nabu.science
MethodologyAboutContact

 

  • Snapshot
  • Key Claims
  • Strengths & Limitations
  • Commentary
  • From the Authors
  • Author Note / Field Feedback
  • Dimension breakdown
  • Impact Profile
Free · About 5 minutes
Report #NBU-79F5782A·Framework v5.13·Pipeline v0.4·Evaluated 18 Nov 2026·OECD: 3.3 Health sciences·Subfield: Psychiatry and Mental health·Basis: Full text·Type: Systematic review

Smoking and dementia risk: an updated systematic review and meta-analysis of 61 prospective cohort studies and implications for tobacco control

Mario Siervo et al.Mario Siervo, Jacob Brain, Sarah Beasley et al.·2026·EClinicalMedicine·10.1016/j.eclinm.2026.104310 ↗

Primary claim

Current smoking was associated with increased risk of all-cause dementia (RR 1.31 [95% CI 1.23–1.39]; number of studies (k) = 39), Alzheimer's disease (1.40 [1.08–1.80]; k = 15), and vascular dementia (1.52 [1.20–1.93]; k = 9).

Supported · pooled observational associations remain consistent in sensitivity analyses despite substantial heterogeneity and risk of bias
In brief

An updated cohort synthesis confirms smoking–dementia associations and adds dose–response and cessation contrasts, but observational bias limits causal interpretation.

ReliabilityCaveats4 of 4 checks completeRead the 2 entries ↓
Abstract

Background Smoking has been identified as a modifiable dementia risk factor by three Lancet Commissions, yet dementia remains absent from most tobacco control strategies. The magnitude of the association across dementia subtypes, smoking categories, and population characteristics has not been comprehensively synthesised since 2015. Methods PubMed, Embase, CENTRAL, and PsycINFO were searched from January 2013 to 31 August 2026, updating the 2015 review, which contributed 34 studies. Random-effects meta-analyses (Restricted Maximum Likelihood (REML) with Hartung–Knapp–Sidik–Jonkman adjustment) estimated pooled relative risks (RRs) with 95% confidence intervals (CIs). Subgroup analyses with tests for interaction and meta-regression examined effect modification by baseline age, ethnicity, follow-up duration, and adjustment for sex, education and Apolipoprotein E ε4 (APOE ε4) genotype. Risk of bias was assessed using Risk Of Bias In Non-randomized Studies of Exposures (ROBINS-E). Heterogeneity was assessed using Cochran's Q, I 2 and τ 2 and publication bias using funnel plots and Egger's test. Dose–response analysis of smoking and dementia risk was performed. Smoking population attributable fractions (PAF, unweighted and communality-weighted) for the risk of dementia were calculated. This study is registered with PROSPERO, CRD42025644539. Findings Sixty-one cohort studies (approximately 2.8 million participants) were included. Current smoking was associated with increased risk of all-cause dementia (RR 1.31 [95% CI 1.23–1.39]; number of studies (k) = 39), Alzheimer's disease (1.40 [1.08–1.80]; k=15), and vascular dementia (1.52 [1.20–1.93]; k=9). Former smoking was not associated with any outcome, and former smokers had a lower risk of all-cause dementia than current smokers (RR 0.81 [0.75–0.88]). I 2 ranged from 0% to 80%. Egger's test suggested small-study effects only for current smoking and all-cause dementia (P = 0.09). The communality-weighted PAF for smoking and dementia was 2.2% (95% CI 1.7–2.8). Dementia risk increased with smoking intensity as the pooled RR was 1.16 (95% CI 1.09–1.23) at 10 cigarettes per day and rose to 1.34 (95% CI 1.18–1.51) at 20 cigarettes per day. Interpretation Current smoking is consistently associated with higher risk of all dementia outcomes and increases with smoking intensity. Estimates refer to clinically diagnosed dementia syndromes [Diagnostic and Statistical Manual of Mental Disorders/International Classification of Diseases (DSM/ICD)] and not to biologically defined disease. Although causality cannot be established from observational data, these findings support considering dementia and brain health outcomes in national tobacco control strategies and associated campaigns. Future work using objectively verified smoking histories is needed to clarify how age at cessation and time since quitting influence dementia risk and whether these effects interact with total years of active smoking. Funding None.

