Accuracy of a wrist-worn wearable device for monitoring heart rates in hospital inpatients: A prospective observational study
CCCF ePoster library. Kroll R. Nov 1, 2016; 150926; 47
Dr. Ryan Kroll
Dr. Ryan Kroll
Login now to access Regular content available to all registered users.

You may also access this content "anytime, anywhere" with the Free MULTILEARNING App for iOS and Android
Abstract
Rate & Comment (0)
#47

Topic: Retrospective or Prospective Cohort Study

Accuracy of a wrist-worn wearable device for monitoring heart rates in hospital inpatients: A prospective observational study


Kroll, Ryan1; Boyd, J. Gordon1,2; Howes, Daniel2,3; Maslove, David1,2

1
Department of Medicine, Queen’s University, Kingston, Canada; 2Department of Critical Care Medicine, Queen’s University, Kingston, Canada; 3Department of Emergency Medicine, Queen’s University, Kingston, Canada
 



Abstract:

Introduction: Demand for wearable devices is increasing, as interest in personal health tracking among consumers and industry continues to rise1. As the sensing capabilities of wearable devices improve, there is increasing interest in their medical applications. While the use of wearable devices in ambulatory settings is growing, the feasibility of applying this technology to hospital inpatients has not been extensively evaluated2,3. Early Warning Systems (EWS) have been shown to reduce inpatient mortality, cardiac arrests, and Intensive Care Unit (ICU) admissions through early detection of clinical instability4-7. While EWS structures differ by institution, heart rate derangements feature prominently in most algorithms4,8. Heart rate monitoring wearable devices may therefore be useful in hospitalized patients as a means of enhancing routine monitoring and detecting instability.

Objectives: To evaluate the accuracy of heart rate monitoring from a consumer-grade personal fitness tracker (PFT) among hospital inpatients.

Methods: We conducted a prospective observational study of 50 stable ICU patients in a tertiary care centre. Each participant completed 24 hours of heart rate monitoring using a consumer-grade wrist-worn PFT, the Fitbit Charge HRTM. Accuracy of heart rate recordings was compared to gold-standard measurements from continuous electrocardiography (cECG). Additionally, we evaluated the sensitivity and specificity of the PFT in the detection of tachycardia. The above metrics for pulse oximetry derived heart rates (SPO2.R) were also measured as positive controls. 

Results: On a per-patient basis, PFT-derived heart rate values were slightly lower than those derived from cECG. Bland-Altman analysis revealed an average bias of -1.14 beats per minute [bpm], with limits of agreement of 24 bpm. By comparison, SPO2.R recordings produced more accurate values (average bias of +0.15 bpm, limits of agreement of 13 bpm, P < .001 as compared to PFT). See Table 1 for a summary of the statistical analysis findings, and Figure 1 for pooled results. In two patients the PFTs were removed early. PFT device performance was significantly better in patients in sinus rhythm than in those who were not (average bias -0.99 bpm vs -5.02 bpm, P = 0.02). In detecting tachycardia (heart rate > 110), PFT devices had a sensitivity of 59% and specificity of 99%, compared with 89% and 99% respectively for SPO2.R readings.

Conclusions: PFT-derived heart rates were slightly lower than those derived from cECG, though not as accurate as SPO2.R-derived heart rates. While the PFT and SPO2.R were equally specific in detecting tachycardia, the PFT was considerably less sensitive. PFT performance was worse among patients not in sinus rhythm. While this may reflect the PFT’s inability to determine heart rate in states of pulse deficit, further study is needed to characterize this effect. Addressing critical illness in its earliest stages, as it manifests outside the ICU, is invaluable in ultimately improving patient outcomes; further clinical evaluation is indicated to see if PFTs can effectively augment EWS in hospitals to reduce inpatient mortality.


References:
1. Bietz MJ, Bloss CS, Calvert S, Godino JG, Gregory J, Claffey MP, et al. Opportunities and challenges in the use of personal health data for health research. J Am Med Inform Assoc 2015 Sep 2;:ocv118–8. 

2. Petersen C. Patient-generated health data: a pathway to enhanced long-term cancer survivorship. J Am Med Inform Assoc 2015 Dec 29;:ocv184–6. 

3. Chuang C-C, Ye J-J, Lin W-C, Lee K-T, Tai Y-T. Photoplethysmography variability as an alternative approach to obtain heart rate variability information in chronic pain patient. J Clin Monit Comput 2015 Dec; 29(6):801–806.

4. Smith MEB, Chiovaro JC, O'Neil M, Kansagara D, Quiñones AR, Freeman M, et al. Early warning system scores for clinical deterioration in hospitalized patients: a systematic review. Ann Am Thorac Soc 2014 Nov; 11(9):1454–1465. 

5. Mitchell IA, McKay H, Van Leuvan C, Berry R, McCutcheon C, Avard B, et al. A prospective controlled trial of the effect of a multi-faceted intervention on early recognition and intervention in deteriorating hospital patients. Resuscitation 2010; 81(6):658–666.

6. Subbe CP, Kruger M, Rutherford P, Gemmel L. Validation of a modified Early Warning Score in medical admissions. QJM 2001; 94(10):521–526. 

7. Moon A, Cosgrove JF, Lea D, et al.: An eight year audit before and after the introduction of modified early warning score (MEWS) charts, of patients admitted to a tertiary referral intensive care unit after CPR. Resuscitation 2011; 82:150–154.

8. Churpek MM, Yuen TC, Winslow C, Meltzer DO, Kattan MW, Edelson DP. Multicenter Comparison of Machine Learning Methods and Conventional Regression for Predicting Clinical Deterioration on the Wards. Critical Care Medicine 2016 Feb; 44(2):368–374. 

Document: File 1 File 2
    This eLearning portal is powered by:
    This eLearning portal is powered by MULTIEPORTAL
Anonymous User Privacy Preferences

Strictly Necessary Cookies (Always Active)

MULTILEARNING platforms and tools hereinafter referred as “MLG SOFTWARE” are provided to you as pure educational platforms/services requiring cookies to operate. In the case of the MLG SOFTWARE, cookies are essential for the Platform to function properly for the provision of education. If these cookies are disabled, a large subset of the functionality provided by the Platform will either be unavailable or cease to work as expected. The MLG SOFTWARE do not capture non-essential activities such as menu items and listings you click on or pages viewed.


Performance Cookies

Performance cookies are used to analyse how visitors use a website in order to provide a better user experience.


Save Settings