Meat consumption and risk of incident dementia: cohort study of 493,888 UK Biobank participants

Abstract

Background: Worldwide, the prevalence of dementia is increasing and diet as a modifiable factor could play a role. Meat consumption has been cross-sectionally associated with dementia risk, but specific amounts and types related to risk of incident dementia remain poorly understood.

Objective: We aimed to investigate associations between meat consumption and risk of incident dementia in the UK Biobank cohort.

Methods: Meat consumption was estimated using a short dietary questionnaire at recruitment and repeated 24-h dietary assessments. Incident all-cause dementia comprising Alzheimer disease (AD) and vascular dementia (VD) was identified by electronic linkages to hospital and mortality records. HRs for each meat type in relation to each dementia outcome were estimated in Cox proportional hazard models. Interactions between meat consumption and the apolipoprotein E (APOE) ε4 allele were additionally explored.

Results: Among 493,888 participants included, 2896 incident cases of all-cause dementia, 1006 cases of AD, and 490 cases of VD were identified, with mean ± SD follow-up of 8 ± 1.1 y. Each additional 25 g/day intake of processed meat was associated with increased risks of incident all-cause dementia (HR: 1.44; 95% CI: 1.24, 1.67; P-trend < 0.001) and AD (HR: 1.52; 95% CI: 1.18, 1.96; P-trend = 0.001). In contrast, a 50-g/d increment in unprocessed red meat intake was associated with reduced risks of all-cause dementia (HR: 0.81; 95% CI: 0.69, 0.95; P-trend = 0.011) and AD (HR: 0.70; 95% CI: 0.53, 0.92; P-trend = 0.009). The linear trend was not significant for unprocessed poultry and total meat. Regarding incident VD, there were no statistically significant linear trends identified, although for processed meat, higher consumption categories were associated with increased risks. The APOE ε4 allele increased dementia risk by 3 to 6 times but did not modify the associations with diet significantly.

Conclusion: These findings highlight processed-meat consumption as a potential risk factor for incident dementia, independent of the APOE ε4 allele.

Keywords: Alzheimer disease; UK Biobank; dementia; meat consumption; processed meat; vascular dementia.

FIGURE 1

HRs (95% CIs) for the associations between incident all-cause dementia and meat consumption in UK Biobank (n = 493,888). The black squares and horizontal lines represent HRs and 95% CIs respectively in Cox proportional-hazards regressions. The distribution of ticks on the x axis is exponential. Participants were categorized based on the data distribution of baseline meat intakes. Mean daily intakes in each category were calculated from the multiple 24-h dietary assessments which were used to test the linear trend per increment. Minimally adjusted models adjusted for age, gender, ethnicity, education, socioeconomic status. Fully adjusted models additionally adjusted for region, smoking status, physical activity, BMI, sleep duration, stroke history, and family history of dementia, and dietary covariates including vegetables and fruits, total fish, tea and coffee, alcohol drinking, processed meat, unprocessed poultry, and unprocessed red meat were also mutually adjusted for.

FIGURE 2

HRs (95% CIs) for the associations between incident Alzheimer disease and meat consumption in UK Biobank (n = 493,888). The black squares and horizontal lines represent HRs and 95% CI, respectively, in Cox proportional-hazards regressions. The distribution of ticks on the x axis is exponential. Participants were categorized based on the data distribution of baseline meat intakes. Mean daily intakes in each category is calculated from the multiple 24-h dietary assessments which were used to test the linear trend per increment. Minimally adjusted models adjusted for age, gender, ethnicity, education, and socioeconomic status. Fully adjusted models additionally adjusted for region, smoking status, physical activity, BMI, sleep duration, stroke history, family history of dementia, and dietary covariates including vegetables and fruits, total fish, tea and coffee, alcohol drinking; processed meat, unprocessed poultry, and unprocessed red meat were also mutually adjusted for.

FIGURE 3

HRs (95% CIs) for the associations between incident vascular dementia and meat consumption in UK Biobank (n = 493,888). The black squares and horizontal lines represent HRs and 95% CI respectively in Cox proportional-hazards regressions. The distribution of ticks on the x axis is exponential. Participants were categorized based on the data distribution of baseline meat intakes. Mean daily intakes in each category is calculated from the multiple 24-h dietary assessments which were used to test the linear trend per increment. Minimally adjusted models adjusted for age, gender, ethnicity, education, socioeconomic status. Fully adjusted models additionally adjusted for region, smoking status, physical activity, BMI, sleep duration, stroke history, and family history of dementia, and dietary covariates including vegetables and fruits, total fish, tea and coffee, alcohol drinking; processed meat, unprocessed poultry, and unprocessed red meat were also mutually adjusted for.