Abstract
Dementia is a growing global public health concern, with an increasing number of individuals affected due to the aging population. Although chronic inflammation is implicated in cognitive impairment, studies on its association with autoimmune diseases, which are representative chronic inflammatory diseases, are lacking. This study aimed to investigate whether the onset of autoimmune diseases is associated with an increased risk of dementia or cognitive impairment. This study used data from the Korean National Health Insurance Service–Health Screening from 2002 to 2019, with 8,743,801 person-years. The risk of dementia according to the type of autoimmune disease was calculated using the Cox proportional hazards model. The status and risk of cognitive impairment, which were assessed using the Korean Dementia Screening Questionnaire-Prescreening related with autoimmune diseases also estimated. Among participants, 8.3% developed dementia. All autoimmune diseases significantly increased dementia risk (hazard ratio [HR] 1.32; 95% confidence interval [CI] 1.29–1.35), particularly Alzheimer’s disease (HR 1.36; 95% CI, 1.32–1.40), vascular dementia (HR, 1.21; 95% CI, 1.12–1.30), and unspecified dementia (HR, 1.25; 95% CI, 1.18–1.33). Autoimmune diseases also increased the risk of positive dementia screening (odds ratio [OR], 1.18; 95% CI, 1.12–1.24), particularly connective tissue disorders (OR, 1.24; 95% CI, 1.15–1.34). This large-scale cohort study demonstrates a significant association between autoimmune diseases and increased risk of dementia and cognitive impairment. These findings underscore the potential importance of chronic inflammation in dementia pathogenesis and highlight the need for early cognitive screening and intervention strategies in patients with autoimmune diseases.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s11357-025-01904-2.
Keywords: Autoimmune diseases; Dementia in Alzheimer’s disease; Vascular dementia, KDSQ-P; National Health Insurance Service in Korea
Introduction
Cognitive impairment is an acquired disorder characterized by a decline in cognition involving multiple domains, including visuospatial abilities, executive function, memory, language function, and judgment [1]. Dementia, a major neurocognitive disorder, is a specific type of cognitive impairment, such as Alzheimer’s disease (AD) and vascular dementia (VD) [2]. The World Health Organization estimates that approximately 50 million individuals worldwide have dementia. The global annual medical expenditure for dementia has increased by a trillion dollars. Owing to the aging global population, its prevalence is projected to dramatically increase by 2030 and to affect more than 150 million individuals by 2050 [3]. Given the aging population and the profound socioeconomic impact of dementia, identifying modifiable risk factors is essential for early intervention and prevention strategies.
One emerging area of interest is the role of chronic inflammation in the pathogenesis of dementia. Inflammatory processes have been implicated in cognitive impairment through mechanisms such as blood–brain barrier (BBB) dysfunction, microglial activation, and neurodegeneration [4]. Autoimmune diseases, which are characterized by chronic systemic or organ-specific inflammation, share several pathogenic pathways with neurodegenerative disorders [5, 6]. Despite this, the association between autoimmune diseases and dementia remains underexplored.
Autoimmune diseases, including rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), Sjögren’s syndrome, multiple sclerosis (MS), and inflammatory bowel diseases (IBD) such as Crohn’s disease (CD) and ulcerative colitis (UC), have been associated with an increased risk of cognitive impairment [7–11]. Autoimmune diseases are caused by immune system dysregulation, characterized by persistent systemic inflammation that affects all organs. Systemic inflammation and autoantibody-mediated neuronal damage are potential mechanisms linking autoimmune diseases to neurodegenerative processes. For example, inflammatory cytokines such as TNF-α, IL-6, and IL-1β play a critical role in both autoimmunity and neurodegeneration by exacerbating neuroinflammation and promoting amyloid beta (Aβ) deposition, a hallmark of AD [12]. Additionally, microvascular dysfunction seen in autoimmune diseases like systemic sclerosis (SSc) may contribute to cerebrovascular pathology, increasing the risk of VD [13].
Furthermore, recent studies suggest that certain autoimmune diseases exhibit direct neuropsychiatric manifestations. For instance, neuro-Sjögren’s syndrome has been associated with cognitive deficits and memory impairment, sometimes preceding the formal diagnosis of the autoimmune condition [14]. Similarly, MS, a central nervous system (CNS)-targeting autoimmune disorder, leads to chronic neuroinflammation, axonal degeneration, and cognitive decline [15]. The gut-brain axis has also been proposed as a key link between autoimmune diseases and dementia, particularly in conditions such as IBD, where dysbiosis-induced systemic inflammation may affect brain function [16].
While previous research has mainly focused on single autoimmune diseases in relation to dementia [8, 10, 11]. However, recent studies have highlighted shared pathophysiological mechanisms—including systemic inflammation and immune dysregulation—across various autoimmune diseases, supporting the importance of evaluating these conditions collectively rather than individually [17, 18].
To address these gaps, the present study aims to provide a comprehensive assessment of the association between multiple autoimmune diseases and the risk of cognitive impairment and dementia using a large-scale national cohort, thereby generating broader and more reliable epidemiological evidence. Furthermore, this study evaluates the impact of autoimmune diseases on dementia screening results, offering important insights into early detection and potential intervention strategies.
Materials and methods
Data and study participants
This study used data from the National Health Insurance Service–Health Screening Cohort (NHIS–HealS) from 2002 to 2019. The NHIS–HealS comprises nationally representative cohort data collected and managed by the NHIS system in the Republic of Korea. The NHIS provides mandatory public health insurance that offers coverage of medical care services, including national health insurance, medical aid, and long-term care insurance, to all individuals residing in Korea [19, 20]. The NHIS establishes a cohort to provide valuable nationally representative information for public health professionals and policymakers using data from health insurance and health examinations. The NHIS–HealS comprises 514,866 health screening participants who represent a random selection of 10% of the approximately 5.1 million health screening participants in 2002 and 2003. It is a non-identifiable dataset that includes the type of insurance, socioeconomic characteristics, medical and hospital use history, and health examination results from 2002 to 2019 [21].
