Skip to main content
American Journal of Translational Research logoLink to American Journal of Translational Research
. 2026 Apr 25;18(4):3424–3436. doi: 10.62347/IIMX7598

Adherence to sleep restriction therapy and its impact on emotional intelligence in high-stress professionals

Zhaohua Liu 1,*, Xuemei Zhang 2,*, Min Lin 3, Juan Chen 3, Jiacheng Long 4, Ling Wang 3, Deqiu Wen 5
PMCID: PMC13186716  PMID: 42170483

Abstract

Objective: To investigate the relationship between adherence to sleep restriction therapy and emotional intelligence (EI) among Chinese individuals with sleep disorders in high-stress professions. Methods: A retrospective case-control study was conducted using data from the electronic medical records system. A total of 257 patients with sleep disorders who underwent 12 weeks of sleep restriction therapy at The Affiliated Hospital of Southwest Medical University between January 2022 and January 2023 were included. Participants were classified into low EI (n = 76) and high EI (n = 181) groups. Adherence to therapy was the exposure variable. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using binary logistic regression, with subgroup and sensitivity analyses. Results: Higher adherence to sleep restriction therapy was significantly associated with higher EI scores. Compared to fully adherent participants, partially adherent individuals had a 2.39-fold increased risk of lower EI (OR = 2.39, 95% CI: 1.13-5.05), while non-adherent participants had a 5.35-fold elevated risk (OR = 5.35, 95% CI: 1.86-15.38). Subgroup analyses revealed stronger effects in males and individuals aged < 28 years. Males exhibited ORs of 2.02 (95% CI: 0.87-4.66) for partial adherence and 6.75 (95% CI: 2.00-22.77) for non-adherence. Similarly, the < 45-year subgroup showed ORs of 1.36 (95% CI: 0.61-3.03) and 2.99 (95% CI: 1.04-8.56), respectively. Weaker associations were observed in females and those aged ≥ 45 years. Conclusion: Better adherence to sleep restriction therapy is positively associated with EI in high-stress professionals with sleep disorders, particularly among males and younger individuals. Optimizing adherence to therapy is a key strategy for improving emotional regulation in these populations.

Keywords: Sleep restriction therapy, adherence, emotional intelligence, sleep disorders, occupational stress

Introduction

The relationship between sleep and emotional health is an increasingly important area of research, especially in modern occupational environments characterized by high stress levels [1]. Emotional intelligence (EI) refers to the ability to recognize, understand, and manage one’s own emotions as well as those of others [2]. This ability is crucial for effective interpersonal interactions and decision-making under stress [3]. High-stress professions, such as healthcare, law enforcement, aviation, and emergency services, often require high levels of EI to navigate daily challenges [4,5]. However, irregular working hours, extended shifts, and inherent occupational stressors often lead to disrupted sleep patterns [6]. Sleep deprivation and disorders are well-established contributors to EI impairment [7].

Among treatments for sleep disorders, sleep restriction therapy aims to optimize sleep by limiting time in bed to the actual sleep time, thereby consolidating sleep and improving efficiency [8,9]. Emerging evidence suggests that sleep restriction therapy may enhance prefrontal cortex activity through improvements in sleep quality, potentially benefiting cognitive and emotional processing [10]. However, the effect of adherence to sleep restriction therapy on EI in individuals with sleep disorders in high-stress professions is not well understood [11].

This study seeks to explore the association between adherence to sleep restriction therapy and EI among individuals with sleep disorders in high-stress professions. Our findings may contribute to the academic understanding of how adherence to sleep restriction therapy affects EI and provide a theoretical foundation for future research. Furthermore, the results could guide healthcare practitioners in optimizing sleep restriction therapy protocols.

Methods

Study population and design

This retrospective case-control study was conducted using the electronic medical record system at The Affiliated Hospital of Southwest Medical University (Ethics No. KY2026027). A total of 257 patients diagnosed with sleep disorders who underwent 12 weeks of sleep restriction therapy between January 2022 and January 2023 were included. Among them, 181 patients were classified into the high EI group, and 76 into the low EI group.

Additionally, 344 control participants were included: (1) aged 18 to 65 years; (2) engaged in high-stress occupations (e.g., healthcare workers, police officers, firefighters, pilots) [12], which were defined by the International Standard Classification of Occupations (ISCO-08) and validated by the Job Content Questionnaire (JCQ) [13] with scores for psychological job demands in the upper tertile of the reference population; (3) meeting diagnostic criteria for sleep disorders; (4) possessing sufficient cognitive function to understand and comply with the intervention protocol.

Exclusion criteria: (1) severe mental illnesses (e.g., schizophrenia or bipolar disorder) that were unstable or under active treatment; (2) serious physical health conditions, including heart, liver, or kidney diseases; (3) pregnancy or lactation; and (4) inability to complete the full study protocol.

Patients in the low EI group met the diagnostic criteria for both insomnia disorder and low EI. Insomnia disorder was defined by fulfilling all of the following criteria (A-F) [14].

A. Patient or caregiver report of ≥ 1 of the following: (1) difficulty initiating sleep; (2) difficulty maintaining sleep; (3) early morning awakening; (4) reluctance to go to bed at an appropriate bedtime; or (5) difficulty falling asleep without caregiver intervention.

B. Patient or caregiver report of ≥ 1 of the following consequences of nocturnal sleep disturbance: (1) fatigue or discomfort; (2) impaired attention or memory; (3) impaired social, household, occupational, or academic performance; (4) emotional disturbance/irritability; (5) daytime sleepiness; (6) behavioral problems such as hyperactivity, impulsivity, or aggression; (7) decreased motivation or energy; (8) increased susceptibility to errors and accidents; (9) concern or dissatisfaction with sleep.

