Conference PaperPDF Available

Dependence of Heart Rate Variability on Stress Factors of Stress Response Inventory

Authors:

Abstract

Heart rate variability (HRV) analysis is commonly used as a quantitative marker depicting the activity of autonomic nervous system (ANS) related to mental stress. Stress response inventory (SRI) has been devised to score mental and physical symptoms occurred during the past two weeks. SRI is composed of seven stress factors that may influence the status of mental stress levels. In this study, we investigated the relationships of physiological measures and HRV features on ages and stress factors. Physiological measures and HRV features in low (SRI scores: 7.1 plusmn 4.1, n=225) and high stress group (22.5 plusmn 7.4, n=135) were compared with age as the covariate (ANCOVA). Age was reconfirmed as a significant factor influencing physiological measures and most of HRV features. Age was also inversely correlated to stress factor scores. Systolic blood pressure, glucose level, and normalized HF were significantly lower, whereas body temperature, LF/HF, and normalized LF were significantly higher in high stress group. Our results showed that stress levels were associated with ages, physiological measures, and HRV features.
TABLE I. A
GES AND
G
ENDERS OF
P
ARTICIPANTS
Gender
Age
Female Male
Total
20-29 72
33
105
30-39 43
80
123
40-49 45
45
90
50-69 22
20
42
Total
182
178
360
Dependence of Heart Rate Variability on Stress
Factors of Stress Response Inventory
Lizawati Salahuddin, Myeong Gi Jeong, Desok Kim
School of Engineering
Information and Communication University
Daejeon, Korea
lizawati, mgjeong, kimdesok@icu.ac.kr
Seong-Kyeon Lim
1
, Kim Won
2
, Jong-Min Woo
2
1
Department of Neuropsychiatry and
2
Stress Research
Institute, Seoul Paik Hospital
Inje University School of Medicine
Seoul, Korea
healthpsy@naver.comphrenie@dreamwiz.com,
menfi@naver.com
Abstract—Heart rate variability (HRV) analysis is commonly
used as a quantitative marker depicting the activity of autonomic
nervous system (ANS) related to mental stress. Stress Response
Inventory (SRI) has been devised to score mental and physical
symptoms occurred during the past two weeks. SRI is composed
of seven stress factors that may influence the status of mental
stress levels. In this study, we investigated the relationships of
physiological measures and HRV features on ages and stress
factors. Physiological measures and HRV features in low (SRI
scores: 7.1 ± 4.1, n=225) and high stress group (22.5 ± 7.4, n=135)
were compared with age as the covariate (ANCOVA). Age was
reconfirmed as a significant factor influencing physiological
measures and most of HRV features. Age was also inversely
correlated to stress factor scores. Systolic blood pressure, glucose
level, and normalized HF were significantly lower, whereas body
temperature, LF/HF, and normalized LF were significantly
higher in high stress group. Our results showed that stress levels
were associated with ages, physiological measures, and HRV
features.
Keywords-Heart rate variability; autonomic nervous system;
physiological measure; stress; Stress Response Inventory
I.
I
NTRODUCTION
Stress involves alteration in behavior, autonomic function
and the secretion of several hormones such as cortisol,
corticosterone, and adrenal catecholamines [1]. Higher blood
pressure and heart rates during stress reflected the
predominance of sympathetic nervous system activity [2].
Mental stress decreased high frequency of heart rate variability
(HRV) and increased low frequency of HRV [3]. HRV
decreased in subjects with depression, higher hostility and
anxiety [4]. Stress increases susceptibility to negative health
outcomes [5]. Autonomic nervous system (ANS) contributes
to the physiological adaptive process in short durations but
may be damaged if the releases of the mediators such as
adrenaline from the adrenal medulla are not terminated when
they are no longer needed.
Numerous stress questionnaires have been used in clinical
practice and psychiatric researches such as Perceived Stress
Scale (PSS) [6], Life Events and Coping Inventory (LECI) [7],
and Stress Response Inventory (SRI) [8]. PSS measures the
degree to which situations in one’s life are considered as
stressful. A number of questions may be included to scale
feelings, thoughts, and current levels of experienced stress
during the last month [6]. LECI is composed of 125 life event
questions that assess the experience of life stress and 42
questions that ask the use of coping behaviors in 12-14 years
old children [7]. Recently, SRI questionnaire has been devised
to score mental and physical symptoms occurred during the
past two weeks that may influence the status of mental stress
levels. SRI consists of 39 items that focus on the emotional,
somatic, cognitive, and behavioral stress responses. SRI scores
could be categorized into seven stress factors: tension,
aggression, somatization, anger, depression, fatigue, and
frustration [8]. Both PSS and SRI were designed to measure
stress severity in adults. PSS was designed to assess how
unpredictable, uncontrollable, and overloaded respondents find
their lives. Unlike PSS, SRI assesses the stress severity based
on the stress symptoms or the effects of stressors. However,
the relationships of ages, stress factors, physiological measures
and HRV features have not been investigated thoroughly. In
this study, we compared physiological and HRV features in
subjects with high and low stress factors to investigate stress-
related symptoms and their influence on HRV features.
II. M
ETHODS
A. Subjects and Data Acquisition during Baseline Stage
The experiment was carried out in Department of
Neuropsychiatry and Stress Research Institute, Seoul Paik
Hospital, Seoul, Korea and Information and Communications
This research was supported by the grant from Samsung Electronics, Inc.,
Suwon, Korea.