Figure 1 shows the selection flow of study participants and cohort profiles during follow-up periods. We selected two groups of participants from the NHIS–HealS. First, to investigate the incidence of dementia, we excluded participants who were diagnosed with dementia before the diagnosis of autoimmune diseases during the follow-up period (n = 999). This resulted in 514,866 participants. Second, we used data from the 2nd National Screening Program for Transitional Ages (NSPTA) in the NHIS–HealS. The NSPTA targeted Korean citizens aged 40 (1st) and 66 (2nd) years who fell within key transition periods in their lives [22]. Importantly, the 2nd NSPTA screened for cognitive impairment using the Korean Dementia Screening Questionnaire-Prescreening (KDSQ-P). We selected participants who completed the 2nd NSPTA in the NHIS–HealS (n = 79,153). Participants diagnosed with dementia before screening for cognitive impairment (n = 317) were excluded. In total, 78,470 participants were included in the second analysis.
Fig. 1.
Flow diagram of study participants and the National Health Insurance Service–Health Screening Cohort profile
Dementia and cognitive impairment
Medical information was based on medical facility visit history from the NHIS. All diagnoses followed the standardized protocol of the Korean Classification of Diseases and Causes of Death, 4th edition, which corresponds to the International Classification of Diseases, 10th revision (ICD-10) [23]. The dementia group was defined as participants who had ever visited a hospital facility with medical records that included the ICD-10 codes “F00 (dementia in Alzheimer’s disease),” “F01 (vascular dementia),” “F02 (dementia in other diseases classified elsewhere),” or “F03 (unspecified dementia)” during follow-up periods. We regrouped dementia diagnoses into three categories: “dementia in Alzheimer’s disease (F00),” “vascular dementia (F01),” and “unspecified dementia (F02 or F03).”
Cognitive impairment was evaluated using the short form of the KDSQ-P, which includes five questions that assess memory impairment [24]. The KDSQ-P is a brief, validated, informant-based questionnaire used widely in Korea to screen for dementia or cognitive impairment. It assesses memory, language, and daily functional changes through caregiver reports and is advantageous for large-scale community screenings due to its simplicity, cost-effectiveness, and feasibility without clinical expertise. Response options of “never,” “sometimes,” and “frequently” are scored as 0, 1, and 2, respectively, and the cut-off for cognitive impairment is a sum score ≥ 4 (sensitivity = 73.0%, specificity = 81.2%) [25]. Cognitive impairment was defined as a score ≥ 4 on the 2nd NSPTA during the follow-up period.
Autoimmune diseases
Autoimmune diseases were categorized based on the InterLymph classification [26]. This classification included inpatients who had the following ICD-10 codes: Sarcoidosis (D86), Rheumatoid arthritis (M05.1–05.9, M06.0, and M08.0), Systemic lupus erythematosus (M32, except for M32.0), Dermatomyositis/polymyositis (M33), Systemic sclerosis (M34, except for M34.2), Sjögren’s syndrome (M35.0), Bechet’s disease (M35.2), Ankylosing spondylitis (M08.1 and M45), Pernicious anemia (D51.0), Autoimmune hemolytic anemia (D59.1), Immune thrombocytopenic purpura (D69.38), Hashimoto’s thyroiditis (E06.3), Type 1 diabetes (E10), Grave’s disease (E05.0), Multiple sclerosis (G35), Myasthenia gravis (G70.0), Celiac disease (K90.0), ulcerative colitis (K51), Crohn’s disease (K50), and Psoriasis (L40). Based on disease characteristics, we classified them into two groups: connective tissue and organ-specific disorders. Connective tissue diseases refer to autoimmune diseases that cause damage to multiple organs simultaneously and involve multiple systems, such as systemic lupus erythematosus, and systemic sclerosis. Organ-specific disorders refer to autoimmune diseases in which the antigen is limited to a specific organ, such as thyroid disease and diabetes [27].
Covariates
Socioeconomic status included sex, age, and household income level. Age was classified into three groups (< 40, 41–60, and > 61 years). Household income levels (low, low-moderate, high-moderate, and high) were obtained from the baseline characteristics of the initial cohort year. Chronic diseases and health behavioral characteristics were based on the 2nd NSPTA data from the NHIS–HealS during follow-up periods. To assess health status, we used questionnaires to evaluate the past history of chronic diseases, such as hypertension, diabetes, dyslipidemia, cerebrovascular disease, and cardiovascular diseases. Health behavioral status included smoking and drinking statuses. Smoking status was categorized into three groups (never, past, and current). Current smokers were defined as those who had smoked more than 100 cigarettes (five packs of cigarettes) during their lifetime and were currently smoking. Never smokers were those who smoked fewer than 100 cigarettes in their lifetime. This definition aligns with the established criteria commonly employed by the Centers for Disease Control and Prevention and previous epidemiological studies on smoking behavior [28, 29]. Past smokers were defined as individuals who previously smoked but were not current smokers at the time of data collection. Binge drinking was defined as drinking > 7 (men) or > 5 (women) glasses of alcohol-containing drinks more than twice per week [30].