C. Sleep complaints were not attributable solely to inadequate sleep opportunities or inappropriate sleep environments.

D. Sleep disturbances and related daytime symptoms occurred ≥ 3 times a week.

E. Duration of sleep disturbances and related daytime symptoms ≥ 3 months.

F. Sleep-wake difficulties could not be better explained by another sleep disorder.

Low EI was defined as a total score of 60 or lower on the Emotional Intelligence Scale (EIS) [9]. The EIS comprises four dimensions with a total of 16 items: self-emotion appraisal (items 1-4), self-emotion management (items 5-8), self-emotion utilization (items 9-12), and other-emotion appraisal (items 13-16). The scale uses a 5-point Likert rating system (1 = “Strongly Disagree” to 5 = “Strongly Agree”). Higher scores indicate higher EI. The EIS demonstrates strong reliability, with an overall Cronbach’s alpha coefficient of 0.885, and the Cronbach’s alpha coefficients for the four dimensions ranging from 0.826 to 0.904 [15].

Patients in the control group met the same diagnostic criteria for insomnia disorder but had high EI (EIS score > 60) [16]. A schematic overview of the study sample is presented in Figure 1.

Figure 1.

Figure 1

Flow chart of the study sample.

Exposure factors

The primary exposure factor was adherence to sleep restriction therapy. Sleep restriction therapy targeted individualized optimization of bedtime and wake-up time. Participants were required to maintain detailed sleep diaries, recording their bedtime, wake-up time, and actual sleep duration, which were reviewed weekly to calculate sleep efficiency based on the formula: Sleep efficiency = (actual sleep time/total time in bed) × 100%.

Adjustment rules were as follows: (1) If sleep efficiency ≥ 90%, participants’ bedtime advanced by 15 minutes. (2) If sleep efficiency ranged between 80% and 90%, participants’ bedtime remained unchanged. (3) If sleep efficiency < 80%, participants’ bedtime was delayed by 15 minutes. The entire intervention lasted for 12 weeks.

Adherence to sleep restriction therapy was classified into three levels: full adherence (score 8-10), partial adherence (score 5-7), and non-adherence (score 1-4). This adherence score was developed for this study based on prior behavioral intervention adherence research and expert consensus among three board-certified sleep medicine specialists [17]. Each item was rated from 1 (non-adherence) to 5 (complete adherence), with a total possible score of 10. The scale demonstrated good internal consistency in this sample (Cronbach’s α = 0.82). Higher scores indicate greater adherence to sleep restriction therapy. For operational purposes, full adherence was defined as an average of ≤ 2 missed nights per week and ≥ 90% concordance with prescribed time-in-bed schedules. Partial adherence corresponded to 3-5 missed nights per week or 60-89% concordance, while any pattern below these thresholds was classified as non-adherence.

Covariates

Covariates included the Pittsburgh Sleep Quality Index (PSQI), psychological assessment indicators, health status assessment indicators, and other demographic, physiological, and behavioral factors.

The PSQI is an effective tool for assessing sleep quality, comprising 18 scored items across seven components (0-3 points each; total 0-21). Higher scores indicate poorer sleep quality. The scale demonstrates a Cronbach’s alpha coefficient of 0.71, indicating good internal consistency [18].

Psychological assessment indicators included: (1) Self-rating Depression Scale (SDS): Scored 0-100, with higher scores indicating more severe depression (Cronbach’s α = 0.92) [19]; (2) Self-Assessment Anxiety Scale (SAS): A 20-item scale using a 4-point system (1 = not at all/a little, 2 = sometimes, 3 = often, 4 = most/all of the time). Fifteen items are negatively worded, and five are positively worded and reverse-scored. Scores ≥ 50 indicate anxiety, with ranges for severity: 50-59 = mild, 60-69 = moderate, ≥ 70 = severe. The SAS demonstrates high internal consistency (Cronbach’s α = 0.897) [20]. (3) Self-Esteem Scale (SES): A 10-item scale measuring self-worth and self-acceptance using a 4-point Likert scale (1 = strongly disagree to 4 = strongly agree). Items 3, 5, 8, 9, and 10 are reverse-scored. Total scores range from 10 to 40, with ≤ 25 indicating low, 26-32 moderate, and ≥ 33 high self-esteem. The SES demonstrates good reliability (Cronbach’s α = 0.86) [21].

Health status assessment indicators included: (1) Physical Component Summary (PCS): A key component of the Short Form 12 (SF-12), assessing physical functioning, role-physical, bodily pain, general health, vitality, and social functioning. PCS scores (0-100) are derived from standardized algorithms integrating multiple physical health dimensions; higher scores indicate better health (Cronbach’s α = 0.743) [22].

(2) EuroQoL-5 Dimensions, three-level (EQ-5D-3L): This scale assesses health-related quality of life across five dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Higher scores reflect better quality of life (Cronbach’s α = 0.75) [23].

Other covariates included: age (continuous variable), body mass index (BMI), gender (male/female), smoking (yes/no), drinking (yes/no), diabetes (yes/no), hypertension (yes/no), and first-onset of sleep disorders (yes/no).

Statistical analysis

Continuous data were expressed as mean ± standard deviation (SD) or median with interquartile range (IQR), and categorical variables as frequency (percentage). Comparisons of sleep quality dimensions among occupational groups were performed using one-way analysis of variance (ANOVA) or Kruskal-Wallis tests, followed by post-hoc analyses. Between-group comparisons of continuous variables were conducted using unpaired t-tests or Mann-Whitney U tests, and categorical variables were compared using chi-square tests.

Logistic regression analysis was conducted on the full sample to assess the association between adherence to sleep restriction therapy and EI, adjusting for covariates such as age, BMI, gender, smoking, drinking, diabetes, hypertension, first-onset of sleep disorders, PSQI, SDS, SAS, SES, PCS, and EQ-5D-3L scores. Subgroup logistic regression analyses by gender and age groups were performed to explore potential differences across demographic strata.

