Abstract
Background
Excessive sugar intake has been implicated in increased dementia risk; however, existing studies are constrained by small sample sizes and a primary focus on total sugar, with limited investigation into specific sugar subtypes. This study explores the relationship between sugar intake, its subtypes, and the incidence of dementia.
Methods
We analyzed 172,516 participants from the UK Biobank who completed at least one 24-hour dietary recall (Oxford WebQ). Cox proportional hazards models estimated the hazard ratios (HRs) and 95 % confidence intervals (95 % CIs) for total sugar and its subtypes (free sugar, fructose, glucose, sucrose, maltose, lactose, and other sugars) about the risk of dementia. Sex-stratified analyses were also performed.
Results
Higher intakes of total sugar (HR = 1.292, 95 % CI = 1.148–-1.453) and free sugar intake (HR = 1.254, 95 % CI = 1.117–-1.408) were significantly associated with increased dementia risk. Positive associations were also observed for non-milk extrinsic sugars (HR = 1.321, 95 % CI = 1.175–-1.486) and sucrose (HR = 1.291, 95 % CI = 1.147–-1.452). These associations were evident in women, with higher intakes of total sugars, free sugars, glucose, sucrose, and non-milk extrinsic sugars independently linked to increased dementia risk, whereas no significant associations were found in men.
Conclusion
Higher consumption of total sugars, free sugars, sucrose, and non-milk extrinsic sugars confers increased dementia risk, particularly among women.
Keywords: Dementia, Sugar intake, Sugar subtypes, Sex differences, Cox proportional hazards model, UK Biobank
1. Introduction
Dementia is a clinical syndrome that primarily affects cognitive functions, including attention, memory, language, executive function, and processing speed [1]. Given the global acceleration of population aging, the prevalence of dementia is expected to rise significantly. The number of people affected by dementia worldwide is projected to increase from 57.4 million in 2019 to more than 150 million by 2050, representing a 166 % increase. Dementia has become one of the major threats to the health and quality of life of older individuals [2]. Although recent advances in the early treatment of dementia, particularly those targeting amyloid-associated lesions, have provided hope for patients, preventive measures aimed at delaying the onset of dementia remain a critical and cost-effective approach [3].
Sugar, classified into monosaccharides, disaccharides, and polysaccharides based on its chemical structure, is the body’s primary source of energy [4]. The brain, being one of the most energy-demanding organs, relies almost exclusively on glucose to fuel its complex physiological functions [5]. Excessive sugar consumption has been linked to cardiovascular diseases, metabolic disorders, and systemic inflammation [6].
Recent epidemiological studies have found that sugar intake may be closely associated with the development of dementia [7]. Excessive sugar consumption damages neurons in the brain, impairs cognitive function, increases the risk of dementia, and may contribute to cerebrovascular diseases [8]. Additionally, excessive sugar intake is linked to chronic conditions such as diabetes and obesity, which are known risk factors for dementia [[9], [10], [11]]. Based on these findings, we hypothesize that excessive sugar intake and its subtypes may increase the risk of dementia compared to low sugar intake.
However, significant controversies and limitations remain in the current research on the relationship between sugar intake and the risk of dementia [12]. Most previous studies were based on small sample sizes and primarily focused on a single type of sugar, overlooking the complexity of sugar’s presence in food as multiple subtypes. This narrow focus has led to certain limitations and uncertainties in the study results [13,14].
In this study, we examined the relationship between sugar intake and its subtypes with the incidence of dementia using a large dataset from the UK Biobank. This provides new insights into the treatment and prevention of dementia.
2. Methods
2.1. Study design and study population
The UK Biobank (UKB) is a large-scale, population-based cohort study that recruited over 500,000 participants aged 37 to 73 years between 2006 and 2010 in the United Kingdom (http://www.ukbiobank.ac.uk/about-biobank-uk/). The following exclusion criteria were applied in this study: (1) participants who did not complete at least one 24-hour dietary recall; (2) participants who reported implausible energy intakes (<500 or >3,500 kcal/day for women; <800 or >4,000 kcal/day for men); (3) individuals with missing data on covariates; and (4) participants diagnosed with dementia at baseline.
After data cleaning and screening, a total of 172,516 eligible participants were included in the final analysis. The UK Biobank received ethical approval from the National Information Governance Board for Health and Social Care and the National Health Service (NHS) North West Centre for Research Ethics Committee (Ref: 11/NW/0382). All participants provided informed consent via electronic signature.
2.2. Sugar intake measurement
Dietary data in the UK Biobank were collected using the Oxford WebQ, a web-based questionnaire designed to assess participants’ dietary intake over the previous 24 h. For each food or beverage item, consumption was calculated by multiplying the reported number of servings by a standard portion size. Built-in algorithms analyzed food composition and calculated specific nutrient intakes, including total sugar intake [15], [16], [17]. The validity of the Oxford WebQ has been demonstrated through comparisons with traditional interviewer-led 24-hour dietary recalls, particularly when multiple questionnaires are completed [18].
In this study, data on total sugar intake (g/day) and sugar subtypes (free sugar, fructose, glucose, sucrose, maltose, lactose, and other sugars) were extracted and analyzed. Participants were categorized into tertiles based on their daily intake of each sugar type, forming three groups: Q1 (lowest intake), Q2, and Q3 (highest intake), with Q1 serving as the reference group for analysis.
