Impact of Shift Work on Dementia: a Systematic Review and Dose-Response Meta-Analysis

Yun Gao,Xiaoli Fu,Hongtao Hu,T. Li,Ling Yuan,J. Zhang,Yile Wu,Maggie Haitian Wang,Ke Yang,Yong Li,Fulan Hu,Ming Zhang,Liang Sun,Hao Wen,Ruili Guan,Peng Gao,Wen Lin Chai,Yang Zhao,Dongsheng Hu
DOI: https://doi.org/10.1016/j.puhe.2023.07.029
IF: 4.984
2023-01-01
Public Health
Abstract:Although shift work has been reported as having a link to dementia, evidence remains inconsistent, and a comprehensive dose-response meta-analysis of the association is still lacking. We therefore conducted this meta-analysis to explore the association between shift work and the risk of dementia. Systematic review and dose-response meta-analysis. PubMed, Embase, and Web of Science databases were systematically searched. Fixed or random-effects models were used to estimate the summary relative risks (RRs) and 95% confidence intervals (95% CIs). Generalized least squares regression was used to estimate dose-response associations, and restricted cubic splines were used to examine possible linear or non-linear associations. Five articles (10 studies) with 72,999 participants and 23,067 cases were eventually included in the meta-analysis. The summary RRs and 95% CIs of dementia risk with shift work and night shift work versus daytime work were 1.13 (95% CI: 1.05–1.21, I2 = 46.70%) and 1.13 (95% CI: 1.03–1.24, I2 = 9.20%), respectively. The risk of dementia increased by 1% (RR = 1.01, 95% CI: 1.01–1.02, I2 = 41.3%) with each 1-year increase in the duration of shift work. We found a non-linear dose-response association between the duration of shift work and the risk of dementia (Pnon-linearity = 0.006). Though the shape of the curve was steeper with the duration of shift work <7 years, the increase was more gradual after 7 years. Our findings suggest that shift work may be a risk factor for future dementia and that controlling the length of shift work is a feasible measure that may contribute to prevent dementia.
What problem does this paper attempt to address?