Sustaining a traumatic brain injury (TBI) is associated with an increased risk of developing both dementia and Alzheimer's disease, according to a nationwide study of 2.8 million people over 36 years, published in The Lancet Psychiatry.
The study found that people sustaining a TBI were 24% more likely to be diagnosed with dementia than those without a history of TBI over the study period (5.3% of participants with dementia had a history of TBI vs 4.7% of participants without dementia; table 2). The risk of dementia increased with the number of TBIs and severity of injury, and even a single mild TBI (concussion) was linked with a higher risk of dementia (17% increased risk; 4.3% of participants with dementia had at least one mild TBI vs 4.0% of participants without dementia).
This study is one of the first to have a sufficient sample size and follow-up time to assess the effect of TBI in younger adults on long-term dementia risk.
"Individuals with a history of traumatic brain injury, including those with less severe injuries have an increased risk of developing dementia, even decades after the injury", says Jesse Fann, Professor of psychiatry and behavioural sciences at the University of Washington School of Medicine, in Seattle, USA, who led the study. "However, it's important to emphasise that although the relative risk of dementia is increased after traumatic brain injury, the absolute risk increase is low. Our findings do not suggest that everyone who suffers a traumatic brain injury will go on to develop dementia in later life." 
Dementia affects 47 million people worldwide, a number expected to double in the next 20 years. Every year, more than 50 million people worldwide experience a TBI.  TBI occurs when an external force such as a bump or blow to the head disrupts the normal function of the brain. Leading causes include falls, motor vehicle accidents, and assaults.
Previous research has suggested a link between TBI, including mild TBI (concussion) and subsequent dementia. Whether TBI among veterans and in contact sports such as football and boxing increases the risk of dementia has been hotly debated. Research so far has produced conflicting results, and the relationship between TBI and dementia remains poorly understood due to limitations of previous studies including: small sample sizes, short follow-up periods, limited information about the number, severity, and timing of head impacts, and lack of comparison to non-TBI trauma controls.
To address these methodological issues, Danish and American researchers identified a population-based cohort of nearly 2.8 million adults to examine the long-term risk of dementia in people with a history of TBI compared to those without TBI and those with non-TBI physical trauma (fracture not involving head or spine). The researchers tracked all diagnoses of TBI from Danish national registries from 1977 to 2013 and modelled the association between time since TBI, severity of TBIs, number of TBIs, and person's sex on the risk of dementia. They adjusted for other factors that could affect the risk of dementia including diabetes, heart disease, depression, and substance abuse. However, they were not able to account for some potential confounders such as education, high blood pressure, smoking, and obesity, which could result in the association between TBI and dementia being over or underestimated.
Over 36 years, 132,093 individuals (4.7%) had at least one TBI diagnosis, most (85%) were mild in severity. Between 1999 and 2013, 126,734 people (4.5%) aged 50 or older were diagnosed with dementia.
Compared with individuals without a history of TBI, the risk of dementia after the age of 50 increased consistently with the number of TBIs--22% higher risk with one TBI (4.6% participants with dementia vs 4.1% of participants without dementia), 33% higher with two (0.5% vs 0.4%) or three TBIs (0.1% vs 0.1%), 61% higher with 4 TBIs (?0.1% vs ?0.1%), and nearly three times the likelihood of dementia with five or more TBIs (?0.1% vs ?0.1%).
The researchers also found that a single severe TBI increased the risk of dementia by 35% (0.8% of participants with dementia had at least one severe TBI vs 0.1% of participants without dementia), whilst one mild TBI increased the risk by 17% (4.3% vs 4.0%).
Importantly, the younger the individual sustaining a TBI the higher the risk of subsequent dementia, when taking time since TBI into account. For example, individuals having a TBI in their 20s were 63% more likely to develop dementia about 30 years later compared to those who didn't sustain a TBI in their 20s (overall dementia rate 0.55 per 1000 person years vs 0.34 per 1000 person-years); whereas individuals sustaining a TBI in their 30s were 37% more likely to develop dementia 30 years later compared with those without a TBI in their 30s (1.67 per 1000 person-years vs 1.22 per 1000 person-years; figure 2).  The findings also show that men with a history of TBI had a slightly higher risk of developing dementia than women.
The association between TBI and dementia held true even when comparing people with a history of TBI to those with non-TBI fractures not involving the skull or spine.
According to Professor Fann, "Shedding light on risk factors for dementia is one of the most important tasks in health research. Our analysis raises some very important issues, in particular that efforts to prevent traumatic brain injury, especially in younger people, may be inadequate considering the huge and growing burden of dementia and the prevalence of TBI worldwide. Our findings suggest that improved traumatic brain injury prevention programmes may have an opportunity to reduce the burden of dementia worldwide." 
The authors note some limitations, including that the study included people taken from one country with a fairly similar ethnic population, so the findings cannot be generalised to all ethnic groups in other countries. They also note that they did not include TBIs treated by general practitioners, so the data might not have captured some less severe TBIs.
Writing in a linked Comment, Professor Carol Brayne from the Cambridge Institute of Public Health, University of Cambridge, Cambridge, UK says, "Now we need to tease out what is happening in terms of TBI, wider spectrum exposures, and how these occur across different ages, by gender, and also by community within societies. The association of TBI with different causes and how these change across time needs policy attention, as it is likely that prevention needs to be considered at societal, community, and local levels."
NOTES TO EDITORS
This study was funded by the Lundbeck Foundation.
 Quotes direct from author and cannot be found in text of Article.
 Person-years were used as the indicator for years at risk for dementia.
IF YOU WISH TO PROVIDE A LINK FOR YOUR READERS, PLEASE USE THE FOLLOWING, WHICH WILL GO LIVE AT THE TIME THE EMBARGO LIFTS: http://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(18)30065-8/fulltext