2.3. Outcome ascertainment
The primary outcome variable for this study, referred to as all-cause dementia (hereafter “dementia”), was defined by an algorithm provided by the UK Biobank. This algorithm determined the date of dementia diagnosis based on linked data from hospital admission records, death registries, and self-reported medical conditions at baseline. Additionally, first-occurrence data from primary care visits were included to complement the algorithm-defined dementia outcomes. To enhance the accuracy of dementia diagnosis, individuals diagnosed with dementia at baseline were excluded. Follow-up time was defined as the period from the first dietary assessment to the earliest of the following events: the first dementia diagnosis, loss to follow-up, death, or the end of the study review [19,20]. Dementia diagnoses were made using the International Classification of Diseases (ICD-10) codes F00-F03, G30, G310, G311, and G318.
2.4. Covariates
Covariates were selected based on previously established risk factors for dementia and included sex (female/male), baseline age, race (White/non-White, including mixed race, Black, Asian, and other ethnic groups), body mass index (BMI), Townsend Deprivation Index (TDI), education level (Degree level or professional education, Other levels, and Missing value), smoking status, alcohol consumption, use of vitamin or mineral supplements, physical activity level (assessed using the International Physical Activity Questionnaire [IPAQ] and categorized as low, medium, or high), and medical conditions including diabetes, hypertension, and obesity [21].
2.5. Data analysis
Baseline characteristics were summarized according to tertiles of total sugar intake. Continuous variables were presented as means with standard deviations (SD), while categorical variables were reported as counts and percentages.
A Cox proportional hazards regression model was employed to assess the association between sugar intake and dementia risk. Sugar intake was categorized into tertiles (Q1 to Q3), with Q1 serving as the reference group. Covariates were adjusted in three models: Model 1 included sociodemographic factors (age, sex, ethnicity, education level, and TDI); Model 2 additionally incorporated behavioral factors, including BMI, smoking status, alcohol consumption, IPAQ score, and supplement use; and Model 3 further adjusted for medical conditions, including diabetes, cancer, hypertension, and cardiovascular disease. The proportional hazards (PH) assumption was thoroughly evaluated for all Cox regression models included in our study. Specifically, we assessed this assumption using Schoenfeld residuals via the cox.zph function in R.
Hazard ratios (HRs) and 95 % confidence intervals (CIs) were calculated, and p-values for trend were computed using the median value of each tertile as a quasi-continuous variable in the model. Stratified analyses were conducted by sex to explore potential sex-specific associations. To further assess whether sex modifies the association between sugar intake and dementia risk, we included interaction terms between sex and sugar intake variables in the Cox proportional hazards models. We tested the significance and consistency of these interactions across different model specifications and subsamples to evaluate the potential effect modification by sex and the robustness of the interaction effects.
All statistical analyses were performed using the R programming language, with a significance threshold set at p < 0.05. Given that ten independent tests were conducted, Bonferroni correction was applied, resulting in an adjusted significance level of 0.005 (0.05 / 10). Only results with p-values < 0.005 were considered statistically significant.
3. Results
3.1. Baseline characteristics of UK Biobank participants
Based on tertiles of total sugar intake (Q1 to Q3, representing progressively higher intake), the characteristics of the study population are summarized in Table 1. The number of participants in the Q1, Q2, and Q3 groups was 57,506, 57,505, and 57,505, respectively. Notably, the mean age of participants increased across tertiles: 55.0 ± 8.0 years in Q1, 56.1 ± 7.9 years in Q2, and 56.6 ± 7.9 years in Q3, suggesting a trend of higher mean age with increasing sugar intake. Regarding sex distribution, females were fewer than males in the Q1 group, whereas females outnumbered males in both the Q2 and Q3 groups.
Table 1.
Baseline characteristics of the study population.
| Characteristics | Total sugar intake |
||
|---|---|---|---|
| Q1 | Q2 | Q3 | |
| Number | 57,506 | 57,505 | 57,505 |
| Age | 55.0 (8.0) | 56.1 (7.9) | 56.6(7.9) |
| Sex | |||
| Female | 27,990 (48.7) | 32,294 (56.2) | 32,130 (55.9) |
| Male | 29,516 (51.3) | 25,211 (43.8) | 25,375 (44.1) |
| Racial | |||
| white | 55,298 (96.2) | 55,443 (96.4) | 54,750 (95.2) |
| Non-white | 2208 (3.8) | 2062 (3.6) | 2755 (4.8) |
| TDI | −1.50 (2.9) | −1.70 (2.8) | −1.63 (2.9) |
| BMI | 27.28 (4.7) | 26.65 (4.5) | 26.65 (4.5) |
| IPAQ | |||
| Low | 12,208 (21.2) | 10,369 (18.0) | 8996 (15.6) |
| Moderate | 24,816 (43.2) | 24,947 (43.4) | 23,458 (40.8) |
| High | 20,482 (35.6) | 22,189 (38.6) | 25,051 (43.6) |
| Education level | |||
| Degree/Professional | 31,538 (54.8) | 32,070 (55.8) | 31,058 (54.0) |
| Other | 21,824 (38.0) | 21,257 (37.0) | 21,763 (37.8) |
| Missing value | 4144 (7.2) | 4178 (7.3) | 4684 (8.1) |
| Alcohol intake, Yes | 56,493 (98.2) | 55,947 (97.3) | 54,861 (95.4) |
| Smoking, Yes | 27,685 (48.1) | 24,280 (42.2) | 23,013 (40.0) |
| Supplementation, Yes | 24,584 (42.8) | 29,014 (50.5) | 31,471 (54.7) |
| Medical history | |||
| Hypertension, Yes | 15,633 (27.2) | 15,093 (26.2) | 15,920 (27.7) |
| CVD,Yes | 10,042 (17.5) | 9857 (17.1) | 10,708 (18.6) |
| Cancer, Yes | 4945 (8.6) | 5414 (9.4) | 5657 (9.8) |
| Diabetes, Yes | 4654 (8.1) | 3485 (6.1) | 3330 (5.8) |
Notes: Data is presented as mean (SD) for continuous variables and as frequency (percentage) for categorical variables. BMI, body mass index; TDI, Townsend Deprivation Index; IPAQ, International Physical Activity Questionnaire.
