Oxford’s Big Data Institute and Nuffield Department of Population Health report that daily step counts may help identify who will later be diagnosed with Parkinson’s disease, with lower activity patterns acting as an early marker of the condition.
Parkinson’s disease (PD) ranks as the second most prevalent neurodegenerative disorder and as the most rapidly growing, with an estimated 9.4 million cases in 2020 compared with 5.2 million in 2004. In the phases of the disease preceding a clinical diagnosis, subtle motor dysfunction and other early signs start to manifest as much as a decade before formal recognition. Signals in that phase offer clues for understanding disease development and for identifying potential modifiable risk factors.
Previous studies have linked lower self-reported physical activity with a higher risk of incident Parkinson’s disease. The progressive and prolonged path in development of Parkinson’s disease complicates the reading of causality into this risk, since underlying disease may already be present at baseline.
Daily step counts offer an appealing alternative since they provide a simple, objective proxy for physical activity and can be monitored closely with common consumer wearables or phones.
In the study, “Daily steps are a predictor of, but perhaps not a risk factor for Parkinson’s disease: findings from the UK Biobank,” published in npj Parkinson’s Disease, researchers used UK Biobank wrist accelerometer data to examine how accelerometer-derived daily step counts relate to incident PD and how that association changes across successive follow-up windows.
UK Biobank enrolled 502,536 adults in the United Kingdom between 2006 and 2010. A subset later joined the physical activity monitoring study and wore a research grade wrist accelerometer for up to seven days between 2013 and 2015. Final analyses included recordings from 94,696 individuals.
Moving data
Participants recorded an average of 9,446 daily steps. People in the highest step-count quintile, defined as 12,369 or more daily steps, tended to be younger and to have lower body mass index than those in the lowest group, defined as fewer than 6,276 daily steps.
Incident Parkinson’s disease cases showed an interesting pattern in relation to time. Among individuals who later developed PD, median daily step counts were higher when more years had passed between accelerometer wear and the first recorded diagnosis, yet step counts in these incident cases still remained lower than those of participants who never developed PD.
During a median follow-up of 7.9 years, investigators observed 407 incident PD cases, with a median 5.2 years from accelerometer wear to diagnosis.
Step-based risks
Analyses revealed participants walking more than 12,369 steps per day had a 59% lower risk of PD compared with those walking fewer than 6,276 steps per day.
Analyses then shifted from categories to a continuous perspective. Each additional 1,000 median daily steps was associated with an 8% lower risk of incident PD.
Extended to models that also adjusted for body mass index, depression, type 2 diabetes, constipation, bladder dysfunction, and accelerometer-derived sleep duration, pointed to health-related PD risk factors with links to physical activity and possible roadmarks on the causal pathway.
Subgroup analyses stratified by age group, sex, body mass index categories, and history of depression showed no significant differences in the association between daily steps and incident PD.
Study investigators then examined how the association between step counts and PD shifted over follow-up intervals to address reverse causation more directly. Strongest associations appeared in the earliest interval.
During the first two years of follow-up, every additional 1,000 daily steps (from the baseline assessment) was associated with a hazard ratio of 0.83, with 55 incident PD cases in that window. For follow-up periods longer than six years, the hazard ratio per 1,000 additional daily steps trended towards a less distinguishable difference, eventually reaching thresholds for statistical non-significance.
A signal rather than a risk factor
Results from this large prospective analysis indicate that higher daily step counts are associated with lower PD incidence when follow-up remains relatively short, offering clinicians a signal for early and focused monitoring. Findings support the labeling of low physical activity as a marker for PD rather than as a risk factor leading to the disease.