A validation study of a smartphone application for heart rate variability assessment in asymptomatic adults.
Bruno Bavaresco Gambassi, Victor Ribeiro Neves, Einstein Zeus Alves de Brito, Danilo Sobral da Silva Fernandes, Camila Almeida Sá, Raquel Moraes da Rocha Nogueira, Fabiano de Jesus Furtado Almeida, Pedro Augusto de Araújo Cavalcanti, Daniela Conceição Gomes Gonçalves e Silva, Danilo Sobreira Neto, Fabrício Olinda de Souza Mesquita, Tânia Maria Gaspar Novais, Vânia de Fátima Matias de Souza, Antonio Carlos Oliveira, Richard Diego Leite, Paulo Adriano Schwingel
Abstract
BACKGROUND: ) were calculated to determine the degree of association between the two electronic devices. Two-way repeated measures analysis of variance and Bland-Altman analysis were used to calculate the measurement consistency and agreement, respectively, between the two methods. Effect size was also used to estimate the magnitude of the differences. RESULTS: = 0.999; Confidence Interval [CI] at 95%: 0.999-0.999; P < 0.0001) and during orthostatic challenge (r = 0.988; 95% CI: 0.988-0.989; P < 0.0001). A negative bias of -0.526 milliseconds (95% limits of agreement [LoA] from -4.319 to 3.266 milliseconds) was observed in supine position between ECG and the smartphone application. On the other hand, a positive bias of 0.077 milliseconds (95% LoA from -10.090 to 10.240 milliseconds) during the orthostatic challenge was observed. CONCLUSIONS: Our results cross-validated a freeware smartphone application with the ECG-derived reference condition for asymptomatic adults at rest in the supine position and during orthostatic challenge.