Key Claims

Primary finding · Supported

Current smoking was associated with increased risk of all-cause dementia (RR 1.31 [95% CI 1.23–1.39]; number of studies (k) = 39), Alzheimer's disease (1.40 [1.08–1.80]; k = 15), and vascular dementia (1.52 [1.20–1.93]; k = 9).

pooled observational associations remain consistent in sensitivity analyses despite substantial heterogeneity and risk of bias

Summary, Findings
Secondary · Tentative

Former smoking was not associated with any [dementia] outcome, and former smokers had a lower risk of all-cause dementia than current smokers (RR 0.81 [0.75–0.88]).

observational cessation comparisons face selection and sick-quitter bias, while null former-versus-never estimates do not establish equivalent risks

Abstract, Findings (and Results, Cessation versus continued smoking; Table 1)
Quote verified
Secondary · Supported

Dementia risk increased with smoking intensity as the pooled RR was 1.16 (95% CI 1.09–1.23) at 10 cigarettes per day and rose to 1.34 (95% CI 1.18–1.51) at 20 cigarettes per day.

a 12-study observational synthesis supports the modelled gradient, with exposure conversion dependent on an assumed 40-year smoking duration

Summary, Findings
Quote verified
Secondary · Supported

In the nine cohorts reporting both [subtypes], the ratio of the VaD to the AD relative risk for current smoking was 1.11 (95% CI 0.88–1.40; P = 0.31)

paired estimates directly support the reported ratio within nine cohorts, with clinical subtype misclassification limiting interpretation

Results, Comparison of subtypes (and Summary, Added value of this study)
Quote verified
Secondary · Tentative

The communality-weighted PAF for smoking and dementia was 2.2% (95% CI 1.7–2.8).

population attribution assumes causality and shared Commission inputs, with the interval capturing only relative-risk uncertainty

Abstract, Findings; Results, PAF calculation
Quote verified

Derived from the full evaluation — not a separate score.

Strengths · Limitations

Strengths

#methodological rigour

Sensitivity analyses test the pooled associations

REML models with Hartung–Knapp–Sidik–Jonkman adjustment and prediction intervals make uncertainty visible. Current-smoking estimates remain similar under alternative estimators, risk-of-bias restrictions, and one-publication-per-cohort analyses.

↳ Methods, Statistical analysis; Table 1; Results, Sensitivity analyses

#contribution

Within-cohort contrasts extend the earlier synthesis

The review adds former-versus-current smoking comparisons across 29 cohorts and paired dementia-subtype comparisons across nine cohorts. These comparisons address questions not resolved by simply comparing separately pooled estimates.

↳ Results, Comparison of subtypes; Cessation versus continued smoking

#positioning

Interpretation addresses mortality and diagnostic limitations

The Discussion explains how competing death, survivor selection, and illness-related quitting can shape observed associations. It also confines subtype interpretation to clinically diagnosed syndromes rather than biomarker-defined pathology.

↳ Discussion, cessation, competing-risk, and clinical-syndromic diagnosis paragraphs

Limitations

#methodological rigour

Null estimates do not establish equivalent risks

The conclusion describes former smokers' risks as similar to never smokers', but no equivalence analysis is reported. The former-smoking vascular-dementia interval of 0.93–1.20 still permits meaningful excess risk.

↳ Table 1, Former vs never; Discussion, final paragraph

#methodological rigour

Evidence certainty and exposure harmonisation remain limited

No included study was rated low risk of bias, and 61% were high or very high risk, without a formal certainty-of-evidence grading. The dose-response analysis also converts pack-years using an assumed common 40-year smoking duration.

↳ Results, Risk of bias; Methods, Statistical analysis

#reporting

Study counts and citations need reconciliation

The reported 61 included studies and five non-poolable studies imply 56, not the stated 57 meta-analysed studies. Citation mismatches, including the ROBINS-E reference, further hinder source tracing.

↳ Results, Study selection and characteristics; Methods, Risk of bias assessment; References 61 and 63

Does this match your read? Add your assessment ↓

Nabu Commentary

The review confirms established current-smoking associations while extending the evidence through within-cohort cessation and subtype contrasts and dose-response modelling. Its registered procedures, prediction intervals, and sensitivity analyses support the synthesis, but the conclusion's description of former smokers as having similar risks to never smokers exceeds what nonsignificant comparisons establish. Observational bias, the absence of low-risk studies, and assumed smoking duration in exposure conversion constrain interpretation. The Discussion substantively addresses competing mortality and clinical diagnosis, although the PAF is conditional on causal assumptions rather than independent causal evidence.

Evaluate another paper

Free · About 5 minutes · Research preview

Author Note / Field Feedback

Nabu’s assessment, alongside the field’s view.

Are you an author of this paper?

Dimension breakdown

Evidence Score

Out of 5
012345
3.3±0.2

Contribution
3.8

The review consolidates an established association while adding within-cohort cessation and subtype contrasts and dose-response estimates. Its principal current-smoking estimates remain close to the 2015 synthesis, limiting the extent of the advance.