Statistical analyses
Baseline characteristics (sex, age, household income level, and autoimmune diseases) were described according to dementia that was newly onset during follow-up periods using the chi-square test in the initial cohort. The incidence and hazard ratio (HR) with a 95% confidence interval (CI) of dementia according to the type of autoimmune disease were estimated using the Cox proportional hazards model. To address the issue of multiple comparisons when multiple autoimmune diseases and dementia associations are analyzed, we applied a false discovery rate (FDR) correction using the Benjamini–Hochberg procedure. The risk of screening positive for cognitive impairment according to the type of autoimmune disease was calculated using logistic regression, adjusted for sex, household income level, chronic diseases (hypertension, diabetes, dyslipidemia, cerebrovascular and cardiovascular diseases), and lifestyle factors (smoking and drinking). The mean KDSQ-P score was also estimated according to the type of autoimmune disease. All analyses were performed using SAS, version 9.4 (SAS Institute, Cary, NC, USA).
Ethical consideration
This study was conducted in accordance with the ethical standards of the Declaration of Helsinki (1964) and its subsequent amendments. The NHIS–HealS data were anonymized before being released to the authors. All participants provided written informed consent to the NHIS. This study was approved by the Institutional Review Board (IRB) of Gil Medical Center, Gachon University (IRB number: GCIRB2021-380).
Results
The characteristics of the NHIS-NSC participants stratified by dementia status are shown in Table 1. This study included 514,866 participants (279,125 men and 235,741 women). During the follow-up periods, 8.3% (42,708 individuals) of the participants were diagnosed with dementia. Among those with dementia, significantly higher proportions were observed in women (n = 26,359 [11.2%]), the elderly aged > 60 years (n = 31,274 [26.3%]), those with low household income level (n = 15,944 [10.2%]), and those diagnosed with any autoimmune disease (n = 8304 [11.3%]).
Table 1.
The characteristics of study participants of the National Health Insurance Service–Health Screening according to dementia
| Characteristics at baseline | Total participants, n (% of column) |
Dementia during follow-up periods, n (% of row) | P-value | |
|---|---|---|---|---|
| No | Yes | |||
| Total participants | 514,866 (100.0) | 472,158 (91.7) | 42,708 (8.3) | |
| Sex | < 0.0001 | |||
| Male | 279,125 (54.2) | 262,776 (94.1) | 16,349 (5.9) | |
| Female | 235,741 (45.8) | 209,382 (88.8) | 26,359 (11.2) | |
| Age (years) | < 0.0001 | |||
| ≤ 40 | 25,350 (4.9) | 25,274 (99.7) | 76 (0.3) | |
| 41–60 | 370,602 (72.0) | 359,244 (96.9) | 11,358 (3.1) | |
| ≥ 61 | 118,914 (23.1) | 87,640 (73.7) | 31,274 (26.3) | |
| Household income level | < 0.0001 | |||
| Low | 156,054 (30.3) | 140,110 (89.8) | 15,944 (10.2) | |
| Low-moderate | 130,270 (25.3) | 119,634 (91.8) | 10,636 (8.2) | |
| High-moderate | 136,668 (26.5) | 126,768 (92.8) | 9900 (7.3) | |
| High | 91,874 (17.8) | 85,646 (93.2) | 6228 (6.8) | |
| Autoimmune diseases | < 0.0001 | |||
| No | 441,590 (85.8) | 407,186 (92.2) | 34,404 (7.8) | |
| Yes | 73,276 (14.2) | 64,972 (88.7) | 8304 (11.3) | |
Table 2 shows the cases and risks for dementia, stratified by autoimmune disease classification, connective tissue and organ-specific disorders, using the Cox proportional hazards model after adjusting for confounding variables. All types of autoimmune diseases were significantly associated with an increased risk of any types of dementia (hazard ratio [HR], 1.32; 95% confidence interval [CI], 1.29–1.35). Dementia in AD showed the highest risk (HR, 1.36; 95% CI, 1.32–1.40), followed by VD (HR, 1.21; 95% CI, 1.12–1.30) and unspecified dementia (HR, 1.25; 95% CI, 1.18–1.33). When stratified by autoimmune disease classification, of all dementia cases, 3653 individuals had connective tissue disorders. Connective tissue disorders were significantly associated with all types of dementia (HR, 1.22; 95% CI, 1.18–1.27). The highest risk was observed in dementia in AD (HR, 1.25; 95% CI, 1.20–1.30), followed by VD (HR, 1.16; 95% CI, 1.05–1.29), and unspecified dementia (HR, 1.16; 95% CI, 1.07–1.26). In the other group of 4651 cases with organ-specific disorders, a statistically significant association was found with any type of dementia (HR, 1.35; 95% CI, 1.31–1.39), dementia in AD (HR, 1.38; 95% CI, 1.34–1.44), VD (HR, 1.20; 95% CI, 1.10–1.32), and unspecified dementia (HR, 1.29; 95% CI, 1.19–1.39).
Table 2.