TABLE II. A
S
IMPLIFIED
S
TRESS
R
ESPONSE
I
NVENTORY
I
TEMS
C
ATEGORIZED INTO
S
EVEN
S
TRESS
F
ACTORS
Stress Factor
Questions
My body trembles.
I feel tense.
Tension
My head hurts or it feels heavy.
Aggression
I act violently (such as reckless driving, cursing,
fighting).
I suffer from indigestion.
My stomach hurts.
Somatization
I feel dizzy.
My voice is louder than it usually is.
Anger
I easily get impatient.
I have lost my self-confident.
I have lost incentive to do anything.
I have no future in my current work.
I often stare blankly.
I feel bored.
I am useless (or unworthy).
Depression
I don’t like moving any part of my body.
I am easily fatigued.
Fatigue
I feel totally exhausted.
My chest feels tight.
Everything bothers me.
I feel on edge.
Frustration
My heart throbs.
University, Daejeon, Korea. This study was approved by the
Institutional Review Board (IRB) and all subjects signed the
consent form that allowed the use of data. All subjects visited
the clinic or laboratory for physical check up. Subjects who
had psychopathic treatment history were excluded from the
experiment. Four hundred two normal subjects aged from 21
to 69 years (37.3 ± 10.7) participated in the study originally
(Table I).
A simplified version of original SRI questionnaire was
devised by one of the authors (JMW) and used in this study.
The simplified version of SRI questionnaire was composed of
22 questions (Table II) that have been categorized into seven
stress factors as in the original SRI [8]. Each question was
scored in a five-point Likert scale format: ‘Not at all’ (0),
‘Somewhat’ (1), ‘Moderately’ (2), ‘Very much (3), or
‘Absolutely(4). SRI questionnaires were filled up before the
physiological and heartbeat measurements.
Height, body fat, body temperature (at the forehead), blood
pressure and blood glucose levels were measured. Subjects
were seated in the comfortable chair and rested for five minutes
prior to the heartbeat measurement. Three minute records of
heartbeat were then recorded right after the resting stage.
Finger photoplethysmography sensor (Freeze-Framer®,
Institute of HeartMath LLC, Boulder Creek, CA) was used to
produce heartbeat (R peak) interval records.
B. Data Processing and Feature Calculation
Thirty eight R peak interval records containing motion
artifacts were discarded from further analysis. Four records
were removed from the analysis due to the lack of blood
pressure and blood glucose data. This reduced the usable
records to 360 (Table I). In each record, the following HRV
metrics were calculated: mean heart rate (Mean HR), mean R
peak intervals (Mean RR), standard deviation of R peak
intervals (SDNN), root mean square of standard deviation
(RMSSD), and % R peak intervals with difference in
successive R peak intervals greater than 50 ms (PNN50) as
time domain features; HRV index (HRV Index, bin width of
8.0 ms), triangular interpolation of R peak intervals histogram
(TINN), and stress index (SI) [9] as geometrical analysis
features; and very low frequency (VLF), low frequency (LF),
high frequency (HF), the ratio of LF to HF (LF/HF),
normalized LF (LFnu), and normalized HF (HFnu) as
frequency domain features [10].
C. Data Analysis
To assess the association with stress factors, individual SRI
scores were grouped into their corresponding stress factors
(Table II) to calculate stress factor scores. Dependence on ages
of stress factor scores as well as physiological measures and
HRV features was evaluated using Pearson’s correlation
analysis (StatGraphics Plus V4.1, Manugistics, Inc., Rockville,
MD). Dependence on stress factors of physiological measures
and HRV features was evaluated using multiple regression
analysis (StatGraphics Plus V4.1).
The subjects were divided into low and high stress group
using k-means cluster analysis (based on square Euclidean)
with stress factor scores as the variables (StatGraphics Plus
V4.1). Physiological measures and HRV features in these two
groups were compared, with age as the covariate using analysis
of covariance (ANCOVA) (Minitab V15, Minitab Inc., State
College, PA).
III. R
ESULTS
A. Relationships of Physiological Measures, HRV Features,
and Stress Factors with Ages
Table
III summarizes the relationships of physiological
measures, HRV features, and stress factors with ages. Body fat
content, blood pressures, blood glucose level, mean RR, and SI
were positively correlated, whereas body temperature, mean
HR, SDNN, CV, RMSSD, PNN50, HRV Index, TINN, VLF,
LF, and HF were inversely correlated with ages (Pearson’s
correlation, p < 0.05). All stress factors scores were negatively
correlated with ages (p < 0.001).
Tension, depression and frustration were the stress factors
that were frequently associated with body fat, body
temperature, and HRV features (data not shown). Depression
factors were positively associated with body fat, body
temperature, SI, and HFnu; negatively associated with CV,
VLF, LF, LF/HF, and LFnu (multiple regression, p < 0.05).
Tension factors were positively associated with LF, LF/HF,
and LFnu; negatively associated with body fat, SDNN, and
HFnu (multiple regression, p < 0.05). Frustration factors were
positively associated with SI, LF/HF, and LFnu; negatively
associated with RMSSD, HRV Index, TINN, HF, and HFnu.
None of the individual SRI scores were correlated with
diastolic blood pressure, mean HR, and mean RR (multiple
regression, p > 0.05).
Cluster analysis classified 225 subjects as low stress factor
(total SRI scores: 7.1 ± 4.1) and 135 as high stress factor group
(22.5 ± 7.4).