3.2. Sugar intake and dementia
The PH assumption showed p-values > 0.05, and the residual plots demonstrated approximately flat trend lines over time, providing further support for the validity of the PH assumption. The detailed results of the PH assumption tests are provided in Supplementary Table 1.
In a series of analyses examining the effects of total and free sugar intake on dementia incidence, three models (Model 1, Model 2, and Model 3) were constructed for in-depth evaluation. The results indicated that the highest intake levels of both total and free sugars were significantly associated with increased risk of dementia. Specifically, the hazard ratio (HR) and 95 % confidence interval (CI) for the highest total sugar intake group were 1.292 (1.148, 1.453). Similarly, the highest free sugar intake group had an HR of 1.254 (1.117, 1.408). Moreover, the highest intake groups of non-milk extrinsic sugars and sucrose were also significantly associated with elevated dementia risk, with HRs (95 % CIs) of 1.321 (1.175, 1.486) and 1.291 (1.147, 1.452), respectively. In this study, only the results from Model 3 are presented in Table 2, as they best represent the final specification and key findings of the analysis.
Table 2.
Associations between sugar intake and dementia.
| Sugar | Quartile | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|---|
| HR (95 %CI) | P-Value | HR (95 %CI) | P-Value | HR (95 %CI) | P-Value | ||
| Total sugars | Q1 | Ref | |||||
| Q2 | 0.953 (0.840, 1.080) | 4.49 × 10–1 | 0.965(0.850, 1.095) | 5.77 × 10–1 | 0.980(0.863, 1.112) | 7.50 × 10–1 | |
| Q3 | 1.266(1.127, 1.422) | 6.85 × 10–5 | 1.291(1.148, 1.452) | 2.08 × 10–5 | 1.292(1.148, 1.453) | 2.02 × 10–5 | |
| Free sugar | Q2 | 1.033(0.916, 1.165) | 5.99 × 10–1 | 1.037(0.919, 1.170) | 5.56 × 10–1 | 1.073(0.951, 1.211) | 2.50 × 10–1 |
| Q3 | 1.224(1.091, 1.374) | 5.90 × 10–4 | 1.228(1.094, 1.379) | 4.96 × 10–4 | 1.254(1.117, 1.408) | 1.31 × 10–4 | |
| Fructose | Q2 | 0.865 (0.764, 0.979) | 2.19 × 10–2 | 0.875(0.772, 0.991) | 3.57 × 10–2 | 0.879(0.776, 0.995) | 4.22 × 10–2 |
| Q3 | 1.115(0.993, 1.252) | 6.49 × 10–2 | 1.138(1.012, 1.280) | 3.06 × 10–2 | 1.150(1.023, 1.294) | 1.97 × 10–2 | |
| Glucose | Q2 | 0.936(0.827, 1.059) | 2.94 × 10–1 | 0.946(0.835, 1.070) | 3.76 × 10–1 | 0.955(0.844, 1.081) | 4.70 × 10–1 |
| Q3 | 1.133(1.009, 1.273) | 3.50 × 10–2 | 1.155(1.026, 1.299) | 1.67 × 10–2 | 1.167(1.037, 1.313) | 1.04 × 10–2 | |
| Lactose | Q2 | 0.901(0.797, 1.019) | 9.56 × 10–2 | 0.907(0.803, 1.026) | 1.21 × 10–1 | 0.920(0.814, 1.040) | 1.82 × 10–1 |
| Q3 | 1.101(0.981, 1.235) | 1.03 × 10–1 | 1.114(0.993, 1.251) | 6.65 × 10–2 | 1.117(0.995, 1.253) | 6.14 × 10–2 | |
| Maltose | Q2 | 0.973(0.866, 1.093) | 6.44 × 10–1 | 0.974(0.866, 1.095) | 6.56 × 10–1 | 0.967(0.861, 1.088) | 5.79 × 10–1 |
| Q3 | 1.000(0.889, 1.125) | 9.96 × 10–1 | 0.995(0.884, 1.119) | 9.30 × 10–1 | 0.982(0.873, 1.104) | 7.60 × 10–1 | |
| Intrinsic and milk sugars | Q2 | 0.969(0.855, 1.098) | 6.20 × 10–1 | 0.982(0.867, 1.113) | 7.79 × 10–1 | 0.986(0.870, 1.118) | 8.24 × 10–1 |
| Q3 | 1.145(1.017, 1.290) | 2.55 × 10–2 | 1.170(1.037, 1.320) | 1.06 × 10–2 | 1.157(1.025, 1.305) | 1.81 × 10–2 | |
| Non-milk extrinsic sugars | Q2 | 1.086(0.961,1.226) | 1.84 × 10–1 | 1.094(0.968, 1.1236) | 1.49 × 10–1 | 1.132(1.002, 1.279) | 4.65 × 10–2 |
| Q3 | 1.285(1.144,1.444) | 2.30 × 10–5 | 1.294(1.151, 1.454) | 1.50 × 10–5 | 1.321(1.175, 1.486) | 3.11 × 10–6 | |
| Sucrose | Q2 | 1.088(0.962, 1.230) | 1.79 × 10–1 | 1.099(0.972, 1.243) | 1.31 × 10–1 | 1.108(0.980, 1.254) | 1.02 × 10–1 |
| Q3 | 1.278(1.137, 1.436) | 4.05 × 10–5 | 1.290(1.147, 1.451) | 2.18 × 10–5 | 1.291(1.147, 1.452) | 2.21 × 10–5 | |
| Other Sugars | Q2 | 1.024(0.908, 1.154) | 7.04 × 10–1 | 1.028(0.912, 1.160) | 6.50 × 10–1 | 1.051(0.931, 1.185) | 4.21 × 10–1 |