Results from the Cox proportional hazards model for dementia according to autoimmune diseases
| Autoimmune disease classification, cases (%) | Cases (%), hazard ratio (95% confidence interval) | |||
|---|---|---|---|---|
| All dementia | Dementia in Alzheimer’s disease | Vascular dementia | Unspecified dementia | |
|
All autoimmune diseases, 73,276 (14.23) |
8,304 (1.61), 1.32 (1.29–1.35) |
6,039 (1.17), 1.36 (1.32–1.40) |
852 (0.17), 1.21 (1.12–1.30) |
1,413 (0.27), 1.25 (1.18–1.33) |
|
Connective tissue disorders, 28,617 (5.56) |
3,653 (0.71), 1.22 (1.18–1.27) |
2,672 (0.52), 1.25 (1.20–1.30) |
375 (0.07), 1.16 (1.05–1.29) |
606 (0.12), 1.16 (1.07–1.26) |
|
Organ-specific disorders, 44,659 (8.67) |
4,651 (0.90), 1.35 (1.31–1.39) |
3,367 (0.65), 1.38 (1.34–1.44) |
477 (0.09), 1.20 (1.10–1.32) |
807 (0.16), 1.29 (1.19–1.39) |
All results were adjusted for age, sex, and household income level
The specific associations between each autoimmune disease and the risk of dementia are presented in the Supplementary Table. Among the connective tissue disorders, RA, dermatomyositis/polymyositis, and Sjögren’s syndrome showed statistically significant associations with AD (HR, 1.21, 1.44, and 1.14, respectively). In contrast, SSc was significantly associated with vascular and unspecified dementias (HR, 3.43 and 2.75, respectively). Moreover, ankylosing spondylitis was significantly associated with all types of dementia (AD [HR, 1.32], VD [HR, 1.29], unspecified dementia [HR, 1.32]). Among the organ-specific disorders, Graves’ disease, UC, and CD were significantly associated with AD, (HR, 1.28, 1.24, and 1.32, respectively). MS was associated with AD and unspecified dementia. Type 1 diabetes mellitus (T1DM) was significantly associated with all types of dementia (AD [HR, 1.79], VD [HR, 1.62], and unspecified dementia [HR, 1.58]). After applying the FDR correction to account for multiple statistical comparisons, the associations between autoimmune diseases and all dementia were re-evaluated. Significant associations remained robust for RA, SSc, Sjögren’s syndrome, ankylosing spondylitis, T1DM, Graves’ disease, MS, UC, and CD. Conversely, associations for SLE, dermatomyositis/polymyositis, Behçet’s disease, pernicious anemia, autoimmune hemolytic anemia, immune thrombocytopenic purpura, Hashimoto’s thyroiditis, myasthenia gravis, celiac disease, and psoriasis were no longer statistically significant after FDR adjustment. Conversely, no new statistically significant diseases were identified.
To further explore the association between autoimmune diseases and cognitive impairment, we analyzed participants of the 2nd NSPTA according to the results of the dementia screening test. Participants’ characteristics are presented in Table 3. The total number of participants was 78,470, and 12,316 (15.7%) tested positive in the dementia screening test. The participants were categorized based on sex, household income level, hypertension, diabetes, dyslipidemia, cerebrovascular disease, cardiovascular disease, smoking and drinking habits, and autoimmune diseases. Among the specific categorized factors, groups of women with diabetes, dyslipidemia, cerebrovascular disease, cardiovascular disease, and autoimmune diseases showed a significant difference between the participants testing positive and negative in the screening test, except for household income level and drinking habit.
Table 3.
The baseline characteristics of the study participants of the 2nd National Screening Program for Transition Ages according to autoimmune diseases
| Total participants, n (% of column) |
Cognitive impairment, n (% of row) | P-value | ||
|---|---|---|---|---|
| No | Yes | |||
| Total participants | 78,470 (100.0) | 66,154 (84.3) | 12,316 (15.7) | |
| Sex | < 0.0001 | |||
| Male | 39,472 (50.3) | 34,188 (86.6) | 5284 (13.4) | |
| Female | 38,998 (49.7) | 31,966 (82.0) | 7032 (18.0) | |
| Household income level | 0.1486 | |||
| Low | 25,494 (32.5) | 21,527 (84.4) | 3967 (15.6) | |
| Low-moderate | 20,786 (26.5) | 17,443 (83.9) | 3343 (16.1) | |
| High-moderate | 17,139 (21.8) | 14,355 (83.8) | 2784 (16.2) | |
| High | 15,051 (19.2) | 12,829 (85.2) | 2222 (14.8) | |
| Hypertension | 0.0004 | |||
| No | 45,788 (58.4) | 38,422 (83.9) | 7366 (16.1) | |
| Yes | 32,682 (41.6) | 27,732 (84.8) | 4950 (15.2) | |
| Diabetes | < 0.0001 | |||
| No | 65,524 (83.5) | 55,441 (84.6) | 10,083 (15.4) | |
| Yes | 12,946 (16.5) | 10,713 (82.7) | 2233 (17.3) | |
| Dyslipidemia | < 0.0001 | |||
| No | 69,796 (89.0) | 59,031 (83.7) | 10,765 (15.4) | |
| Yes | 8674 (11.0) | 7123 (80.8) | 1551 (17.9) | |
| Cerebrovascular disease | < 0.0001 | |||
| No | 75,324 (96.0) | 63,835 (84.7) | 11,489 (15.3) | |
| Yes | 3146 (4.0) | 2319 (73.7) | 827 (26.3) | |
| Cardiovascular disease | < 0.0001 | |||
| No | 72,333 (92.2) | 61,276 (84.7) | 11,062 (15.3) | |
| Yes | 6132 (7.8) | 4878 (79.6) | 1254 (20.4) | |
| Smoking status | < 0.0001 | |||
| Never | 53,869 (68.6) | 45,163 (83.8) | 8706 (16.2) | |
| Past | 15,833 (20.2) | 13,497 (85.3) | 2,336 (14.7) | |
| Current | 8768 (11.2) | 7494 (85.5) | 1274 (14.5) | |
| Alcohol consumption | 0.3084 | |||
| Never or social | 69,720 (88.8) | 58,810 (84.4) | 10,910 (15.6) | |
| Binge drinking | 8750 (11.2) | 7344 (83.9) | 1406 (16.1) | |
| Autoimmune diseases | < 0.0001 | |||
| No | 65,437 (83.4) | 55,467 (84.8) | 9970 (15.2) | |
| Yes | 13,033 (16.6) | 10,687 (82.0) | 2346 (18.0) | |
Table 4 presents the logistic regression analysis results of the positive cognitive impairment screening tests among participants with autoimmune diseases according to confounding factors. Fully adjusted models revealed that the risk of cognitive impairment significantly increased with any type of autoimmune diseases (odds ratio [OR], 1.18; 95% CI, 1.12–1.24). According to autoimmune disease classification, the ORs (95% CIs) of connective tissue and organ-specific disorders were 1.24 (1.15–1.34) and 1.11 (1.04–1.18), respectively.
