Table 5.

Summary of mediation analyses

	Association with shift worka	Incident CVD		Fatal CVD
		Association with outcomeb	% mediated (95% CI)	Association with outcomeb	% mediated (95% CI)
Totalc			52.3		14.1
MET-min per week	+	0		0
Hours of TV viewing	0	+		+
Portions of red-meat intake	+	0		0
No oily-fish intake	0	+		0
Frequent processed-meat intake	0	0		0
Portions of fruit/vegetable intake	+	0		0
Units of alcohol intake	0	0		0
Current smoker	+	+	14.1 (7.2–65)	+	8.3 (3.2–43)
Short sleeper	+	+	6.2 (2.4–41)	0
Sleep disturbance	+	+	6.1 (2.2–35)	0
Frequency of social visits	0	0		−
Obesity	+	+	4.9 (1.7–32)	+	2.3 (−8.5, 16)
Central obesity	+	+	4.9 (1.4–23)	+	2.9 (−10.5, 25)
Grip strength	−	−	0.0 (−4.3, 3.0)	−	0.2 (−0.8, 2)
Systolic blood pressure	−	+		+
LDL cholesterol	0	+		0
Lipoprotein(a)	0	+		+
HbA1c	+	+	10.7 (5.0–41)	+	5.8 (2.4–33)
Cystatin C	+	+	5.5 (1.5–28)	+	3.1 (−12, 19)
GGT	0	+		+

⁺positive association; − negative association; 0 no significant association; CVD, cardiovascular disease.

^aSummarized from Supplementary Table S2 (available as Supplementary data at IJE online); multiple linear or logistic-regression models with the potential mediator as the dependent variable and shift work as the independent variable. Adjusted for each other and for age, sex, education, deprivation, hours of work per week, duration of current job, walking/standing at work and heavy manual/physical work.

^bSummarized from Supplementary Table S3 (available as Supplementary data at IJE online); Cox regression models with potential mediators as independent variables. Adjusted for each other and for shift work, age, sex, education, deprivation, hours of work per week, duration of current job, walking/standing at work and heavy manual/physical work.

^cTotal %s mediated were the sum of %s from all significant mediators. This is based on all mediators being mutually adjusted in analyses.