Jesse D Cook1, Michael L Prairie2, David T Plante2. 1. Department of Psychiatry, School of Medicine and Public Health, University of Wisconsin-Madison, Madison, WI, USA. Electronic address: jdcook4@wisc.edu. 2. Department of Psychiatry, School of Medicine and Public Health, University of Wisconsin-Madison, Madison, WI, USA.
Abstract
BACKGROUND: Sleep disturbance is a common and important component of affective illness. Fitness activity trackers are emerging as alternative means to estimate sleep in psychiatric patients; however, their ability to quantify sleep in mood disorders has not been empirically evaluated. Thus, this study sought to evaluate the utility of the Fitbit Flex (FBF) to estimate sleep in patients with major depressive disorder (MDD) relative to gold standard polysomnography (PSG) and a widely-used actigraph (Actiwatch-2; AW-2). METHODS: Twenty-one patients with unipolar MDD wore the FBF and AW-2 during in-laboratory PSG. Bland-Altman analysis compared sleep variables among devices. Epoch-by-epoch analysis further evaluated sensitivity, specificity, and accuracy for the FBF and AW-2 relative to PSG. RESULTS: The FBF demonstrated significant limitations in quantifying sleep and wake, relative to PSG. In the normal setting, the FBF significantly overestimated sleep time and efficiency, and displayed poor ability to correctly identify wake epochs (i.e. low specificity). In the sensitive setting, the FBF significantly underestimated sleep time and efficiency relative to PSG. Performance characteristics of the FBF were more similar to the AW-2 in the normal compared to sensitive setting. LIMITATIONS: Participants were young to middle aged and predominantly female, which may limit generalizability of findings. Study design also precluded ability to assess longitudinal performance of FBF. CONCLUSIONS: The FBF is not an adequate substitute for PSG when quantifying sleep in MDD, and the settings of the device sizably impact its performance relative to PSG and other standard actigraphs. The limitations and capabilities of the FBF should be carefully considered prior to clinical and research implementation.
BACKGROUND:Sleep disturbance is a common and important component of affective illness. Fitness activity trackers are emerging as alternative means to estimate sleep in psychiatricpatients; however, their ability to quantify sleep in mood disorders has not been empirically evaluated. Thus, this study sought to evaluate the utility of the Fitbit Flex (FBF) to estimate sleep in patients with major depressive disorder (MDD) relative to gold standard polysomnography (PSG) and a widely-used actigraph (Actiwatch-2; AW-2). METHODS: Twenty-one patients with unipolar MDD wore the FBF and AW-2 during in-laboratory PSG. Bland-Altman analysis compared sleep variables among devices. Epoch-by-epoch analysis further evaluated sensitivity, specificity, and accuracy for the FBF and AW-2 relative to PSG. RESULTS: The FBF demonstrated significant limitations in quantifying sleep and wake, relative to PSG. In the normal setting, the FBF significantly overestimated sleep time and efficiency, and displayed poor ability to correctly identify wake epochs (i.e. low specificity). In the sensitive setting, the FBF significantly underestimated sleep time and efficiency relative to PSG. Performance characteristics of the FBF were more similar to the AW-2 in the normal compared to sensitive setting. LIMITATIONS: Participants were young to middle aged and predominantly female, which may limit generalizability of findings. Study design also precluded ability to assess longitudinal performance of FBF. CONCLUSIONS: The FBF is not an adequate substitute for PSG when quantifying sleep in MDD, and the settings of the device sizably impact its performance relative to PSG and other standard actigraphs. The limitations and capabilities of the FBF should be carefully considered prior to clinical and research implementation.
Authors: Luciane de Souza; Ana Amélia Benedito-Silva; Maria Laura Nogueira Pires; Dalva Poyares; Sergio Tufik; Helena Maria Calil Journal: Sleep Date: 2003-02-01 Impact factor: 5.849
Authors: Kyu Young Chae; Daniel F Kripke; J Steven Poceta; Farhad Shadan; Shazia M Jamil; John W Cronin; Lawrence E Kline Journal: Sleep Med Date: 2008-12-21 Impact factor: 3.492
Authors: Kelly Glazer Baron; Jennifer Duffecy; Mark A Berendsen; Ivy Cheung Mason; Emily G Lattie; Natalie C Manalo Journal: Sleep Med Rev Date: 2017-12-20 Impact factor: 11.609
Authors: Madeleine J George; Joshua G Rivenbark; Michael A Russell; Leonard Ng'eno; Rick H Hoyle; Candice L Odgers Journal: J Res Adolesc Date: 2019-09
Authors: Christopher M Depner; Philip C Cheng; Jaime K Devine; Seema Khosla; Massimiliano de Zambotti; Rébecca Robillard; Andrew Vakulin; Sean P A Drummond Journal: Sleep Date: 2020-02-13 Impact factor: 5.849
Authors: Andrew G Kubala; Bethany Barone Gibbs; Daniel J Buysse; Sanjay R Patel; Martica H Hall; Christopher E Kline Journal: Behav Sleep Med Date: 2019-08-27 Impact factor: 2.964