New study compares Kaia Health musculoskeletal digital therapy with live physical therapists

Results show Kaia Health computer vision technology as accurate as physical therapists in suggesting exercise corrections

  • Peer-reviewed, prospective cohort study published in the Journal of Medical Internet Research (JMIR) is first to compare smartphone-based musculoskeletal (MSK) digital therapy with physical therapists (PTs)
  • Findings demonstrate the ability of Kaia Health computer vision technology—formerly known as Motion Coach—to provide consistency and equity of care across users without sensors or telehealth physical therapy (telePT)
  • Employers, health plans, and their members can benefit from computer vision technology with lower costs, higher engagement, and improved outcomes
  • Results validate a digital-first approach to MSK pain as a convenient, scalable, less costly approach that complements in-person or remote physical therapy appointments

The benefits of digital-first care, including digital health apps and remote telehealth interactions, are increasingly apparent for employers and providers addressing chronic MSK conditions in their populations. One of the biggest remaining questions has been about the quality of digital physical therapy compared with a live PT: How effectively can virtual care complement in-person or telePT?

Researchers are now helping to answer that question with a new, peer-reviewed study published in the highly respected JMIR. In the study App-based feedback for rehabilitation exercise correction in patients with knee or hip osteoarthritis: prospective cohort study, patients with chronic osteoarthritis of the knee or hip performed a series of exercises while being simultaneously evaluated by Kaia Health computer vision technology and live PTs.[1]

Results found that corrections triggered by Kaia Health computer vision technology were consistent with in-person, real-time feedback from PTs participating in the study. This groundbreaking research—the first ever to compare smartphone-based MSK digital therapy with live PTs—has important implications for health plans, employers, and benefits managers seeking to lower MSK care costs while improving outcomes.

A demo of Kaia Health computer vision technology

The power of computer vision technology

Computer vision technology is a core component of the Kaia Health app, providing automated feedback by tracking exercise form through a user’s smartphone or tablet camera. The technology is already included in the Kaia Health product and complements its US network of PTs. JMIR study results demonstrate the benefits of this accessible, accurate, and digital-first approach to MSK.

For most patients, computer vision technology can be just as accurate and effective as an in-person approach, with added benefits specific to the Kaia Health solution of no sensors, no additional hardware or telePT video setup, and 24/7 availability. This results in a digital-first approach that is more convenient and scalable as well as less costly than in-person or telePT.

“A common misconception is that physical exercise must be accompanied, whether in person or virtual, by a human physical therapist. This is simply not the case,” said Kaia Health US Chief Medical Officer Justin Yang, MD, MPH. “But we’re also not a replacement for PT when in-person care is necessary. The Kaia Health app and its computer vision capability facilitates provider integration, complements brick-and-mortar care pathways, and augments the traditional MSK care experience. We’re seeing that technology can have a profound impact on a person dealing with pain and reduce the number of times the patient relies first on in-person or telePT. This has great potential to save time and money while enhancing access to care, equality of care, and consistency of care. These are distinct benefits that can lead to better adherence and improved outcomes.”

Dr. Yang noted that the results of the study are not that surprising. This is because the Kaia Health app has been developed by medical and computer vision experts and rigorously evaluated in studies published by peer-reviewed journals.

By continually improving upon its computer vision capabilities, Kaia Health promotes equality of care through consistent exercise feedback for all patient subgroups. This level of care is consistent across gender, age, body mass index (BMI), and location of pain and applies to all investigated exercises based on difficulty, pose, and complexity. The study further demonstrated that Kaia Health’s accessible, PT-grade digital therapy can be as effective as human intervention every time exercises are being done.

In previous peer-reviewed research, the Kaia Health app 12-week program has been demonstrated to alleviate pain to the same degree as 6 sessions of traditional in-person physical therapy combined with online education. Furthermore, the Kaia Health app has demonstrated the ability to significantly outperform standard-of-care treatment for back pain across all medical outcomes, including pain, anxiety, depression, stress, body function, and well-being.

Computer vision technology provides direct feedback on exercise form and suggests corrections.

“This is the first study that clearly demonstrates the potential of a technology that can suggest corrections in the setting of therapeutic exercises for MSK pain care. Our computer vision technology is now the only technology in the MSK market to provide exercise correction and be validated for its accuracy,” said Kaia Health President, CEO, and Founder Konstantin Mehl. “This builds trust with users, employers, and health plans by integrating with our new Gateway care pathway, which enables our team to review and adjust the member clinical pathway for optimal results. The Kaia Health app is available through off-the-shelf smartphones or tablets, without the need for additional hardware, making it widely available to a broad patient population.”

"These findings validate the ability of [computer vision] technology to detect form during exercise and provide audiovisual feedback to users with preexisting musculoskeletal pain conditions."

—JMIR computer vision vs PT study

In the JMIR study, Kaia Health computer vision technology was able to provide consistent feedback to users across different exercises, body parts, and difficulty levels, as well as across genders, age, BMI, and pain areas. In telePT offerings, the PT is not actually present during exercise. And sensor solutions only track and show 1 angle without providing feedback and suggestions to improve form.

Key JMIR study highlights[1]:

  • 24 participants were enrolled and performed the assessments
  • The interrater agreements showed an accuracy of 0.828 (Kaia Health app vs PTs) vs 0.833 (PT1 vs PT2), P <0.001 and confirmed noninferiority of computer vision technology to PTs
  • There was no detectable difference in levels of interrater agreement between computer vision technology vs live PTs and PTs themselves in any of the 6 exercises, regardless of exercise pose or exercise difficulty levels
  • The interrater agreement for suggesting corrections during therapeutic exercises observed in the current study did not differ between 2 PTs among each other and PTs and the computer vision technology
  • These results confirm the ability of the computer vision technology to detect form during exercise and provide valid feedback to users with preexisting musculoskeletal disorders. This finding was valid for all investigated exercises and subgroups.

As the world’s most popular MSK platform with 60 million lives covered worldwide, Kaia Health continues to advance next-generation MSK digital therapy and build clinical validation of our computer vision technology. Kaia Health computer vision technology is now the only digital therapeutic with the demonstrated ability to detect and correct exercise form similar to a PT.

Schedule a meeting

To learn more about computer vision technology and how our app’s real-time feedback on MSK exercises can boost the effectiveness of chronic pain therapy for your employer or health plan population, schedule a brief intro call with the Kaia Health team.

Reference

1. Biebl JT, Marzena R, Strobel M, et al. Comparison of app-based correction feedback to exercise correction by physiotherapists in patients with osteoarthritis of the knee and/or hip: A prospective cohort study. J Med Internet Res 2021;23(7):e26658. doi: 10.2196/26658

Further Reading