Heart Rate Variability in Insomnia and Poor Sleep Quality versus Controls: An Updated Meta-Analysis of Case-Control and Cross-Sectional Studies Using Standardised Mean Differences

Abstract

Introduction: Chronic insomnia is linked to autonomic dysregulation and to elevated cardiovascular and cerebrovascular risk. Reduced heart rate variability (HRV), particularly its vagal component, has been proposed as a peripheral marker of the central hyperarousal of insomnia, but the quantitative evidence is inconsistent and the only prior pooled estimate used an unconventional method. This study aimed to provide an updated, conventionally pooled estimate of the HRV difference between adults with insomnia or poor sleep and controls, and to test whether it depends on insomnia severity.

Methods: PubMed/MEDLINE was searched (cross-checked against Scopus and Web of Science) for case-control and cross-sectional studies comparing HRV between adults with insomnia or poor sleep and non-insomnia controls. Six studies with verified, openly accessible, extractable per-group data were pooled. Hedges’ g standardised mean differences were combined under a random-effects model (DerSimonian-Laird), with restricted maximum likelihood and the Hartung-Knapp adjustment as robustness checks. The primary outcome was composite vagal HRV; secondary outcomes were SDNN, the LF/HF ratio and mean heart rate. A pre-specified severity subgroup, leave-one-out, prediction interval, Egger test and Newcastle-Ottawa appraisal were performed. The review was not registered.

Results: Six studies (484 with insomnia/poor sleep; 365 controls) were pooled. Composite vagal HRV was lower in insomnia/poor sleep (g = -0.58, 95% CI -1.19 to 0.03; I² = 93.7%). The effect was concentrated in clinically diagnosed insomnia (g = -1.04) and near-null in poor sleep (g = -0.10). Removing one elite-athlete outlier yielded a significant, homogeneous estimate (g = -0.29, 95% CI -0.56 to -0.01, p = 0.039; I² = 59%). SDNN, LF/HF and heart rate did not differ. The estimate bracketed the previously published value (SMD -0.41).

Conclusion: Vagal HRV is reduced in insomnia and poor sleep, with a clinically meaningful effect confined to diagnosed insomnia disorder. Low vagal HRV is a candidate autonomic marker linking insomnia to vascular risk, although high heterogeneity and low certainty warrant cautious interpretation.