TABLE III. S
TATISTICALLY
S
IGNIFICANT
F
EATURES THAT
D
ISTINGUISH
T
WO
G
ROUPS WITH
L
OW VERSUS
H
IGH
S
TRESS
F
ACTORS
(ANCOVA
WITH AGE AS COVARIATE
,
P
<0.05)
Measures
Subject with Low
Stress Factor
Scores (n=225)
Subjects with High
Stress Factor
Scores (n=135)
Body Temperature 36.36 ± 0.352
36.43 ± 0.629
Systolic blood pressure
120.0 ± 14.12
115.5 ± 13.39
Glucose level 98.38 ± 17.20
91.70 ± 12.10
LF/HF 1.751 ± 1.715
1.862 ± 1.526
LFnu 53.43 ± 18.74
56.94 ± 16.94
HFnu 45.82 ± 18.93
42.31 ± 17.15
TABLE IV. R
ELATIONSHIPS
OF
P
HYSIOLOGICAL
M
EASURES
,
HRV
F
EATURES
,
AND
S
TRESS
F
ACTORS WITH
A
GES
(S
IMPLE REGRESSION
,
N
=360)
Measurements/Stress Factors
Correlation with age (ρ
ρρ
ρ)
Body fat 0.255**
Body temperature -0.304**
Systolic blood pressure 0.322**
Diastolic blood pressure 0.244**
Glucose level 0.392**
Mean HR -0.138**
Mean RR 0.139**
SDNN -0.332**
CV -0.432**
RMSSD -0.241**
PNN50 -0.253**
HRV Index -0.402**
TINN -0.398**
SI 0.421**
VLF -0.233**
LF -0.270**
HF -0.155**
LF/HF -0.023**
LFnu -0.069**
HFnu 0.063**
Tension -0.236**
Aggression -0.234**
Somatization -0.190**
Anger -0.232**
Depression -0.259**
Fatigue -0.245**
Frustration -0.246**
*p < 0.05
**p < 0.005
B. Physiological Measures and HRV Features in Subjects
with Low and High Stress Factors Group
Using ANCOVA with age as covariate, several
physiological measures and HRV features were found to be
significantly different in the low and high stress factor group
(Table IV). Systolic blood pressure, glucose level, and HFnu
were significantly lower, whereas body temperature, LF/HF,
and LFnu were significantly higher in high stress factor group.
IV. C
ONCLUSIONS AND
D
ISCUSSION
Previous studies have found that HRV declines with ages
[11, 12]. In the present study, age was newly found to be
correlated with geometrical features such as HRV index, TINN,
and SI at moderate levels (-0.398 ~ 0.421) (Table III). In
addition, all physiological measures were found to be
dependent on ages although at low levels in our subjects
(0.244~0.392). Normalized HRV features such as LF/HF,
LFnu, and HFnu did not show significant dependence on ages
(Table III).
Physiological measures and HRV features were correlated
with the stress factor scores in the SRI questionnaire (data not
shown). In brief, tension, depression and frustration were the
main stress factors associated with HRV features. Tension and
frustration factors were positively associated with the
sympathetic activity (LF and LFnu) and negatively associated
with parasympathetic activity (HFnu). Conversely, depression
factors were negatively associated with sympathetic activity
and positively with parasympathetic activity. During the
experience of negative emotions such as anger, frustration, or
anxiety, heart rhythms are known to become disordered,
indicating less synchronization in the reciprocal action between
the parasympathetic and sympathetic branches of the ANS [13].
Since age was a strong factor influencing HRV features,
we sough to rule out age-dependent features and seek out the
relationship between stress factors and HRV features. Cluster
analysis using stress factor scores was useful to identify high
stress group in our case. High stress group showed higher
LF/HF (reflects the predominance of sympathetic over
parasympathetic activity) and LFnu (mainly influenced by
sympathetic activity), whereas lower HFnu (mainly influenced
by parasympathetic activity) compared to the low stress factors
group (Table IV). Significant association of stress factors with
HRV features suggested that the questionnaire items in our
simplified version of SRI are useful to classify subjects into
high and low stress group. In addition, our results indicated
that further investigation is warranted for stress factors and
their relationships with body temperature systolic blood
pressure, and blood glucose level.
A
CKNOWLEDGMENT
We would like to thank all participants and students at ICU
who participated in the experiment. All correspondence should
be addressed to Desok Kim, PhD, School of Engineering,
Information and Communications University, 103-6, Munji
Dong, Yuseong Gu, Daejeon, 305-732, Korea; phone: 82-42-
866-6156; fax: 82-42-866-6110 and Jong-Min Woo, MD, PhD,
Inje University Paik Hospital, Sanggye 7 Dong, Nowon Gu,
Seoul, Korea 139-707; phone: 82-2-2270-0063; fax: 82-2-
2270-0334.
R
EFERENCES
[1] L.D. Van der Kar and M.L. Blair, “Forebrain pathways mediating stress
induced hormone secretion,” Frontiers in Neuroendocrinology, 1999, vol.
20, pp. 41–48.
[2] T. Ritvanen, V. Louhevaara, P. Helin, S. Vaisanen, and O. Hanninen,
“Responses of the autonomic nervous system during periods of
perceived high and low work stress in younger and older female
teachers,” Applied Ergonomics, June 2005, vol. 37, pp. 311-318.
[3] L. Bernardi, J. Wdowczyk-Szulc, C. Valenti, S. Castoldi, C. Passino, G.
Spadacini, and P. Sleight, “Effects of controlled breathing, mental
activity, and mental stress with or without verbalization on heart rate
variability,”Journal of the American College of Cardiology, May
2000,vol. 35, no. 6, pp. 1462-1469.
