Does aspirin or non-aspirin non-steroidal anti-inflammatory drug use prevent colorectal cancer in inflammatory bowel disease?

Table 1 Characteristics of studies included in the analysis

Author	Design	Population	Definition of IBD	Drugexposure	Exclusioncriteria	No. of patients	OR (95%CI)1	Adjustment/matching	NOS quality assessment
Bansal et al[42], 1996	Case-control	US veterans affairs	Clinical database	NSAID associated diagnosis	Not specified	11446	0.84 (0.65-1.09)	Adjusted for age, sex and ethnicity	6
Eaden et al[38], 2000	Case-control	UK Hospital	Clinical, pathological and radiological records	Prescribed 5-10 years before diagnosis	Colorectal surgery, IBD diagnosed at time of cancer diagnosis	206	0.80 (0.21-2.98) Aspirin	Non-adjusted	4
Van Staal et al[14], 2005	Nested case-control	UK general practice	Clinical records	Prescribed in the 6 mo prior to diagnosis	Colorectal surgery, previous history of CRC	700	1.52 (0.7-3.25) (Aspirin)	Non-adjusted	5
							0.80 (0.38-1.66) (NA-NSAID’s)
Velayos et al[39], 2006	Case-control	US Hospital	Clinical, pathological and endoscopic records	2 records of use in notes	Previous CRC, IBD diagnosed at same time as CRC, incomplete data	376	0.3 (0.1-0.8) (Aspirin)	Matched on gender, duration of disease and extent of disease	8
							0.1 (0.03-0.5) (NA-NSAID)
Terdiman et al[40], 2007	Case-control	US insurance claims	Clinical records	Prescribed in the year before diagnosis	Colorectal surgery	1536	0.97 (0.74-1.28)	Non-adjusted	5
Tang et al[43], 2010	Retrospective cohort	US Hospital	Clinical database	Ever used	No colonic involvement of IBD	48	0.29 (0.03-2.75)	Non-adjusted	5
Samadder et al[41], 2011	Case-control	N. Israel community	Patient questionnaires	Weekly for > 3 yr	Previous history of CRC	60	0.49 (0.07-3.32)	Matched for age, gender and ethnicity	6
Baars et al[44], 2011	Case-control	Netherlands nationwide pathology	Pathology reports	Ever used	IBD diagnosed at the same time as CRC	551	1.96 (0.72-5.36)	Non-adjusted	6
Rubin et al[45], 2013	Case-control	US Hospital	Pathology reports	Not specified	Incomplete records	200	1.84 (0.75-2.5)	Non-adjusted	5

¹OR for colorectal cancer (CRC) chemoprotective effect of non-aspirin non-steroidal (NA-NSAID) use in patients in inflammatory bowel disease (IBD) unless otherwise stated. For cohort studies was used only for the Tang et al[43] study. The scale for case-control studies was used for the other studies. NOS: Newcastle-Ottawa Score.