More active people with Parkinson's did better over time on thinking and movement tests in a 294-person study

People with Parkinson’s disease are routinely told that staying active is good for them. What has been less clear is whether activity goes with a slower course of the disease over years, and whether that holds for thinking and memory as well as for movement.

A study published on October 7 in the journal Neurology followed 294 people with Parkinson’s, for up to five years on thinking and seven on movement, and found that it did, on both counts. The more active participants fared better over time. The differences in any one year were small, though, and the study cannot say whether activity slowed the disease or whether people with milder disease were simply able to do more.

What the Parkinson’s study found

The summary in the paper is short. Higher levels of physical activity were associated with slower cognitive and motor progression in PD. PD is Parkinson disease, cognitive refers to thinking and memory, and motor refers to movement.

The lead author, Dong-Woo Ryu of the University of Cincinnati and The Catholic University of Korea, stressed the thinking side. “Previous research has shown associations between physical activity and movement symptoms. Our study found physical activity may have dual benefits, with associations with slower declines in movement and cognition,” Ryu said.

The University of Cincinnati’s announcement was careful about what that means. The study does not prove that more physical activity causes slower progression of the disease; it only shows an association between the two.

The paper’s own starting point was modest too. Evidence for the effect of physical activity on disease progression in Parkinson disease (PD) remains limited, it begins.

How physical activity and progression were measured

The participants came from a long-running research program. The researchers analyzed data from individuals with mild-to-moderate PD enrolled in the Cincinnati Cohort Biomarker Program between January 2019 and February 2025. A total of 294 individuals with PD (mean age, 66.4 years; 33.0% female) were included.

Activity was reported by the participants. Physical activity was assessed using standardized interviews, with people reporting weekly estimated hours of physical activity at work, at home and during hobbies or recreational activities. That is broader than exercise. Moderate activity included brisk walking, doubles tennis, light swimming and household activities such as washing windows, vacuuming or mowing the lawn. Vigorous activity included jogging, fast bicycling, team sports and singles tennis as well as shoveling and carrying heavy loads.

From those answers the team worked out three things for each person: total time spent active, a volume measure that gave extra weight to vigorous activity, and time spent on vigorous activity alone.

For the main comparison, people were split into thirds by total time. The low group had up to four hours per week of activity, the medium activity group had four to eight hours, and the high group had more than eight hours.

Progression was tracked with two standard tests, repeated every year. One was a short test of thinking. The cognitive test had a score range of zero to 30, with lower scores indicating worse thinking and memory. The other was a rating of movement symptoms, on which higher scores are worse. The Parkinson’s symptom assessment had a score range of zero to 132, with an average score of 23 at the start of the study.

Coauthor Alberto Espay pointed to the repeated testing as the study’s strength. “Because participants have undergone standardized annual motor and cognitive evaluations and detailed assessments of physical activity, we could examine trajectories extending for up to seven years rather than taking a snapshot at a single point in time,” Espay said. Thinking was followed for up to five years and movement for up to seven.

How big the differences in progression were

On the thinking test, the three groups separated in order. The least active third had an average estimated annual decrease in cognitive scores of 0.478, the medium group had a decrease of 0.090, and the high group had a decrease of just 0.007. The least active third, in other words, lost about half a point a year on a 30-point test, and the most active third lost almost nothing.

The movement scores went the same way. There the low group had an average estimated annual score increase of 0.143, while the middle group’s score fell by 0.225 points a year and the most active group’s by 0.783. An increase is a worsening. So the least active group got slightly worse each year, and the two more active groups were rated slightly better. Improvement on a symptom scale in a progressive disease usually reflects treatment as well as the disease, and the comparison that matters here is between the groups.

The formal comparison set the top third against the bottom third. After adjusting for age, sex and disease duration, the high group showed slower cognitive decline and slower worsening of movement symptoms when compared with the low group. On thinking, the gap was 0.47 points a year, with a range from 0.20 to 0.74. On movement it was 0.93 points a year, with a range from 0.02 to 1.84.

Two things stand out in those figures. The first is that the movement estimate only narrowly excludes no difference at all. The second is scale. A gap of under one point a year, on a 132-point scale where participants started at an average of 23, is small. For comparison, a Dutch exercise trial set its bar at 3.5 points on a version of the same scale, the difference that was deemed a-priori clinically relevant, meaning large enough to matter to patients.

Ryu acknowledged this. “While our study found small differences between activity levels, these differences may become more meaningful over time,” Ryu said. Taken at face value, the movement gap in the new study would take about four years to reach the size that trial treated as meaningful.

What kind of physical activity seemed to matter

The team tried to pull apart how long people were active from how hard they worked. All three measures were linked to slower decline when looked at one by one. But when they were set against each other, intensity dropped out. In the paper’s words, vigorous physical activity was no longer associated with outcomes after adjustment for total physical activity time.

