Chronotype may be related to pain, sleep quality, depression, and quality of life in people with tension-type headache.

This cross-sectional observational study was conducted at SANKO University Sani Konukoğlu Hospital Health Practice and Research Centre. The study adhered to the ethical principles of the Declaration of Helsinki and received approval from the SANKO University Non-Invasive Research Ethics Committee (approval no: 04/2021/16). Reporting of the study followed the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) checklist.

Adult patients diagnosed with TTH according to the International Classification of Headache Disorders-III (ICHD-III) were consecutively recruited from the neurology outpatient clinic of the hospital between June 2021 and June 2022 [11]. A total of 77 patients meeting the inclusion criteria and providing consent were included in the study. Patients were excluded if they were under 18 years of age, were receiving medication for any psychiatric disorder, had a diagnosed sleep disorder (e.g., obstructive sleep apnea, restless leg syndrome), used analgesic medication regularly for any reason, had a history of medication-overuse headache, or had received antidepressant treatment within the preceding month. A post hoc power analysis was performed based on the observed PSQI group means and standard deviations (Evening chronotype: 9.63 ± 5.58, Intermediate chronotype: 7.98 ± 4.05, Morning chronotype: 5.14 ± 3.06). For the three groups with a total of 77 participants, the calculated effect size was Cohen's f = 0.366, indicating that approximately 11.8% of the total variance was explained by group differences (η² = 0.118), which corresponds to a medium-to-large effect. Given this effect size and an alpha level of 0.05, the statistical power (1 − β) was estimated to be approximately 0.82. Therefore, the current sample size provided sufficient power to detect differences in the observed magnitude [12,13].

Data were collected through in-person interviews using the following instruments:

Sociodemographic data form: This structured form collected information on demographic characteristics (age, sex, education level, height, body weight, body mass index (BMI), smoking and alcohol use), clinical features (headache duration, frequency, duration of attacks, precipitating factors, analgesic use and response, known comorbidities, and neurological findings), and work schedule. Headache frequency was recorded as ≤4 or >4 attacks per month.

Visual Analogue Scale (VAS): Pain intensity was self-rated by the participants on a 0–10 scale, where 0 indicates no pain and 10 indicates the worst imaginable pain [14,15].

Morningness–Eveningness Questionnaire (MEQ): The MEQ is a self-reported questionnaire used to assess individual preferences for morningness and eveningness. Chronotype of the participants was classified based on total MEQ scores. Respondents indicate their preferred times for sleep, wake, and activity, reflecting their internal circadian rhythm rather than their actual schedule. The questionnaire consists of 19 items, each scored on a 4–5 point multiple-choice scale. Total scores range from 16 to 86. Scores ≤ 41 indicate an evening type, scores ≥ 59 indicate a morning chronotype, and scores between 42 and 58 indicate an intermediate chronotype [16]. The Turkish adaptation of the tool showed a test–retest reliability coefficient of 0.84 [17].

These items illustrate the preferential nature of the MEQ and the method used to classify participants as morning, intermediate, or evening chronotypes.

Hospital Anxiety and Depression Scale (HADS): The HADS is a 14-item self-report questionnaire developed by Zigmond and Snaith to assess anxiety and depressive symptoms (7 items each) [18]. Each item is scored from 0 to 3, yielding a total score of 0–21 for either anxiety or depression. Scores of 8–10 indicate a moderate level of symptoms, while scores above 11 suggest a clinically significant level of symptoms that may correspond to a clinical diagnosis [18]. Validity and reliability of the Turkish version of the scale were demonstrated by Aydemir et al., with Cronbach's alpha coefficients of 0.85 for the anxiety subscale and 0.77 for the depression subscale [19].

Pittsburgh Sleep Quality Index (PSQI): The PSQI is a widely used self-report questionnaire that assesses sleep quality and disturbances over the previous month [20]. It consists of 19 items generating seven component scores: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleep medication, and daytime dysfunction. Each component is scored from 0 to 3. The component scores are summed to yield a total score ranging from 0 to 21. The original developers reported an internal consistency of Cronbach's alpha = 0.83, a test–retest reliability of 0.85 for the global scale, a sensitivity of 89.6%, and a specificity of 86.5% [20]. The Turkish version developed by Ağargün et al. demonstrated good internal consistency (Cronbach's alpha = 0.80). The total PSQI score clearly distinguishes between good sleepers (total score ≤ 5) and poor sleepers (total score > 5) [21]. The questionnaire has been validated in various clinical populations, including patients with major depressive disorder, sleep disorders, cancer, and fibromyalgia [22].

