Table 2.
Associations of current shift work (n = 189,488) and morningness-eveningness preference (n = 169,926) on adjusted odds of type 2 diabetes and adjusted mean difference in HbA1c
| Type 2 diabetes | HbA1c (mmol/mol) | |||||
|---|---|---|---|---|---|---|
| Type 2 diabetes case/control subjects, n/n | Sex- and age-adjusted OR (95% CI) | Multivariable-adjusted OR (95% CI) | N | Sex- and age- adjusted β (95% CI) | Multivariable-adjusted β (95% CI) | |
| Shift work | ||||||
| Day workers | 4,047/154,792 | Reference | Reference | 146,993 | Reference | Reference |
| Shift work without nights | 475/14,863 | 1.26 (1.15–1.39) | 0.99 (0.88–1.12) | 14,110 | 0.34 (0.28–0.40) | 0.14 (0.08–0.20) |
| Sometimes night shift work | 284/8,434 | 1.33 (1.17–1.50) | 1.01 (0.87–1.17) | 8,005 | 0.48 (0.40–0.56) | 0.24 (0.16–0.32) |
| Usual night shift work | 80/2,171 | 1.48 (1.18–1.86) | 1.12 (0.84–1.49) | 2,069 | 0.44 (0.29–0.60) | 0.20 (0.05–0.35) |
| Always night shift work | 156/4,186 | 1.47 (1.24–1.73) | 1.01 (0.82–1.24) | 3,979 | 0.75 (0.64–0.86) | 0.38 (0.27–0.49) |
| Morningness-eveningness preference | ||||||
| Definite morning | 1,272/42,097 | Reference | Reference | 39,976 | Reference | Reference |
| More morning than evening | 1,482/60,064 | 0.86 (0.80–0.93) | 0.93 (0.85–1.03) | 57,127 | −0.13 (−0.18 to 0.09) | −0.02 (−0.07 to 0.03) |
| More evening than morning | 1,268/48,593 | 0.96 (0.89–1.04) | 1.02 (0.93–1.13) | 46,267 | −0.04 (−0.09 to 0.01) | 0.04 (0–0.09) |
| Definite evening | 497/14,653 | 1.30 (1.17–1.45) | 1.29 (1.13–1.47) | 13,886 | 0.12 (0.05–0.19) | 0.14 (0.07–0.21) |
Associations are across shift work and morningness-eveningness preference categories in employed UK Biobank participants of European descent. Prevalent type 2 diabetes associations are sex- and age-adjusted ORs (95% CI), then further adjusted for BMI (continuous) and other previously established covariates (family history of type 2 diabetes [yes/no], Townsend deprivation index [continuous], alcohol consumption [never, once/week, two to three times/week, four to six times/week, or daily], physical activity [continuous, METs], hypertension [yes/no], hypertension medication use [yes/no], hypercholesterolemia [yes/no], and lipid-lowering medication intake [yes/no]). HbA1c associations are restricted to participants with no prevalent type 2 diabetes. HbA1c associations are sex- and age-adjusted βs (95% CI) in mmol/mol, then further adjusted for BMI and other previously established covariates. In all analyses, day workers or definite morning participants serve as reference group. Boldface type indicates P < 0.05.