Table 4.
Results from the logistic regression model for positive screening of cognitive impairment according to the types of autoimmune diseases
| Autoimmune disease classification, (cases) | Odds ratio (95% confidence interval) for positive screening of cognitive impairment |
|---|---|
| All autoimmune diseases (13,033) | 1.18 (1.12–1.24) |
| Connective tissue disorders (5019) | 1.24 (1.15–1.34) |
| Organ-specific disorders (8014) | 1.11 (1.04–1.18) |
All results were adjusted for sex, household income level, chronic diseases (hypertension, diabetes, dyslipidemia, cerebrovascular and cardiovascular diseases), and lifestyle (smoking and drinking habits)
The presence or absence of autoimmune diseases and their classifications is depicted in Fig. 2A, which shows the average KDSQ-P scores of the study participants. The participants with autoimmune diseases had significantly higher KDSQ scores than did those without autoimmune diseases, and the difference was particularly pronounced in connective tissue disorders.
Fig. 2.
A Mean KDSQ-P scores by autoimmune disease status, B Mean of the Korean Dementia Screening Questionnaire-Prescreening according to autoimmune diseases among the positive screen group. *Statistically significant difference compared with the non-autoimmune disease group
In Fig. 2B, the average KDSQ-P scores of the study participants who scored ≥ 4 points were classified based on the diagnosis of autoimmune diseases. Those with autoimmune diseases had significantly higher KDSQ scores, especially for connective tissue disorders, which had the highest scores observed.
Discussion
In this study, we found that autoimmune diseases were significantly associated with up to a 38% increased risk of dementia and cognitive impairment. Although this results represent a moderate risk contribution relative to well‐established factors such as hypertension and diabetes, the substantial prevalence of autoimmune disorders indicates that over the 30% risk elevation could not negligible at the population level [31, 32].
Specifically, patients with connective tissue disorders, such as RA, SLE, and Sjögren’s syndrome driven by chronic systemic inflammation and neurovascular damage, showed an increased risk of Alzheimer’s disease. Organ-specific autoimmune disorders, including T1DM and MS, which involve metabolic dysregulation and direct neuroimmune interactions, were strongly associated with both Alzheimer’s and vascular dementia. These findings support the role of autoimmune and inflammatory mechanisms in dementia. Additionally, our study found that individuals with autoimmune diseases were more likely to screen positive for cognitive impairment, reinforcing their link to early cognitive decline. These findings suggest that autoimmune diseases contribute not only to clinical dementia onset but also to preclinical cognitive dysfunction, underscoring the importance of early detection.
The role of chronic inflammation in dementia pathogenesis is well established, and our findings further support this link by highlighting the involvement of autoimmune disorders. Elevated levels of proinflammatory cytokines such as TNF-α, IL-6, and IL-1β may exacerbate neuroinflammation, disrupt the BBB, and facilitate Aβ deposition, all of which are key mechanisms in AD [12] (Fig. 3). Additionally, microvascular complications observed in SSc may contribute to cerebrovascular dysfunction, increasing the risk of VD [13]. Systemic autoimmune diseases can directly affect the nervous system. Extraglandular neurological manifestations of Sjögren’s syndrome, called neuro-Sjögren’s syndrome, are increasingly recognized [33]. Moreover, systemic inflammation is linked to vascular dementia. Our findings suggest that SSc may contribute via fibrosis-induced ischemia and reduced cerebral blood flow [34]. Notably, our study reveals a close correlation between ankylosing spondylitis and the occurrence of all types of dementia. Gene set enrichment analysis in patients with ankylosing spondylitis indicated considerable overexpression of COX7B and ATP5J, which are critical hippocampal genes in the pathogenesis of AD [35]. In a subset of patients with ankylosing spondylitis, TNF-α can increase Aβ production through the upregulation of both β-secretase expression and chronic neuronal TNF-α expression, resulting in extensive neuronal cell death [36].
Fig. 3.