Timing was the other question. People were asked about the past decade as well as the present. Current activity still tracked with slower decline after accounting for previous 10-year physical activity.

The authors drew the two results together. The beneficial associations of moderate-to-vigorous physical activity may be related to activity duration and ongoing engagement. Ryu put it in plainer terms. “The possible benefits of physical activity for people with Parkinson’s disease may depend on how long they stay active and whether they continue exercising over time,” Ryu said.

On intensity, a randomized trial points in a different direction. In that study, 128 were randomly assigned to 1 of 3 groups (high-intensity exercise, moderate-intensity exercise, or control). All had been recently diagnosed and were not yet on medication. The exercise was on a treadmill, four days a week for six months. Movement scores barely changed in the high-intensity group and worsened by about three points in the group that did no added exercise. The high-intensity group, but not the moderate-intensity group, reached the predefined nonfutility threshold compared with the control group. Reaching a nonfutility threshold means the result was promising enough to justify a larger trial.

The two studies measured different things, structured treadmill sessions in one and the whole of daily activity in the other. They do not settle between them whether effort or hours counts for more.

How this fits with earlier Parkinson’s research

Two earlier studies both drew on another large research cohort to ask a similar question, and both found a similar link.

The first followed 380 people with early Parkinson’s. It reported that activity levels decline longitudinally in PD, which means they fall as the years pass. It also found that higher self-reported physical activity is associated with less disease progression. Its authors suggested that interventions to increase physical activity in early PD could potentially modify the disease course.

The second looked at how activity held up over time. That analysis ultimately included 237 patients with PD. Its conclusion was that a sustained increase in overall regular physical activity levels in patients with early PD was associated with slower decline of several clinical parameters. Those authors were direct about the limits of the approach. The study was observational in nature instead of interventional, they wrote, so causal relationships between the variables could not be assessed. Of studies that measure activity only once, at the start, they warned that their results may merely reflect differences in disease traits.

Trials can show cause, and for symptoms they have. In the Dutch trial, 130 people cycled on a stationary bike at home or did stretching for six months. The authors concluded that aerobic exercise can be done at home by patients with Parkinson’s disease with mild disease severity and it attenuates off-state motor signs. Aerobic exercise is the kind that raises the heart rate, attenuates means lessens, and off-state means measured when medication has worn off. They added that future studies should establish long-term effectiveness and possible disease-modifying effects.

A Cochrane review pulled together 154 trials of many kinds of exercise, from dance to strength training. Most helped with movement symptoms, and the type mattered less than expected. The reviewers concluded that exercise was important while the exact exercise type might be secondary. The trials were small, though. The number of participants per study was small (mean 51, range from 10 to 474). Effects were measured up to six weeks after the intervention, which says little about the years-long course of the disease. The reviewers wrote that larger, well-conducted studies are needed to increase confidence in the evidence.

Practice guidance reflects that state of knowledge. A guideline from the American Physical Therapy Association gives aerobic exercise its strongest rating for reducing movement symptoms, and is plain about what is not known. Though it is not clear whether the effects of aerobic exercise are disease modifying, the potential to reduce motor disease severity with aerobic exercise warrants early intervention. Disease modifying means changing the underlying course of an illness, as opposed to easing its symptoms.

What the Parkinson’s activity study cannot show

The paper lists its main weaknesses in a single line. Limitations include the observational design, self-reported physical activity, and potential selection bias.

Which way the link runs. Parkinson’s makes movement harder, and activity tends to fall as it advances. People whose disease is milder or slower may find it easier to stay active. If so, slow progression would produce high activity, and not the other way round. The analysis adjusted for age, sex and how long people had had the disease, which does not remove that possibility.

Reported activity. Nobody wore a monitor. A limitation of the study was that physical activity was reported by participants, and some people may not have reported their activity accurately.

Who took part. Selection bias means the people studied may not be typical. Participants had mild-to-moderate disease and kept returning to a research clinic for yearly testing. Two thirds were men. The results may not apply to people with advanced Parkinson’s.

Room to decline. Ryu raised a point about the thinking test. “Since participants generally had good thinking and memory skills at the start of the study, our results should be interpreted with caution, and further studies are needed,” Ryu said.

Small margins. The yearly differences were modest, and the movement result was only just distinguishable from none.

The study was supported by a grant from the Gardner Family Foundation and the Precision Medicine in Neurodegenerative Diseases Fund.

Which kind and amount of exercise suits a particular person with Parkinson’s is assessed by their doctor.

The study adds up to seven years of follow-up to the case that activity and a gentler course of Parkinson’s go together, and it leaves the question of cause to trials that last as long.