The relation between female emotional laborers’ sleep quality and autonomic nervous system activity was investigated.
Thirty-three subjects’ heart rate variability (HRV) data and results of self-reported scale on sleep, depression, anxiety and suicidality, were collected. Subjects were classified into good sleeper (GS) and poor sleeper (PS) groups relying on sleep quality. Changes of HRV between working time and resting time in each group were compared.
The PS group showed significantly lower difference in root mean square of successive differences (RMSSD), percentage of successive normal-to-normal intervals that differ by more than 50 ms (pNN50), and natural logarithm high-frequency (LnHF) when they were working as compared to when they were resting, which means decreased function of the parasympathetic nervous system (PNS). Repeated measures analysis of covariance showed that the group effect was significant only for LnHF, with score of depression scale as a covariate.
Female emotional laborers who complain of sleep difficulty may have decreased function of the PNS. Therefore, good sleep quality is essential for maintaining and promoting mental and physical health of women engage in emotional labor.
As women and girls are expected to be much more emotionally aware and better at providing emotional support than men and boys, the proportion of women is higher in work related to emotional labor. The survey on emotional laborers’ mental health conducted by Ansan Community Mental Health Center in 2018 reported that 84.5% of the respondents were female [
Emotional labor refers to the form of emotional regulation in which individuals adjust their posture and facial expressions to meet organizational requirements [
Emotional dissonance can be reflected by a change in autonomic nervous system (ANS) activities assessed in terms of heart rate variability (HRV) [
Sleep difficulty has been associated with a variety of physical and mental health-related outcomes [
We hypothesized that sleep difficulty in emotional laborers is associated with impairment of emotional regulation and burnout, leading to ANS dysregulation. Discovering the associations between insomnia and ANS dysregulation can provide evidence for emotional regulation difficulty in emotional laborers.
Thirty-three female customer service representatives were recruited from 2 reservation call centers at Korea University Ansan Hospital and Anam Hospital. Participation in the research was entirely voluntary, and confidentiality was guaranteed.
The subjects were requested to provide personal information such as age, education level completed, marital state, living status, and tenure. In addition, the participants were asked to fill out self-reported questionnaires that assessed their psychological state. The participants were required to wear a wristband during working time and resting time on 2 workdays.
The questionnaires included the Athens Insomnia Scale-8 (AIS-8), Center for Epidemiological Studies Depression Scale (CES-D), Generalized Anxiety Disorder 7-item (GAD-7), and Suicide Behaviors Questionnaire-Revised (SBQ-R). The AIS-8 measures the intensity of insomnia. The AIS-8 contains 8 items and is scored on a 4-point numerical rating scale: 0=no problem, 1=minor problem, 2=marked problem, and 3=very serious problem. Total scores range from 0 to 24, with a higher score indicating poorer sleep. A cut-off score of 8 was used in the present study [
HRV data for this study were obtained using the E4 wristband (Empatica, Boston, MA, USA) which is validated research instrument for heart rate (HR) and HRV [
Participants were respectively classified into poor sleeper (PS) and good sleeper (GS) groups based on the AIS-8 cut-off score. The general characteristics of the participants in PS and GS groups were compared using Mann-Whitney U-test or chi-square test. HRV variables between working time and resting time were compared within each group. Comparisons were obtained using the Wilcoxon signed rank test. Lastly, we compared the value of the HRV parameters between PS and GS. We examined the effect of sleep difficulty on the differences of the HRV parameters between working time and resting time by performing an analysis of covariance (ANCOVA) with repeated measures. The between-subjects factor was group (GS vs. PS) and the within-subjects factor was time (work vs. rest). The CES-D score was used as the covariates because this variable has been shown to be related with the AIS score. All data were statistically analyzed by SPSS for Windows version 25 (IBM Corp., Armonk, NY, USA).
The study protocol was reviewed and approved by the Institutional Review Board from Korea University (KUIRB-2019-0168-01).
The present study included 33 female subjects. The mean±standard deviation (SD) age was 35.52±7.11 years (range 22–47 years). Among the subjects, 45.4% (n=15) were married, and 54.5% (n=18) were singles. The majority (87.8% [n=29]) of the subjects were living with someone else, whereas 12.1% (n=4) were living by themselves. The mean duration of employment was 4 years (SD=4.93). Mean scores and SDs of each psychological scale are summarized in
Subjects were classified based on the AIS-8 cut-off score into good sleeper (GS) and poor sleeper (PS) groups. The GS group consisted of 17 subjects (51.5%) and the PS group of 16 subjects (48.5%). The average AIS-8 scores were 4.24±4.92 in the GS group and 9.68±1.62 in the PS group.