The innate immune cells in peripheral blood, including neutrophils, NK cells, monocytes, and macrophages, infiltrate the brain and produce pro-inflammatory cytokines. Both Aβ and pro-inflammatory mediators can amplify peripheral inflammation and endothelial aging, leading to damage to the BBB and promoting the entry of Aβ through the BBB into the brain, while accelerating the deposition of Aβ in the brain. In the brain, infiltrated peripheral innate immune cells and inflammatory mediators can stimulate the activation of astrocytes and microglia, produce inflammatory cytokines, promote the deposition of Aβ in the brain, and lead to neuronal cell death, synaptic degeneration, inflammation, and gliosis, further exacerbating neurodegeneration and ultimately leading to dementia
Cognitive dysfunction is increasingly recognized as an important comorbidity and complication of T1DM. Summary data-based Mendelian randomization analysis indicated that the expression of three target genes related to antidiabetic drugs, ETFDH, GANC, and MGAM, was associated with an increased risk of autoimmune diseases [37]. In addition, abnormal glucose metabolism in diabetes itself plays a key role in the pathophysiological changes in AD by inducing oxidative stress and mitochondrial dysfunction [38, 39]. MS is the neurodegenerative disease marked by immune cell infiltration, microglial activation, astrocyte proliferation, and neuronal damage [15]. Inflammasome activation in monocytes and microglia triggers proinflammatory cytokines, such as IL-1β, TNF-α, and IL-6 [40]. Activated neutrophils accumulate in the blood and migrate to the CNS, leading to progressive damage to the BBB [41]. In addition, mitochondrial respiratory chain deficiency and transport abnormalities in MS plays a major role in neurodegeneration in AD [42]. The gut–brain axis involves intestinal inflammatory and neural pathways [16]. PARK7/DJ-1 plays against neurodegeneration, maintains the gut microbiome, and regulates intestinal inflammation [43]. Autoimmune diseases and UC share the hub genes PPARG and NOS2, which influence macrophage and microglial polarization [44]. Moreover, changes in intestinal microbial metabolites, such as metabolic disorders of bile acids, are observed in IBD, which aggravate the development of AD [45].
Furthermore, our study demonstrated that patients with autoimmune diseases had a significantly increased risk of developing cognitive dysfunction, as assessed by a prescreening test designed to identify individuals at risk of dementia. With Korea’s rapidly aging population and rising dementia prevalence [46, 47], early detection is increasingly important. The KDSQ-P demonstrated high validity and reliability for the early screening of dementia, distinguishing between stages of cognitive dysfunction [15, 48]. Compared to the Mini-Mental State Examination (MMSE), a widely used cognitive test administered directly to individuals, KDSQ-P relies on indirect assessments from informants. While MMSE evaluates cognitive performance through direct tools, KDSQ-P observes daily functional changes via informant reports, detecting subtle deficits that direct tests may miss. Validation studies have demonstrated that a higher KDSQ-P score correlates with lower MMSE scores, indicating its effectiveness in identifying cognitive decline [25]. Accurate screening minimizes misdiagnosis and ensures appropriate resource allocation. Early detection enables timely intervention, potentially slowing disease progression and improving outcomes and quality of life. Our findings indicate that autoimmune diseases are associated not only with an increased risk of clinically diagnosed dementia but also with cognitive impairment, potentially representing an intermediate or precursor state preceding dementia. The significant associations between autoimmune diseases and cognitive decline suggests autoimmune-mediated neuroinflammation may trigger subtle cognitive changes, ultimately progressing to overt dementia. Previous studies have demonstrated that chronic inflammation and autoimmunity accelerate neurodegeneration, contributing to preclinical cognitive decline [15, 49]. Therefore, cognitive impairment in these patients could serve as an early indicator or intermediate phenotype predicting subsequent dementia risk. These findings, considered together, highlight the importance of monitoring cognitive function in patients with autoimmune diseases. Future longitudinal studies should explore whether early detection and intervention can mitigate the dementia progression and support targeted prevention.
Our results are consistent with large-scale cohort studies reporting an increased risk of dementia in patients with autoimmune diseases, particularly VD and in middle-aged individuals with rheumatic conditions [7–10], although these studies had limited consideration of the full spectrum of immune-mediated diseases. These findings reinforce the hypothesis that chronic immune dysregulation contributes to neurodegeneration. However, some prior studies have reported inconsistent results, possibly due to short follow-up periods, small sample sizes, and lack of adjustment for cognitive changes [11]. Our study addresses these limitations by analyzing a broad spectrum of autoimmune conditions, including both connective tissue and organ-specific types, incorporating cognitive screening data, and applying robust statistical methods to enhance validity.
To the best of our knowledge, this is the first nationwide population-based study evaluating the subtype of autoimmune diseases associated with risk of onset and potential dementia. Despite its strengths, there are several limitations to consider. Given the observational design, causal inference cannot be established because residual and unmeasured confounding and possible reverse causation cannot be excluded. To more definitively assess causality, further studies are necessary using methods such as longitudinal designs, interventional trials, and Mendelian randomization analyses are necessary. Furthermore, dementia was defined based on the medical records of the patients who visited a hospital facility during the follow-up periods. Because South Korea operates a fee-for-service health system, presumptive diagnoses may be recorded as principal diagnoses, a common challenge in NHIS-based studies. However, previous researches indicated that ICD-10-based definitions can reliably capture clinically significant dementia cases in the NHIS, particularly severe diseases requiring specialist care such as dementia [21, 50–52]. While this method enabled the inclusion of a broad spectrum of dementia subtypes, some degree of misclassification is still possible. Rare autoimmune diseases may have been underestimated due to diagnostic limitations at initial visits. To overcome this, we included duplicate cases involving multiple autoimmune conditions; however, the model performed poorly. This reflects a well-known challenge in autoimmune disease research, highlighting the need for improved methods to address low prevalence in limited datasets. In addition, ourstudy could not account for disease severity or treatment modalities (e.g., immunosuppressants, biologics,corticosteroids). These factors are important because they may substantially modify dementia risk. Future research should incorporate such clinical variables to allow a more comprehensive assessment. Regarding the analytical approach, we grouped heterogeneous autoimmune diseases based on similar inflammatory mechanisms. While this integrative approach supported our hypothesis, it may be interpretatively misleading, given the heterogeneity of autoimmune diseases in pathogenesis, inflammatory profiles, and clinical presentation. Future studies should investigate disease- or pathway-specific associations to better identify which autoimmune mechanisms are most strongly linked to dementia, thereby enhancing our understanding of underlying pathophysiology and informing targeted prevention. Finally, although we applied the FDR method to correct for multiple statistical tests across autoimmune diseases and dementia types, the risk of type I error (false-positive findings) remains. Therefore, some observed associations, particularly those with marginal statistical significance after correction, should be interpreted cautiously and require confirmation from future studies. Moreover, our study did not include explicit competing risk analyses. Although our Cox models accounted for censoring events such as death, they did not address competing risks such as mortality from non-dementia or severe diseases potentially precluding dementia. This may result in an over- or underestimation of dementia risk. Future studies should incorporate detailed individual-level data on competing events to more accurately evaluation.
