Table 3.

GRADE tables for dementia progression, aspirin vs. control

Certainty assessment							No. of patients		Effect of aspirin vs control			Confidence in efficacy statement
No. of studies	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Aspirin	No aspirin	Relative (95% CI)	Absolute (95% CI)	Efficacy statement
MMSE decline (follow up: mean 24 months)
1	Randomised trials	Seriousa	Not serious	Seriousb	Seriousc	Inconsistency with below	83	86	-	MD 0.2 pts. less decline (2.4 more to 2.6 less)	Aspirin has no significant effect on MMSE decline	⨁◯◯◯ Very low
							4.8 pt. decline	5.0 pt. decline
Rapid MMSE decline (follow-up: mean 24 months; assessed with above median decline of 2 pts. in first year and 4 pts. in 2 years)
1	Observational studies	Seriousd	Not serious	Seriouse	Not serious	Strong association; inconsistency with above	73	87	OR 0.34 (0.11 to 0.88)	262 fewer rapid decliners per 1000 (from 31 fewer to 458 fewer)	Aspirin protects against rapid MMSE decline	⨁◯◯◯ Very low
							Adjusted model	Adjusted model

CI, confidence interval; MD, mean difference; OR, odds ratio

^aCochrane risk of bias tool highlighted risk from deviation from intended treatment (includes lack of blinding) and missingness

^bIncludes only those with Alzheimer’s disease and without high vascular risk

^cAlthough this measure of difference does not have pre-specified line of clinical effect, the authors considered that > 2.5-pt. difference on MMSE was significant, and therefore, this was imprecise

^dNewcastle-Ottawa assessment scale highlighted concern re-ascertainment of exposure, no detail on those followed up vs. not followed up and no mention of assessor blinding

^eIncludes only those with Alzheimer’s disease who were followed up