| Q3 | 1.189(1.057, 1.337) | 3.95 × 10–3 | 1.190(1.058, 1.339) | 3.75 × 10–3 | 1.180(1.049, 1.328) | 5.83 × 10–3 |
Notes:
Model 1: Adjusted for sociodemographic factors, including age, sex, ethnicity, education level, and TDI.
Model 2: Additionally adjusted for behavioral factors, including BMI, smoking status, alcohol consumption, IPAQ, and supplement use.
Model 3: Further adjusted for medical conditions, including diabetes, cancer, hypertension, and cardiovascular disease.
Abbreviations: HR, hazards ratio; CI, confidence interval; Ref, reference; Q, quartile.
3.3. Stratified analyses
We stratified the population by sex and observed significant associations between sugar intake and dementia risk among females. Specifically, higher intakes of total sugars (HR = 1.516, 95 % CI: 1.250–1.838), free sugars (HR = 1.410, 95 % CI: 1.177–1.690), glucose (HR = 1.341, 95 % CI: 1.104–1.630), sucrose (HR = 1.485, 95 % CI: 1.226–1.798), and non-milk extrinsic sugars (HR = 1.494, 95 % CI: 1.241–1.798) were significantly associated with increased dementia risk. In contrast, no significant associations were found between total or subtype-specific sugar intake and dementia risk in males. However, after incorporating the sugar × sex interaction terms into the Cox models, the interaction effects were not statistically significant, indicating no meaningful difference in the associations between sexes. These findings are presented in Fig. 1, Fig. 2, while the interaction results are provided in Fig. S1.
Fig. 1.
Association between sugar intake and the risk of dementia in the female subgroup
Notes: Blue dots indicate the hazard ratio (HR), and the connecting line represents the 95 % confidence interval (CI).
Fig. 2.
Association between sugar intake and the risk of dementia in the male subgroup
Notes: Red dots indicate the hazard ratio (HR), and the connecting line represents the 95 % confidence interval (CI).
4. Discussion
This study examined the association between sugar intake and dementia risk using data from a large UK Biobank cohort. Higher consumption of total sugars, free sugars, non-milk extrinsic sugars, and sucrose was independently associated with an increased risk of dementia. Stratified analyses revealed significant associations between dementia risk and the intake of total sugars, free sugars, glucose, sucrose, and non-milk extrinsic sugars, particularly among females. These results align with previous research and support the hypothesis that a high-sugar diet constitutes a significant risk factor for dementia [22].
To our knowledge, this is the first prospective cohort study in a general population to systematically investigate the dose-response relationship between total sugar intake, its subtypes—including free sugars, fructose, glucose, sucrose, lactose, and other sugars—and incident dementia. Current evidence on dietary sugars and dementia risk remains limited and primarily focuses on composite dietary patterns. For example, Zhang et al. reported a positive association between high-sugar dietary patterns and dementia risk in a prospective cohort study, but their analysis did not differentiate among sugar subtypes, leaving unclear whether specific sugars (e.g., free sugars, fructose, glucose, sucrose, maltose, milk extrinsic sugars) contribute differently to dementia risk [23]. In contrast, our study not only reaffirmed the significant association between high sugar intake and dementia risk but also made an important analytical advance by carefully distinguishing sugar subtypes—including free sugars, fructose, and sucrose—and examining their distinct effects on dementia risk. This novel approach overcomes previous limitations and establishes a robust foundation for further mechanistic research.
