Table 2. Risk of breast cancer associated with aspirin, non-aspirin nonsteroidal anti-inflammatory drugs (NSAID), paracetamol and steroid use duration using the 2 years lag-time analysis.

	Cases (n=3708)	Controls (n=20 000)	Odds ratioa (95% CI)
Aspirin
No useb	3524	18 866
Current use	122	792	0.84 (0.69, 1.02)
0–0.9 years	57	349	0.87 (0.66, 1.16)
1–1.9 years	15	124	0.65 (0.38, 1.12)
2–3.9 years	31	193	0.88 (0.60, 1.30)
4+ years	19	126	0.87 (0.53, 1.41)
Past use	62	342	0.96 (0.73, 1.26)
Non-aspirin NSAIDs
No useb	1884	10 138
Current use	802	4444	0.98 (0.89, 1.08)
0–0.9 years	606	3346	0.98 (0.88, 1.09)
1–1.9 years	62	360	0.93 (0.70, 1.23)
2–3.9 years	67	377	0.97 (0.74, 1.28)
4+ years	67	361	1.05 (0.80, 1.38)
Past use	1022	5418	1.01 (0.93, 1.11)
Paracetamol
No useb	2244	11 762
Current use	744	4345	0.88 (0.79, 0.97)
0–0.9 years	499	2760	0.92 (0.83, 1.03)
1–1.9 years	70	393	0.92 (0.70, 1.19)
2–3.9 years	87	606	0.72 (0.57, 0.92)
4+ years	88	586	0.76 (0.60, 0.97)
Past use	720	3893	0.95 (0.86, 1.04)
Oral steroids
No useb	3452	18 567
Current use	124	704	0.96 (0.79, 1.17)
0–0.9 year	96	484	1.08 (0.87, 1.35)
1–1.9 years	12	60	1.04 (0.56, 1.95)
2–3.9 years	11	70	0.93 (0.49, 1.76)
4+ years	5	90	0.31 (0.12, 0.76)
Past use	132	729	0.98 (0.81, 1.19)

^aEstimates are adjusted for age, calendar year, BMI, alcohol intake, smoking status, HRT use, prior benign breast disease, and all the variables in the table using logistic regression.

^bReference category.