Get Newsletter
AlzRisk Paper Detail
Back

Reference: Côté, 2012
Cohort: Canadian Study of Health and Aging
Risk Factor: Non-Steroidal Anti-Inflammatory Drugs


Average Follow-up Time Detail
Among 10,263 participants enrolled at baseline, participants were excluded due to prevalent dementia (n=1,132), incomplete risk factor questionnaire (n=1,365), death (n= 1702), loss to follow-up (n=788). After further excluding 360 participants diagnosed with CIND at baseline, the total analytic sample included 4,916 participants. Initial assessment of study subjects (CSHA-1) occurred between 1991 and 1992. The first follow-up (CSHA-2) occurred between 1996 and 1997, and the second follow-up (CSHA-3) was between 2001 and 2002.

Exposure Detail
Information about risk factors, including NSAID use, was obtained via a structured in-person interview and a self-administered questionnaire. No information on dose was requested. The investigators reported results separately for current use of NSAIDs vs. not using any NSAID, current use of non-ASA NSAIDs vs. not using any NSAID, and current use of ASA (salicylates with and without barbiturates) vs. not using any NSAID. This entry pertains to results on the current use of non-ASA NSAIDs vs. not using any NSAID.

The investigators compared AD risk in two groups: the group of participants who used non-ASA NSAIDs at baseline ("Current use") and the reference group of participants who did not use any NSAID at baseline ("Not using").

Ethnicity Detail
The investigators do not provide data on ethnicity.
The CSHA is based on a representative, nationwide sample of Canadians.

Age Detail
Entry criteria required all participants to be 65+ in 1989-1990.

Screening and Diagnosis Detail
Screening Method:
3MSEModified Mini-Mental State Examination (Teng 1987)

AD Diagnosis:
NINCDS ADRDA National Institute of Neurological and Communicative Diseases and Stroke/Alzheimer's Disease and Related Disorders Association Criteria (McKhann 1984)

Covariates & Analysis Detail
Analysis Type:
Cox proportional hazards regression

Investigators performed three main analyses. The first examined the association between the use of NSAIDs and the incidence of all-cause dementia and AD. This analysis included participants who screened negative for dementia, participants not diagnosed with CIND, and participants without cognitive impairment at CSHA-1. The second analysis excluded participants diagnosed with CIND at baseline. The third analysis examined the association between use of NSAIDs and incidence of CIND. This entry pertains to the second analysis, excluding participants diagnosed with CIND at baseline. However, the results from the first analysis were very similar, both with minimal association of current non-ASA NSAID (vs not using non-ASA NSAID) with AD risk (HR: 1.09; 95% CI: 0.79-1.49).

The Cox model with delayed entry and age as the time scale was used and the proportional hazards model was tested.

Several categorizations of NSAIDs were analyzed, including salicylates excluding analgesic preparations containing caffeine and barbiturates, salicylates including barbiturates, and nonsalicylate agents, and selective and nonselective β-amyloid (1-42) suppressing NSAIDs.

Using incident cases at the 5-year first follow-up of the Canadian Study of Health and Aging, the investigators previously reported a 35% reduced risk of AD with NSAIDs using a case-control approach (Lindsay et al. 2002).

AD Covariates:
Aage
Eeducation
Ggender
ALCalcohol intake
AOSantioxidative Supplements
CVRFcardiovascular risk factors
CIcomorbidity index
MGMigraine
OAOsteoarthritis
PAphysical activity
SMsmoking status

TD Covariates:
Aage
Eeducation
Ggender
ALCalcohol intake
AOSantioxidative Supplements
CVRFcardiovascular risk factors
CIcomorbidity index
MGMigraine
OAOsteoarthritis
PAphysical activity
SMsmoking status