Dementia Incidence in Patients With Prurigo Nodularis: An Observational Retrospective Cohort Study

Prurigo nodularis (PN) is a chronic neuroinflammatory dermatosis of unknown etiology characterized by severely pruritic, symmetrically-distributed nodules with a negative impact on quality of life [1]. While some consider PN a primary idiopathic disorder, it often arises as a secondary manifestation of underlying conditions, including atopic dermatitis (AD), chronic kidney disease (CKD), human-immunodeficiency-virus (HIV), and thyroid disorders [2]. Although patients with PN are at higher risk for psychiatric comorbidities including depression, anxiety, and suicidal ideation [3] research which investigates associations between PN and neurodegenerative disorders is limited [4]. Moreover, emerging evidence implicating sustained systemic inflammation, characteristic of psoriasis and AD, in the pathogenesis of dementias, particularly Alzheimer's, underscores the necessity of investigating whether inflammatory skin conditions like PN similarly influence the development of dementia [5, 6]. We conducted a large-scale, retrospective-cohort-study to assess the relationship between PN and dementia. We utilized de-identified electronic health record (EHR) data from the TriNetX Research Network (2005–2025) to explore whether patients diagnosed with PN are more likely to develop dementia. Participants were identified using ICD-10-CM codes. Individuals aged 65–95 with ≥ 2 separate instances of PN were identified and compared to a control cohort of individuals without PN; those with prior history of dementia were excluded. Eligible patients were 1:1 propensity-score-matched (PSM) by age-at-index, demographics, body mass index, and common risk factors for dementia and pruritis (Table 1). Outcomes were assessed over 20-years. Incidence and odds ratios (OR) with 95% confidence intervals were calculated to evaluate the likelihood of Alzheimer's, vascular, and nonvascular dementia. A time-to-event analysis was also performed using Cox proportional hazards to estimate risk, reported as hazard ratios (HR) with 95% confidence intervals. 3458 (14.5%) 435,945 (8.9%) 3454 (14.5%) 1818 (7.6%) 178,802 (3.7%) 1,817 (7.6%) 606,659 (12.4%) 780 (3.3%) 80,835 (1.7%) 778 (3.3%) 764 (3.2%) 875 (3.7%) 78,237 (1.6%) 873 (3.7%) 883 (3.7%) 18,329 (0.4%) 10,949 (0.2%) 254 (1.1%) 227 (1.0%) 4483 (18.9%) 599,360 (12.3%) 4480 (18.8%) 4472 (18.8%) 1443 (6.1%) 77,905 (1.6%) 1437 (6.0%) 2937 (12.3%) 268,111 (5.5%) 2.932 (12.3%) 2908 (12.2%) 10 (0.0%) 13 (0.0%) 10 (0.0%) 0 (0.0%) We identified 29,933 patients with PN and 5,288,405 controls who met study criteria. PSM yielded cohorts of 23,776 patients. Patients were predominantly White (61.4%) and female (54.4%) with a mean age-at-index of 67.6 ± 9.0 years. Over a 20-year follow-up period, patients diagnosed with PN were 23% more likely to develop Alzheimer's (521 vs. 427; OR = 1.23 [1.08–1.39]), 30% more likely to develop vascular dementia (377 vs. 292; OR = 1.30 [1.11–1.51]), and 11% more likely to develop nonvascular dementia (1568 vs. 1418; OR = 1.11 [1.03–1.20]) than individuals without PN (Table 2). When a time-to-event analysis was incorporated, findings were significant and consistent with the primary results, with modest elevations in hazard ratios across dementia subtypes (Alzheimer's disease HR = 1.30[1.14–1.48], vascular dementia HR = 1.37 [1.18–1.60], and nonvascular dementia HR = 1.19 [1.11–1.28]). Vascular dementia n (%) 1.46 (1.32–1.62) Nonvascular dementiaa n (%) In this retrospective-cohort-study, PN was associated with a > 11% likelihood of developing dementia, with Alzheimer's and vascular dementia identified as prevalent subtypes. While the absolute difference in dementia cases was modest, findings remained statistically significant after controlling for several major known confounders, including systemic and psychiatric comorbidities common in patients with pruritus. A supplementary time-to-event analysis using Cox models supported these incidence findings, showing similar relative patterns over follow-up without suggesting large time-dependent effects. The association between PN and dementia is likely multifactorial, involving overlapping mechanisms. PN frequently occurs in the context of chronic systemic diseases such as CKD, diabetes, and AD [2], conditions characterized by immune dysregulation and sustained systemic inflammation. These inflammatory processes may contribute to cytokine-mediated neuroinflammation, a key pathway implicated in neurodegeneration [5, 6]. Separately, PN has been strongly linked to neuropsychiatric comorbidities, particularly depression, a well-established risk factor for dementia [2, 3]. Chronic pruritus can contribute to psychological distress, creating a feedback loop that further exacerbates neuropsychiatric vulnerability. Together, these pathways may explain the increased incidence of dementia observed in patients with PN. While this study offers valuable insight, results should be interpreted cautiously given inherent limitations of retrospective database research. Diagnostic coding inaccuracies, surveillance bias, modest effect sizes, and absence of a baseline utilization proxy limit causal inference. Despite adjustment for major confounders, the dataset did not allow differentiation of PN subtypes, and residual confounding may persist due to other unmeasured factors such as disease severity and medication exposure. These hypothesis-generating results contribute to a growing body of evidence emphasizing the complex interplay between chronic cutaneous inflammatory conditions and neuroinflammation in the pathogenesis of neurodegeneration [5]. They underscore the need for further investigation into the etiological mechanisms by which PN associates with dementia, and highlight the potential value of heightened clinical vigilance, with consideration for early cognitive assessment in higher-risk PN patients. M.O. conceptualized, designed, analyzed, drafted, and revised the study. A.O. and C.R. assisted with study design, data curation, and manuscript revisions. C.E. provided oversight of study and assisted with all elements of the study's development. All authors reviewed and approved the final version of the manuscript and agree to be accountable for all aspects of the work. The authors received no specific funding for this work. This study was conducted using de-identified data obtained from the TriNetX research network. Because the data are fully de-identified and aggregated, the study was exempt from institutional review board (IRB) approval and did not require informed consent. Dr. Clinton Enos, corresponding author is an investigator for Amgen. All other authors, M.O., A.O., and C.R. report no disclosures. The data that support the findings of this study are available from the corresponding author upon reasonable request.