ExploreMenu
Get involved

Claim

Opening the claim…

In 2021 to 2023, the coefficient of serum albumin on workday sleep of 5 hours or less, against more than 7 and up to 8 hours, is -0.27 g/L (95% confidence interval -0.57 to 0.04), against a published -1.00 g/L (-1.26 to -0.74): a smaller decrement than published, with an interval that includes zero, and a significant difference from it (corrected p 0.0039, or 0.037 by t test). With the phlebotomy weight in place of the interview weight the paper used, it is -0.21 g/L (-0.53 to 0.11), still significantly different from the published estimate (corrected p 0.0029, or 0.025 by t test). These tests take the published standard error from its interval; against our harmonized estimate on the paper's own cycles, the 2021 to 2023 estimate does not differ significantly (corrected p 0.052). On the paper's own cycles, adding a term for the cycle to its model, which absorbs every difference between its two cycles, gives -0.64 g/L, and putting 2017 to 2018's albumin on 2015 to 2016's analyzer scale by NCHS's bridging equation gives -0.75 g/L (-1.10 to -0.41). · sciencejournal.ai