Last week I discussed my upcoming appointment with Dr. Isaacson at the Weill Cornell Alzheimer’s Prevention Clinic, and the fact that he recommended that I try my best to achieve a body fat percentage under 27%. (And, that I am currently 28.9% body fat according to the InBody 770 body composition scan that I had done.) I wondered what the evidence is for this recommendation, and spent a little time reading the medical literature addressing body weight or body fat and its link to dementia.
I found out that it’s complicated. There have been numerous observational studies that have looked at the body weight or adiposity of participating subjects at the time of enrollment in the study and then followed the subjects over many years to see who developed mild cognitive impairment (MCI) or full blown dementia during the follow up period. These studies try to control for all of the potential confounding factors that could lead to a misleading result (like comorbid health conditions such as hypertension or type 2 diabetes, which are also known to independently affect brain function for the worse), and they can give us a sense of whether there is a link between being overweight and the development of cognitive impairment. But observational studies can never truly tell us whether something actually caused the outcome. Think about it this way: if we observed 100 smokers, counted how many ashtrays were in their homes, and then followed them for 10 years, we might find a link between a greater number of ashtrays and the development of lung cancer. Does this mean that the presence of ashtrays is the causative factor and that by simply removing the ashtrays from the home and only smoking outdoors that you will eliminate your risk of lung cancer? Obviously not, but that’s how an observational study – which simply shows a link between the presence of two things – can be misinterpreted if you try to infer causation from it. Correlation does not equal causation.
This is the problem with many of the studies looking at body fatness and dementia: they are observational. Many show an increased risk of dementia correlated with greater severity of obesity, especially if the obesity is more central, or visceral (meaning that a lot of your body fat is only inside your abdominal cavity underneath the muscle layer rather than outside the muscle), and if the overweight is present at midlife. However, some observational studies of older subjects show a higher incidence of dementia in normal weight people compared to people with severe obesity, leading to a debate about whether body fat is harmful or protective. Many dementia experts point out that people with Alzheimer’s may start to lose weight as a part of their disease process early on, which can lead to folks being lighter when they’re diagnosed with dementia. This would be along the lines of the ashtray example: saying that lower weight causes an increased risk of Alzheimer’s may be like saying that more ashtrays cause lung cancer. The correlation may be true, but the causation isn’t really clear. Another observational study published last summer looked at functional neuroimaging scans using single-photon emission computerized tomography (SPECT) from more than 17,000 people to measure blood flow and brain activity. (Low blood flow to the brain predicts future Alzheimer’s disease, and is also associated with other brain disorders like addiction, depression, ADHD, bipolar disorder, and schizophrenia, among other conditions.) They found that the heavier the subjects got, the less blood flowed to their brains. This was found in many different areas of the brain, including those known to be affected in Alzheimer’s disease:
This is suggestive, but still — it’s an observational study. So we can’t say that being overweight led to the lower brain perfusion. In fact, it could be the other way around: maybe lower blood flow to the brain leads to weight gain! You just can’t tell with an observational study.
This is where we turn to other types of studies to help us. Prospective randomized controlled trials are the gold standard for figuring out causation. If you take similar people and randomly assign them to two or more different treatment groups and then look and see whether outcomes are different, this can actually tell you if it’s likely that the treatment caused an outcome. So in our ashtray example, we would take all of the people with ashtrays in their homes and randomly assign half of them to remove the ashtrays from their homes and only smoke outdoors, while allowing the other half of them (the control group) to keep their ashtrays indoors. Then we would follow them over 10 years and find out that there is no difference in lung cancer rates. This would prove that the ashtrays indoors are not the cause of the lung cancer. Similarly, we could randomly assign people to start keeping ashtrays in their home when they didn’t previously, and we’d find that they don’t gave a greater risk of lung cancer than they did before – proving that adding ashtrays to your home doesn’t cause cancer.
In this way, to figure out whether being overweight actually causes dementia, we would have to take a huge group of normal weight people and make them intentionally become overweight and then find out in 10 or 20 years if they were more likely to develop dementia than the normal weight control group people who were assigned to maintain their weights the whole time. You can imagine that a clinical trial like this will never be done, for a myriad of reasons. But what can be done is taking a group of people with overweight or obesity and asking them to intentionally lose weight and then seeing whether they have a decreased risk of dementia compared to similar overweight people who don’t lose weight (a control group). Well, we don’t quite have that study done yet, but there was a clinical trial in which 80 people with obesity (average BMI 35.5) aged 60 or older who already had a diagnosis of mild cognitive impairment (83.7% were women, and 26.3% were APOE ϵ4 gene carriers) were randomly assigned to either usual medical care (the control group) or a 12 month nutrition counseling program aimed at helping them lose weight through calorie restriction. The investigators found that “BMI decrease was associated with improvements in verbal memory, verbal fluency, executive function, and global cognition, after adjustment for education, gender, physical activity, and baseline tests. This association was strongest in younger seniors (for memory and fluency) and in APOE allele ϵ4 carriers (for executive function). Changes in homeostasis model assessment-estimated insulin resistance, C-reactive protein, leptin and intake of energy, carbohydrates, and fats were associated with improvement in cognitive tests.” Their conclusion: people with obesity and mild cognitive impairment who intentionally lose weight through diet improve their cognitive function.
Similarly, a systematic review and meta analysis of 12 different clinical trials found that intentional weight loss improved cognitive function in people with obesity.
These studies still don’t prove that losing weight if you’re overweight or obese can prevent Alzheimer’s disease or other dementias (because the outcomes that these studies looked at were performance on cognitive tests, not actual diagnoses of dementia). But they are highly suggestive that this may be so, especially when combined with all of the observational evidence we have that obesity at midlife is associated with a higher risk of dementia in later life.
So. I am going to tow Dr. Isaacson’s line and say that I believe achieving a healthy body weight (that is to say, healthy level of body fat – especially visceral body fat) is likely to be important for brain health, for preserving my mind as best I can. I’m curious where he got his target body fat percentage of <27% from, and will try to remember to ask him that question when I see him in person in June. In the meantime, I am newly certified as a life and weight loss coach through the Life Coach School and happy to help you with your own weight loss struggles as I continue to work on my own goal of reaching 135 pounds! Just send me an email at [email protected] if you could use my one-on-one help. I will also eventually offer a self-directed online course to help people with implementing the lifestyle changes needed for weight loss and/or brain health — look for that later this year!
Project 135 stats:
Starting weight: 159.6 (May 2020)
Week 1: 157.2
Week 2: 155.6
Week 3: 155.4
Week 4: 153.8
Week 8: 150.8
Week 12: 150.8
Week 17: 155.2 (after sugar binges)
Week 19: 153.8
Week 25: 146.6
Week 26: 148.2 (after Halloween candy)
Week 28: 146.6
Week 29: 148.8 (after Thanksgiving)
Week 30: 148.4
Week 31: 149.0
Week 32: 152.2
Week 33: 153.0 (after Christmas)
Week 37: 153.8 (January 2021: Started intuitive eating experiment)
Week 38.5: 160.8 (on Feb 1, 2021 -> the experiment failed! Go back to NSNF)
Week 40: 156.4
Week 42: 151.6
Week 44: 153.0
Week 45: 154.4
Week 46: 152.4
Week 48: 151.0
Total weight loss: 8.6 pounds (5.4%)
Be well, everyone!
Love, Jen