Because the sample size in the GS group was small (n=17), the Wilcoxon signed rank test was used to compare HRV parameters between working time and resting time in the GS group. SDNN (difference=-2.06, p=0.039), NN50 (difference=-3.62, p<0.001), and LnLF (difference=-3.00, p=0.003) were significantly different. No significant difference was observed in RMSSD (difference=-0.21, p=0.831), pNN50 (difference=-0.45, p=0.653) and LnHF power (difference=-1.40, p=0.162) (
As was the case for the GS group, Wilcoxon signed rank test was used to compare HRV parameters between working time and resting time for the PS group because of the small sample size (n=16). All HRV parameters turned out to be significantly different for the PS group: SDNN (difference=-3.26, p=0.001), RMSSD (difference=-3.05, p=0.002), NN50 (difference=-3.52, p<0.001), pNN50 (difference=-2.84, p=0.004), LnLF (difference=-2.74, p=0.006) and LnHF (difference=-3.36, p=0.001) (
No significant differences were observed in age, marital status, living status, education duration, tenure, and SBQ-R scores between PS and GS groups. Conversely, significant differences were observed in CES-D and GAD-7 between both groups (
In growing recognition of the consequences of sleep problems, particularly for the working population, many researchers have suggested that sleep difficulty is associated with a variety of health problems [
Sleep deprivation may cause changes in the neurotransmitter and neuroendocrine activity mediated by the autonomic sympatho-adrenal system and the hypothalamic-pituitary-adrenal axis, which is similar to a stress response. Hyperarousal has been hypothesized to contribute to the development, maintenance, and 24-hour systemic sequelae of insomnia [
People who suffer from sleep difficulty may have a poorly functioning PNS, which is associated with a variety of physical and mental health-related outcomes [
The present study has several limitations. First, the sample size was not sufficiently large to ensure the validity of results. It might be the reason why there was no group-by-time interaction effect, although group comparison showed significant difference. Thus, another study with a larger cohort may be necessary to confirm the results. Second, this research was conducted with employees working at a call center who may not be true representatives of other occupational groups who work in the emotional labor field. Therefore, to validate the results, further studies that include various occupational groups are needed. Third, confounders such as cigarette smoking, alcohol consumption, body mass index, and habitual physical activity were not considered. In addition, the female hormone may have affected the study results. For more accurate results, adjustments should be made for various confounders. Lastly, the extent to which sleep difficulty precedes changes in PNS activity or whether sleep difficulty is a consequence of suppressed PNS activity remain unclear. The causality of the association between sleep disturbance and the PNS activity remains to be established.
The current study examined the relationship between sleep difficulty and ANS dysregulation in female emotional laborers. The HRV parameters associated PNS decreased during working time in female customer service representatives who had sleep difficulty. Sleep health care is essential to protect emotional laborers’ mental health. If they have sleep problem, it can be helpful to guide them into mental health care programs such as cognitive-behavioral treatment for insomnia [
The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.
The authors have no potential conflicts of interest to disclose.
Conceptualization: Ho-Kyoung Yoon. Data curation: June Kang, Ho-Kyoung Yoon. Formal analysis: Boram Chae, June Kang, Ho-Kyoung Yoon. Investigation: June Kang, Ho-Kyoung Yoon. Methodology: June Kang, Ho-Kyoung Yoon. Project administration: June Kang, Ho-Kyoung Yoon. Resources: June Kang, Ho-Kyoung Yoon. Supervision: Cheolmin Shin, Young-Hoon Ko, Ho-Kyoung Yoon. Writing—original draft: Boram Chae, Ho-Kyoung Yoon. Writing—review & editing: Boram Chae, Ho-Kyoung Yoon.
This study was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (No. 2020R1A2C1008072), and the Korea University Grant.
Comparison of LnHF in GS and PS groups during working time and resting. *p<0.01; †p<0.05. GS, good sleepers; PS, poor sleepers; LnHF, natural logarithm high-frequency; n.s., not significant.
Sociodemographic and psychological characteristics of subjects
Variables | Categories | Total (N=33) |
GS (N=17) |
PS (N=16) |
χ2 or u |
---|---|---|---|---|---|
N or M±SD | N (%) or M±SD | N (%) or M±SD | |||
Age | 35.52±7.11 | 37.82±6.79 | 32.47±6.58 | u=89.5 | |
Marital status | Unmarried | 18 (54.5) | 8 (47.1) | 10 (62.5) | χ2=0.79 |
Married | 15 (45.4) | 9 (52.9) | 6 (37.5) | ||
Living status | Live alone | 4 (12.1) | 1 (5.9) | 3 (18.8) | χ2=1.28 |
Live with others | 29 (87.8) | 16 (94.1) | 13 (81.3) | ||
Education duration (yr) | 14.18±1.53 | 14.59±1.54 | 13.73±1.49 | u=94.5 | |
Tenure (yr) | 4.00±4.93 | 4.77±4.92 | 3.35±5.08 | u=89.0 | |
AIS-8 | 6.88±3.41 | 4.24±2.33 | 9.80±1.61 | u=0.00 |
|
CES-D | 12.24±10.67 | 5.82±5.25 | 19.06±10.61 | u=35.5 |
|
GAD-7 | 4.09±3.76 | 2.59±3.64 | 5.80±3.21 | u=46.0 |
|
SBQ-R | 3.61±1.89 | 3.12±1.32 | 4.20±2.34 | u=101.50 |
p<0.05.