[4] R.P. Sloan, P.A. Shapiro, E. Bagiella, M.M. Myers, and J.M. Gorman,
“Cardiac autonomic control buffers blood pressure variability responses
to challenge: A psychophysiologic model of coronary artery disease,”
Psychosomatic Medicine, 1999, vol. 61, pp. 58-68.
[5] B.S. McEwen, “Allostasis and allostatic load: Implications for
neuropsychopharmacology,” Neuropsychopharmacology, 2000, vol. 22,
pp. 108-124.
[6] S. Cohen, T. Kamarck, and R. Mermelstein, “A global measure of
perceived stress,” Journal of Health and Social Behavior, 1983, vol. 24,
pp. 386-396.
[7] J.E. Dise-Lewis, “The life events and coping inventory: An assessment
of stress in children,” Psychosomatic Medicine, 1988, vol. 50, no. 5, pp.
484-499.
[8] K.B. Koh, J.K. Park, C.H. Kim, and S. Cho, “Development of the Stress
Response Inventory and its application in clinical practice,”
Psychosomatic Medicine, 2001, vol. 63, pp. 668-678.
[9] R.M. Bayevsky, et al.. (2002, April, 11) HRV Analysis under the usage
of different electrocardiography systems (Methodical recommendations).
Available:
http://www.hrvcongress.org/second/education/guidelines.html
[10] Task Force of the European Society of Cardiology and the North
American Society of Pacing and Electrophysiology, “Heart rate
variability: Standards of measurement, physiological interpretation, and
clinical use,” European Heart Journal, March 1996, vol. 17, pp. 1043-
1065.
[11] D. Vigo, S. Guinjoan, M. Scaramal, L. Nicola Siri, and D. Cardinali,
“Wavelet transform shows age-related changes of heart rate variability
within independent frequency components,” Autonomic
Neuroscience, 2005, vol. 123, no. 1-2, pp. 94-100.
[12] K. Umetani, D. H. Singer, M.D., R. McCraty, and M. Atkinson,
“Twenty-four hour time domain heart rate variability and heart rate:
Relations to age and gender over nine decades,” Journal of the American
College of Cardiology, 1998, vol. 31 no. 3, pp. 593-601.
[13] R. McCraty, and D. Tomasino, “Heart rhythm coherence feedback: A
New tool for stress reduction, rehabilitation, and performance
enhancement,” Proceedings of the First Baltic Forum on Neuronal
Regulation and Biofeedback, Riga, Latvia, November 2–5, 2004.
... Self-reported stress responses were assessed using Visual Analogue Scales (VAS) on which participants were asked to report their positive and negative affect on a scale from 'not at all' (1) to 'a lot' (100). As evidenced by Salahuddin et al. (2007) and Schlotz et al. (2008), self-reported positive and negative affect can serve as a subjective measure for assessing stress exposures. More specifically, participants reported in which degree they were experiencing feelings of tension, frustration, boredom, anxiety, sadness and angriness. ...
... Physiological stress responses were measured by heart rate (HR) and High Frequency (HF) HRV. HF-HRV reflects vagally-mediated parasympathetic activity and is a commonly used measure of HRV in the stress literature (see for example Salahuddin et al., 2007). HR and HF-HRV were assessed using ECG signals that were recorded at a sample rate of 1000 Hz with a Porti 16-channelamplifier (TMSi) and the software Polybench 1.2. ...
Article
Full-text available
Objectives Previous studies demonstrated that self-compassion may generate positive effects on adults’ mental health through its impact on stress responses. As adolescence is characterized by elevated levels of stress, self-compassion may be particularly relevant for this age group. The aim of this study was to assess the immediate effects of a brief training in self-compassion on adolescents’ stress recovery following a validated stress induction. Methods Fifty-three adolescents between 11 and 18 years old (64% girls) were randomly assigned to a self-compassion group or a control group prior to undergoing a three-phase experiment (i.e., baseline, stress induction, and instruction phase). Adolescents in the self-compassion group received a brief training in self-compassion before the start of the experiment and were asked to use the learned technique during the instruction phase. Adolescents in the control group did not receive a training and were provided with neutral instructions during the instruction phase. Physiological stress outcomes (i.e., salivary cortisol, heart rate, and heart rate variability) and self-reported stress outcomes (i.e., self-reported affect) were compared between groups. Results The main results revealed no clear differences between both groups pertaining physiological and self-reported stress responses. Conclusions The current findings could not provide evidence for the beneficial effects of a brief self-compassion training among adolescents, and even suggest that it may have detrimental effects on the physiological stress response. Findings are discussed within a developmental framework and important considerations for further research are noted.
... HF and LF on the other hand can be assessed on 1 to 2 min windows respectively [2]. Their ability to reflect the overall cardiac health and the state of the autonomic nervous system (ANS) has been proven by many studies [13,1], in different contexts including stress [14,15] and sleep [16]. ...
... and high frequency (HF) 0.15-0.4Hz bands and their ratio, i.e., LF/HF, are the validated metrics for interpreting stress [196][197][198] . The LF/HF ratio is interpreted as the ratio of the activity of the sympathetic nervous system to that of the parasympathetic nervous system (sympathovagal balance) 178 . ...
