Summary Background: The interaction between physical activity and sleep with cardiovascular disease remains poorly understood, despite both being key risk factors. This study investigated the independent and joint associations of device-measured physical activity and sleep duration with incident major adverse cardiovascular events (MACE). Methods: Prospective analysis of UK Biobank participants who wore a wrist-based accelerometer for seven days. Open-source machine learning algorithms derived daily step count and overnight sleep duration. The outcome was incident MACE (cardiovascular death, non-fatal myocardial infarction or stroke, or revascularisation procedure), identified through electronic health record linkage. Cox proportional hazards models were used to examine independent and joint associations of median daily step count (low 11,000) and median overnight sleep duration (short 7.5 hours) with incident MACE. Findings: Among 88,012 participants (mean age 62.2 years SD 7.8), 3,817 were diagnosed with MACE during follow-up (median 7.9 years IQR 7.3-8.4). Fewer daily steps and short sleep duration were independently associated with a higher risk of MACE, but there was no evidence of an interaction between step count and sleep duration (Pinteraction = 0.42). Compared with the reference group - participants with high step count and intermediate sleep duration - the highest risk of MACE was observed in participants with both low step count and short sleep duration (HR: 1.84, 95% CI: 1.62-2.10). Interpretation: The results of this study show that higher levels of physical activity do not fully attenuate the higher risk of cardiovascular disease associated with short sleep duration, reinforcing the importance of sufficient levels of both physical activity and sleep for the prevention of cardiovascular disease. Funding: Wellcome Trust (223100/Z/21/Z). Copyright: For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission.
Brocklebank et al. (Wed,) studied this question.