Sleep Characteristics and Cancer-Related Outcomes: An Umbrella Review of Systematic Reviews and Meta-Analyses of Observational Studies.

Abstract

Sleep is closely related to various diseases. Several meta-analyses have provided evidence of sleep and cancer, and yet the credibility of this evidence has not been comprehensively quantified. Thus, we conducted an umbrella review to quantify the evidence for systematic reviews and meta-analyses of observational studies on sleep characteristics (sleep duration, sleep quality, napping, bedtime, and wake-up time) and cancer-related outcomes. PubMed, Web of Science (Core Collection), and Embase databases were searched from inception until 29 July 2022. Assessment of Multiple Systematic Reviews, version 1, was used to evaluate the methodological quality of each eligible systematic review or meta-analysis. For each association, the summary effect with a 95% confidence interval was evaluated by fixed and random effects models. The 95% prediction interval, heterogeneity, small-study effects, and excess significance bias were also evaluated. Evidence of the associations from systematic reviews and meta-analyses was ranked based on the established criteria of published literature as convincing, highly suggestive, suggestive, weak, or non-significant. The umbrella review identified thirty meta-analyses on the aforementioned associations from six articles. The methodological quality of five articles was high or moderate. Suggestive evidence was found for associations between long sleep duration and a 21% increased risk of colorectal cancer, a 9% increased all-cancer mortality and a 65% increased mortality of lung cancer, and associations between short sleep duration and a 21% increased mortality of lung cancer. Additionally, the evidence of associations between short sleep duration and lung cancer mortality was upgraded to convincing, and between long sleep duration and lung cancer mortality was upgraded to highly suggestive, among the population reporting 24 h sleep duration. Abnormal sleep duration might be linked to several adverse cancer-related outcomes.