Lu · PloS one 2009 · Systematic review and meta-analysis of prospective cohort studies · n=?

Diabetes and the risk of multi-system aging phenotypes: a systematic review and meta-analysis.

Cited 396 times in the scientific literature.

Level 3 - non-randomized controlled study

Systematic review and meta-analysis of prospective cohort studies

PubMed 19127292 · doi:10.1371/journal.pone.0004144 · record verified 2026-08-28

What was done

MEDLINE and PsycINFO databases were searched through November 2007 along with bibliography checks for population-based, prospective cohort studies examining baseline diabetes mellitus (DM) in relation to geriatric outcomes. Two reviewers independently extracted study population, follow-up length, DM ascertainment, outcome measures, and adjusted covariates.

What was found

Fifteen studies evaluated cognitive dysfunction: DM was associated with faster cognitive decline, with pooled adjusted risk ratios (RRs) of 1.47 (95% CI, 1.25 to 1.73) for all-cause dementia, 1.39 (95% CI, 1.16 to 1.66) for Alzheimer's disease, and 2.38 (95% CI, 1.79 to 3.18) for vascular dementia. Four of 5 studies found a significant association between DM and faster mobility decline and incident disability. Two studies in older women linked DM to falls (with higher recurrent fall risk in insulin users), and 1 study linked DM to urinary incontinence. Two studies did not find DM to be an independent predictor of incident depression.

Why it matters

This review aggregates prospective evidence showing that diabetes substantially increases the risk of multiple geriatric conditions, particularly vascular and non-vascular dementia, highlighting the multisystem impact of diabetes on aging.

Limits

The underlying evidence consists entirely of observational cohort studies, precluding causal conclusions. Non-cognitive geriatric outcomes (falls, incontinence, depression) relied on very small numbers of studies (1–2 each). The abstract does not report the total participant sample size, heterogeneity metrics, or numerical risk estimates for non-cognitive endpoints.

Cited by