Does having a coach make healthy habits work better for your brain?
Yes. In a two-year trial of 1,065 older adults across 11 Latin American countries, people given a structured, supervised lifestyle program improved their thinking scores about 55% faster than people given health advice and left to follow it alone.
Both groups got better. That is the first thing worth saying. Everyone in this trial was at high risk for dementia, and everyone’s cognitive scores went up over two years. The difference was speed. The coached group climbed at 0.31 standard deviations per year on a global thinking test. The self-guided group climbed at 0.20 per year. The gap of 0.11 per year is small in a single year and meaningful when you stack two years on top of each other.
This trial, called LatAm-FINGERS, matters for a second reason. Most dementia prevention research has been done in wealthy, mostly White populations in Northern Europe and North America. This one ran in Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, the Dominican Republic, Ecuador, Mexico, Peru and Uruguay. Nearly six in ten participants identified as Mestizo. The question was not just “does this work,” but “does this work when the food, the culture and the health system are different.”
What the data show
Researchers screened 1,719 people and enrolled 1,065 who were aged 60 to 77, had cardiovascular risk factors, scored 6 or higher on a dementia risk score, and were already performing below expectations on cognitive testing. Mean age was 67.5 years and 795 of them, or 75%, were women. Half were assigned to the structured program and half to the flexible one.
On the main thinking test, the structured group gained 0.31 SD per year and the flexible group gained 0.20 SD per year. The between-group difference of 0.11 SD per year is very likely between 0.06 and 0.15 SD per year, and a gap that size would be very unlikely if the program had no real effect (p<0.0001). Stretched across the full two years, that works out to roughly a 0.22 SD advantage for the coached group.
People also stuck with the coached version better. Dropout was 15.2% in the structured group versus 20.2% in the flexible group, an absolute difference of 5 percentage points, or about 50 fewer dropouts per 1,000 people over two years. That difference was statistically significant, but only just (p=0.042). Overall, 877 of 1,065 participants, or 82.3%, finished the full two years.
Dr. Kumar’s Take
The adherence numbers matter as much as the cognitive numbers. One group got a structured multidomain lifestyle program with supervised support and monitoring. The other got health advice. The structured group improved faster, and more of its members stayed in the trial.
That fits a familiar pattern in health habits. People rarely fail because they do not know what to do. They struggle when nobody is walking alongside them. Telling a 68-year-old to exercise more and eat better is nearly free. Building them a supervised program with a person who checks in is expensive, and this trial suggests the expensive version is the one that moves the needle.
I want to be honest about the size of the effect. A 0.11 SD per year difference on a composite score is not the same as preventing dementia. Nobody in this trial was followed long enough to know whether these gains translate into fewer diagnoses years later. What the trial does establish is that a program built in Finland can be rebuilt across 11 Latin American countries and still work.
How strong is the evidence?
This was a randomised trial, which is the strongest design available for questions like this. Assignment was random and balanced by study site, and the people scoring the cognitive tests did not know which group anyone was in. That masking matters, because it removes the most obvious way wishful thinking creeps into results.
The main weakness is unavoidable. Participants knew whether they were getting supervised sessions or only health advice, and you cannot blind someone to whether a coach is showing up. Part of the coached group’s gain may come from expectation and from simply practicing tests more often. Two years is also a short window for a disease that develops over decades.
Safety and side effects
There were 478 adverse events overall, 412 in the structured group and 66 in the flexible group. Muscle and joint symptoms led the list, affecting 21% of the structured group versus 2% of the flexible group. Upper respiratory infections followed at 9% versus under 1%, and 6% of the structured group had COVID-19 infections.
Serious adverse events occurred in 9% of the structured group and 5% of the flexible group, an absolute difference of 4 percentage points, or about 40 more per 1,000 people. None were judged related to the intervention. Eight people died during the trial, three in the structured group and five in the flexible group, and none of those deaths were related to the intervention either.
Practical Takeaways
- If you are over 60 with high blood pressure, diabetes or other heart risk factors, a structured program that covers several lifestyle areas at once, with supervised support, is the approach with trial evidence behind it, not any single habit on its own.
- Build in accountability rather than relying on willpower, whether that means a trainer, a class, a walking group or a standing appointment, since supervision was the main difference between the two groups here.
- Expect sore muscles and joints when you start, since that was by far the most common side effect, and scale the exercise up gradually instead of starting at full intensity.
- Build in regular check-ins on your progress, because monitoring was part of the structured program that did better.
Related Studies and Research
- Too much salt may speed up memory loss in older men
- Creatine and cognitive function in young and older adults
- Small Changes in Sleep, Exercise, and Diet Linked to Lower Risk of Dying Early
FAQs
What does a 0.11 SD per year difference actually mean for someone’s daily life?
A standard deviation is a way of measuring how spread out scores are in a group. A difference of 0.11 SD per year is modest at the individual level and would not be something you would notice in a single year. Over the two years of the trial it added up to roughly 0.22 SD. The trial did not run long enough to show whether the gap keeps growing after that. Researchers use composite scores like this because they are more stable than any single memory test, but they are a research measure, not a clinical diagnosis.
Why did both groups improve if only one got the real program?
Everyone in the trial got health advice, attention and repeated cognitive testing, so the comparison group was not a do-nothing group. Taking the same test repeatedly also produces practice effects, where scores rise partly because people get familiar with the test. On top of that, people who volunteer for a two-year dementia prevention trial tend to be motivated to change their habits regardless of which group they land in. This is why the between-group difference is the number that matters, not the improvement in either group alone.
Do I need a formal program, or can I do this on my own?
You can do all of it on your own, and the flexible group in this trial did improve. The trial’s point is that a structured program with supervised support and monitoring worked better than health advice alone and kept more people engaged. If you are self-directing, the fixable weakness is structure, so put the exercise on a calendar, join a group class, and check your progress on a schedule rather than when you remember.
Does this apply to people who already have memory problems?
This trial enrolled people at risk, not people already diagnosed. Participants were 60 to 77, had cardiovascular risk factors and scored below expectations on testing, but they did not have dementia. The findings say nothing about whether this approach helps someone who already has Alzheimer’s disease or another form of dementia. If you have noticeable memory problems, the right first step is a proper evaluation, not a lifestyle program.
Bottom Line
LatAm-FINGERS shows that a structured, culturally adapted lifestyle program improved thinking scores faster than health advice given without support, at 0.31 versus 0.20 SD per year in 1,065 at-risk older adults across 11 Latin American countries. The improvement was modest in size but statistically clear, and the program kept more people engaged over two years. The larger finding may be that dementia prevention research can be rebuilt for populations it has mostly ignored, and still work.