Sensitivity analyses were performed using both unadjusted (crude) and partially adjusted logistic regression models to assess the robustness of the association between adherence to sleep restriction therapy and EI. In the partially adjusted model, questionnaire scores (PSQI, SDS, SAS, SES, PCS, and EQ-5D-3L) were included as covariates. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated for each adherence category (full adherence, partial adherence, non-adherence).

To explore mediation, mean sleep efficiency over 12 weeks was tested as a mediator of the adherence-EI relationship. Scatter plots and linear regression lines were plotted for each adherence group to visualize the association between sleep efficiency and EI. The strength of associations across adherence groups was compared based on regression slopes.

Longitudinal sleep efficiency trends over 12 weeks were analyzed using linear mixed-effects models, with time and EI group as fixed effects and participants as random effects.

To assess the moderating effects of psychological states on intervention efficacy, multiple linear regression models with interaction terms among depression, anxiety, and self-esteem scores were constructed and visualized using heatmaps.

All statistical analyses were conducted using SPSS version 22 and R version 3.0.2, with a two-sided significance level set at P < 0.05.

Results

Baseline characteristics of participants

No statistically significant differences were observed between the two groups in terms of age, BMI, gender, smoking status, drinking status, education level, marital status, occupation type, and average monthly income (all P > 0.05) (Table 1).

Table 1.

Baseline characteristics of participants

Characteristic High EI Group (n = 181) Low EI Group (n = 76) t/χ2 P Value
Age (years) 40.21 ± 9.17 41.53 ± 9.34 1.046 0.296
BMI (kg/m2) 25.32 ± 3.15 25.87 ± 3.31 1.244 0.214
Gender (male/female) 98/83 42/34 0.027 0.869
Education Level [n/(%)] 0.299 0.861
    Primary School 5 (2.8) 3 (3.9)
    Secondary School 31 (17.1) 12 (15.8)
    College 145 (80.1) 61 (80.3)
Marital Status [n/(%)] 0.615 0.735
    Married 121 (66.8) 48 (63.2)
    Single 46 (25.4) 20 (26.3)
    Divorced 14 (7.7) 8 (10.5)
Occupation Type 0.084 0.994
    Healthcare Workers 55 (30.4) 22 (28.9)
    Police Officers 32 (17.7) 14 (18.4)
    Firefighters 39 (21.6) 16 (21.1)
    Pilots 55 (30.4) 24 (31.6)
Average Monthly Income (RMB) 0.168 0.983
    < 3000 16 (8.8) 7 (9.2)
    3000-5999 62 (34.3) 25 (32.9)
    6000-9000 64 (35.4) 26 (34.2)
    > 9000 39 (21.6) 18 (23.7)
Smoking (yes/no) 44/137 20/56 0.115 0.734
Drinking (yes/no) 31/150 12/64 0.069 0.793
Diabetes [n/(%)] 11 (6.1) 9 (11.8) 2.479 0.115
Hypertension [n/(%)] 25 (13.8) 16 (21.1) 2.093 0.148
Adherence [n (%)] 9.174 0.010
    Full Adherence 128 (70.7) 40 (52.6)
    Partial Adherence 41 (22.7) 24 (31.6)
    Non-Adherence 12 (6.6) 12 (15.8)
First-onset of Sleep Disorders 10 (5.5) 9 (11.8) 3.120 0.077
PSQI score 6.73 ± 2.14 8.16 ± 2.34 4.741 < 0.001
EIS Total Score 71.26 ± 8.33 46.87 ± 11.42 16.836 < 0.001
SDS Scale 42.63 ± 8.12 45.38 ± 7.01 2.579 0.010
SAS Score 46.64 ± 7.16 49.22 ± 6.04 2.752 0.006
SES Score 31.76 ± 2.19 30.37 ± 2.26 4.605 < 0.001
PCS Score 48.67 ± 10.61 44.55 ± 11.93 2.736 0.007
EQ-5D-3L Score 57.39 ± 8.63 53.36 ± 8.77 3.408 < 0.001

Note: BMI, body mass index; PSQI, Pittsburgh sleep quality index; EIS, Emotional Intelligence Scale; SDS, Self-rating Depression Scale; SAS, Self-rating anxiety scale; SES, Self-Esteem Scale; PCS, Physical Component Summary; EQ-5D-3L, Euro Qol five-dimension three-level questionnaire; EI, emotional intelligence.

Association between adherence to sleep restriction therapy and EI

Logistic regression analysis revealed a significant positive association between adherence to sleep restriction therapy and EI. Compared with full adherence, partial adherence was associated with a 2.41-fold higher risk of low EI (OR = 2.41; 95% CI: 1.09-5.29), while non-adherence was associated with a 6.50-fold higher risk (OR = 6.50; 95% CI: 2.14-19.73) (Figure 2).

Figure 2.

Figure 2

ORs for sleep adherence for EI. Adjust: Age, Body Mass Index, Gender, Smoking, Drinking, Diabetes, Hypertension, whether sleep disorders occur for the first time, PSQI score, SDS Score, SAS Score, SES Score, PCS Score, EQ-5D-3L Score. OR, odds ratio; EI, emotional intelligence; PSQI, Pittsburgh sleep quality index; SDS, Self-rating Depression Scale; SAS, Self-rating anxiety scale; SES, Self-Esteem Scale; PCS, Physical Component Summary; EQ-5D-3L, Euro Qol five-dimension three-level questionnaire.

Subgroup analysis

Gender subgroup

The association between therapy adherence and EI was significant in males but not in females. Compared with fully adherent males, partially adherent and non-adherent males showed an increasing trend of low EI risk, with ORs of 2.02 (95% CI: 0.87-4.66) and 6.75 (95% CI: 2.00-22.77), respectively.

