3% and 37.5%, and the negative predictive value was between 99.8% and 99.9%.

Furthermore, the authors found the positive predictive value and negative predictive value increased for individuals aged 65 years and older with hypertension (positive predictive value, 20.5% to 39.2%; negative predictive value, 99.8% to 99.9%).

"Future research should address the accuracy of these applications when screening other high-risk population groups, their ability to help monitor chronic AFib and, ultimately, their associations with patient-important outcomes," the authors conclude.

Clinical Topics: Arrhythmias and Clinical EP, Cardiovascular Care Team, Prevention, Atrial Fibrillation/Supraventricular Arrhythmias, Hypertension

Keywords: Atrial Fibrillation, Cell Phone, Heart Rate, Sensitivity and Specificity, Monitoring, Physiologic, Tumor Necrosis Factor Ligand Superfamily Member 13, Hypertension

< Back to Listings