Highlights
– In two independent cohorts (UCSF Memory and Aging Center and NACC), depression, anxiety, and PTSD were associated with earlier Alzheimer disease (AD) symptom onset.
– Within the UCSF MAC cohort, depression and anxiety were more common in early-onset AD, and PTSD was associated with the largest shift in age at onset (mean ~6.8 years earlier).
– The burden of psychiatric comorbidity showed a dose-dependent relationship with age at AD onset: 1 condition ≈ 1.5 years earlier, 2 conditions ≈ 3–3.3 years earlier, ≥3 conditions up to ~7.7 years earlier.
Background
Alzheimer disease (AD) is the leading cause of dementia worldwide and a major contributor to disability, morbidity, and healthcare costs. While age remains the dominant risk factor for sporadic AD, pre-existing psychiatric illnesses have emerged in epidemiologic and clinic-based studies as potential modifiers of dementia risk, timing, and presentation. Depression and anxiety have long been observed in association with later-life cognitive decline and dementia; whether they represent prodromal manifestations of neurodegeneration, independent risk factors, or both remains a critical question with implications for prevention and clinical care.
Study design
Investigators analyzed retrospective clinical data from two large AD cohorts to quantify psychiatric comorbidity prevalence and to examine associations with age at symptom onset.
Populations and data sources
– University of California, San Francisco Memory and Aging Center (UCSF MAC): n = 1,500 patients with a clinical diagnosis of AD, approximately half meeting criteria for early-onset AD (commonly defined as symptom onset <65 years).
– National Alzheimer’s Coordinating Center (NACC): >8,000 participants with AD diagnoses drawn from multiple Alzheimer Disease Research Centers across the United States; used as an external comparator for replication and generalizability.
Case ascertainment and psychiatric exposure measurement
Psychiatric conditions (including depression, anxiety, posttraumatic stress disorder [PTSD], bipolar disorder, and schizophrenia) were identified retrospectively from patient charts and initial clinic visit notes. The UCSF cohort evaluated lifetime histories, while NACC analyses focused on presence of current depressive or anxiety symptoms at assessment.
Outcomes and analysis
Primary outcome: age at AD symptom onset (patient- or informant-reported). Statistical comparisons assessed differences in age at onset according to psychiatric diagnoses and by number of psychiatric conditions. Analyses report absolute differences in years and statistical significance (P values); the primary published report also compared prevalence rates between early-onset and late-onset AD cases within the UCSF cohort.
Key findings
The study by Eijansantos et al. (2025) reports several clinically relevant findings from the UCSF MAC cohort with replication in the NACC dataset.
Prevalence of psychiatric conditions (UCSF MAC)
– Depression history: 43% of patients.
– Anxiety history: 32% of patients.
– Bipolar disorder: 1%.
– PTSD: 1%.
– Schizophrenia: 0.4%.
Higher rates among early-onset AD
Participants with early-onset AD had significantly higher rates of depression (P <.001) and anxiety (P <.001) compared with those with late-onset AD in the UCSF cohort.
Association with age at AD symptom onset
Compared with those without the specified psychiatric conditions, UCSF MAC participants with:
- Depression were on average 2.2 years younger at AD symptom onset (P <.001).
- Anxiety were on average 3.0 years younger at AD symptom onset (P <.001).
- PTSD were on average 6.8 years younger at AD symptom onset (P <.05).
In the NACC cohort, individuals with current depressive or anxiety symptoms were on average 2.1 years younger at AD onset compared with those without symptoms (P <.001), providing external consistency for the UCSF observations.
Psychiatric burden and dose–response relationship
Within UCSF MAC, an increasing number of psychiatric diagnoses correlated with progressively earlier AD symptom onset:
- One psychiatric condition: ~1.5-year reduction in age at onset (P <.001).
- Two conditions: ~3.3-year reduction in age at onset (P <.001).
- Three or more conditions: ~7.7-year reduction in age at onset (P <.001).
NACC data showed a similar pattern for one (1.5 years earlier, P <.001) and two disorders (3.2 years earlier, P <.001) though NACC analyses used current symptomatology rather than lifetime histories.
Secondary observations
– Bipolar disorder and schizophrenia were infrequent in these cohorts, limiting statistical power to evaluate their associations robustly.
– The largest single-condition effect estimate was seen for PTSD, although prevalence was low, and confidence intervals were not provided in the summary; replication in larger PTSD-enriched samples is needed.
Expert commentary: interpretation, strengths, and limitations
Interpretation and biological plausibility
The observed associations align with a body of literature suggesting that mood and anxiety disorders are linked to increased risk of cognitive decline and dementia. Potential mechanisms include chronic neuroinflammation, dysregulation of the hypothalamic–pituitary–adrenal (HPA) axis, glucocorticoid-mediated hippocampal injury, cerebrovascular disease mediated by behavioral and metabolic comorbidities, and lifestyle factors (physical inactivity, social isolation, poor sleep) that both exacerbate psychiatric disorders and accelerate neurodegeneration. An alternative (and not mutually exclusive) explanation is that psychiatric symptoms in midlife or later life can represent prodromal manifestations of incipient AD pathology.
Strengths
- Large clinical cohort (UCSF MAC) with detailed phenotyping and a replication dataset (NACC) that supports generalizability across research centers.