Age subgroup

A stronger association was observed in participants aged < 45 years: partial adherence (OR = 1.36, 95% CI: 0.61-3.03) and non-adherence (OR = 2.99, 95% CI: 1.04-8.56) were associated with increased low EI risk. No significant association was found in participants aged ≥ 45 years.

Combined stratification

Non-adherent males had a much higher low EI risk (OR = 6.75, 95% CI: 2.00-22.77) than non-adherent females (OR = 1.22, 95% CI: 0.29-5.17).

The low EI risk in non-adherent participants aged < 45 years (OR = 2.99, 95% CI: 1.04-8.56) was higher than that in those aged ≥ 45 years (OR = 4.63, 95% CI: 0.88-24.33, P > 0.05).

These results suggest potential moderating effects of age and gender (Tables 2 and 3).

Table 2.

Association between sleep restriction therapy adherence and EI in different gender subgroups

Male group Female group


No. of cases/No. of total participants No. of controls/No. of total participants OR (95% CI) No. of cases/No. of total participants No. of controls/No. of total participants OR (95% CI)
Full adherence 19/87 68/87 Ref. 21/81 60/81 Ref.
Partial adherence 14/39 25/39 2.02 (0.87-4.66) 10/26 16/26 1.79 (0.70-4.55)
Non-adherence 9/14 5/14 6.75 (2.00-22.77) 3/10 7/10 1.22 (0.29-5.17)

Note: EI, emotional intelligence; OR, odds ratio; CI, confidence Interval.

Table 3.

Association between sleep restriction therapy adherence and EI in different age subgroups

Young adults group (< 45 years) Middle-aged and older adults group (≥ 45 years)


No. of cases/No. of total participants No. of controls/No. of total participants OR (95% CI) No. of cases/No. of total participants No. of controls/No. of total participants OR (95% CI)
Full adherence 28/118 90/118 Ref. 12/50 38/50 Ref.
Partial adherence 12/41 29/41 1.36 (0.61-3.03) 12/24 12/24 3.08 (1.09-8.70)
Non-adherence 8/17 9/17 2.99 (1.04-8.56) 4/7 3/7 4.63 (0.88-24.33)

Note: EI, emotional intelligence; OR, odds ratio; CI, confidence Interval.

Dynamic changes in sleep efficiency and its association with EI

Sleep efficiency showed an overall upward trend over time across all groups, indicating improvement possibly due to intervention or natural progression. Participants with higher EI had higher baseline sleep efficiency and greater improvement over time, reflecting stronger responsiveness to intervention (Figure 3).

Figure 3.

Figure 3

Dynamic change in sleep efficiency over 12 weeks and association with EI scores. Horizontal axis: time (weeks 1-12); vertical axis: sleep efficiency (%). Yellow represents high EI and purple represents low EI. Each line indicates the trajectory of sleep efficiency over time for an individual participant. Yellow lines (high EI) show higher baseline values and steeper upward trends; purple lines (low EI) show lower baselines with smaller or more variable improvements. Three trend lines with grey confidence intervals correspond to the EI group legend on the right: Low EI (dark blue solid line) shows the slowest improvement; Medium EI (dark blue dashed line) shows moderate improvement; High EI (light grey dotted line, with confidence interval) shows the greatest and most stable improvement. EI, emotional intelligence.

Verification of the mediating effect of sleep efficiency

Across all groups, sleep efficiency was positively associated with EI scores, as indicated by upward-sloping scatter points and regression lines. The association strength varied by adherence level: Full adherence: Highest sleep efficiency (~85-92.5%) and EI scores (most > 100, some > 105), with a strong positive correlation; Partial adherence: Moderate sleep efficiency (~85-87.5%) and EI scores (~100), with a moderate positive correlation; Non-adherence: Lowest sleep efficiency (~82.5-85%) and EI scores (~95-97), with a weak but positive correlation.

These results indicate that full adherence is associated with higher sleep efficiency (Figure 4).

Figure 4.

Figure 4

Results of the mediation effect verification of sleep efficiency. X-axis: “12-week Mean Sleep Efficiency (%)” -Average sleep efficiency over 12 weeks; Y-axis: “EI Score (EIS)” -Emotional Intelligence score measured by the EIS; Full Adherence (red points/lines): Participants with the highest adherence to sleep monitoring (e.g., fewest missed nights); Partial Adherence (blue points/lines): Participants with moderate adherence; Non-Adherence (green points/lines): Participants with the lowest adherence. EI, emotional intelligence; EIS, Emotional Intelligence Scale.

Occupational type and sleep response patterns

Distinct sleep quality patterns were observed across occupational groups. Healthcare workers scored close to 3 in Subjective Sleep Quality and Sleep Latency, indicating poor perceived sleep quality and difficulty falling asleep. Police officers scored notably higher in “Use of Sleeping Medication”, indicating that their sleep problems had progressed to a stage requiring pharmacological intervention. Their scores for “Sleep Duration” and “Sleep Efficiency” were moderate, reflecting average sleep length and quality.

Firefighters exhibited relatively balanced scores across all dimensions, with no single prominent sleep issue. However, all scores were above 0, indicating some degree of sleep disturbance.

Pilots scored high in “Daytime Dysfunction” and “Sleep Disturbances”, reflecting poor daytime functioning and frequent sleep interruptions. In contrast, their scores for “Subjective Sleep Quality” were relatively low, indicating fewer self-reported sleep complaints (Figure 5).

Figure 5.

Figure 5

Radar plot of occupational type and sleep response patterns. Axis scale: 0-3 representing PSQI component scores; higher scores indicate more severe sleep problems in that dimension. PSQI, Pittsburgh sleep quality index.

