. Author manuscript; available in PMC: 2020 Jul 1.

Published in final edited form as: Cancer Epidemiol Biomarkers Prev. 2019 May 29;28(7):1177–1186. doi: 10.1158/1055-9965.EPI-18-1018

Table 2.

Geometric means of light exposure and urinary melatonin levels across shift work categories^1,2,3

	Day-shift workers	Rotating-shift workers
		Day/off shifts	Night shifts

	Geometric means (95% CI)	Geometric means (95% CI)	Geometric means (95% CI)
Light exposure

Average light exposure over 24 hours (in lux)	128.1 (87.1–188.3)^a,b	50.5 (24.0–106.6)^a,c	84.7 (55.5–129.3)^b,c

Average light exposure over 24 hours, by time period (in lux)
7:00–10:59	171.5 (133.6–220.1)^a,b	72.1 (45.3–114.8)^a	68.0 (48.4–95.5)^b
11:00–14:59	272.7 (208.6–356.6)^a,b	123.0 (74.5–203.2)^a,c	30.2 (17.9–51.0)^b,c
15:00–18:59	204.0 (152.5–272.8)^b	109.6 (68.6–175.1)	64.5 (45.4–91.6)^b
19:00–22:59	28.7 (22.3–36.8)^a	14.6 (10.4–20.6)^a,c	36.8 (27.6–48.9)^c
23:00–6:59	3.0 (2.2–4.0)^a,b	1.6 (1.0–2.4)^a,c	42.9 (31.8–58.0)^b,c

Duration of light exposure at intensity >20 lux (in hours)
7:00–10:59	3.6 (3.5–3.8)^a,b	2.9 (2.6–3.1)^a,c	2.4 (2.2–2.6)^b,c
11:00–14:59	3.8 (3.7–4.0)^a,b	3.3 (3.0–3.5)^a,c	2.3 (2.0–2.5)^b,c
15:00–18:59	3.4 (3.2–3.6)^a,b	3.0 (2.7–3.3)^a	2.7 (2.5–3.0)^b
19:00–22:59	1.9 (1.7–2.2)^b	1.6 (1.3–1.9)^c	2.9 (2.6–3.1)^b,c
23:00–6:59	0.8 (0.6–0.9)^b	1.1 (0.7–1.5)^c	6.1 (5.6–6.6)^b,c

Melatonin levels

Total urinary melatonin excretion over 24 hours (in ng)	22,822 (20,267–25,698)	20,618 (17,699–24,018)	20,938 (18,500–23,697)

Urinary melatonin concentration over 24 hours by time period (in ng/mg-creatinine)
7:00–10:59	18.6 (16.5–21.1)	20.7 (17.6–24.2)	16.5 (14.2–19.0)
11:00–14:59	8.4 (7.4–9.5)	7.6 (6.1–9.5)	9.1 (7.7–10.7)
15:00–18:59	6.7 (5.8–7.6)	6.3 (5.4–7.4)	7.0 (5.9–8.1)
19:00–22:59	9.4 (8.4–10.6)	9.3 (7.8–11.2)	8.9 (7.7–10.3)
23:00–6:59	32.9 (29.3–37.0)^b	30.0 (25.6–35.3)^c	20.9 (18.4–23.7)^b,c

CI=confidence interval

Data from rotating-shift workers was divided into categories of “night shifts” or “day/off shifts” based on whether participants were working (or not working) night shifts during each day of the study.

Results were obtained using ANOVA with Bonferroni correction.

Statistically significant differences for group comparisons are indicated as follows:

for day-shift workers versus rotating-shift workers on day/off shifts

for day-shift workers versus rotating-shift workers on night shifts

for rotating-shift workers on day/off shifts versus rotating-shift workers on night shifts.