Short Form 36 (SF-36): The SF-36 is a self-assessment questionnaire developed to evaluate health-related quality of life [23]. It consists of 36 items across eight dimensions: physical functioning (10 items), social functioning (2 items), role limitations due to physical problems (4 items), role limitations due to emotional problems (3 items), mental health (5 items), energy/vitality (4 items), bodily pain (2 items), and general health perceptions (5 items) The assessment considers experiences over the past four weeks and is widely used to compare general and specific populations, evaluate the effects of interventions, and monitor patients over time.

Validity and reliability of the Turkish SF-36 questionnaire have been demonstrated by Kocyigit et al. [24]. The Turkish SF-36 demonstrated high internal consistency, with Cronbach's alpha values for the subscales ranging from 0.792 to 0.992. Item-total correlations were: physical functioning 0.436–0.840, role limitations (physical) 0.887–0.895, bodily pain 0.861–0.958, general health perceptions 0.564–0.892, vitality 0.702–0.841, social functioning 0.949–0.952, role limitations (emotional) 0.396–0.473, and mental health 0.456–0.824. Convergent validity was supported by significant correlations between SF-36 subscales and the Turkish HADS, confirming the scale's sensitivity to both physical and mental health dimensions [24].

Dependent Variables (Outcomes): The primary outcomes of interest were pain intensity, assessed using the VAS, and bodily pain, measured by the SF-36 Bodily Pain (BP) subscale. These outcomes were selected as they represent the core clinical manifestations of TTH and are widely used in headache and chronic pain research.

Independent Variables (Predictors): The primary explanatory variable was chronotype, assessed by the MEQ score. MEQ scores were treated both as a numeric variable and as categorical groups (evening, intermediate, and morning chronotypes) for descriptive and inferential analyses. Secondary predictors (covariates) included sex, HADS-A score (anxiety), and HADS-D score (depression).

Headache-Specific Variables: These variables included duration of pain (in months), duration of TTH attacks (hours), frequency of TTH attacks (number per month), circadian pattern of attacks, and analgesic use and response. Frequency of TTH attacks was further dichotomized as ≤4 or >4 per month based on prior literature indicating that ≥4 attacks per month represents a clinically relevant threshold for recurrent TTH [25]. These variables were collected for descriptive purposes and were not treated as explanatory variables in the regression models.

Descriptive Sociodemographic and Clinical Variables: Sociodemographic and clinical characteristics, including age, sex, BMI, education level, smoking status, alcohol use, work schedule (shift work), comorbidities, were collected to describe the sample.

Outcomes for Regression Models: Two separate linear regression models were constructed: Model 1: Outcome = SF-36 BP score Model 2: Outcome = VAS score

Continuous variables were reported as mean ± standard deviation (SD) when normally distributed and as median [25th–75th percentile] when non-normally distributed. Group differences in normally distributed continuous variables (e.g., BMI) were evaluated using one-way analysis of variance ANOVA with Tukey's post hoc tests for pairwise comparisons. Non-normally distributed or ordinal variables (e.g., frequency of analgesic use and analgesic response) were analyzed using the Kruskal–Wallis test. Categorical variables, including the circadian pattern of headache attacks, were compared using contingency tables, with Fisher's exact test applied when expected cell counts were low.

Primary analyses were conducted using linear regression models to examine associations between predictors and outcomes. Dependent variables (outcomes) included VAS and SF-36 BP scores. These outcomes were chosen as they represent the core clinical manifestations of TTH and are widely used in headache and chronic pain research. The primary explanatory variable was MEQ score (chronotype), treated as a numeric variable, while sex, HADS-A score, and HADS-D score were included as covariates to account for potential confounding effects. The models did not include MEQ score as a moderator or mediator. These variables were included in the regression models because SF-36 BP score represents the pain component of quality of life, VAS score reflects subjective pain intensity, and prior literature indicates that sex, anxiety, and depression are associated with pain perception and chronotype differences in headache and chronic pain populations [8,9]. Age was evaluated in preliminary analyses but was not included as a covariate because it did not show significant associations with the outcomes. Standardized beta coefficients (β) were reported to allow direct comparison of the relative magnitude of associations across predictors. Separate models were constructed for each outcome (Model 1: SF-36 BP; Model 2: VAS). Model performance was evaluated using R² and Akaike Information Criterion (AIC), with lower AIC values indicating better fit while penalizing additional variables.

