The risk of Alzheimer's disease and cognitive impairment characteristics in eight mental disorders: A UK Biobank observational study and Mendelian randomization analysis.

INTRODUCTION

The cognitive impairment patterns and the association with Alzheimer's disease (AD) in mental disorders remain poorly understood.

METHODS

We analyzed data from 486,297 UK Biobank participants, categorizing them by mental disorder history to identify the risk of AD and the cognitive impairment characteristics. Causation was further assessed using Mendelian randomization (MR).

RESULTS

AD risk was higher in individuals with bipolar disorder (BD; hazard ratio [HR] = 2.37, P < 0.01) and major depressive disorder (MDD; HR = 1.63, P < 0.001). MR confirmed a causal link between BD and AD (OR= 1.098), as well as obsessive-compulsive disorder (OCD) and AD (OR= 1.050). Cognitive impairments varied, with BD and schizophrenia showing widespread deficits, and OCD affecting complex task performance.

DISCUSSION

Observational study and MR provide consistent evidence that mental disorders are independent risk factors for AD. Mental disorders exhibit distinct cognitive impairment prior to dementia, indicating the potential different mechanisms in AD pathogenesis. Early detection of these impairments in mental disorders is crucial for AD prevention.

HIGHLIGHTS

This is the most comprehensive study that investigates the risk and causal relationships between a history of mental disorders and the development of Alzheimer's disease (AD), alongside exploring the cognitive impairment characteristics associated with different mental disorders. Individuals with bipolar disorder (BD) exhibited the highest risk of developing AD (hazard ratio [HR] = 2.37, P < 0.01), followed by those with major depressive disorder (MDD; HR = 1.63, P < 0.001). Individuals with schizophrenia (SCZ) showed a borderline higher risk of AD (HR = 2.36, P = 0.056). Two-sample Mendelian randomization (MR) confirmed a causal association between BD and AD (OR= 1.098, P < 0.05), as well as AD family history (proxy-AD, OR= 1.098, P < 0.001), and kept significant after false discovery rate correction. MR also identified a nominal significant causal relationship between the obsessive-compulsive disorder (OCD) spectrum and AD (OR= 1.050, P < 0.05). Individuals with SCZ, BD, and MDD exhibited impairments in multiple cognitive domains with distinct patterns, whereas those with OCD showed only slight declines in complex tasks.