Skip to main content
. 2014 Sep 11;111(9):1852–1859. doi: 10.1038/bjc.2014.473

Table 3. Aspirin use and UADT cancer risk by demographic and behavioural factors.

Characteristics OR (95% CI) Test for heterogeneity P-value
Gender
M 1.02 (0.86, 1.22)a
F 0.72 (0.56, 0.91) P=0.019
Age (years)
18–55 0.78 (0.46, 1.32)b
56–65 0.91 (0.68, 1.21)
66–75 0.96 (0.77, 1.19)
76+ 0.81 (0.64, 1.02) P=0.269
Carstairs deprivation category
1 (Least deprived) 0.90 (0.45, 1.77)
2 0.86 (0.53, 1.41)b
3 0.94 (0.69, 1.40)
4 0.87 (0.68, 1.10)
5 0.77 (0.55, 1.08)
6 1.14 (0.78, 1.69)
7 (Most deprived) 0.85 (0.57, 1.27) P=0.438
Body mass index
30.0–54.0 1.08 (0.77, 1.50)b
25.0–29.9 1.32 (1.02, 1.71)
18.5–24.9 0.88 (0.66, 1.17)
15.5–18.4 0.75 (0.13, 4.26)      P=0.164
Missing 0.65 (0.51, 0.84)
Smoking
Never 1.00 (0.76. 1.32)b
Ever 0.88 (0.74, 1.05)      P=0.108
Missing 0.54 (0.37, 0.79)
Alcohol consumption
No or low(within recommended limits) 0.97 (0.82, 1.15)b
High (above recommended limits) 0.92 (0.62, 1.37)      P=0.978
Missing 0.74 (0.56, 0.97)

Abbreviations: CHD=coronary heart disease; CI=confidence interval; F=female; M=male; NSAID=non-steroidal anti-inflammatory drug; OR=odds ratio; UADT=upper aerodigestive tract cancer.

a

From conditional logistic regression adjusted for deprivation, CHD, stroke and other NSAIDs.

b

From unconditional logistic regression adjusted for age, gender, deprivation, CHD, stroke and other NSAIDs taking into account aggregation within medical practices.