High midlife fitness tracked 31% lower dementia risk, independent of coronary artery calcium

A preventive checkup at the Cooper Clinic in Dallas can include two very different heart tests. One is a treadmill session that continues until the patient can go no further. The other is a scan that counts calcium deposits in the coronary arteries, a buildup that reflects the cumulative impact of risk factors on the artery wall. Only the treadmill measures something a person can train.

Researchers at Tel Aviv University and the Cooper Institute in Dallas matched 10,933 of those patients, tested at an average age of 57, to Medicare claims data. Writing in a journal of the American College of Cardiology, JACC: Advances, they report that lower fitness and higher calcium were independently associated with later dementia. Those with high fitness for their age and sex had about a third lower risk of dementia than the least fit. A calcium score of 100 or more went with an elevated risk, roughly a quarter higher than for lower scores.

Earlier studies, including work on this same clinic population, had tied midlife fitness to a lower incidence of dementia decades later. Calcium was the open question. Evidence linking it to dementia was limited and largely preliminary, and no one knew whether the two tests carried the same warning or two separate ones. The authors believe theirs is among the first cohort studies to jointly evaluate them against dementia risk.

Why treadmill fitness is the half of the dementia link you can train

Treadmill fitness, which researchers call cardiorespiratory fitness, is scored in METs. One MET, the energy or oxygen used while sitting quietly, is the resting baseline, so a score of 10 means working at ten times that rate. This cohort averaged 10.1 METs at maximal workload, with the least fit near 7 and the highly fit near 11. The low, moderate and high fitness categories were age- and sex-specific, so each person was ranked against peers.

The treadmill score is closely linked to habitual physical activity: those rated highly fit reported more than three times the weekly exercise of the least fit. The link with dementia was not confined to athletic levels, either. Dementia risk decreased steadily across most of the observed range, and the researchers’ model put each one-MET step up at an estimated 9% lower risk.

To climb that scale, the Physical Activity Guidelines for Americans set the benchmark: adults need 150 minutes of moderate-intensity physical activity a week, such as brisk walking, or 75 minutes of vigorous activity. By the CDC’s rule of thumb, moderate means you can talk, but not sing.

Whether getting fitter would lower anyone’s dementia risk is a question this study was not designed to answer. Nor does the study make a case for booking a calcium scan for the brain’s sake. The American Heart Association presents the scan as a way to decide whether preventive treatment, such as statin therapy, may be helpful. The test is not meant for people with a very low risk of heart disease.

How treadmill tests and calcium scans were matched to 868 dementia diagnoses

The participants, aged 40 to 80, were typically self-referred or referred by their employers, and none had a history of heart attack, stroke, heart procedures, cancer or dementia. Dementia diagnoses were drawn from their Medicare claims, from 1999 to 2019. The tracking window opened a median of 8.9 years after the clinic visit and lasted a median of 5.8 years. Over that stretch, 868 of them developed all-cause dementia, meaning dementia of any type, at an average age of 76.5.

The analysis allowed for the competing risk of death, since someone who dies first can never be diagnosed. It adjusted for age, exam year, education, body mass index, smoking, blood sugar, cholesterol and blood pressure, and weighed fitness and calcium against each other.

Both fitness and calcium were measured with instruments: a treadmill test pushed to the limit, and scans scored by a method shown to be highly reproducible. The cases, in the hundreds, were recorded long after those measurements. Still, a study that simply follows people has no way to hand out fitness or calcium at random. The result is a well-measured association, strong enough to take seriously, that cannot show either measure drives dementia.

Fitness and coronary calcium barely overlapped, and each tracked dementia on its own

In raw terms, the dementia rate was about 15 a year per 1,000 people among those with low fitness and about 11 among those with high fitness. After adjustment, high fitness went with a 31% lower risk of dementia than low fitness. Only 909 participants fell in the least-fit category, though, and the middle category could not be clearly told apart from it.

Calcium told its own story. Participants scoring 100 or more had about 16 cases a year per 1,000, compared with those with lower scores at about 10. Once fitness and the other factors were accounted for, that score carried a 24% higher risk, and dementia risk generally increased as calcium categories climbed.

The two measures were close to strangers. The share with a high calcium score fell only from 31.9% in the low-fitness group to 26.8% in the high-fitness group. There was no evidence of interaction, meaning neither measure changed what the other predicted. Among the 597 participants with calcium scores of 1,000 or more, dementia rates still decreased across fitness categories. The least fit had about 26 cases a year per 1,000, the highly fit about 14. Numbers that small leave wide margins of error.

The authors see two separate strands of heart health, functional fitness and cumulative vascular injury: the capacity a body has in hand, and the damage its blood vessels have absorbed. Whether coronary calcium contributes directly to cognitive decline or merely signals shared risk is still unclear. The authors suggest it is one manifestation of systemic vascular disease, a sign of wear throughout the circulation. Reduced blood flow, stiff arteries and inflammation have been implicated in the amyloid and tau protein buildup seen in Alzheimer’s, and fitness may mitigate many of these adverse processes. Accounting for strokes and mini-strokes along the way barely changed either estimate, suggesting that mechanisms beyond overt strokes may contribute. No brain scans were done, so none of this was measured directly.

What billing codes and a single treadmill test leave open about exercise and dementia

The authors list the weak points themselves. Dementia was identified using Medicare administrative claims, which may result in misclassification, with some cases missed or wrongly coded. Detection may also have been lopsided. People with more calcium “may have had more frequent health care encounters and, therefore, more opportunities to receive dementia-related diagnostic codes.” That could push the calcium estimate up, although participants generally had access to health care, which may limit the effect.

The cohort is predominantly white, well educated and relatively healthy, and about 70% men, so the numbers may not carry over to everyone. There was no evidence that the associations differed by sex, but women made up only 30% of the group, and their estimates were too imprecise to lean on. Each measure was assessed once in this analysis, so nothing here tracks what happens when fitness changes. Cause and effect could also run backward, since the processes leading to dementia begin decades before symptoms and could sap fitness long before a diagnosis. The authors argue that the gap between testing and Medicare tracking reduces concerns about reverse causation, which falls short of ruling it out.

Against earlier work, the fitness finding looks familiar. Large studies of US veterans found roughly 8% to 10% lower dementia risk per 1-MET higher fitness level, close to the estimate here. The calcium evidence had been shakier. In the Multi-Ethnic Study of Atherosclerosis, the link faded once stroke and heart disease were accounted for. The Rotterdam Study tied dementia to calcification in the aorta and carotid arteries, but not in the coronary arteries.

Clinical trials are further behind. The National Institute on Aging says no clear link between physical activity and preventing Alzheimer’s has been documented, citing a review that found encouraging but inconclusive evidence for exercise. Exercise already reduces the risk of diabetes, high blood pressure and stroke, conditions that can contribute to age-related cognitive decline.

Those benefits are reason enough to build fitness while the dementia question stays open.