Although a significant association between sugar intake and dementia risk was observed, the underlying mechanisms remain incompletely understood. Existing studies suggest several pathways through which a high-sugar diet may increase dementia risk, including insulin resistance, oxidative stress, inflammatory responses, and alterations in the gut microbiome [24], [25], [26], [27]. Insulin resistance, a hallmark of metabolic dysfunction commonly associated with high sugar consumption, has been increasingly implicated in the development and progression of dementia. Chronic consumption of high-sugar diets may lead to systemic insulin resistance, which not only impairs peripheral metabolic regulation but also disrupts insulin signaling in the brain. This disruption leads to synaptic dysfunction, reduced neuronal plasticity, and ultimately, cognitive decline. Neuronal insulin signaling plays a critical role in maintaining cognitive function, and its impairment may hinder amyloid-β (Aβ) clearance, promote abnormal tau phosphorylation, and induce synaptic damage—hallmarks of Alzheimer’s disease and related dementias. Although the present study did not directly examine these biological pathways, the observed associations at the population level are consistent with existing mechanistic research. These findings provide indirect support for the role of metabolic dysregulation in dementia and contribute additional epidemiological evidence regarding the potential impact of dietary sugar intake on cognitive health [28,29]. Furthermore, high sugar consumption may exacerbate neuronal damage by increasing oxidative stress. Oxidative stress contributes to mitochondrial dysfunction, lipid peroxidation, and neuronal apoptosis, all of which are central features of neurodegeneration. It may also damage the cerebral vascular endothelium, disrupt the blood–brain barrier, and initiate neuroinflammatory responses, thereby accelerating cognitive decline [30,31]. A high-sugar diet may also alter gut microbiota composition by increasing harmful bacteria, which can affect brain function via systemic circulation and exacerbate neuropathological changes [32,33]. Despite these insights, further experimental and mechanistic studies are needed to fully elucidate how sugar intake influences dementia risk.
Furthermore, our stratified analyses revealed a sex-specific association between sugar intake and dementia risk. Previous research shows that women differ from men in neuroendocrine regulation, metabolic profiles, and behavior, potentially increasing their vulnerability to impaired glucose metabolism and neurodegenerative diseases [34,35]. For instance, estrogen decline after menopause may exacerbate the harmful effects of a high-sugar diet on hippocampal glucose metabolism by reducing insulin sensitivity [36]. Metabolomics studies in Alzheimer’s disease (AD) mouse models have found that female mice display more severe impairments in glucose clearance, whereas male mice sometimes show improved clearance at certain ages [37]. Epidemiological data also indicate that women generally consume more sugar daily than men, and chronic sugar exposure may promote abnormal tau protein phosphorylation through the advanced glycation end products–receptor for advanced glycation end products (AGEs–RAGE) pathway, aggravating neurodegeneration [38,39]. These sex-related biological differences provide valuable insights into how sugar intake influences dementia risk, underscoring sex as a critical factor in glucose metabolism and dementia development. Further analysis of the sex × sugar intake interaction did not reach statistical significance, indicating that although stratified analyses revealed significant effects in females, the moderating role of sex in this association remains statistically inconclusive. This outcome may be due to the stringent significance threshold (p < 0.005) employed in this study to mitigate the risk of false positives arising from multiple testing. Although some stratified analyses showed significant associations, the interaction test did not meet this strict criterion, suggesting that sex differences in the effect warrant confirmation through larger sample sizes or independent cohorts.
Despite these important findings, several limitations warrant consideration and suggest directions for future research. First, dietary assessment limitations must be acknowledged. Although combining the Oxford WebQ and 24-hour dietary recalls improved data reliability, sugar intake was ultimately self-reported, which is subject to systematic underreporting of high-sugar foods due to social desirability bias [40]. Despite data adjustments (e.g., portion size correction, frequency conversion, nutrient algorithm calculations), changes in dietary habits during follow-up (e.g., voluntary sugar reduction) may not have been fully captured. Future studies integrating objective biomarkers, such as urinary glucose metabolites, could enhance exposure measurement accuracy. Second, due to the non-random availability of dietary recall data in the UK Biobank, concerns about the generalizability of the study findings arise. Participants who completed the dietary assessments may differ systematically from those who did not, potentially introducing selection bias. Additionally, a “healthy volunteer” selection bias within the UK Biobank cohort may have contributed to an underestimation of the observed positive associations, as participants tend to be healthier than nonparticipants [41]. Moreover, the majority of participants in the UK Biobank were White British, which may limit the generalizability of our findings to more ethnically diverse populations [23]. Third, this study did not directly investigate key mechanistic mediators linking sugar intake to dementia. Although sugar subtypes were distinguished for the first time, crucial molecular markers like advanced glycation end products (AGEs) were not measured [42,43], limiting subtype-specific neurotoxicity analysis. The absence of insulin resistance indices (e.g., HOMA-IR) and systemic inflammatory markers (e.g., IL-6, CRP) also precluded a comprehensive multi-omics framework [44]. Fourth, although participants with baseline dementia were excluded, individuals with other neurological disorders, such as stroke and Parkinson’s disease, were neither excluded nor adjusted for in the analyses, which may have introduced residual confounding. This represents a limitation of our study. Future research should consider addressing this issue to better clarify the relationship between exposure and dementia risk. Fifth, to ensure data quality, participants with physiologically implausible energy intakes—defined as less than 500 or more than 3500 kcal/day for women, and less than 800 or more than 4000 kcal/day for men—were excluded based on established criteria. Extremely low or high reported energy intakes often reflect reporting errors, recall bias, or technical issues during data entry. Including such implausible dietary records can distort dose–response relationships and increase measurement error, thereby reducing the statistical power of the analysis. Future studies using objective biomarkers of dietary intake could help overcome these limitations.