Subgroup analysis of EI by occupation

Subgroup analysis by occupation revealed significant differences in sleep quality, with only PSQI scores reaching statistical significance. Notably, the mean PSQI scores of all four occupational groups exceeded the clinical threshold for sleep disturbance (≥ 5), indicating a high prevalence of sleep problems in these high-stress professions, with varying severity due to occupational characteristics: Firefighters: Highest PSQI score (7.54 ± 2.23, most severe impairment), likely due to 24-hour on-call duties and work-related physical/psychological stress; Pilots: Second-highest score (7.27 ± 2.41), attributed to circadian rhythm disruption from transmeridian flights and high-altitude operational pressures; Healthcare workers: Score of 7.18 ± 2.23, affected by shift work and emotional exhaustion from compassion fatigue; Police officers: Lowest score (6.62 ± 2.25, still above threshold), possibly due to fewer nocturnal tasks and more regular schedules compared to other professions (Table 4).

Table 4.

Subgroup analysis of EI by occupation

Group Pilots Firefighters Police officers Healthcare workers F P_value
PSQI score 7.27 ± 2.41 7.54 ± 2.23 6.62 ± 2.25 7.18 ± 2.23 2.326 0.0378
EIS score 65.20 ± 13.92 61.90 ± 14.11 64.68 ± 13.20 63.75 ± 16.28 1.756 0.2181
SDS score 43.93 ± 9.63 43.19 ± 6.11 42.65 ± 7.29 43.66 ± 7.42 0.316 0.8136
SAS score 46.97 ± 6.62 48.24 ± 6.25 48.11 ± 7.15 46.86 ± 7.50 0.667 0.5731
SES score 31.11 ± 2.45 31.28 ± 1.94 31.63 ± 2.26 31.43 ± 2.36 0.607 0.6113
PCS score 45.10 ± 9.75 48.20 ± 13.75 49.58 ± 11.02 47.83 ± 10.64 1.931 0.1251
EQ-5D-3L score 55.94 ± 8.01 54.23 ± 9.43 56.62 ± 9.09 57.31 ± 9.06 0.784 0.9575

Note: EI, Emotional Intelligence; PSQI, Pittsburgh Sleep Quality Index; EIS, Emotional Intelligence Scale; SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; SES, Self-Esteem Scale; PCS, Pain Catastrophizing Scale; EQ-5D-3L, EuroQol 5-Dimensions 3-Levels questionnaire.

Moderating effect of psychological status on intervention outcomes

Heatmap analysis showed that the linear association between individual psychological factors and intervention outcomes was weak but trend-clear: negative emotions were inversely correlated with intervention efficacy, while positive psychological traits were positively correlated. Moreover, the combined effects of multiple psychological factors were more pronounced than single-variable effects - overlapping negative emotions amplified adverse impacts, while self-esteem’s “buffering effect” against negative emotions was weak. These findings highlight the need for tailored interventions targeting individual psychological states and cumulative negative emotions (Figure 6).

Figure 6.

Figure 6

Heatmap of moderating effects of psychological status on intervention efficacy. EIS, Emotional Intelligence Scale; SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; SES, Self-Esteem Scale. A: Direct associations between psychological variables and EIS (warm colors = positive correlation; cool colors = negative correlation). EIS ↔ SDS, weak negative correlation (-0.18), indicating higher depression associated with poorer intervention outcomes; EIS ↔ SAS, weak negative correlation (-0.14), indicating higher anxiety associated with poorer outcomes; EIS ↔ SES, weak positive correlation (0.17), indicating higher self-esteem associated with better outcomes. B: Moderating effects of psychological interaction terms on EIS (NA = self-association, constant = 1, not meaningful; dark color = negative correlation; light color = positive correlation). SDS × SAS, strongest negative correlation (-0.22), indicating co-occurrence of depression and anxiety has a greater adverse impact than either factor alone; SDS × SES (-0.11) and SAS × SES (-0.05), indicating limited buffering effect of self-esteem on negative emotions, with interaction terms still negatively correlated and stronger than the positive correlation of self-esteem alone.

Sensitivity analysis

In the crude model, partial and non-adherence were both associated with increased low EI risk compared with full adherence. The association remained significant in the adjusted model (controlling for questionnaire scores): Partial adherence: 2.42-fold higher low EI risk (OR = 2.42, 95% CI: 1.15-5.09); Non-adherence: 5.59-fold higher low EI risk (OR = 5.59, 95% CI: 2.02-15.47). Consistent results across models confirm the robustness of the findings (Table 5).

Table 5.

Association between sleep restriction therapy adherence and EI

Model 1 Model 2


OR (95% CI) P OR (95% CI) P
Adherence
    Full Adherence 1.00 (Reference) 1.00 (Reference)
    Partial Adherence 1.87 (1.01-3.47) 0.046 2.42 (1.15-5.09) 0.020
    Non-adherence 3.20 (1.33-7.68) 0.009 5.59 (2.02-15.47) < 0.001

Note: Model 1: Crude model; Model 2: Adjust: PSQI score, SDS Score, SAS Score, SES Score, PCS Score, EQ-5D-3L Score. PSQI, Pittsburgh Sleep Quality Index; SDS, Self-Rating Depression Scale; SAS, Self-Rating Anxiety Scale; SES, Self-Esteem Scale; PCS, Pain Catastrophizing Scale; EQ-5D-3L, EuroQol 5-Dimensions 3-Levels questionnaire; EI, Emotional Intelligence; OR, odds ratio; CI; confidence Interval.

Discussion

Using binary logistic regression and subgroup analyses, this study identified a significant positive association between adherence to sleep restriction therapy and EI among individuals with sleep disorders in high-stress professions.