- Quantification of psychiatric burden (number of conditions) allowed demonstration of a dose–response relationship with age at onset.
- Clinically actionable outcome (age at symptom onset) rather than only long‑term incidence makes the findings directly relevant to clinicians managing patients with psychiatric histories.
Key limitations and potential biases
- Retrospective ascertainment of psychiatric diagnoses from clinical notes is vulnerable to misclassification, incomplete capture of severity/duration, and differential documentation practices across providers and centers.
- Temporality is uncertain: the studies cannot definitively separate psychiatric disorders as causal risk factors from neuropsychiatric prodromes of AD pathology. Early AD can present with mood and anxiety symptoms prior to clear cognitive decline.
- Referral and ascertainment bias: UCSF MAC is a tertiary memory clinic where patients with atypical or early-onset disease and complex psychiatric histories may be overrepresented. NACC adds breadth but retains research-center referral patterns.
- Residual confounding: lifestyle factors, socioeconomic status, vascular risk factors, medication exposures (including psychotropic drugs), and genetic risk (e.g., APOE ε4) may influence both psychiatric illness and dementia onset but were not fully addressed in the summary data presented.
- Low prevalence of bipolar disorder, schizophrenia, and PTSD in these cohorts reduces precision for these diagnoses and raises the possibility that estimates for rarer disorders are unstable.
Clinical implications and practical recommendations
For clinicians working in neurology, psychiatry, geriatrics, and primary care, several practical takeaways emerge:
- Systematically assess psychiatric history in patients presenting to memory clinics or those at risk for cognitive decline. A detailed lifetime psychiatric history (age at onset, course, treatment response, and current symptom burden) can inform prognostic conversations and risk stratification.
- Recognize psychiatric comorbidity as a potential marker of increased susceptibility to earlier AD onset. While causal inference is not established, high psychiatric burden may prompt closer longitudinal cognitive monitoring and earlier consideration of biomarker assessment where clinically indicated.
- Pursue aggressive management of modifiable contributors to both psychiatric morbidity and cognitive decline — vascular risk factor control, treatment of sleep disorders, optimization of social support, and evidence-based treatment of mood and anxiety disorders — recognizing that these interventions may benefit overall brain health even if their effect on AD onset timing is not yet proven.
- Be cautious in attributing new or worsening psychiatric symptoms in midlife or later life solely to primary psychiatric illness; consider neurodegenerative causes when symptoms are atypical, treatment-resistant, or accompanied by subtle cognitive or behavioral changes.
Research implications and next steps
The findings highlight several priority areas for future research:
- Prospective, population-based studies that capture psychiatric diagnoses (with standardized instruments), onset timing, duration, severity, and treatment exposure to examine temporality and potential causal pathways.
- Integration of biomarker data (amyloid, tau PET; CSF or plasma biomarkers; MRI measures of atrophy and cerebrovascular disease) to determine whether psychiatric disorders predict earlier clinical expression of AD pathology versus independent pathways to cognitive impairment.
- Studies in PTSD-enriched populations (e.g., military veterans) to validate the observed large effect estimates and explore mechanisms such as repeated stress exposures and neuroendocrine dysregulation.
- Randomized or pragmatic trials testing whether optimized psychiatric care and mitigation of modifiable risk factors can delay cognitive decline or delay symptom onset in at-risk individuals.
Conclusion
The study by Eijansantos et al. adds to growing evidence that psychiatric disorders — notably depression, anxiety, and PTSD — are associated with earlier clinical onset of Alzheimer disease and that cumulative psychiatric burden correlates with progressively earlier symptom emergence. Although causal pathways remain to be clarified, these data support thorough psychiatric assessment in memory-clinic populations, heightened vigilance for cognitive changes in patients with significant psychiatric histories, and interdisciplinary approaches that address psychiatric and vascular risk factors as part of brain health strategies.
Funding and clinicaltrials.gov
Funding sources were reported in the original publication (see reference). No clinicaltrials.gov registration applies to this retrospective cohort analysis; prospective interventional studies addressing psychiatric risk modification would require appropriate trial registration.
References
1) Eijansantos E, Allen IE, de Leon J, Grasso S, Rogers N, Bogley R, Paramo A, Ehrenberg AJ, Montembeault M, Sturm V, Spina S, Grinberg LT, Seeley WW, Rankin KP, Kramer JH, Rosen HJ, Rabinovici GD, Gorno-Tempini ML, Miller BL, Perry DC, Miller ZA. Burden of psychiatric disease inversely correlates with Alzheimer’s age at onset. Alzheimers Dement. 2025 Oct;21(10):e70677. doi: 10.1002/alz.70677. PMID: 41131662; PMCID: PMC12549220.
2) Ownby RL, Crocco E, Acevedo A, John V, Loewenstein D. Depression and risk for Alzheimer disease: systematic review, meta-analysis, and metaregression analysis. Arch Gen Psychiatry. 2006 May;63(5):530-538. doi:10.1001/archpsyc.63.5.530.
3) Diniz BS, Butters MA, Albert SM, Dew MA, Reynolds CF 3rd. Late-life depression and risk of vascular dementia and Alzheimer’s disease: Systematic review and meta-analysis of community-based cohort studies. Br J Psychiatry. 2013 May;202(5):329-335. doi:10.1192/bjp.bp.112.118307.
