“People as young as 30 who are obese may be at greater risk [of dementia],” The Independent reports.
This UK study examined a set 14-year period (1998 to 2011) and looked at whether NHS hospital records documenting obesity in adults above the age of 30 were associated with subsequent hospital or mortality records documenting dementia in the remaining years of the study.
Overall there was actually no significant association between obesity and dementia in later life.
When the researchers broke down the data into 10-year age bands (30s, 40s, 50s and 60s) they found that people in these age groups had increased risk of dementia. However, it must be remembered that the researchers were not looking at lifetime dementia diagnoses, but only looking at diagnoses in the remaining years of the study. Very few people in the younger age groups would have developed dementia over the following few years.
For example, the study found more than a trebled risk of dementia for people with obesity in their 30s, but this was based on only 19 people who developed dementia during the remaining years of the study. Calculations based on small numbers may be less reliable and should be given less "weight".
As expected the greatest number of subsequent dementia diagnoses occurred in people who were 70 or above when obesity was assessed, and obesity did not increase dementia risk in these people.
Aside from any dementia link or not, overweight and obesity are well established to be associated with a variety of chronic diseases and a healthy weight should be the aim.
Where did the story come from?
The study was carried out by two researchers from the University of Oxford and was funded by the English National Institute for Health Research.
The study was published in the peer-reviewed Postgraduate Medical Journal.
The UK media failed to report the various limitations of this research. This includes the lack of a significant association with dementia overall for the total cohort.
And while significant associations for people between the ages of 30 and 60 were found, these are based on only very small numbers who developed dementia during the study so may be less reliable.
As said, the links between vascular dementia specifically and obesity do seem to be more apparent, but this is to be expected.
It is also not clear in the study where the 50% increased risk for people in middle age comes from.
What kind of research was this?
This was a retrospective cohort study that aimed to examine how obesity in middle age may be associated with the risk of subsequent dementia.
The researchers say the worldwide prevalence of dementia in 2010 was around 35.6 million cases, which was estimated to double to 65.7 million by 2030.
Meanwhile we are in the midst of an obesity epidemic, with the World Health Organization reporting that in 2008 just over a third of all adults were overweight (BMI over 25kg/m²) while 10% of men and 14% of women were obese (BMI over 30kg/m²).
As the researchers say, with the rapidly increasing burden of dementia, it is important to identify which modifiable risk factors are associated. The researchers say there is growing evidence that mid-life obesity is associated with “dementia” overall.
Dementia is just the general term for problems with memory and thinking, which has different causes. Alzheimer’s disease is the most common cause of dementia, which is associated with characteristic symptoms and changes in the brain (the formation of protein plaques and tangles). The causes of Alzheimer’s are not fully understood, with increasing age and genetic factors being the most well established. Overweight and obesity are not currently established as risk factors for Alzheimer’s disease.
Meanwhile, vascular dementia – the second most common cause – has the same risk factors as cardiovascular disease, so there would be a plausible link between obesity and this type of dementia.
This study simply examined a set 14 year period (1998 to 2011) and looked at whether hospital re-ords documenting obesity in adults of different ages, was associated with subsequent documentation of dementia in the remaining years of the study.
What did the research involve?
This study used Hospital Episode Statistics (HES) data, which includes data for all hospital admissions including day cases in NHS hospitals in England between April 1998 and December 2011. They also linked with the Office for National Statistics (ONS) to identify deaths up to December 2011.
The researchers identified a cohort of people with obesity by looking for the first admission or day care visit where obesity was recorded as a diagnosis (according to the International Classification of Diseases [ICD] codes). They identified a comparison control cohort without obesity who had received day care or hospital admission for various medical, surgical conditions or injuries. They only included adults in the obesity and comparison groups who were aged 30 or older and did not have an admission for dementia at the same time as, or before, the date of admission when obesity was recorded.
For the obesity and comparison groups they searched the HES and ONS databases for all subsequent hospital care or deaths from dementia (according to ICD codes). The researchers say they subdivided admissions into those specifically documented to be due to Alzheimer’s disease or vascular dementia, and separately examined men and women.
They grouped obesity and comparison groups into 10-year age bands at the time obesity was first recorded, then compared their risk of dementia in the subsequent years. Adjustment was made for sex, time period of the study, region of residence and deprivation score.
What were the basic results?
There were 451,232 adults in the obesity cohort, 43% of whom were male (number in the comparison cohort not specifically reported).
Overall compared to controls, for the total cohort of all adults aged 30 or above, there was no statistically significant association between a hospital record of obesity and subsequent record of dementia in the remaining years of the study (relative risk [RR]0.98, 95% confidence interval [CI] 0.95 to 1.01).