All statistical tests were two-tailed, and p-values < 0.05 were considered statistically significant. All analyses were conducted using R software version 3.6.0 (R Core Team, 2019; R Foundation for Statistical Computing, Vienna, Austria).

A total of 77 participants, including 63 women and 14 men aged 18 to 52, were included in the study. Fifty-five participants (71.4%) had an intermediate chronotype, 14 (18.2%) had a morning chronotype, and 8 (10.4%) had an evening chronotype ().

No significant differences in age, sex, education level, smoking, alcohol use, shift work, comorbidities, pain duration, duration or frequency of TTH attacks were observed across chronotype groups.

A one-way ANOVA indicated a significant effect of chronotype on BMI (F(2, 74) = 3.52,= 0.035, η= 0.087) (). Post-hoc comparisons showed that participants with an evening chronotype had a significantly lower BMI than those with an intermediate chronotype (= 0.033) and a trend toward lower BMI relative to morning chronotypes (= 0.053). No significant difference was observed between the morning and intermediate chronotypes. p p p 2

A contingency table analysis of the circadian pattern of TTH attacks across chronotype groups showed that 37.5% (= 3) of participants with an evening chronotype exhibited a circadian pattern, compared with 74.5% (= 41) of those with an intermediate chronotype and 78.6% (= 11) of those with a morning chronotype. Fisher's exact test indicated that this association was not significant (= 0.089) (). The timing of peak headache intensity varied by chronotype. Among evening chronotypes, peak intensity was evenly distributed across early morning (00:00–06:00, 33%), morning (06:00–12:00, 33%), and afternoon (12:00–18:00, 33%) intervals. For participants with a morning chronotype, peak intensity was most frequently reported in the evening (18:00–00:00, 45%), followed by the afternoon (12:00–18:00, 36%), with fewer reports in the early morning (9%) and morning (9%) intervals. In the intermediate chronotype group, peak intensity was most common in the afternoon (12:00–18:00, 34%) and evening (18:00–00:00, 29%), and less frequent in the morning (22%) and early morning (15%) intervals. n n n p

Analysis of analgesic use frequency across chronotype groups showed that among participants with an evening chronotype, most reported no use (62.5%), while rare use, use once every few weeks, or once a week were equally reported (12.5% each). In the intermediate chronotype group, the majority reported no use (56.4%), followed by rare use (18.2%), occasional use once every few weeks or once a week (1.8% each), use once every few days (9.1%), once daily (3.6%), and several times daily (9.1%). Among participants with a morning chronotype, most reported no use (71.4%), followed by rare use (14.3%), and occasional use once every few weeks or once every few days (7.1% each). Overall, the majority of participants (59.7%) reported no analgesic use. A Kruskal–Wallis test revealed no significant differences among the chronotype groups ().

The most commonly reported analgesic response was "occasional" (Evening chronotype: 75.0%; Intermediate chronotype: 58.2%; Morning chronotype: 50.0%). Kruskal–Wallis test showed no significant differences across chronotype groups ().

A Kruskal–Wallis test indicated a significant difference in SF-36 Role limitations due to physical problems (RP) scores across chronotype groups (ε= 0.0852). Post-hoc Dwass–Steel–Critchlow–Fligner comparisons showed that the evening chronotypes scored significantly lower than the morning chronotypes (W = 3.47,= 0.038), with no significant differences between the evening and intermediate chronotypes (W = 2.37,= 0.214) or intermediate and morning (W = 2.32,= 0.229) chronotypes (). 2 p p p

Linear regression analysis using SF-36 BP score as the dependent variable, MEQ score (chronotype) as the primary predictor, and sex, HADS-A, and HADS-D scores as covariates yielded a significant model, with all model assumptions satisfactorily met (Model 1,).

A separate linear regression model using VAS score as the dependent variable and MEQ score, sex, HADS-A, and HADS-D scores as predictors was also significant. MEQ score (chronotype) was not a significant predictor, whereas sex was significant, with higher VAS scores observed in females (Model 2,).

shows the correlation matrix that illustrates the relationships between chronotype, anxiety, depression, pain and quality of life. Supplementary Table S1