. 2003 Jul 19;327(7407):128. doi: 10.1136/bmj.327.7407.128

Table 1.

Characteristics of cohort studies evaluating role of NSAIDs and aspirin in preventing Alzheimer's disease. All were community studies

							Adjusted relative risk or odds ratio (95% CI)
Study	No	Age (years)	Diagnosis of Alzheimer's disease	NSAID assessment	Variable adjustment	Event rate (cases/person years) with NSAIDs v non-NSAIDs	NSAID	Aspirin
In't Veld¹	6989	>55	Clinical investigation	Prescription database	Age, sex, smoking, education, diabetes, antihypertensives, acid blockers	184/29 359 v 210/16 715	0.86 (0.66 to 1.09)	1.3 (0.97 to 1.74)
Zandi²	3227	>65	Interviews and clinical investigation	Patient interviews	Age, sex, APOE gene, education	79/7048 v 22/3017	0.67 (0.40 to 1.06)	0.82 (0.54 to 1.23)
Stewart²¹	1686	<65	Clinical investigation	Patient interviews	Age, sex, education, year of cohort entry	Only adjusted relative risks presented	0.46 (0.24 to 0.86)	0.85 (0.53 to 1.37)
Fourrier⁴	516	>65	MMSE scores†	Patient interviews	Age, education	Only adjusted relative risks presented	2.84 (0.99 to 8.10)	—
Henderson²³	588	80^*	Interviews and clinical investigation	Patient interviews	Age, sex, education, stroke, APOE gene, arthritis medication	Only adjusted relative risks presented	1.66 (0.64 to 4.32)	1.79 (0.72 to 4.45)
Breitner 1995²⁴	205	NS	Interviews and autopsy	Patient interviews	Age, sex, acid blockers, insulin	Only adjusted relative risks presented	0.19 (0.02 to 1.49)	0.37 (0.17 to 0.79)

NS=not stated, all older adults. APOE=apolipoprotein E. ^*Mean age. †Folstein mini-mental state examination.