Child abuse was linked to higher dementia risk, and lifestyle factors explained less than a fifth of it
In more than 5,000 older adults in England, those who said they had been abused as children were diagnosed with dementia more often over about 11 years of follow-up. Education, habits, heart health and mental health together accounted for roughly 18% of that link, almost all of it through depression and anxiety. An observational study, so it shows links, not causes.
Abuse in childhood has long shadows. Earlier research has linked it to depression, heart disease and, more recently, a higher risk of dementia, the loss of memory and thinking skills severe enough to disrupt daily life. If that link is real, a practical question follows: does it run through things that can still be changed in adulthood, like smoking, exercise, blood pressure or mental health?
A team at University College London looked at this in a large, nationally representative study of older people in England. Child abuse was linked to an 80% higher risk of dementia, but the usual modifiable risk factors explained only about 18% of that link, and most of that share ran through depression and anxiety. Because this is an observational study, it shows associations; it cannot prove that abuse caused the dementia.
Why look at mediators?
A mediator is a step on the path between a cause and an effect. If child abuse raised dementia risk mainly because it made smoking or high blood pressure more likely, then tackling those in adulthood might remove much of the extra risk. Earlier studies mostly tested such factors one at a time, or lumped abuse together with other hard childhood experiences such as divorce or poverty. This study focused on abuse alone and tested several groups of mediators together, because they are bound up with each other (depression often goes together with smoking, smoking with heart disease, and so on).
What they did
The English Longitudinal Study of Ageing has interviewed adults aged 50 and over every two years since 2002. In 2007 participants gave a detailed life history, which included whether, before age 16, they had been sexually assaulted, seriously physically attacked, or had physically abusive parents.
The team followed 5,448 people who did not have dementia at that point for an average of 10.9 years. Dementia was identified from a reported doctor's diagnosis, a questionnaire filled in by someone close to the participant, or a drop in test scores combined with trouble with everyday tasks.
They grouped possible mediators, measured between 2002 and 2007, into four sets:
- education (the age at which people left school);
- health behaviours: smoking, frequent drinking, physical inactivity;
- heart and blood-vessel health: high blood pressure, obesity, diabetes, high cholesterol;
- mental health: depression or anxiety.
All analyses were adjusted for age, sex and signs of the family's circumstances in childhood (such as the number of books at home and the parents' jobs).
What they found
- 7.4% of participants reported child abuse, and 10.2% of the whole group developed dementia during follow-up.
- Abuse was linked to a higher dementia risk. In the main analysis, those who reported abuse had an 80% higher risk of being diagnosed with dementia than those who did not.
- Mental health was the clearest link in the chain. People who reported abuse had more than twice the odds of depression and almost three times the odds of anxiety. Depression and anxiety, in turn, were each linked to a roughly 58% higher risk of dementia.
- The other factors added little. Links between abuse and smoking, inactivity, obesity or diabetes were weak or uncertain, and abuse was not linked to how long people stayed in education.
- Together, the four groups explained about 18% of the link; depression and anxiety alone accounted for roughly 15 to 17%. In the model, shifting all four sets of factors in people who had been abused lowered their dementia risk by only about 8%.
What to keep in mind
- Association, not cause. The authors note that the mediation method rests on "strong and untestable assumptions"; hidden confounding, measurement errors and reverse causation remain possible.
- Abuse was reported decades later, by people in their fifties or older. Some may not have wanted to disclose it, and early memory problems could also lead to under-reporting; both would tend to make links look weaker than they are. The severity and timing of the abuse were not known.
- Depression and anxiety may partly be early signs of dementia rather than steps towards it, so the authors ask readers to interpret the mediation estimates cautiously.
- Some mediators were measured crudely. Drinking was recorded as how often, not how much, and most mediators were self-reported and reduced to yes/no categories.
- Who took part. The people analysed had higher incomes and more education than those left out, and the study could not measure post-traumatic stress, adversity in adult life or long-term poverty.
- When only complete records were used, without filling in missing data statistically, the estimates became very imprecise.
Why it matters
Most dementia-prevention advice focuses on blood pressure, exercise, smoking and similar factors in midlife and later. These matter for everyone, but this study suggests they account for only a small part of the extra dementia risk carried by people who were abused as children. The share that was explained ran mostly through mental health, which points, in the authors' view, to trauma-informed mental health care earlier in life. It echoes an earlier study on childhood abuse and slower thinking, where inflammation also explained only a sliver of the link. Most of the pathway from abuse to dementia remains unexplained, and the authors call for research into biological and social mechanisms.