In conclusion, our study provides compelling evidence that autoimmune diseases significantly increase the risk of dementia and cognitive impairment, reinforcing the critical role of chronic inflammation in neurodegenerative processes. The inclusion of dementia screening assessments in autoimmune disease management may offer valuable opportunities for early intervention and improved patient outcomes.
Supplementary Information
Below is the link to the electronic supplementary material.
(DOCX 22.5 KB)
Funding
This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (No. RS-2025-00520480).
Data Availability
All rights to data of the Korean National Health Insurance Service–Health Screening are reserved by Korean National Health Insurance Service. The authors do not have permission to share data.
Declarations
Ethical approval
All participants provided written informed consent to the NHIS of the Republic of Korea. This study was approved by the Institutional Review Board (IRB) of Gil Medical Center, Gachon University (IRB number: GCIRB2021-380).
Competing interests
The authors declare no competing interests.
Footnotes
Wanhyung Lee and Xiaoxue Ma are co-first authors.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Wanhyung Lee and Xiaoxue Ma contributed equally to this work.
References
- 1.Park SH, et al. The characteristics of cognitive impairment and their effects on functional outcome after inpatient rehabilitation in subacute stroke patients. Ann Rehabil Med. 2017;41(5):734–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Slot RE, et al. Subjective cognitive decline and rates of incident Alzheimer’s disease and non–Alzheimer’s disease dementia. Alzheimers Dement. 2019;15(3):465–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Patterson C. World Alzheimer Report 2018—The State of the Art of Dementia Research: new frontiers. Alzheimer’s Disease International (ADI). London:2018.
- 4.Oláh C, et al. Cognitive dysfunction in autoimmune rheumatic diseases. Arthritis Res Ther. 2020;22(1):78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Charisis S, et al. Diet inflammatory index and dementia incidence: a population-based study. Neurology. 2021;97(24):e2381-91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Saeedi M, Rashidy-Pour A. Association between chronic stress and Alzheimer’s disease: therapeutic effects of saffron. Biomed Pharmacother. 2021;133:110995. [DOI] [PubMed] [Google Scholar]
- 7.Wotton CJ, Goldacre MJ. Associations between specific autoimmune diseases and subsequent dementia: retrospective record-linkage cohort study. UK. J Epidemiol Community Health. 2017;71(6):576–83. [DOI] [PubMed] [Google Scholar]
- 8.Lin T-M, et al. Autoimmune rheumatic diseases increase dementia risk in middle-aged patients: a nationwide cohort study. PLoS ONE. 2018;13(1):e0186475. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Li X, et al. Dementia and Alzheimer’s disease risks in patients with autoimmune disorders. Geriatr Gerontol Int. 2018;18(9):1350–5. [DOI] [PubMed] [Google Scholar]
- 10.Zhang Y-R, et al. Immune-mediated diseases are associated with a higher incidence of dementia: a prospective cohort study of 375,894 individuals. Alzheimers Res Ther. 2022;14(1):130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lu K, et al. Association between autoimmune rheumatic diseases and the risk of dementia. BioMed Res Int. 2014;2014(1):861812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Trzeciak P, Herbet M, Dudka J. Common factors of Alzheimer’s disease and rheumatoid arthritis—pathomechanism and treatment. Molecules. 2021;26(19):6038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wan S, et al. Plasma inflammatory biomarkers in cerebral small vessel disease: a review. CNS Neurosci Ther. 2023;29(2):498–515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Módis LV, et al. Central nervous system involvement in primary Sjögren’s syndrome: narrative review of MRI findings. Diagnostics. 2022;13(1):14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rossi B, et al. Common peripheral immunity mechanisms in multiple sclerosis and Alzheimer’s disease. Front Immunol. 2021;12:639369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Houser MC, Tansey MG. The gut-brain axis: is intestinal inflammation a silent driver of Parkinson’s disease pathogenesis? NPJ Parkinsons Dis. 2017;3(1):3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Jauregui W, et al. Shared pathophysiology of inflammatory bowel disease and psoriasis: unraveling the connection. Cureus. 2024;16(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Thurman JM, Yapa R. Complement therapeutics in autoimmune disease. Front Immunol. 2019;10:672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cheol Seong S, et al. Data resource profile: the national health information database of the National Health Insurance Service in South Korea. Int J Epidemiol. 2017;46(3):799–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Shin DW, Cho B, Guallar E. Korean national health insurance database. JAMA Intern Med. 2016;176(1):138–138. [DOI] [PubMed] [Google Scholar]
- 21.Seong SC, et al. Cohort profile: the national health insurance service-national health screening cohort (NHIS-HEALS) in Korea. BMJ Open. 2017;7(9):e016640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kim HS, et al. National screening program for transitional ages in Korea: a new screening for strengthening primary prevention and follow-up care. J Korean Med Sci. 2012;27:S70–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lee W, Kang M-Y, Yoon J-H. Cancer incidence among air transportation industry workers using the national cohort study of Korea. Int J Environ Res Public Health. 2019;16(16):2906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kim A, et al. Reliability and validity of a short form of the Korean dementia screening questionnaire-cognition. J Clin Neurol. 2020;16(1):145–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Yang D-W, et al. (2002) The development and validation of Korean dementia screening questionnaire (KDSQ). J Korean Neurol Assoc. 2022;135–141.