5. Conclusions
The present study confirmed a significant association between a high-sugar diet and dementia risk through prospective cohort analysis, revealing notable sex differences. Specifically, the association was more pronounced among women. These findings underscore the importance of developing targeted dietary intervention strategies and suggest that public health policies should prioritize stricter restrictions on high-sugar diets, particularly for the female population. Future research should further investigate the biological mechanisms underlying the relationship between sugar intake and dementia risk, incorporating multi-omics approaches and neuroimaging data to establish a more comprehensive scientific foundation for dementia prevention and treatment.
Statement of ethics
Ethical approval of the UKB study was granted by the National Health Service National Research Ethics Service (reference 11/NW/0382).
Funding
The Natural Science Basic Research Plan in Shaanxi Province of China [2021JCW-08].
CRediT authorship contribution statement
Yue Che: Writing – original draft, Methodology, Conceptualization. Wenming Wei: Writing – original draft, Visualization. Tingting Mao: Visualization. Lina Qin: Writing – original draft. Hanchi Wang: Methodology. Yijia Li: Methodology. Weixuan Da: Conceptualization. Jin Feng: Supervision. Li Liu: Methodology, Conceptualization. Bolun Cheng: Methodology, Formal analysis. Huan Liu: Supervision, Data curation. Yan Wen: Supervision. Yumeng Jia: Supervision, Data curation. Feng Zhang: Writing – review & editing, Methodology, Funding acquisition.
Declaration of competing interest
The authors have stated that they have no conflict of interest.
Acknowledgements
This study was conducted using the UK Biobank Resource (application number 46478). We thank the research subjects who participated in this study.
Footnotes
Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.tjpad.2025.100311.
Appendix. Supplementary materials
Data availability
The data supporting the findings of this study are available from the UK Biobank. Access to the UK Biobank resource is available to qualified researchers through an application process via their official website (http://www.ukbiobank.ac.uk/about-biobank-uk/).
References
- 1.Zhang Y.-R., Xu W., Zhang W., Wang H.-F., Ou Y.-N., Qu Y., et al. Modifiable risk factors for incident dementia and cognitive impairment: an umbrella review of evidence. J Affect Disord. 2022;314:160–167. doi: 10.1016/j.jad.2022.07.008. [DOI] [PubMed] [Google Scholar]
- 2.Nichols E., Steinmetz J.D., Vollset S.E., Fukutaki K., Chalek J., Abd-Allah F., et al. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health. 2022;7(2):e105–e125. doi: 10.1016/S2468-2667(21)00249-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hammond T.C., Xing X., Wang C., Ma D., Nho K., Crane P.K., et al. β-amyloid and tau drive early Alzheimer’s disease decline while glucose hypometabolism drives late decline. Commun Biol. 2020;3(1):352. doi: 10.1038/s42003-020-1079-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schaefer S.M., Kaiser A., Eichner G., Fasshauer M. Association of sugar intake from different sources with incident dementia in the prospective cohort of UK Biobank participants. Nutr J. 2023;22(1):42. doi: 10.1186/s12937-023-00871-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mergenthaler P., Lindauer U., Dienel G.A., Meisel A. Sugar for the brain: the role of glucose in physiological and pathological brain function. Trend Neurosci. 2013;36(10):587–597. doi: 10.1016/j.tins.2013.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhang X., Tong T., Chang A., Ang T.F.A., Tao Q., Auerbach S., et al. Midlife lipid and glucose levels are associated with Alzheimer’s disease. Alzheimer Dement. 2023;19(1):181–193. doi: 10.1002/alz.12641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cao F., Yang F., Li J., Guo W., Zhang C., Gao F., et al. The relationship between diabetes and the dementia risk: a meta-analysis. Diabetol Metab Syndr. 2024;16(1):101. doi: 10.1186/s13098-024-01346-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Muth A.-K., Park S.Q. The impact of dietary macronutrient intake on cognitive function and the brain. Clin Nutrit. 2021;40(6):3999–4010. doi: 10.1016/j.clnu.2021.04.043. [DOI] [PubMed] [Google Scholar]
- 9.Cao D., Lu H., Lewis T.L., Li L. Intake of sucrose-sweetened water induces insulin resistance and exacerbates memory deficits and amyloidosis in a transgenic mouse model of Alzheimer disease. J Biolog Chem. 2007;282(50):36275–36282. doi: 10.1074/jbc.M703561200. [DOI] [PubMed] [Google Scholar]
- 10.Robison L.S., Salinero A.E., Abi-Ghanem C., Mansour F.M., Kelly R.D., Tyagi A., et al. High-fat diet exacerbates cognitive decline in mouse models of Alzheimer’s disease and mixed dementia in a sex-dependent manner. J Neuroinflammation. 2022;19(1):110. doi: 10.1186/s12974-022-02466-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kim W.J., Lee S.J., Lee E., Lee E.Y., Han K. Risk of incident dementia according to Glycemic status and comorbidities of hyperglycemia: a nationwide population-based cohort study. Diabet Care. 2022;45(1):134–141. doi: 10.2337/dc21-0957. [DOI] [PubMed] [Google Scholar]