Thesis
Full-text available
Over 151 million people visit US Emergency Departments (EDs) annually. The diverse nature and overwhelming volume of patient visits make the ED one of the most complicated settings in healthcare to study. ED overcrowding is a recognized worldwide public health problem, and its negative impacts include patient safety concerns, increased patient length of stay, medical errors, patients left without being seen, ambulance diversions, and increased health system expenditure. Additionally, ED crowding has been identified as a leading contributor to patient morbidity and mortality. Furthermore, this chaotic working environment affects the well-being of all ED staff through increased frustration, workload, stress, and higher rates of burnout which has a direct impact on patient safety. This research takes a step-by-step approach to address these issues by first forecasting the daily and hourly patient arrivals, including their Emergency Severity Index (ESI) levels, to an ED utilizing time series forecasting models and machine learning models. Next, we developed an agent-based discrete event simulation model where both patients and physicians are modeled as unique agents for capturing activities representative of ED. Using this model, we develop various physician shift schedules, including restriction policies and overlapping policies, to improve patient safety and patient flow in the ED. Using the number of handoffs as the patient safety metric, which represents the number of patients transferred from one physician to another, patient time in the ED, and throughput for patient flow, we compare the new policies to the current practices. Additionally, using this model, we also compare the current patient assignment algorithm used by the partner ED to a novel approach where physicians determine patient assignment considering their workload, time remaining in their shift, etc. Further, to identify the optimal physician staffing required for the ED for any given hour of the day, we develop a Mixed Integer Linear Programming (MILP) model where the objective is to minimize the combined cost of physician staffing in the ED, patient waiting time, and handoffs. To develop operations schedules, we surveyed over 70 ED physicians and incorporated their feedback into the MILP model. After developing multiple weekly schedules, these were tested in the validated simulation model to evaluate their efficacy in improving patient safety and patient flow while accounting for the ED staffing budget. Finally, in the last phase, to comprehend the stress and burnout among attending and resident physicians working in the ED for the shift, we collected over 100 hours of physiological responses from 12 ED physicians along with subjective metrics on stress and burnout during ED shifts. We compared the physiological signals and subjective metrics to comprehend the difference between attending and resident physicians. Further, we developed machine learning models to detect the early onset of stress to assist physicians in decision-making.
... HF and LF, in contrast, can be assessed in shorter time periods (1 and 2 min windows, respectively) [1]. Their correlation with the autonomic nervous system (ANS) and the overall cardiac health has been proven by many studies [31,32] in different contexts, including stress [33]. ...
Article
Full-text available
Thanks to wearable devices joint with AI algorithms, it is possible to record and analyse physiological parameters such as heart rate variability (HRV) in ambulatory environments. The main downside to such setups is the bad quality of recorded data due to movement, noises, and data losses. These errors may considerably alter HRV analysis and should therefore be addressed beforehand, especially if used for medical diagnosis. One widely used method to handle such problems is interpolation, but this approach does not preserve the time dependence of the signal. In this study, we propose a new method for HRV processing including filtering and iterative data imputation using a Gaussian distribution. The particularity of the method is that many physiological aspects are taken into consideration, such as HRV distribution, RR variability, and normal boundaries, as well as time series characteristics. We study the effect of this method on classification using a random forest classifier (RF) and compare it to other data imputation methods including linear, shape-preserving piecewise cubic Hermite (pchip), and spline interpolation in a case study on stress. Features from reconstructed HRV signals of 67 healthy subjects using all four methods were analysed and separately classified by a random forest algorithm to detect stress against relaxation. The proposed method reached a stable F1 score of 61% even with a high percentage of missing data, whereas other interpolation methods reached approximately 54% F1 score for a low percentage of missing data, and the performance drops to about 44% when the percentage is increased. This suggests that our method gives better results for stress classification, especially on signals with a high percentage of missing data.
... The study concluded that the HR and HRV of the subject change when facing a mental stressor and hence can be used as a potential biomarker for the assessment of human stress. Another study about the association of mental stress with HRV is presented in (Salahuddin et al., 2007b). Correlation between the perceived stress face by college students and the HRV is explored in (Lombardo and Vick, 2019). ...
Preprint
Full-text available
This paper presents a comprehensive review of methods covering significant subjective and objective human stress detection techniques available in the literature. The methods for measuring human stress responses could include subjective questionnaires (developed by psychologists) and objective markers observed using data from wearable and non-wearable sensors. In particular, wearable sensor-based methods commonly use data from electroencephalography, electrocardiogram, galvanic skin response, electromyography, electrodermal activity, heart rate, heart rate variability, and photoplethysmography both individually and in multimodal fusion strategies. Whereas, methods based on non-wearable sensors include strategies such as analyzing pupil dilation and speech, smartphone data, eye movement, body posture, and thermal imaging. Whenever a stressful situation is encountered by an individual, physiological, physical, or behavioral changes are induced which help in coping with the challenge at hand. A wide range of studies has attempted to establish a relationship between these stressful situations and the response of human beings by using different kinds of psychological, physiological, physical, and behavioral measures. Inspired by the lack of availability of a definitive verdict about the relationship of human stress with these different kinds of markers, a detailed survey about human stress detection methods is conducted in this paper. In particular, we explore how stress detection methods can benefit from artificial intelligence utilizing relevant data from various sources. This review will prove to be a reference document that would provide guidelines for future research enabling effective detection of human stress conditions.
... The approaches to stress detection can be roughly classified into: 1) those performed in the ambulatory setting during a relatively short period of time, and 2) those that are performed during the long term when the participant continue his/her normal life activities. The signals reflecting the ANS activity, can be divided into physiological, such as, for example, electro-dermal activity (EDA) [4], heart rate (HR), heart rate variability (HRV) [15], and levels of cortisol [13], [20], and behavioural, such as smartphone activity statistics [16], and annotated geolocation. It is clear that experiment settings define the set of signals that must be considered as more reliable for that experiment. ...
