Study finds multiple psychiatric diagnoses linked with substantially higher dementia risk

People with more than one psychiatric diagnosis were far more likely to be diagnosed with dementia during follow-up than those with a single mental health condition, a French study published in BMJ Mental Health reported, prompting calls for targeted screening while researchers caution the findings are observational.

The team at Paul Brousse Hospital examined medical records for 3,688 patients treated in the psychiatry department of Bicêtre Hospital between August 2009 and October 2023. The average age at baseline was 45 and follow-up lasted at least five years. Patients had been diagnosed with one or more common mental health disorders including depression, anxiety, psychosis, substance misuse, personality disorder or bipolar disorder.

Just over 70% of patients (2,608) had a single psychiatric diagnosis, 21.5% (789) had two, 6% (226) had three and 2% (65) had four or more. After adjusting for age and other diagnosed health conditions that could confound results, the investigators reported that having two psychiatric conditions was associated with roughly double the odds of a later dementia diagnosis compared with having one. Having three diagnoses was associated with more than a fourfold increase in odds, and four or more diagnoses with an approximately 11-fold increase.

The researchers highlighted the combination of a mood disorder — such as depression or bipolar disorder — and anxiety as particularly associated with increased dementia odds, reporting an increase in odds of up to 89.6%. In their paper the authors wrote that the "co-occurrence of psychiatric disorders" showed a "strong association" with an increased posterior probability of developing dementia and suggested that identifying patients with multiple psychiatric diagnoses could inform earlier detection strategies.

The study team also noted recent advances in dementia diagnostics, including biomarker development for blood and cerebrospinal fluid, and suggested that integrating such tools into care for high-risk patients could enhance detection and management.

Researchers and commentators stressed important limitations. The study was observational and cannot establish causation, and all participants were drawn from a single psychiatric department, which the authors said limits generalisability. They also acknowledged the analysis did not account for some potentially influential factors, such as family history of dementia and the duration of psychiatric illness.

Independent experts welcomed the research as adding useful information about risk patterns but reiterated the need for caution in interpreting associations as causal. They called for replication in broader, population-based samples and for studies that can examine timing, underlying biology and whether treating psychiatric conditions reduces later dementia risk.

The findings add to a growing body of research linking individual psychiatric disorders with elevated dementia risk. Large-scale studies and reviews have previously associated depression, anxiety and bipolar disorder with higher rates of later cognitive impairment, but this analysis is among the first to quantify how combinations of diagnoses relate to subsequent dementia diagnoses.

Dementia poses a major and growing public health burden. In the United Kingdom, organisations estimate that nearly one million people currently live with dementia, and modelling by University College London researchers has projected that figure could rise to about 1.7 million within two decades as the population ages. A recent analysis published in The Lancet identified multiple modifiable and nonmodifiable risk factors for dementia, and experts have emphasised prevention and early detection as priorities.

The study authors recommended developing targeted screening strategies for psychiatric patients with multiple diagnoses and suggested clinicians consider heightened monitoring for cognitive changes in this group. They also called for further research using biomarkers and broader datasets to better characterise risk and guide interventions.

Until such evidence is available, clinicians and patients are likely to weigh these findings alongside established clinical guidance. Public health authorities and clinicians may consider the study when designing services for mental health and cognitive care, but the observational nature of the data means it should not be interpreted as proving that psychiatric disorders directly cause dementia.