Digital Measures Track Progression in Early Parkinson’s Disease Over 36 Months in the WATCH-PD Study
In the 36-month WATCH-PD study, composite digital gait and speech measures demonstrated the most consistent and significant tracking of early Parkinson's disease progression, outperforming other digital metrics that were confounded by learning effects or symptomatic treatments.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Parkinson's disease is a condition that slowly changes how the body moves and how the mind works. For a person living with it, the world can feel like it is moving too fast while their own body struggles to keep up. Symptoms often start subtly: a hand might shake slightly, a step might become shorter, or a voice might grow quieter. Over time, these small shifts can grow into significant challenges. For decades, doctors have relied on asking patients questions and watching them perform simple tasks in an office to track how the disease is progressing. While this method is standard, it has a blind spot. It captures only a snapshot of a person's condition at a single moment, often missing the slow, steady changes that happen between visits. Furthermore, human observation can vary from one doctor to another, and it is difficult to notice the tiny, incremental declines that signal the disease is moving forward.
To solve this, scientists have begun turning to the technology people already carry in their pockets and on their wrists. Smartphones and smartwatches can record movement, sound, and activity continuously, turning everyday life into a stream of data. This approach offers a way to see the disease not as a series of isolated moments, but as a flowing story. The question researchers have been asking is whether these digital tools can actually detect the slow march of Parkinson's disease over several years, long enough to see if new treatments are working to slow it down.
A team of researchers recently completed a long-term study to answer this question. They followed a group of people with early-stage Parkinson's disease for three years, using a remote setup where participants used their own smartphones and smartwatches every day. The goal was to see which digital measurements could reliably track the disease as it progressed. The study involved 43 participants who were monitored over 36 months. During this time, the participants performed specific tasks on their phones, such as walking, speaking, and tapping their fingers, while their watches passively recorded their movements and tremors. The researchers looked at a wide variety of data points, from how fast someone walked to how long they paused while speaking, to see which ones showed a clear pattern of change over time.
The results revealed that the digital tools were indeed sensitive enough to catch the disease's progression, but not all measurements were equally useful. The most consistent signal came from how people walked. The researchers created a combined score based on seven different aspects of walking, including arm swing, step length, and how long it took to turn around. Over the three years, this composite walking score showed a clear worsening. Specifically, the time it took for participants to turn around increased significantly, and their walking speed and arm swing decreased slightly. This pattern was most obvious in the participants who had not yet started taking medication for their condition, but the trend was visible across the group. The study found that these digital walking measures were often better at detecting change than the traditional scores doctors use in the clinic.
Speech also emerged as a key area of change. The digital analysis of voice recordings showed that the participants' speech became measurably different over the three years. One specific change was an increase in the pauses between words while reading aloud. This digital speech score worsened over time, suggesting that changes in how people speak can be an early and reliable sign of the disease moving forward. Like the walking measures, the digital speech tools were able to detect these subtle shifts better than the self-reported questions or the doctor's ratings found in standard clinical exams.
However, not every digital measurement told a story of progression. Some measures, like the number of times a person tapped their finger or how quickly they completed a cognitive puzzle, showed an initial improvement that later stabilized. The researchers believe this was a "learning effect," where participants simply got better at the tasks because they had practiced them many times over the three years. Once this initial learning curve flattened out, these tasks stopped showing the disease's progression. Similarly, measures of daily activity, such as the total number of steps taken per day, did not show a clear decline. This was likely because many participants started taking medication during the study, which helped them stay active and masked the natural slowing of the disease.
The study also looked at tremors, the shaking that is a hallmark of Parkinson's. The smartwatches recorded the amount of time participants spent shaking. In the first year, the time spent with tremors increased, but by the end of the three years, it had returned to baseline levels for those on medication. This suggests that while digital tools can track tremors, the results are heavily influenced by whether a person is taking medication, making it harder to use tremor data alone to measure long-term disease progression in a mixed group of treated and untreated patients.
The researchers concluded that for future studies aiming to test new drugs that might slow down Parkinson's, the best digital tools are likely to be those that measure walking and speech. These two areas provided the clearest, most consistent evidence of the disease getting worse over time, even when other factors like medication or practice effects were at play. The study also highlighted the value of a remote approach, where people could participate from their own homes, maintaining high levels of engagement over a long period. This method proved that it is possible to gather high-quality, continuous data without requiring patients to travel to a clinic every few weeks. By focusing on the subtle, real-world changes in how people walk and speak, these digital tools offer a promising new way to understand the long-term journey of Parkinson's disease.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.