Stress Watch: The Use of Heart Rate and Heart Rate Variability to Detect Stress: A Pilot Study Using Smart Watch Wearables

Taryn Chalmers, Blake Anthony Hickey, Phillip J. Newton, Chin‐Teng Lin, David Sibbritt, Craig S. McLachlan, Roderick Clifton‐Bligh, John W. Morley, Lal SSensors 2021 · DOI

TL;DR

This pilot study explores the potential of using smartwatches to detect stress by analyzing heart rate (HR) and heart rate variability (HRV). While not finding universal stress patterns, the research highlights the importance of considering individual baseline stress and anxiety levels for accurate stress detection, paving the way for more personalized wearable stress monitoring.

What Problem Does This Paper Address?

Stress is a ubiquitous human experience, but chronic or inappropriate stress responses can lead to significant health issues, including mental health decline. Early intervention is crucial, and real-time monitoring of stress could provide individuals with immediate biofeedback for self-management. However, a robust physiological algorithm for stress detection, particularly one that accounts for individual differences, is lacking for wearable technologies. This paper aims to address this by investigating if HRV changes can predict stress response quality and if baseline stress levels influence these changes, contributing to the development of such an algorithm.

Methods

The study employed a cross-sectional design with two cohorts: 30 medical students and 30 participants from the general population. Participants underwent a controlled laboratory experiment involving the Trier Social Stress Test (TSST) to induce acute stress.

  • Physiological Data Collection:
    • ECG: 3-lead electrocardiogram (ECG) was used to record heart electrical activity.
  • HRV Analysis: Raw ECG data was processed using Kubios HRV Premium software to extract frequency domain HRV parameters: Very Low Frequency (VLF), Low Frequency (LF), High Frequency (HF), and Total Power (TP), along with the LF:HF ratio. Data was collected over 5-minute intervals and averaged to ensure stationarity for frequency domain analysis.
  • Heart Rate (HR): Second-by-second HR data was recorded using a FitBit Versa 2 wearable device.
  • Blood Pressure (BP): Automated blood pressure measurements were taken before and after the stress task.
  • Psychometric Data Collection:
    • Lifestyle Questionnaire: Screened for factors that could influence HRV.
  • General Health Questionnaire (GHQ): Assessed general psychological wellbeing.
  • Depression Anxiety Stress Scale (DASS): Measured symptom severity of stress, depression, and anxiety.
  • Statistical Analysis:
    • Descriptive statistics were used to characterize the participants.
  • Paired t-tests were employed to compare HRV parameters between resting and stress phases.
  • A Holm-Bonferroni correction was applied to account for multiple comparisons.
  • Partial Pearson's correlation analysis was used to examine associations between HRV parameters, HR, and DASS scores, controlling for age and BMI.

Datasets

The study did not introduce a new benchmark dataset. Instead, it collected physiological and psychometric data from 60 human participants.

Key Insights and Ideas

  • Individual Baseline Matters: The most significant takeaway is that baseline stress and anxiety levels are crucial for accurately interpreting stress-induced HRV changes. Simply looking at HRV shifts during stress without considering an individual's pre-existing state can lead to misinterpretations.
  • Medical Students vs. General Population: Interestingly, the general population group reported significantly higher scores on stress, depression, and anxiety subscales of the DASS compared to medical students. This contrasts with common literature suggesting medical students experience higher stress, hinting at potential differences in self-regulation or reporting.
  • HRV and Stress - A Complex Relationship: The study found that while both groups showed increases in LF and HF during the stress task, the specific patterns and significance varied between the medical student and general population cohorts. This suggests that a "one-size-fits-all" approach to HRV-based stress detection might be insufficient.
  • FitBit for HR, ECG for HRV: The study utilized a FitBit for HR data, acknowledging that HRV data is not yet commercially available from such devices. This highlights the current reliance on more clinical-grade ECG for detailed HRV analysis, while consumer wearables are primarily used for HR monitoring.

Key Results

  • Heart Rate: Both resting and stress phase HR were significantly higher in the medical student population compared to the general population. Both groups showed a significant increase in average heart rate during the stress task.
  • Heart Rate Variability:
    • General Population: Showed significant increases in LF, HF, and the LF:HF ratio during the stress task.
  • Medical Students: Showed significant increases in VLF, LF, and HF during the stress task.
  • Common Finding: Both groups exhibited an increase in LF and HF during the stress task.
  • Psychometric Data: The general population reported significantly higher stress, depression, and anxiety scores than medical students.
  • Correlations: Anxiety (DASS subscale) was significantly correlated with baseline VLF and, during the stress task, with LF and the LF:HF ratio.

Limitations and Future Work

  • Small Cohort Size: The study acknowledges that its small cohort size (n=60) limits the generalizability of the findings.
  • Environment Bias: The laboratory setting might have introduced a bias, potentially benefiting the medical student group who are more familiar with clinical environments, leading to lower perceived stress.
  • Ubiquitous Patterns Not Found: The study failed to identify universal patterns of HRV and HR changes during stress, indicating the need for personalized approaches.

Future Work: * Larger cohort studies are needed to validate these findings and improve generalizability. * Further research with stratified anxiety scores based on DASS is recommended to better evaluate correlations with HRV. * Development of more sophisticated algorithms that incorporate baseline individual characteristics (anxiety, stress levels) is crucial for accurate stress detection using wearables. * The authors suggest that the utilization of HRV as a stress measure should always consider resting (baseline) anxiety and stress states for accurate assessment of additive acute stress effects.

Summary updated

Abstract

Stress is an inherent part of the normal human experience. Although, for the most part, this stress response is advantageous, chronic, heightened, or inappropriate stress responses can have deleterious effects on the human body. It has been suggested that individuals who experience repeated or prolonged stress exhibit blunted biological stress responses when compared to the general population. Thus, when assessing whether a ubiquitous stress response exists, it is important to stratify based on resting levels in the absence of stress. Research has shown that stress that causes symptomatic responses requires early intervention in order to mitigate possible associated mental health decline and personal risks. Given this, real-time monitoring of stress may provide immediate biofeedback to the individual and allow for early self-intervention. This study aimed to determine if the change in heart rate variability could predict, in two different cohorts, the quality of response to acute stress when exposed to an acute stressor and, in turn, contribute to the development of a physiological algorithm for stress which could be utilized in future smartwatch technologies. This study also aimed to assess whether baseline stress levels may affect the changes seen in heart rate variability at baseline and following stress tasks. A total of 30 student doctor participants and 30 participants from the general population were recruited for the study. The Trier Stress Test was utilized to induce stress, with resting and stress phase ECGs recorded, as well as inter-second heart rate (recorded using a FitBit). Although the present study failed to identify ubiquitous patterns of HRV and HR changes during stress, it did identify novel changes in these parameters between resting and stress states. This study has shown that the utilization of HRV as a measure of stress should be calculated with consideration of resting (baseline) anxiety and stress states in order to ensure an accurate measure of the effects of additive acute stress.
BibTeX
@inproceedings{chalmers2021stress,
  author = {Taryn Chalmers and Blake Anthony Hickey and Phillip J. Newton and Chin‐Teng Lin and David Sibbritt and Craig S. McLachlan and Roderick Clifton‐Bligh and John W. Morley and Lal S},
  title = {{Stress Watch: The Use of Heart Rate and Heart Rate Variability to Detect Stress: A Pilot Study Using Smart Watch Wearables}},
  booktitle = {Sensors},
  year = {2021},
  doi = {10.3390/s22010151},
}
Figure