Our findings align with previous research suggesting that improved sleep quality enhances cognitive function and emotional regulation [24]. The mechanism underlying this relationship is likely related to the restorative effects of sleep on neural circuits that govern emotional processing, particularly within the prefrontal cortex and amygdala, which may be optimized through sleep restriction therapy to improve EI [25,26]. However, it is important to note that our discussion of mechanisms involving the prefrontal cortex and amygdala remains speculative, as these biological constructs were not directly measured in this study. Therefore, these interpretations should be considered hypothetical and require further validation. Sensitivity analyses confirmed the robustness of the association, showing that non-adherence remained strongly linked to lower EI, emphasizing the independent role of adherence. While previous studies have focused on the physiological consequences of sleep deprivation, our research uniquely explores adherence to sleep restriction therapy in occupational contexts and its impact on EI.

Subgroup analyses based on gender and age revealed important variations. The stronger association in males may be related to testosterone’s modulation of prefrontal-amygdala connectivity during emotional regulation, although this proposed mechanism remains hypothetical without direct measurement of testosterone levels in our study. Similarly, the attenuated effect in individuals aged ≥ 45 years could reflect age-related declines in prefrontal cortex plasticity, limiting the efficacy of sleep interventions [27,28]. Regarding the subgroup analysis for participants aged ≥ 45 years, the wide confidence intervals indicate imprecision, likely due to the small sample size in this subgroup. Therefore, the conclusion that no significant association was observed should be interpreted cautiously, acknowledging the limited statistical power to detect an effect in this subgroup.

Moreover, our subgroup analysis by occupation revealed that while all four high-stress occupational groups exhibited clinically significant sleep disturbances (PSQI ≥ 5), the severity varied significantly. Firefighters reported the most impaired sleep, followed by pilots and healthcare workers, with police officers being the least affected. This gradient in sleep impairment aligns with the distinct occupational stressors inherent to each profession, such as 24-hour on-call duties, circadian disruption from transmeridian flights, shift-work schedules, and the relative regularity of police work. These findings emphasize that high-stress professions are heterogeneous, with specific stressors differentially impacting sleep quality [29]. Thus, the effectiveness of sleep restriction therapy in improving EI may vary depending on occupation-specific sleep disruption patterns. For instance, the potential for adherence to improve EI could be greatest in professions where sleep disruption is most severe and directly tied to unpredictable work schedules (e.g., firefighters, pilots), as improving sleep in these contexts may yield the most substantial gains in emotional regulation. This observation reinforces the need for occupation-tailored strategies to optimize adherence, suggesting that a one-size-fits-all approach may be suboptimal.

Clinically, our findings highlight adherence to sleep restriction therapy as a critical determinant of EI in high-stress professionals with sleep disorders. Higher adherence to therapy was associated with higher EI scores, enabling individuals to exhibit stronger empathic responses and improved emotion perception during social interactions. These attributes are particularly critical in professions such as healthcare and law enforcement, where frequent and intensive interpersonal interactions are required. In such roles, practitioners must accurately interpret and respond to others’ emotional cues to foster trust, facilitate communication, and provide effective service and support [30,31].

Translating these findings into clinical practice, our study suggests that assessing EI could serve as a risk stratification tool in sleep clinics catering to high-stress professionals. Specifically, EI screening might help identify individuals at higher risk for non-adherence to sleep restriction therapy, allowing for targeted support. However, the cost-effectiveness and practical implementation of such screening require further investigation before widespread adoption. Furthermore, based on the association between EI and adherence, personalized interventions could be developed. For instance, individuals with lower EI might benefit from enhanced psychological support components integrated into sleep restriction therapy, such as motivational interviewing techniques focused on emotional barriers to adherence, or mindfulness-based exercises to improve emotional awareness and regulation.

Technology offers promising avenues for scaling personalized approaches. Artificial intelligence (AI)-driven tools, such as automated sleep diary apps with real-time reminders, personalized feedback on adherence patterns, and integrated emotional check-ins, could provide scalable support. These technologies may help address challenges in adherence tracking (e.g., accuracy of entries) by offering objective monitoring and timely interventions. Future studies should investigate the efficacy of AI-driven monitoring apps in providing real-time feedback to improve sleep restriction therapy adherence, particularly for individuals with lower EI.

High adherence ensures consistent implementation of sleep strategies, which is essential for achieving higher therapeutic efficacy. It is well-established that insufficient sleep impairs emotional regulation and heightens emotional reactivity, primarily due to altered neural circuitry [32]. By improving adherence to sleep restriction therapy, sleep-related disruptions can be effectively reduced, facilitating better emotional comprehension and regulation [33]. Strategies to enhance adherence include patient education, personalized treatment planning, regular follow-up, psychological support, technology-assisted monitoring, and leveraging family and social networks. These approaches may significantly improve adherence to sleep restriction therapy, leading to better therapeutic outcomes.

Despite these contributions, several limitations must be acknowledged. First, as a case-control study, our findings may be susceptible to recall bias. Second, the study population was restricted to high-stress professionals, leaving the effects of sleep restriction therapy adherence on EI in lower-stress settings unexplored. Third, our measure of adherence, though practical, relied heavily on sleep diary completion frequency and self-reported consistency with prescribed schedules. This approach may not fully capture the complexity of behavioral adherence, and the lack of verification procedures (e.g., telephone follow-up for diary accuracy or correlation with objective actigraphy) may have introduced measurement error. Fourth, the absence of objective sleep architecture data, such as polysomnography (PSG)-derived measures of slow-wave sleep and rapid eye movement sleep, prevents our study from elucidating the specific neurophysiological pathways that may mediate the relationship between adherence and EI. Finally, while we propose the potential of AI-driven tools for future interventions, our current study does not provide a specific technical pathway or feasibility verification plan for such technologies. Future research should address these limitations by employing prospective, randomized controlled designs, incorporating objective sleep measures (PSG, actigraphy), implementing robust adherence verification protocols, and developing and testing AI monitoring frameworks in diverse occupational groups.

