, new research showed.
Incident dementia risk was more than three times higher in those who experienced just one episode of delirium, with each additional episode linked to a further 20% increase in dementia risk. The association was strongest in men.
Patients with delirium also had a 39% higher mortality risk than those with no history of delirium.
“We have known for a long time that delirium is dangerous, and this provides evidence that it’s even more dangerous than perhaps we had appreciated,” said study investigator Emily H. Gordon, PhD, MBBS, a geriatrician and senior lecturer at the University of Queensland, Brisbane, Australia.
“But we also know delirium is preventable. There are no excuses anymore; we really need to work together to improve the hospital system, to implement what are known to be effective interventions,” she added.
The findings were published online in The BMJ.
Close Matching
Prior studies that suggested an association between delirium and dementia were relatively small with short follow-up and varied in their adjustment for confounders. They also didn’t account for the competing risk for death, researchers noted.
Investigators used a linked New South Wales (NSW) statewide dataset that includes records of care episodes from all NSW hospitals as well as personal, administrative, clinical, and death information.
The study included an eligible sample of 626,467 older adults without dementia at baseline with at least one hospital admission between 2009 and 2014. For these patients, researchers calculated a hospital frailty risk score and collected other information including primary diagnosis and mean length of hospital stay and stay in the intensive care unit. From diagnostic codes, they categorized patients into no delirium and delirium groups and determined the number of delirium episodes.
Investigators matched patients in the delirium group to patients with no delirium according to characteristics with potential to confound the association between delirium and risk for dementia, including age, gender, frailty, reason for hospitalization, and length of stay in hospital and intensive care.
The matched study sample included 55,211 (mean age, 83 years) each in the delirium and the no delirium groups. Despite matching, the length of hospital stay for the index episode was longer for the delirium group than the no delirium group (mean, 9 days vs 6 days).
The primary outcomes were death and incident dementia, determined via diagnostic codes. During a follow-up of 5.25 years, 58% of patients died, and 17% had a new dementia diagnosis.
Among patients with at least one episode of delirium, the rate of incident dementia was 3.4 times higher than in those without delirium. After accounting for the competing risk for death, incident dementia risk remained three times higher among the delirium group (hazard ratio [HR], 3.00; 95% CI, 2.91-3.10).
This association was stronger for men than women (HR, 3.17 and 2.88, respectively; P = .004).
Sex Differences
The study is thought to be the first to identify a difference between sexes in dementia risk and delirium, Dr. Gordon said. It’s possible delirium in men is more severe in intensity or lasts longer than in women, or the male brain is, for whatever reason, more vulnerable to the effects of delirium, said Dr. Gordon. But she stressed these are only theories.
Investigators also found a mortality rate 1.4 times higher in the delirium group versus those without delirium, equating to a 39% increased risk for death (HR, 1.39; 95% CI, 1.37-1.41). The risk was similar for men and women (interaction P = .62).
When researchers categorized delirium by number of episodes, they found each additional episode was associated with a 10% increased risk for death (HR, 1.10; 95% CI, 1.09-1.12).
In addition to its large size, long follow-up, and close matching, what sets this new study apart from previous research is it accounted for the competing risk for death, said Dr. Gordon.
“This is really important because you’re not going to get dementia if you die, and in this population, the rate of death is incredibly high,” she said. “If we just assume people who died didn’t get dementia, then that screws the results.”
Causal Link?
For those who experienced at least one episode of delirium within the first 12 months, each additional episode of delirium was associated with a 20% increased risk for dementia (HR, 1.20; 95% CI, 1.18-1.23).
That dose-response association suggests a causal link between the two, Dr. Gordon said.
“The number one way to prove causality is to do a randomized controlled trial,” which isn’t feasible with delirium, she said. “By demonstrating a dose-response relationship suggests that it could be a causal pathway.”
Exact mechanisms linking delirium with dementia are unclear. Delirium might uncover preexisting or preclinical dementia, or it might cause dementia by accelerating underlying neuropathologic processes or de novo mechanisms, the authors noted.
Study limitations included the potential for residual confounding from unmeasured variables in the matching criteria. Delirium and dementia diagnoses depended on clinical coding of medical information recorded in the administrative dataset, and under-coding of dementia during hospitalization is well-recognized.
Although the study controlled for length of stay in hospital and in intensive care, this may not have fully captured differences in severity of medical conditions. Data about the duration and severity of delirium episodes were also unavailable, which limited the dose-response analysis.
Commenting on the findings, Christopher Weber, PhD, Alzheimer’s Association as director of Global Science Initiatives, said the results are consistent with other research on the association between delirium and incidents of dementia.
The increased risk for dementia following delirium in males is “an interesting finding,” said Dr. Weber. “This suggests a need for more research to understand the impact of sex differences in delirium, as well as research to see if preventing incidents of delirium could ultimately reduce rates of dementia.”
The study received support from the National Health and Medical Research Council: Partnership Centre for Health System Sustainability. Dr. Gordon and Dr. Weber reported no relevant conflicts of interest.
A version of this article appeared on Medscape.com.