Chapter
Full-text available
The problem of stress detection and classification has attracted a lot of attention in the past decade. It has been tackled with mainly two different approaches, where signals were either collected in ambulatory settings, which can be limited to the period of presence in the hospital, or in continuous mode in the field. A sensor-based continuous measurement of stress in daily life has a potential to increase awareness of patterns of stress occurrence. In this work, we first present a data-flow infrastructure suitable for two types of studies that conforms with the data protection requirements of the ethics committee monitoring the research on humans. The detection and binary classification of stress events is compared with three different machine learning models based on the features (meta-data) extracted from physiological signals acquired in laboratory conditions and ground-truth stress level information provided by the subjects themselves via questionnaires associated with these features. The main signals considered in current classification are electro-dermal activity (EDA) and blood volume pulse (BVP) signals. Different models are compared and the best configuration yields an F1F_1 score of 0.71 (random baseline: 0.48). The importance on prediction of phasic and tonic EDA components is also investigated. Our results also pave the way for further work on this topic with both machine learning approaches and signal processing directions.
... Hz bands and their ratio i.e. LF/HF, are the validated metrics for interpreting stress (Hall et al., 2004;Salahuddin et al., 2007;Sloan et al., 1994). The LF/HF ratio is interpreted as the ratio of the activity of the sympathetic nervous system to that of the parasympathetic nervous system (sympathovagal balance) (Malik et al., 1996). ...
Article
Emergency department physicians work in complex care settings with frequent exposure to stressful conditions. Burnout in physicians is increasing each year, and one of the most prone groups are emergency medicine providers. This research sought to understand if a difference in stress levels and burnout rate exists between attending and resident physicians working in an academic Level 1 trauma center emergency department on the same shift. Twelve attending and resident physicians from Greenville Memorial Hospital in Greenville, SC, participated in the study. Stress levels were estimated using physiological measures, including heart rate variability and electrodermal activity. Burnout scores and workload index were collected using the Maslach Burnout Inventory-Human Services Survey, and the NASA-TLX survey. Over 100 hours of physiological data were collected, and 42 events were compared across the two physician groups. Attending physicians showed a higher heart rate variability for the entire shift, RMSSD (44.2 vs. 35.4, p = .033) and during the trauma events, RMSSD (47.0 vs. 35.2, p = .001), LF/HF ratio (1.7 vs. 2.5, p = .001), implying lower stress levels. Furthermore, attending physicians recorded a lower workload index (35.1 vs. 49.2, p=.004). However, no significant differences were observed in the burnout scores.
Article
Full-text available
Background: Mental workload is one of the contributing factors to human errors in road accidents or other potentially adverse incidents. Objective: This research probes the effects of mental workload on the electroencephalographic (EEG) and electrocardiogram (ECG) of subjects in visual monitoring tasks, based on which a comprehensive evaluation model for mental workload is established effectively. Methods: Three degrees of mental workload were obtained by monitoring tasks with different levels of difficulty. 20 healthy subjects were selected to take part in the research. Results: The subjective scores showed a significant increase with the increase of task difficulty, meanwhile the reaction time (RT) increased and the accuracy decreased significantly, which proved the validity of three degrees of mental workload induced. For the EEG parameters, a significant decrease of θ energy was found in Frontal, Parietal and Occipital with the increase of level of mental workload, as well as a significant decrease of α energy in Frontal, Central and Occipital, meanwhile a significant increase of β energy occurred in Frontal and Occipital. There was a significant decrease of α/θ in Occipital, and significant increases of θ/β and (α+β)/θ in Frontal, Central and Occipital, meanwhile (α+θ)/β and WPE decreased significantly in Frontal and Occipital. Among the ECG parameters, it was shown that Mean RR, RMSSD, HF_norm and SampEn decreased significantly with the increase of task difficulty, while LF_norm and LF/HF showed significant increases. These EEG indictors in Occipital and ECG indictors were chosen and constituted a multidimensional original sample. Principal Component Analysis (PCA) was used to extract the principal elements and decreased the dimension of sample space in order to simplify the calculation, based on which an effective classification model with accuracy of 80% was achieved by support vector machine (SVM). Conclusion: This study demonstrates that the proposed algorithm can be applied to mental workload monitoring.
Article
Full-text available
This study sought to define the effects of age and gender effects on the normal range of time domain heart rate variability (HRV) over nine decades in healthy subjects. Low HRV is considered an independent marker of mortality risk. However, the age-related decline in HRV may limit its predictive value, particularly in the elderly. Delineation of the range of HRV in healthy subjects over the life span is needed. Gender-related differences in HRV also need clarification. We determined, according to decade, 24-h heart rate (HR) and HRV of 260 healthy subjects (10 to 99 years old; 112 male, 148 female) by means of five standard time domain measures: standard deviation of all normal sinus RR intervals over 24 h (SDNN), standard deviation of the averaged normal sinus RR intervals for all 5-mm segments (SDANN), mean of the standard deviations of all normal sinus RR intervals for all 5-min segments (SDNN index), root-mean-square of successive normal sinus RR interval difference (rMSSD) and the percentage of successive normal sinus RR intervals >50 ms (pNN50). 1) HRV decreased with aging, the pattern of change being measure dependent. HRV (SDNN and SDANN) decreased only very gradually, reaching 60% of baseline (second-decade values) by the tenth decade. With the SDNN index, HRV decreased linearly with aging, reaching 46% of baseline by the tenth decade. Using pNN50 and rMSSD, HRV decreased most rapidly, reaching 24% and 47% of baseline, respectively, by the sixth decade and then stabilized. 2) Using the SDNN index, rMSSD and pNN50, HRV of subjects >65 years old fell below published cutpoints for increased risk of mortality in 25%, 12% and 4%, respectively. 3) At age <30 years, HRV for all measures was lower in female than male subjects. Gender differences decreased at age >30 years and disappeared at age >50 years. 4) HR also declined with aging but much more slowly. HR at age <50 years was faster in female than in male subjects. Gender differences disappeared thereafter. 1) Using all measures, HRV of healthy subjects declines with aging, with measure-dependent patterns. 2) Using the SDNN index, rMSSD and pNN50, HRV of healthy subjects, particularly those >65 years old, may decrease to below levels associated with increased risk of mortality. 3) Gender influences HRV. Gender differences in HRV are age and measure dependent. 4) Age and gender also affect heart rate.
