Should AI Be Used to Prompt People in Pain to Exercise?
/By Crystal Lindell
We all know that being too sedentary contributes to poor health. It might even make your pain worse. But many of us still spend way too much time sitting in one place watching TV, scrolling on our phones, or working at a computer.
“Individuals with certain chronic pain disorders face a paradox: they are at increased risk for sedentary behavior due to pain-related interference, even though significant gains in symptom relief and health benefits from increasing their PA (physical activity) for symptom management have been demonstrated,” says Ipek Ensari, PhD, who studies artificial intelligence (AI) and human health.
Ensari and her colleagues at the Icahn School of Medicine at Mount Sinai wondered if wearable devices could be used to help women with chronic pelvic pain stand up and move around more.
Their study, recently published in the journal Women’s Health, wanted to see if AI could predict when patients would have prolonged periods of inactivity.
They recruited 134 women with chronic pelvic pain, primarily from endometriosis, along with 61 healthy women for a control group. Participants wore Fitbit devices for up to 90 days, generating minute-by-minute information about their physical activity, heart rate, steps, and sleep patterns.
"Our goal was to determine whether everyday wearable devices could serve as an early-warning system for prolonged sitting in women living with chronic pelvic pain," Ensari explained in a news release.
"Rather than offering generic advice after the fact, we wanted to determine whether we could anticipate these moments and support people with simple, well-timed prompts that fit naturally into their daily lives."
The research team gathered data from each participant and developed personalized forecasting models to predict their activity levels one hour ahead of time. The goal was to identify 15-minute periods when a participant could stand up and move around, which the researchers called an “exercise snack.”
Those short activity breaks may be more realistic and achievable than longer exercise periods, especially for people with chronic pain. The work also challenged the assumption that AI focused on health needs to be complex. Relatively simple AI models forecast prolonged periods of sitting just as well as complex ones.
"We were surprised by how well the simplest models performed," said lead author Jannes Jegminat, PhD, a former postdoctoral research fellow at the Icahn School of Medicine. "More complex AI is not always better. Lightweight, interpretable models can accurately forecast sedentary behavior while being practical enough to run directly on a person's own device, which also helps protect privacy."
Unfortunately, the research did not include any further investigation into whether or not participants would actually get up for a walk when prompted. And there seems to have been no consultation with pain patients on whether this technology is something they’d want and use.
The researchers admit that further study is needed to see if the AI prompts could actually help patients.
"This study suggests that predicting prolonged sitting is feasible, even if the individual has chronic conditions that might impact their daily routine," said Ensari. "The next step is determining whether delivering personalized movement prompts based on those predictions actually helps reduce sedentary time, improves symptoms, and enhances quality of life. Those questions will require prospective clinical trials."
It’s unclear how the technology would be applied to periods when someone needs to be sedentary, such as when they are working, eating, watching a favorite show, or in too much pain to move.
The last thing I want to see on my AI watch as I’m driving late to work is a notification telling me to go for a walk instead.
It’s becoming clear that anything and everything related to artificial intelligence is the new trendy topic for research, and it’s likely the topic is now also yielding better funding opportunities. But as researchers continue to study the ways AI can help us medically, they would do well to remember to ask patients for their input as well.