AIS-8, Athens insomnia scale-8; CES-D, Center for Epidemiological Studies-Depression Scale; GAD-7, Generalized Anxiety Disorder 7-item; SBQ-R, Suicide Behaviors Questionnaire-Revised; GS, good sleepers; PS, poor sleepers
Changes in heart rate variability measures between work and rest in good sleepers
Domain | Parameter | M±SD |
Diff (Work-Rest) | p | |
---|---|---|---|---|---|
Work | Rest | ||||
Time | SDNN (ms) | 87.38±17.82 | 93.13±15.38 | -2.06 | 0.039 |
RMSSD (ms) | 87.80±24.97 | 87.62±26.29 | -0.21 | 0.831 | |
NN50 (beats) | 7,480.69±1929.31 | 1,971.29±460.16 | -3.62 | <0.001 | |
pNN50 (%) | 44.73±11.14 | 45.64±10.87 | -0.45 | 0.653 | |
Frequency | LnLF | 8.10±0.43 | 8.35±0.25 | -3.00 | 0.003 |
LnHF | 7.86±0.54 | 7.94±0.56 | -1.40 | 0.162 |
SDNN, standard deviation of normal-to-normal; RMSSD, root mean square of successive differences; NN50, the number of successive normal-to-normal interval pairs that differ more than 50 msec; pNN50, percentage of successive normal-to-normal intervals that differ by more than 50 ms; LnLF, natural logarithm low-frequency; LnHF, natural logarithm high-frequency
Changes in heart rate variability measures between work and rest in poor sleepers
Domain | Parameter | M±SD |
Diff (Work-Rest) | p | |
---|---|---|---|---|---|
Work | Rest | ||||
Time | SDNN | 85.58±21.87 | 97.32±19.27 | -3.26 | 0.001 |
RMSSD | 90.31±26.84 | 97.21±27.18 | -3.05 | 0.002 | |
NN50 | 7,375.52±1,909.97 | 2,151.00±494.13 | -3.52 | <0.001 | |
pNN50 | 45.03±12.68 | 49.93±12.42 | -2.84 | 0.004 | |
Frequency | LnLF | 8.00±0.48 | 8.33±0.38 | -2.74 | 0.006 |
LnHF | 7.87±0.65 | 8.11±0.56 | -3.36 | 0.001 |
SDNN, standard deviation of normal-to-normal; RMSSD, root mean square of successive differences; NN50, the number of successive normal-to-normal interval pairs that differ more than 50 msec; pNN50, percentage of successive normal-to-normal intervals that differ by more than 50 ms; LnLF, natural logarithm low-frequency; LnHF, natural logarithm high-frequency; Diff, difference
Effect of the sleep quality on change of heart rate variability measures
Source | Type III Sum of Squares | df | Mean SS | F | Sig |
---|---|---|---|---|---|
SDNN | 593.83 | 1 | 593.83 | 9.16 | 0.005 |
SDNN*CES-D | 37.05 | 1 | 37.05 | 0.57 | 0.46 |
SDNN*group | 175.55 | 1 | 175.55 | 2.71 | 0.110 |
Residual | 1,944.12 | 30 | 64.80 | ||
RMSSD | 39.98 | 1 | 39.98 | 0.84 | 0.37 |
RMSSD*CES-D | 1.61 | 1 | 1.61 | 0.03 | 0.86 |
RMSSD*group | 105.93 | 1 | 105.93 | 2.22 | 0.15 |
Residual | 1,432.31 | 30 | 47.74 | ||
NN50 | 116,094,891.26 | 1 | 116,094,891.26 | 96.45 | 0.00 |
NN50*CES-D | 1,543,690.16 | 1 | 1,543,690.16 | 1.28 | 0.27 |
NN50*group | 1,528,816.26 | 1 | 1,528,816.26 | 1.27 | 0.27 |
Residual | 36,111,184.31 | 30 | 1,203,706.14 | ||
pNN50 | 46.90 | 1 | 46.90 | 2.68 | 0.11 |
pNN50*CES-D | 0.28 | 1 | 0.28 | 0.02 | 0.90 |
pNN50*group | 43.34 | 1 | 43.34 | 2.47 | 0.13 |
Residual | 525.48 | 30 | 17.52 | ||
LnLF | 0.91 | 1 | 0.91 | 16.03 | 0.00 |
LnLF*CES-D | 0.15 | 1 | 0.15 | 2.58 | 0.12 |
LnLF*group | 0.14 | 1 | 0.14 | 2.46 | 0.13 |
Residual | 1.70 | 30 | 0.06 | ||
LnHF | 0.19 | 1 | 0.19 | 8.46 | 0.01 |
LnHF*CES-D | 0.01 | 1 | 0.01 | 0.43 | 0.52 |
LnHF*group | 0.10 | 1 | 0.10 | 4.30 | 0.05 |
Residual | 0.67 | 30 | 0.02 |
SDNN, standard deviation of normal-to-normal; RMSSD, root mean square of successive differences; NN50, the number of successive normal-to-normal interval pairs that differ more than 50 msec; pNN50, percentage of successive normal-to-normal intervals that differ by more than 50 ms; LnLF, natural logarithm low-frequency; LnHF, natural logarithm high-frequency; CES-D, Center for Epidemiological Studies-Depression Scale