Results, Comparison of subtypes; Cessation versus continued smoking; Dose-response analysis; Discussion, comparison with Zhong et al.
Methodological Rigour
2.5

Registered review procedures, ROBINS-E assessment, adjusted random-effects inference, and sensitivity analyses support the synthesis. The conclusion nevertheless interprets nonsignificant former-versus-never estimates as similar risks without equivalence testing, while dose harmonisation and absent formal certainty grading add uncertainty.

Methods, Statistical analysis; Results, Sensitivity analyses; Table 1; Discussion, final paragraph
Reporting
3.9

The manuscript connects methods, pooled estimates, and interpretation and explicitly distinguishes clinical dementia syndromes from biological disease. Count and citation errors impede traceability, and wording around heterogeneous results is occasionally compressed.

Abstract, Interpretation; Results, Study selection and characteristics; Table 1; Discussion
Positioning
3.9

The Discussion connects competing mortality, reverse causation, sick-quitter effects, and diagnostic limitations to interpretation. Treatment of dose-conversion and PAF assumptions is less complete, and high risk of bias does not establish that pooled estimates are conservative.

Discussion, cessation, competing-risk, PAF, and limitations paragraphs

Reliability

Caveats4 of 4 checks


Results checkNoted

One reported value does not reconcile, but the key claims do not rest on it.

Reporting and transparencyNoted

The ROBINS-E citation points to an unrelated reference. Registration, an ethics exemption statement, competing interests, AI assistance, and availability of data and code upon request are disclosed.

Conflict of interest: declared · Ethics approval: declared · Preregistration: declared · Data availability: declared · Code availability: declared

Reference IntegrityClean

107 of 107 checkable references verified

111 references in manuscript 4 are books, websites or datasets — counted, but not index-checkable

References not verified automatically are resolved manually within 24 hours.

Post-Publication RecordClean

No retraction or expression of concern found in Crossref (includes Retraction Watch).

Last checked: 18 Nov 2026

Sources: Crossref (incl. Retraction Watch) ✓

Impact Profile

Impact areaHealth · Neurological · Alzheimer's disease and other dementias
Type of advanceSynthesis
Route to useSpecified
AdvanceConsolidates
BeneficiariesLarge
CitationsToo early
Policy · patentNot yet traced
ReachedNot reachedNot assessable or unverified
ReadinessValidated
Foundational
Controlled
Piloted
Validated
Implementable
FoundationalImplementable

The association is pooled from 61 independent prospective cohorts across multiple world regions and roughly 2.8 million participants, constituting real-world population data from many external settings.

From the paper
“Incorporating this message into campaigns, pictorial warnings and cessation programmes could broaden their reach and motivational appeal”

The paper names the actors (national tobacco control strategies) and the specific decisions and settings where its findings would be applied, namely public health campaigns, pictorial cigarette warnings and cessation programmes.

Pathway

National tobacco control bodies and health departments would add dementia and brain health messages to public campaigns, pictorial warnings and cessation programmes.

Barrier

Causality cannot be established from observational data, and 61% of included studies were rated high or very high risk of bias, weakening the causal certainty policymakers require.

Evaluated 18 Nov 2026 · GBD 2023 · glm_v1_3_3 · impact_profile_v1_3_3

Report an issue →

AI-generated, human-governed. Something look off? Contact us to request a review.

Keep reading

Related evaluations

Other papers Nabu has evaluated blind, starting with the same field.

  1. Effect of intensive blood pressure control on cognitive outcomes and brain structures in adults with type 2 diabetes: prespecified analysis of secondary outcomes of BPROAD randomised controlled trial

    Wang, Xu, Xu, et al. · 2026 · Clinical medicine

    Primary claim

    no statistically significant difference in incidence of a composite of all cause dementia or mild cognitive impairment was observed among adults with type 2 diabetes receiving standard treatment of SBP and intensive treatment

    Well-supported
    Evidence Score
    4.1±0.1
    Reliability
    Caveats
  2. Decreased awareness of cognitive decline is associated with multimodal Alzheimer's disease biomarkers in cognitively unimpaired individuals

    2026 · Clinical medicine

    Primary claim

    decreased ACD was associated with greater AD pathological burden

    Supported
    Evidence Score
    3.2±0.2
    Reliability
    Caveats
  3. The potential of low-field MRI for global dementia care

    2026 · Clinical medicine

    Primary claim

    Studies demonstrate reliable detection of core Alzheimer disease markers, including hippocampal atrophy and white matter hyperintensities, on portable low-field systems, with promising correspondence to conventional high-field MRI findings.

    Tentative
    Evidence Score
    3.7±0.1
    Reliability
    Caveats

How the rubric works →