Sleep patterns, genetic predisposition, and risk of chronic liver disease: A prospective study of 408,560 UK Biobank participants

Little is known about the role that combined sleep behaviors play in the association with chronic liver disease (CLD) risk. We included 408,560 participants initially free of CLD from the UK Biobank. A healthy sleep pattern was defined by early chronotype, sleep duration of 7–8 h/day, no insomnia, no snoring, and no excessive daytime sleepiness. Cox regression models were used to examine the association of healthy sleep pattern with incident CLD and their interaction with PNPLA3 genetic risk. During a median 12.5 years of follow-up, we documented 10,915 incident all-cause CLD cases, including 388 viral hepatitis, 4782 non-alcoholic fatty liver disease (NAFLD), 1356 cirrhosis, 973 alcoholic liver disease, and 725 liver cancer cases. Compared to participants with a healthy sleep score of 0–1, the hazard ratio (HR) (95 % confidence interval [CI]) for those with a sleep score of 5 was 0.54 (0.49, 0.60) for CLD, 0.52 (0.30, 0.90) for viral hepatitis, 0.47 (0.41, 0.55) for NAFLD, 0.57 (0.43, 0.75) for cirrhosis, 0.32 (0.23, 0.44) for alcoholic liver disease, and 0.53 (0.37, 0.77) for liver cancer. Healthy sleep pattern and PNPLA3 genetic risk exerted significant additive effects on CLD risk (relative excess risk due to the interaction: 0.05; attributable proportion due to the interaction: 13 %). Measurement error was unavoidable for self-reported data on sleep behaviors. Our analyses provide evidence that healthy sleep pattern was inversely associated with the development of CLD, and participants with higher genetic risk were more likely to develop CLD when exposed to the unhealthy sleep pattern.