- 12.Liu L., Volpe S.L., Ross J.A., Grimm J.A., Van Bockstaele E.J., Eisen H.J. Dietary sugar intake and risk of Alzheimer’s disease in older women. Nutr Neurosci. 2022;25(11):2302–2313. doi: 10.1080/1028415X.2021.1959099. [DOI] [PubMed] [Google Scholar]
- 13.Chen H., Chen J., Cao Y., Sun Y., Huang L., Ji J.S., et al. Sugary beverages and genetic risk in relation to brain structure and incident dementia: a prospective cohort study. Am J Clin Nutr. 2023;117(4):672–680. doi: 10.1016/j.ajcnut.2023.01.015. [DOI] [PubMed] [Google Scholar]
- 14.Haroon N.N., Austin P.C., Shah B.R., Wu J., Gill S.S., Booth G.L. Risk of dementia in seniors with newly diagnosed diabetes: a population-based study. Diabet Care. 2015;38(10):1868–1875. doi: 10.2337/dc15-0491. [DOI] [PubMed] [Google Scholar]
- 15.Schaefer S.M., Kaiser A., Eichner G., Fasshauer M. Association of sugar intake from different sources with cardiovascular disease incidence in the prospective cohort of UK Biobank participants. Nutr J. 2024;23(1):22. doi: 10.1186/s12937-024-00926-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Liu Z., Huang H., Xie J., Hou L., Xu C. Different dietary carbohydrate component intakes and long-term outcomes in patients with NAFLD: results of longitudinal analysis from the UK Biobank. Nutr J. 2023;22(1):67. doi: 10.1186/s12937-023-00897-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kaiser A., Schaefer S.M., Behrendt I., Eichner G., Fasshauer M. Association of all-cause mortality with sugar intake from different sources in the prospective cohort of UK Biobank participants. Br J Nutr. 2023;130(2):294–303. doi: 10.1017/S0007114522003233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wark P.A., Hardie L.J., Frost G.S., Alwan N.A., Carter M., Elliott P., et al. Validity of an online 24-h recall tool (myfood24) for dietary assessment in population studies: comparison with biomarkers and standard interviews. BMC Med. 2018;16(1):136. doi: 10.1186/s12916-018-1113-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhang X., Hu W., Wang Y., Wang W., Liao H., Zhang X., et al. Plasma metabolomic profiles of dementia: a prospective study of 110,655 participants in the UK Biobank. BMC Med. 2022;20(1):252. doi: 10.1186/s12916-022-02449-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Qureshi D., Collister J., Allen N.E., Kuźma E., Littlejohns T. Association between metabolic syndrome and risk of incident dementia in UK Biobank. Alzheimer Dement. 2024;20(1):447–458. doi: 10.1002/alz.13439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lacy M.E., Gilsanz P., Karter A.J., Quesenberry C.P., Pletcher M.J., Whitmer R.A. Long-term glycemic control and dementia risk in type 1 diabetes. Diabet Care. 2018;41(11):2339–2345. doi: 10.2337/dc18-0073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sun Q., Yang Y., Wang X., Yang R., Li X. The association between sugar-sweetened beverages and cognitive function in middle-aged and older people: a meta-analysis. J Prevent Alzheimer Dis. 2022;9(2):323–330. doi: 10.14283/jpad.2021.71. [DOI] [PubMed] [Google Scholar]
- 23.Zhang S., Xiao Y., Cheng Y., Ma Y., Liu J., Li C., et al. Associations of sugar intake, high-sugar dietary pattern, and the risk of dementia: a prospective cohort study of 210,832 participants. BMC Med. 2024;22(1):298. doi: 10.1186/s12916-024-03525-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Yeh S.H.-H., Shie F.-S., Liu H.-K., Yao H.-H., Kao P.-C., Lee Y.-H., et al. A high-sucrose diet aggravates Alzheimer’s disease pathology, attenuates hypothalamic leptin signaling, and impairs food-anticipatory activity in APPswe/PS1dE9 mice. Neurobiol Aging. 2020;90:60–74. doi: 10.1016/j.neurobiolaging.2019.11.018. [DOI] [PubMed] [Google Scholar]
- 25.Reddy V.P., Zhu X., Perry G., Smith M.A. Oxidative stress in diabetes and Alzheimer’s disease. Bierhaus A, editor. JAD. 2009;16(4):763–774. doi: 10.3233/JAD-2009-1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gardener S.L., Rainey-Smith S.R., Martins R.N. Diet and inflammation in Alzheimer’s disease and related chronic diseases: a review. J Alzheimers Dis. 2016;50(2):301–334. doi: 10.3233/JAD-150765. [DOI] [PubMed] [Google Scholar]
- 27.Navalón-Monllor V., Soriano-Romaní L., Silva M., De Las Hazas M.-C.L., Hernando-Quintana N., Suárez Diéguez T., et al. Microbiota dysbiosis caused by dietetic patterns as a promoter of Alzheimer’s disease through metabolic syndrome mechanisms. Food Funct. 2023;14(16):7317–7334. doi: 10.1039/D3FO01257C. [DOI] [PubMed] [Google Scholar]
