Sleep disturbances in tension-type headache: a systematic review and meta-analysis

Clinical Snapshot

Key Findings

• P Value: PSQI MD vs controls: p<0.01; OR for poor sleep vs controls: p=0.08 (non-significant); depression correlation FDR q=0.005

• Effect Size: Mean PSQI score 9.48 (above the poor sleep threshold of 5); mean difference vs non-headache controls: MD=3.77 (95% CI 2.32–5.22); OR for poor sleep quality vs controls: 2.39 (95% CI 0.83–6.86). Insomnia prevalence 37%; excessive daytime sleepiness 13%; high OSA risk or PSG-defined OSA 26%. Depression-sleep correlation r=0.51 (FDR q=0.005)

• Primary Outcome: Pooled prevalence of poor sleep quality (PSQI ≥5) in adults with tension-type headache: 67% (95% CI 53–79%; I²=97%)

• Nnt Or Sensitivity: Not applicable (observational prevalence meta-analysis). Mean sleep duration 6.61 hours; no significant difference from controls for sleep duration. Estimates for insomnia, EDS, and OSA described as imprecise due to limited contributing studies.

• Confidence Interval: Primary prevalence: 95% CI 53–79%; PSQI MD: 95% CI 2.32–5.22; OR: 95% CI 0.83–6.86

Clinical Application

Routine PSQI administration in TTH clinical assessment is feasible, low-cost, and takes approximately 5–10 minutes. A mean PSQI of 9.48 in TTH patients substantially exceeds the clinical threshold of 5, supporting its use as a screening tool. The strong association with depression (r=0.51) suggests concurrent depression screening (e.g., PHQ-9) is warranted. OSA screening tools (e.g., STOP-BANG) may be appropriate given the 26% OSA risk prevalence, though estimates are imprecise. In Australian primary care, TTH is managed predominantly within the RACGP framework, which emphasises non-pharmacological approaches including sleep hygiene, stress management, and physiotherapy. The PSQI is not routinely administered in Australian GP consultations but is available without cost or licensing restrictions. PBS-listed treatments for TTH (simple analgesics, amitriptyline for chronic TTH) do not specifically target sleep, though amitriptyline has sedating properties that may incidentally benefit sleep quality. TGA-approved sleep aids (e.g., low-dose doxepin, melatonin) are not specifically indicated for TTH-associated sleep disturbance. The strong depression comorbidity signal is clinically relevant given Australia's high burden of comorbid pain and mood disorders; RACGP guidelines support integrated assessment of mood in chronic pain conditions. Referral pathways to sleep medicine for OSA evaluation are available through the MBS, though the imprecise OSA prevalence estimate limits the strength of this recommendation. Adults presenting to primary care or neurology/headache clinics with a diagnosis of tension-type headache (episodic or chronic) per ICHD criteria. Findings are most applicable to clinic-attending patients; community-based TTH prevalence of sleep disturbance is likely lower given recruitment-setting heterogeneity.

Abstract

BACKGROUND: Tension-type headache (TTH) is the most common primary headache, yet accompanying sleep disturbances have not been quantitatively synthesized. METHODS: PubMed, Embase, Web of Science, and the Cochrane Library were searched from inception to January 3, 2026, for observational studies reporting sleep outcomes in adults with TTH. The primary outcome was the prevalence of poor sleep quality assessed with the Pittsburgh sleep quality index (PSQI). Pooled estimates were calculated using random-effects models. Study quality was assessed with Joanna Briggs Institute checklists. RESULTS: Twenty-seven studies were included. Pooled prevalence of poor sleep quality in TTH was 67% (95% CI 53-79%; I2 = 97%), stable across sensitivity and trim-and-fill analyses. Heterogeneity appeared partly related to recruitment settings. In five studies with non-headache controls, the odds of poor sleep quality were elevated but not significant (OR = 2.39; 95% CI 0.83-6.86; p = 0.08). The pooled mean PSQI score was 9.48 and significantly higher than controls (MD = 3.77; 95% CI 2.32-5.22; p < 0.01). Mean sleep duration was 6.61 h and did not differ significantly from controls. Insomnia was reported in 37%, excessive daytime sleepiness in 13%, and high OSA risk or PSG-defined OSA in 26%, with imprecise estimates. PSQI scores correlated robustly with depression (r = 0.51; FDR q = 0.005), whereas correlations with headache frequency and intensity were nominal and exploratory. CONCLUSIONS: Poor subjective sleep quality is common in TTH, particularly among patients presenting to clinical settings, and is associated with depressive symptoms, with exploratory associations with headache burden, supporting consideration of sleep assessment in the clinical evaluation of TTH.