Skip to main content
. Author manuscript; available in PMC: 2015 Sep 1.
Published in final edited form as: Epidemiology. 2014 Sep;25(5):697–706. doi: 10.1097/EDE.0000000000000137

Table 2.

Risk of breast cancer associated with indicators of indoor light-at-night among 106,731 study participants: adjusted hazard ratios (HRs) and 95% confidence intervals (95% CI) estimated from Cox proportional hazard models.a

Indoor light-at-night Category No. Casesb HR (95% CI) Test for (trend)
Uses light Non-user 4,869 1.0
Yes 226 1.03 (0.90–1.18) -
Duration (months) Non-user 4,869 1.0
0–3 97 1.09 (0.89–1.33)
4–6 19 0.92 (0.59–1.45)
7–9 11 0.79 (0.44–1.43)
10+ 80 1.08 (0.87–1.35) P = 0.62
Days/week Non-user 4,869 1.0
1–3 122 1.14 (0.95–1.37)
4–5 25 0.87 (0.59–1.30)
6+ 60 0.97 (0.75–1.25) P = 0.97
Hours/day Non-user 4,869 1.0
1–2 66 1.01 (0.79–1.29)
3–4 43 1.07 (0.80–1.45)
5–6 43 1.03 (0.76 to 1.40)
7+ 56 1.06 (0.81–1.38) P = 0.55
Summary c Non-user 4,869 1.0
Light 45 1.17 (0.87–1.57)
Medium 109 0.99 (0.82–1.20)
Heavy 44 1.13 (0.84–1.52) P = 0.53
a

adjusted for: age strata (1-year), race/birthplace, family history of breast cancer, age at menarche, pregnancy history, breast feeding history, physical activity, BMI, alcohol consumption, menopausal status and hormone therapyuse combination, smoking status, and smoking pack years, neighborhood SES and urbanization.

b

Number of cases does not always sum to total (n=5,095) due to missing/unknown values for light-at-night usage.

c

Summary : light = uses light 0–3 months and 1–3 days/week and 1–2 hours/night; heavy = uses light 10+ months and 5+ days/week and 7+ hours/day; medium = all other combinations of monthly, weekly, and hourly usage.