Conclusion

This study demonstrates a significant positive correlation between adherence to sleep restriction therapy and EI in high-stress professionals with sleep disorders, particularly among males and individuals under 45 years old. The results suggest that enhancing adherence to sleep restriction therapy may be a key strategy for improving EI in these populations. This finding provides important guidance for future clinical interventions and research targeting psychological resilience in occupational health contexts.

Acknowledgements

This study was supported by Health Commission of Sichuan Province Medical Science and Technology Program (25CXTD18) and 2025 Research Project on Chronic Disease Management Issued by National Center for Capacity Building and Continuing Education, National Health Commission Project (GWJJMB202510010026).

Disclosure of conflict of interest

None.

References

  • 1.Philip P, Taillard J, Micoulaud-Franchi JA. Sleep restriction, sleep hygiene, and driving safety: the importance of situational sleepiness. Sleep Med Clin. 2019;14:407–412. doi: 10.1016/j.jsmc.2019.07.002. [DOI] [PubMed] [Google Scholar]
  • 2.Fossier KB. Emotional intelligence. Radiol Technol. 2022;93:396–403. [PubMed] [Google Scholar]
  • 3.Toriello HV, Van de Ridder JMM, Brewer P, Mavis B, Allen R, Arvidson C, Kovar-Gough I, Novak E, O’Donnell J, Osuch J, Ulrich B. Emotional intelligence in undergraduate medical students: a scoping review. Adv Health Sci Educ Theory Pract. 2022;27:167–187. doi: 10.1007/s10459-021-10079-2. [DOI] [PubMed] [Google Scholar]
  • 4.Malik P, Rangel M, VonBriesen T. Why the utilization of ready-to-administer syringes during high-stress situations is more important than ever. J Infus Nurs. 2022;45:27–36. doi: 10.1097/NAN.0000000000000451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Nagata T, Ito R, Nagata M, Odagami K, Kajiki S, Fujimoto K, Matsuda S, Mori K. The differences of the economic losses due to presenteeism and treatment costs between high-stress workers and non-high-stress workers using the stress check survey in Japan. J Occup Health. 2022;64:e12346. doi: 10.1002/1348-9585.12346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kyle SD, Madigan C, Begum N, Abel L, Armstrong S, Aveyard P, Bower P, Ogburn E, Siriwardena A, Yu LM, Espie CA. Primary care treatment of insomnia: study protocol for a pragmatic, multicentre, randomised controlled trial comparing nurse-delivered sleep restriction therapy to sleep hygiene (the HABIT trial) BMJ Open. 2020;10:e036248. doi: 10.1136/bmjopen-2019-036248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Dugue M, Sirost O, Dosseville F. A literature review of emotional intelligence and nursing education. Nurse Educ Pract. 2021;54:103124. doi: 10.1016/j.nepr.2021.103124. [DOI] [PubMed] [Google Scholar]
  • 8.Jurgelis M, Boardman JM, Coxon JP, Drummond SPA, Chong TTJ. Sleep restriction reduces cognitive but not physical motivation. Nat Sci Sleep. 2022;14:2001–2012. doi: 10.2147/NSS.S368335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Maurer LF, Ftouni S, Espie CA, Bisdounis L, Kyle SD. The acute effects of sleep restriction therapy for insomnia on circadian timing and vigilance. J Sleep Res. 2021;30:e13260. doi: 10.1111/jsr.13260. [DOI] [PubMed] [Google Scholar]
  • 10.Maurer LF, Sharman R, Espie CA, Kyle SD. The effect of sleep restriction therapy for insomnia on REM sleep fragmentation: a secondary analysis of a randomised controlled trial. J Sleep Res. 2024;33:e13982. doi: 10.1111/jsr.13982. [DOI] [PubMed] [Google Scholar]
  • 11.MacCann C, Jiang Y, Brown LER, Double KS, Bucich M, Minbashian A. Emotional intelligence predicts academic performance: a meta-analysis. Psychol Bull. 2020;146:150–186. doi: 10.1037/bul0000219. [DOI] [PubMed] [Google Scholar]
  • 12.Roth T, Coulouvrat C, Hajak G, Lakoma MD, Sampson NA, Shahly V, Shillington AC, Stephenson JJ, Walsh JK, Kessler RC. Prevalence and perceived health associated with insomnia based on DSM-IV-TR; International statistical classification of diseases and related health problems, tenth revision; and research diagnostic criteria/international classification of sleep disorders, second edition criteria: results from the America insomnia survey. Biol Psychiatry. 2011;69:592–600. doi: 10.1016/j.biopsych.2010.10.023. [DOI] [PubMed] [Google Scholar]
  • 13.Karasek R, Brisson C, Kawakami N, Houtman I, Bongers P, Amick B. The Job Content Questionnaire (JCQ): an instrument for internationally comparative assessments of psychosocial job characteristics. J Occup Health Psychol. 1998;3:322–55. doi: 10.1037//1076-8998.3.4.322. [DOI] [PubMed] [Google Scholar]
  • 14.Ferini-Strambi L. Insomnia disorder. Minerva Med. 2025;116:309–322. doi: 10.23736/S0026-4806.25.09690-9. [DOI] [PubMed] [Google Scholar]
  • 15.Husain W, Inam A, Wasif S, Zaman S. Emotional intelligence: emotional expression and emotional regulation for intrinsic and extrinsic emotional satisfaction. Psychol Res Behav Manag. 2022;15:3901–3913. doi: 10.2147/PRBM.S396469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Katrushova L, Yalanska S, Rudenko L, Katrushov O. Peculiarities of the process of psychological adaptation of foreign students of ukrainian higher education institutions of medical profile, role of emotional intelligence in the socialization process. Wiad Lek. 2019;72:1930–1934. [PubMed] [Google Scholar]