Article
This article reports the development of the Life Events and Coping Inventory (LECI), an instrument that assesses the experience of life stress and the use of coping behaviors in 12-14 year old children. Six hundred eighty-one child Ss participated in item generation, item rating, reliability procedures, and validity procedures for the LECI. Indices of life stress were correlated at a statistically significant level (p less than 0.01) with measures of state and trait anxiety, psychosomatic symptoms, depression and a variety of behavior problems. A broad range of coping strategies was evidenced, with strategies that provide distraction from the stress event the most useful. The results are discussed in light of current theoretic literature on stress and well-being.
Article
This paper presents evidence from three samples, two of college students and one of participants in a community smoking-cessation program, for the reliability and validity of a 14-item instrument, the Perceived Stress Scale (PSS), designed to measure the degree to which situations in one's life are appraised as stressful. The PSS showed adequate reliability and, as predicted, was correlated with life-event scores, depressive and physical symptomatology, utilization of health services, social anxiety, and smoking-reduction maintenance. In all comparisons, the PSS was a better predictor of the outcome in question than were life-event scores. When compared to a depressive symptomatology scale, the PSS was found to measure a different and independently predictive construct. Additional data indicate adequate reliability and validity of a four-item version of the PSS for telephone interviews. The PSS is suggested for examining the role of nonspecific appraised stress in the etiology of disease and behavioral disorders and as an outcome measure of experienced levels of stress.
Article
Exposure to hostile conditions initiates the secretion of several hormones, including corticosterone/cortisol, catecholamines, prolactin, oxytocin, and renin, as part of the survival mechanism. Such conditions are often referred to as "stressors" and can be divided into three categories: external conditions resulting in pain or discomfort, internal homeostatic disturbances, and learned or associative responses to the perception of impending endangerment, pain, or discomfort ("psychological stress"). The hormones released in response to stressors often are referred to as "stress hormones" and their secretion is regulated by neural circuits impinging on hypothalamic neurons that are the final output toward the pituitary gland and the kidneys. This review discusses the forebrain circuits that mediate the neuroendocrine responses to stressors and emphasizes those neuroendocrine systems that have previously received little attention as stress-sensitive hormones: renin, oxytocin, and prolactin. Anxiolytic drugs of the benzodiazepine class and other drugs that affect catecholamine, GABAA, histamine, and serotonin receptors alter the neuroendocrine stress response. The effects of these drugs are discussed in relation to their effects on forebrain neural circuits that regulate stress hormone secretion. For psychological stressors such as conditioned fear, the neural circuits mediating neuroendocrine responses involve cortical activation of the basolateral amygdala, which in turn activates the central nucleus of the amygdala. The central amygdala then activates hypothalamic neurons directly, indirectly through the bed nucleus of the stria terminalis, and/or possibly via circuits involving brainstem serotonergic and catecholaminergic neurons. The renin response to psychological stress, in contrast to those of ACTH and prolactin, is not mediated by the bed nucleus of the stria terminalis and is not suppressed by benzodiazepine anxiolytics. Stressors that challenge cardiovascular homeostasis, such as hemorrhage, trigger a pattern of neuroendocrine responses that is similar to that observed in response to psychological stressors. These neuroendocrine responses are initiated by afferent signals from cardiovascular receptors which synapse in the medulla oblongata and are relayed either directly or indirectly to hypothalamic neurons controlling ACTH, prolactin, and oxytocin release. In contrast, forebrain pathways may not be essential for the renin response to hemorrhage. Thus current evidence indicates that although a diverse group of stressors initiate similar increases in ACTH, renin, prolactin, and oxytocin, the specific neural circuits and neurotransmitter systems involved in these responses differ for each neuroendocrine system and stressor category.
Article
This article presents a model that identifies effects of blood pressure variability (BPV) as a possible mechanism by which psychological/psychiatric factors and health behaviors confer increased risk of coronary artery disease (CAD) and acute coronary syndromes. Recent research in vascular biology and dynamics of coronary artery blood flow suggests that BPV may have pathogenic effects on the coronary endothelium, plaque formation, and plaque stability. Thus, BPV may be a risk factor for cardiovascular disease independent of mean arterial pressure. The model proposes that autonomic control of the heart exerts a buffering or inhibitory influence on oscillations in blood pressure. Established psychological/behavioral risk factors for CAD, such as depression, hostility, and anxiety, as well as physical deconditioning and aging, are associated with diminished autonomic control of the heart, which may disinhibit pathogenic BPV. Together, these data suggest a coherent, testable psychophysiological model of CAD. In this article, we review these data and make recommendations for research to examine the model.
