What is the Role of Data in Telerehabilitation?

This post sumarizes the main findings of the paper: “Towards Supporting Data-Driven Practices in Stroke Telerehabilitation Technology” (pdf available here). This work is co-authored by Mayara Costa Figueiredo, Lucy Dodakian, Cleidson R. B. de Souza, Steven C. Cramer, and Yunan Chen. Published in April 2021 in the PACMHCI Journal, Volume 5, Issue CSCW1. The paper will be presented at CSCW 2021.

If one of your close friends or family members had a stroke, would they have access to enough rehabilitation services to recover? For most people, the answer would be no. Health insurance often only covers a few weeks of physical therapy and out-of-pocket expenses can very quickly become overwhelming. Lack of access to healthcare can prevent stroke survivors from regaining the ability to work, drive, and even cook by themselves.

Using technology for rehabilitation could make it more accessible for many people, improving their recovery and quality of life. This is not a novel idea — systems that support exercises through games, such as the Wii and Kinect, have been widely available for over a decade. There is evidence that they provide substantial benefits to stroke survivors. What is missing for them to become viable treatment options?

If you have had a telehealth appointment before, you probably know that it carries some of the same trade-offs as remote work: it can be more convenient, time-efficient, and cheaper, but the interaction is just not the same as it would be in-person. The same is true for rehabilitation through telehealth or telerehabilitation. In this post, I discuss what we learned about how data collected by a telerehab system can be used to overcome this challenge.

About the study

This project investigated how stroke survivors and healthcare providers use and perceive the data collected by a game-based telerehabilitation system. This system was created by a team of clinical researchers to be used by stroke survivors in their homes. It includes a set of games designed to use the same movements as traditional physical therapy exercises.

In this game-based system, data such as a log of game scores can indirectly measure progress over time. For example, getting higher scores on a game after using the system for three weeks could mean that there was an improvement in motor function.

We ran a focus group with two physical therapists and two occupational therapists. We also interviewed 10 stroke survivors who used the telerehab system for 12 weeks for a clinical study. Participants discussed their experience with the telerehab data while using the system. They also reflected on different visualizations of their own data during the focus group and interviews, such as the line graph below.

Findings

We found that the data could be used to address many of the participants’ information needs. For that reason, telerehab systems should serve as tracking systems. Both stroke survivors and healthcare providers were interested in information about progress, but there were several differences in their perspectives towards the data.

Stroke survivors cared deeply about knowing that they are still making progress and approached the data with this question in mind. They focused on a specific rehabilitation goal they had not reached yet, such as driving, and wanted to know that they were making progress towards that goal. They hoped that the data would be encouraging and preferred not to see data that they would consider negative, such as decline or plateau.

On the other hand, healthcare providers needed information that could help them to update the patients’ treatment plans. For example, determining the right difficulty level and addressing any issues that the patient might be experiencing, such as pain, improper exercise form, or low adherence.

Their information needs are directly linked with the limitations in communication involved in remote interaction: in regular face-to-face interaction, those questions would be answered by the other party.

These results contribute to the future design of effective telerehab systems that will meet users’ information needs. For both patients and healthcare providers, providing meaningful information about rehabilitation progress over time will require understanding how to measure it reliably based on game performance.

In addition, these findings also reflect a potential issue for designing self-management systems that might combine both roles. For example, a telerehab system that does not involve healthcare professionals could require patients themselves to make decisions on data that are personally meaningful to them. While information needs for each role are not inherently contradictory, they might be difficult to integrate when the data are interpreted negatively.

For more details, see the paper (link). I am also happy to talk more about this project, you can reach me at cmarque@iu.edu.