However, when they were then split into 10-year age brackets, there was increased risk of subsequent dementia for people with obesity recorded in the age brackets:
- 30 to 39 (RR 3.48, 95% CI 2.05 to 5.61)
- 40 to 49 (RR 1.74, 95% CI 1.33 to 2.24)
- 50 to 59 (RR 1.48, 95% CI 1.28 to 1.69)
- 60 to 69 (RR 1.39, 95% CI 1.31 to 1.48)
There was no significant association between obesity and dementia for people with obesity between the ages 70 and 79, and an apparent decrease in risk of dementia for people above the age of 80 with obesity.
When they looked by specific type of dementia, there was no clear link between obesity and Alzheimer’s disease. For the full cohort of adults aged 30 or over, obesity actually seemed to decrease the risk of subsequently developing Alzheimer’s disease (RR 0.63, 95% CI 0.59 to 0.67). Then by age group there was an apparent increased risk for those with obesity in the ages 30 to 39 (RR 5.37, 95% CI 1.65 to 13.7); no association for those between the ages 40 and 59; then decreased risk of Alzheimer’s for those with obesity above the age of 60.
Obesity seemed to have a clearer link with risk of vascular dementia. The full cohort of adults aged 30 or over recorded to have obesity had a 14% increased risk of vascular dementia in the subsequent years of the study (RR 1.14, 95% CI 1.08 to 1.19). There were also significantly increased risks for all age groups up to the age of 69. For the 70 to 79 year age group there was no association, and for obese adults over the age of 80, obesity again seemed to decrease the risk.
How did the researchers interpret the results?
The researchers conclude that: “Obesity is associated with a risk of dementia in a way that appears to vary with age. Investigation of the mechanisms mediating this association might give insights into the biology of both conditions.”
As the researchers say: “The dataset spans 14 years and is therefore just a snapshot of people's lifetime experience of obesity.” The study is just looking at a set 14-year period (1998 to 2011) and looking at whether hospital records documenting obesity in adults of different ages, were associated with subsequent documentation of dementia in the remaining years of the study.
Therefore not only is the study looking at a snapshot of obesity in a 14-year period, is also looking at just a snapshot of time in which people could develop dementia in the remaining years of the study. For those in the cohort who were in their 70s or 80s when their obesity was recorded, you may expect that the study could have a better chance of capturing whether those people were ever going to develop dementia in their lifetime. However, for most of the people in the cohort who were between the ages of 30 and 60, their likelihood of developing dementia in the remaining few years of the study is low.
Therefore, this study cannot reliably show whether or not obesity in mid-life is associated with developing dementia, as the follow-up timeframe will not have been long enough for most people.
The main result of this study was that for all adults in the cohort there was no association between a hospital record of obesity and risk of any type of dementia in the subsequent years of the study.
Though the research did then find increased risks for 10-year age bands in the 30s, 40s, 50s and 60s, many of these analyses are based on only small numbers of people who developed dementia in the remaining years of the study.
For example, the highest more than trebled risk of dementia for people with obesity in their 30s was based on only 19 people who developed dementia during the remaining years of the study. An analysis based on such a small number of people has a much higher chance of error.
The 39% increased risk for people with obesity in their 60s was more reliable as this included 1,037 people from this age band who subsequently developed dementia.
But then the pattern is less clear, as for people with obesity in their 70s, of whom the largest number developed dementia (2,215), there was no association between obesity and dementia.
Meanwhile people who were obese in their 80s seemed to have decreased risk of then developing dementia.
Overall this makes a confusing picture from which to obtain any clear understanding of how obesity is associated with dementia. And it seems possible that various confounding hereditary, health and lifestyle factors may be having an influence.
Looking at Alzheimer’s specifically there was no clear link between adult obesity and Alzheimer’s. Therefore the study doesn’t provide evidence of obesity as a modifiable risk factor for the most common type of dementia. The only increased risk was for people with obesity in their 30s, but considering only five people developed Alzheimer’s in the remaining study years, this makes this risk association far from reliable. In fact for people over the age of 60, obesity apparently seems to be protective against Alzheimer’s for some reason. Though again it is highly possible this could be due to confounding from other factors.
As said, vascular dementia – the second most common type – has the same risk factors as cardiovascular disease, so there would be a plausible link between obesity and this type of dementia. And this study does support this, finding for the overall cohort of all adults above the age of 30, obesity was associated with a 14% increased risk of vascular dementia. Therefore, the study generally supports the link between obesity and this vascular condition.
Another point to bear in mind for this study is that, though it benefits from using a large reliable dataset of HES and ONS data which has recorded obesity and dementia based on valid diagnostic codes, it is of course only looking at hospital presentations of both obesity and dementia.
It is therefore unable to capture the large number of people with both of these conditions who may not have accessed hospital care.
Overall, this study contributes to the literature examining how the obesity epidemic may be associated with the growing prevalence of dementia worldwide, however it provides little in the way of conclusive answers.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Independent, 21 August 2014
Daily Mail, 21 August 2014
The Times, 21 August 2014
Daily Express, 21 August 2014
Links to the science
Postgraduate Medical Journal. Published online August 20 2014