- 28.Berlanga-Acosta J., Guillén-Nieto G., Rodríguez-Rodríguez N., Bringas-Vega M.L., García-del-Barco-Herrera D., Berlanga-Saez J.O., et al. Insulin resistance at the crossroad of Alzheimer Disease pathology: a review. Front Endocrinol. 2020;11 doi: 10.3389/fendo.2020.560375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Barone E., Di Domenico F., Perluigi M., Butterfield D.A. The interplay among oxidative stress, brain insulin resistance and AMPK dysfunction contribute to neurodegeneration in type 2 diabetes and Alzheimer disease. Free Radic Biol Med. 2021;176:16–33. doi: 10.1016/j.freeradbiomed.2021.09.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hahm J.R., Jo M.H., Ullah R., Kim M.W., MO Kim. Metabolic stress alters antioxidant systems, suppresses the adiponectin receptor 1 and induces Alzheimer’s like pathology in mice brain. Cells. 2020;9(1):249. doi: 10.3390/cells9010249. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Yang C., He Y., Ren S., Ding Y., Liu X., Li X., et al. Hydrogen attenuates cognitive impairment in rat models of vascular dementia by inhibiting oxidative stress and NLRP3 inflammasome activation. Adv Healthc Mater. 2024;13(20) doi: 10.1002/adhm.202400400. [DOI] [PubMed] [Google Scholar]
- 32.Noble E.E., Olson C.A., Davis E., Tsan L., Chen Y.-W., Schade R., et al. Gut microbial taxa elevated by dietary sugar disrupt memory function. Transl Psychiatry. 2021;11(1):194. doi: 10.1038/s41398-021-01309-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Frausto D.M., Forsyth C.B., Keshavarzian A., Voigt R.M. Dietary regulation of gut-brain axis in Alzheimer’s Disease: importance of microbiota metabolites. Front Neurosci. 2021;15 doi: 10.3389/fnins.2021.736814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Subramaniapillai S., Almey A., Natasha Rajah M., Einstein G. Sex and gender differences in cognitive and brain reserve: implications for Alzheimer’s disease in women. Front Neuroendocrinol. 2021;60 doi: 10.1016/j.yfrne.2020.100879. [DOI] [PubMed] [Google Scholar]
- 35.Warren A. The relationship between gender differences in dietary habits, neuroinflammation, and Alzheimer’s disease. Front Aging Neurosci. 2024;16 doi: 10.3389/fnagi.2024.1395825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ahmed F., Kamble P.G., Hetty S., Fanni G., Vranic M., Sarsenbayeva A., et al. Role of estrogen and its receptors in adipose tissue glucose metabolism in pre- and postmenopausal women. J Clin Endocrinol Metabol. 2022;107(5):e1879–e1889. doi: 10.1210/clinem/dgac042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Srivastava H., Lasher A.T., Nagarajan A., Sun L.Y. Sexual dimorphism in the peripheral metabolic homeostasis and behavior in the TgF344-AD rat model of Alzheimer’s disease. Aging Cell. 2023;22(7) doi: 10.1111/acel.13854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Alhusseini N., Ramadan M., Aljarayhi S., Arnous W., Abdelaal M., Dababo H., et al. Added sugar intake among the saudi population. Abd El-Hack ME, editor. PLoS ONE. 2023;18(9) doi: 10.1371/journal.pone.0291136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Bhattacharya R., Alam M.R., Kamal M.A., Seo K.J., Singh L.R. AGE-RAGE axis culminates into multiple pathogenic processes: a central road to neurodegeneration. Front Mol Neurosci. 2023;16 doi: 10.3389/fnmol.2023.1155175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Agarwal P., Ford C.N., Leurgans S.E., Beck T., Desai P., Dhana K., et al. Dietary sugar intake associated with a higher risk of dementia in community-dwelling older adults. Rong S, editor. JAD. 2023;95(4):1417–1425. doi: 10.3233/JAD-230013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Fry A., Littlejohns T.J., Sudlow C., Doherty N., Adamska L., Sprosen T., et al. Comparison of sociodemographic and health-related characteristics of UK biobank participants with those of the general population. Am J Epidemiol. 2017;186(9):1026–1034. doi: 10.1093/aje/kwx246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Patil G., Kulsange S., Kazi R., Chirmade T., Kale V., Mote C., et al. Behavioral and proteomic studies reveal methylglyoxal activate pathways associated with Alzheimer’s disease. ACS Pharmacol Transl Sci. 2022;6(1):65–75. doi: 10.1021/acsptsci.2c00143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Mietelska-Porowska A., Domańska J., Want A., Więckowska-Gacek A., Chutorański D., Koperski M., et al. Induction of brain insulin resistance and Alzheimer’s molecular changes by western diet. IJMS. 2022;23(9):4744. doi: 10.3390/ijms23094744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Huber H, Ashton NJ, Weinhold L, Schmid M, Coenen M, Stoffel-Wagner B, et al. 1518-P: Plasma Biomarker Concentrations of Alzheimer’s Disease Are Associated with Insulin Resistance—Results from a Pilot RCT in Adults with Diabetes Risk. Diabetes. 2023 Jun 20;72(Supplement_1):1518-P. 10.2337/db23-1518-P. [DOI]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data supporting the findings of this study are available from the UK Biobank. Access to the UK Biobank resource is available to qualified researchers through an application process via their official website (http://www.ukbiobank.ac.uk/about-biobank-uk/).