Sleep Positional Therapy across Clinical Disorders: Mechanisms, Evidence and a Position-Prescription Framework

Background: Sleep position is seldom considered in clinical care, although body position changes upper-airway geometry, lung volume, cardiac loading, gastro-oesophageal anatomy and uteroplacental blood flow. It is a low-cost and modifiable factor. Aims: To summarise the clinical evidence and physiological mechanisms for sleep positional therapy across disorders, to grade the certainty of this evidence and to propose a condition-specific position-prescription framework. Methods: This narrative review followed the Scale for the Assessment of Narrative Review Articles (SANRA). PubMed, Google Scholar and Scopus were searched for English-language studies published from January 2000 to August 2026 using combinations of sleep, position and condition terms, including Medical Subject Headings. Each study was assigned a design-based level of evidence (I–V), and the overall certainty for each recommendation was rated as high, moderate, low or very low using principles adapted from GRADE. Results: Moderate-certainty evidence from randomised trials and meta-analyses shows that positional devices reduce supine sleep and the apnoea-hypopnoea index in mild-to-moderate positional obstructive sleep apnoea, although to a lesser extent than continuous positive airway pressure, and that avoiding the right side reduces nocturnal reflux. Low-certainty observational evidence links the supine going-to-sleep position in late pregnancy with stillbirth, and the prone position with sudden unexpected death in epilepsy. Evidence for heart failure and central sleep apnoea is low, and evidence for neurodegeneration and glaucoma is very low, coming mainly from small physiological, cross-sectional and animal studies. Infants are recommended to sleep supine. Conclusion: Sleep position can be used as a condition-specific, adjunctive measure. Positional advice is best supported in positional obstructive sleep apnoea and nocturnal reflux; in other conditions, it should be regarded as precautionary or hypothesis-generating until larger trials with clinical outcomes are available.