- 26.Wang SS, et al. Associations of non-Hodgkin lymphoma (NHL) risk with autoimmune conditions according to putative NHL loci. Am J Epidemiol. 2015;181(6):406–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Conrad N, et al. Autoimmune diseases and cardiovascular risk: a population-based study on 19 autoimmune diseases and 12 cardiovascular diseases in 22 million individuals in the UK. Lancet. 2022;400(10354):733–43. [DOI] [PubMed] [Google Scholar]
- 28.Klemperer EM, et al. Tobacco and nicotine use among US adult “never smokers” in Wave 4 (2016–2018) of the Population Assessment of Tobacco and Health Study. Nicotine Tob Res. 2021;23(7):1199–207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Jha P, et al. 21st-century hazards of smoking and benefits of cessation in the United States. N Engl J Med. 2013;368(4):341–50. [DOI] [PubMed] [Google Scholar]
- 30.Im H-J, et al. Binge drinking and its relation to metabolic syndrome in Korean adult men. Korean J Fam Med. 2014;35(4):173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Cheng G, et al. Diabetes as a risk factor for dementia and mild cognitive impairment: a meta-analysis of longitudinal studies. Intern Med J. 2012;42(5):484–91. [DOI] [PubMed] [Google Scholar]
- 32.Kivipelto M, et al. Midlife vascular risk factors and late-life mild cognitive impairment: a population-based study. Neurology. 2001;56(12):1683–9. [DOI] [PubMed] [Google Scholar]
- 33.Manzo C, et al. Cognitive function in primary Sjögren’s syndrome: a systematic review. Brain Sci. 2019;9(4):85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.McNair S, et al. Neuropsychiatric symptoms in scleroderma. Psychosomatics. 2013. 10.1016/j.psym.2012.09.002. [DOI] [PubMed] [Google Scholar]
- 35.Lari A, et al. Evaluation of the ankylosing spondylitis transcriptome for oxidative phosphorylation pathway: the shared pathway with neurodegenerative diseases. Iran J Allergy Asthma Immunol. 2021;20(5):563–73. [DOI] [PubMed] [Google Scholar]
- 36.Watad A, et al. TNF inhibitors have a protective role in the risk of dementia in patients with ankylosing spondylitis: results from a nationwide study. Pharmacol Res. 2022;182:106325. [DOI] [PubMed] [Google Scholar]
- 37.Meng L, et al. Causal association evaluation of diabetes with Alzheimer’s disease and genetic analysis of antidiabetic drugs against Alzheimer’s disease. Cell Biosci. 2022;12(1):28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Yu L, et al. Aberrant energy metabolism in Alzheimer’s disease. J Transl Intern Med. 2022;10(3):197–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Liao G, et al. Assessing neurobehavioral alterations among E-waste recycling workers in Hong Kong. Saf Health Work. 2024;15(1):9–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Waschbisch A, et al. Pivotal role for CD16+ monocytes in immune surveillance of the central nervous system. J Immunol. 2016;196(4):1558–67. [DOI] [PubMed] [Google Scholar]
- 41.Kooij G, et al. Specialized pro-resolving lipid mediators are differentially altered in peripheral blood of patients with multiple sclerosis and attenuate monocyte and blood-brain barrier dysfunction. Haematologica. 2020;105(8):2056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Barcelos IPd, Troxell RM, Graves JS. Mitochondrial dysfunction and multiple sclerosis. Biology. 2019;8(2):37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Pap D, et al. PARK7/DJ-1 as a therapeutic target in gut-brain axis diseases. Int J Mol Sci. 2022;23(12):6626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Dong L, et al. Shared genes of PPARG and NOS2 in Alzheimer’s disease and ulcerative colitis drive macrophages and microglia polarization: evidence from bioinformatics analysis and following validation. Int J Mol Sci. 2023;24(6):5651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Franzosa EA, et al. Gut microbiome structure and metabolic activity in inflammatory bowel disease. Nat Microbiol. 2019;4(2):293–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Kim YJ, et al. Prevalence and trends of dementia in Korea: a systematic review and meta-analysis. J Korean Med Sci. 2014;29(7):903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Park J, et al. Characteristics of work-related fatal injuries among aged workers in Republic of Korea. Saf Health Work. 2024;15(2):158–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Lee S-J, et al. Screening for normal cognition, mild cognitive impairment, and dementia with the Korean dementia screening questionnaire. Psychiatry Investig. 2018;15(4):384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Blanc F, et al. Cognitive dysfunction and dementia in primary Sjögren’s syndrome. International Scholarly Research Notices. 2013;2013(1):501327. [Google Scholar]
- 50.Choi JY, Lee S, Lee W. The impact of hearing loss on clinical dementia and preclinical cognitive impairment in later life. J Alzheimers Dis. 2021;81(3):963–72. [DOI] [PubMed] [Google Scholar]
- 51.Park KM, et al. Population-based dementia prediction model using Korean public health examination data: a cohort study. PLoS ONE. 2019;14(2):e0211957. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Hwangbo S, et al. Dementia incidence and population-attributable fraction for dementia risk factors in Republic of Korea: a 12-year longitudinal follow-up study of a national cohort. Front Aging Neurosci. 2023;15:1126587. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(DOCX 22.5 KB)
Data Availability Statement
All rights to data of the Korean National Health Insurance Service–Health Screening are reserved by Korean National Health Insurance Service. The authors do not have permission to share data.