  • 17.Matthews EE, Schmiege SJ, Cook PF, Berger AM, Aloia MS. Adherence to cognitive behavioral therapy for insomnia (CBTI) among women following primary breast cancer treatment: a pilot study. Behav Sleep Med. 2012;10:217–229. doi: 10.1080/15402002.2012.666220. [DOI] [PubMed] [Google Scholar]
  • 18.Liu D, Kahathuduwa C, Vazsonyi AT. The Pittsburgh Sleep Quality Index (PSQI): psychometric and clinical risk score applications among college students. Psychol Assess. 2021;33:816–826. doi: 10.1037/pas0001027. [DOI] [PubMed] [Google Scholar]
  • 19.Deluca P, Foley M, Dunne J, Kimergard A. The Severity of Dependence Scale (SDS) for Codeine: preliminary investigation of the psychometric properties of the SDS in an online sample of Codeine users from the UK. Front Psychiatry. 2021;12:595706. doi: 10.3389/fpsyt.2021.595706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Samakouri M, Bouhos G, Kadoglou M, Giantzelidou A, Tsolaki K, Livaditis M. Standardization of the Greek version of Zung’s Self-rating Anxiety Scale (SAS) Psychiatriki. 2012;23:212–220. [PubMed] [Google Scholar]
  • 21.Dacakis G, Erasmus J, Nygren U, Oates J, Quinn S, Sodersten M. Development and initial psychometric evaluation of the self-efficacy scale for voice modification in trans women. J Voice. 2024;38:1251.e21–1251.e31. doi: 10.1016/j.jvoice.2022.03.015. [DOI] [PubMed] [Google Scholar]
  • 22.Haddad C, Sacre H, Obeid S, Salameh P, Hallit S. Validation of the Arabic version of the “12-item short-form health survey” (SF-12) in a sample of Lebanese adults. Arch Public Health. 2021;79:56. doi: 10.1186/s13690-021-00579-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Yapp LZ, Scott CEH, Howie CR, MacDonald DJ, Simpson A, Clement ND. Meaningful values of the EQ-5D-3L in patients undergoing primary knee arthroplasty. Bone Joint Res. 2022;11:619–628. doi: 10.1302/2046-3758.119.BJR-2022-0054.R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Maurer LF, Espie CA, Omlin X, Reid MJ, Sharman R, Gavriloff D, Emsley R, Kyle SD. Isolating the role of time in bed restriction in the treatment of insomnia: a randomized, controlled, dismantling trial comparing sleep restriction therapy with time in bed regularization. Sleep. 2020;43:zsaa096. doi: 10.1093/sleep/zsaa096. [DOI] [PubMed] [Google Scholar]
  • 25.Maurer LF, Espie CA, Omlin X, Emsley R, Kyle SD. The effect of sleep restriction therapy for insomnia on sleep pressure and arousal: a randomized controlled mechanistic trial. Sleep. 2022;45:zsab223. doi: 10.1093/sleep/zsab223. [DOI] [PubMed] [Google Scholar]
  • 26.Xin Q, Yuan RK, Zitting KM, Wang W, Purcell SM, Vujovic N, Ronda JM, Quan SF, Williams JS, Buxton OM, Duffy JF, Czeisler CA. Impact of chronic sleep restriction on sleep continuity, sleep structure, and neurobehavioral performance. Sleep. 2022;45:zsac046. doi: 10.1093/sleep/zsac046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Liu PY, Reddy RT. Sleep, testosterone and cortisol balance, and ageing men. Rev Endocr Metab Disord. 2022;23:1323–1339. doi: 10.1007/s11154-022-09755-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Guo X, Keenan BT, Sarantopoulou D, Lim DC, Lian J, Grant GR, Pack AI. Age attenuates the transcriptional changes that occur with sleep in the medial prefrontal cortex. Aging Cell. 2019;18:e13021. doi: 10.1111/acel.13021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Betson JR, Kirkcaldie MTK, Zosky GR, Ross RM. Transition to shift work: sleep patterns, activity levels, and physiological health of early-career paramedics. Sleep Health. 2022;8:514–520. doi: 10.1016/j.sleh.2022.06.001. [DOI] [PubMed] [Google Scholar]
  • 30.Nesi J. The impact of social media on youth mental health: challenges and opportunities. N C Med J. 2020;81:116–121. doi: 10.18043/ncm.81.2.116. [DOI] [PubMed] [Google Scholar]
  • 31.Moylan K, Doherty K. Expert and interdisciplinary analysis of AI-driven chatbots for mental health support: mixed methods study. J Med Internet Res. 2025;27:e67114. doi: 10.2196/67114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Alfini AJ, Won J, Weiss LR, Nyhuis CC, Shackman AJ, Spira AP, Smith JC. Impact of exercise on older adults’ mood is moderated by sleep and mediated by altered brain connectivity. Soc Cogn Affect Neurosci. 2020;15:1238–1251. doi: 10.1093/scan/nsaa149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Setiawan L, Setiabudy R, Kresno SB, Sutandyo N, Syahruddin E, Jovianti F, Nadliroh S, Mubarika S, Setiabudy R, Siregar NC. Circulating miR-10b, soluble urokinase-type plasminogen activator receptor, and plasminogen activator inhibitor-1 as predictors of non-small cell lung cancer progression and treatment response. Cancer Biomark. 2024;39:137–153. doi: 10.3233/CBM-220222. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from American Journal of Translational Research are provided here courtesy of e-Century Publishing Corporation

RESOURCES