Article
The primary hormonal mediators of the stress response, glucocorticoids and catecholamines, have both protective and damaging effects on the body. In the short run, they are essential for adaptation, maintenance of homeostasis, and survival (allostasis). Yet, over longer time intervals, they exact a cost (allostatic load) that can accelerate disease processes. The concepts of allostasis and allostatic load center around the brain as interpreter and responder to environmental challenges and as a target of those challenges. In anxiety disorders, depressive illness, hostile and aggressive states, substance abuse, and post-traumatic stress disorder (PTSD), allostatic load takes the form of chemical imbalances as well as perturbations in the diurnal rhythm, and, in some cases, atrophy of brain structures. In addition, growing evidence indicates that depressive illness and hostility are both associated with cardiovascular disease (CVD) and other systemic disorders. A major risk factor for these conditions is early childhood experiences of abuse and neglect that increase allostatic load later in life and lead individuals into social isolation, hostility, depression, and conditions like extreme obesity and CVD. Animal models support the notion of lifelong influences of early experience on stress hormone reactivity. Whereas, depression and childhood abuse and neglect tend to be more prevalent in individuals at the lower end of the socioeconomic ladder, cardiovascular and other diseases follow a gradient across the full range of socioeconomic status (SES). An SES gradient is also evident for measures of allostatic load. Wide-ranging SES gradients have also been described for substance abuse and affective and anxiety disorders as a function of education. These aspects are discussed as important, emerging public health issues where the brain plays a key role.
Article
To assess whether talking or reading (silently or aloud) could affect heart rate variability (HRV) and to what extent these changes require a simultaneous recording of respiratory activity to be correctly interpreted. Sympathetic predominance in the power spectrum obtained from short- and long-term HRV recordings predicts a poor prognosis in a number of cardiac diseases. Heart rate variability is often recorded without measuring respiration; slow breaths might artefactually increase low frequency power in RR interval (RR) and falsely mimic sympathetic activation. In 12 healthy volunteers we evaluated the effect of free talking and reading, silently and aloud, on respiration, RR and blood pressure (BP). We also compared spontaneous breathing to controlled breathing and mental arithmetic, silent or aloud. The power in the so called low- (LF) and high-frequency (HF) bands in RR and BP was obtained from autoregressive power spectrum analysis. Compared with spontaneous breathing, reading silently increased the speed of breathing (p < 0.05), decreased mean RR and RR variability and increased BP. Reading aloud, free talking and mental arithmetic aloud shifted the respiratory frequency into the LF band, thus increasing LF% and decreasing HF% to a similar degree in both RR and respiration, with decrease in mean RR but with minor differences in crude RR variability. Simple mental and verbal activities markedly affect HRV through changes in respiratory frequency. This possibility should be taken into account when analyzing HRV without simultaneous acquisition and analysis of respiration.
Article
The purpose of this study was to develop the Stress Response Inventory (SRI), which includes emotional, somatic, cognitive, and behavioral stress responses, and then to use the scale in clinical practice. First, a preliminary survey was conducted using 109 healthy adults to obtain 75 response items. Second, the preliminary questionnaire was completed by 215 healthy subjects. Third, stress responses were compared among 242 patients (71 with anxiety disorder, 73 with depressive disorder, 47 with somatoform disorder, and 51 with psychosomatic disorder) and the 215 healthy subjects. Factor analysis yielded seven subscales: tension, aggression, somatization, anger, depression, fatigue, and frustration. Reliability was computed by administering the SRI to 62 healthy subjects during a two-week interval. Test-retest reliability for the seven subscale scores and the total score was high, ranging between 0.69 and 0.96. Internal consistency was computed, and Cronbach's alpha for the seven subscales ranged between 0.76-0.91 and 0.97 for the total score. Convergent validity was computed by correlating the seven subscales and the total score of the SRI with the total score of the Global Assessment of Recent Stress (GARS) scale, the Perceived Stress Questionnaire (PSQ), and the subscale scores of the Symptom Checklist-90-Revised (SCL-90-R). The correlations were all at significant levels. The sensitivity of the SRI was 0.57, specificity 0.74, and the predictive value positive (PVP) was 0.71. The patient group also scored significantly higher on the six subscale scores and the total score than the control group, with the exception being the aggression subscale. The depressive disorder group was highest in total scores on the SRI among the four patient groups, and showed significantly higher total scores than the anxiety disorder and psychosomatic disorder groups. In total scores on the SRI, female subjects scored significantly higher than males. These results indicate that the SRI is highly reliable and valid, and that it can be utilized as an effective measure of stress for research in stress-related fields. The depressive disorder group showed more prominent stress responses than the anxiety and psychosomatic disorder groups.
Article
The aim of this study was to examine the response of the autonomic nervous system in younger (mean age 31 yrs, n=14) and older (mean age 54 yrs, n=14) healthy female teachers during work periods of perceived high and low stress. In the younger participants, heart rate, cortisol excretion rate and psychosomatic symptoms were significantly higher during the high work stress period. The older participants experienced no decrease in their heart rate and cortisol excretion during the low stress period and they exhibited no significant decrease in blood pressure after the work in the evening during both periods. It may be concluded that the recovery from the stress in the older teachers was insufficient particularly in view of their elevated diastolic blood pressure during the low work stress period. Ergonomic and individually tailored measures in terms of work time control, specific relaxation techniques, and a part-time retirement may improve the stress management of older teachers.