Assessing Pain In Dementia Patients
Dr. Thomas Hadjistavropoulos, Research Chair in Aging and Health at the University of Regina (U of R) believes that the most significant advancements in dementia care over the next decade are more likely to come from technology and engineering than from the traditional health sciences. This has led him in a unique direction, working with biomedical engineers and computer scientists from the University of Toronto, including Dr. Babak Taati, to develop an automated system to detect pain behaviours and assess pain in patients with dementia. Vivian Tran, a doctoral student in clinical psychology at the U of R, and a Saskatchewan Centre for Patient-Oriented Research trainee supervised by Hadjistavropoulos, has brought this new technology out of the lab and into the real world setting of long-term care facilities, where she worked directly with dementia patients, family members and staff.
Determining and assessing pain in dementia patients can be challenging; it can’t be measured like blood pressure, for example. With patients unable to verbally communicate their pain, serious issues can go undetected. Dementia patients may display responsive behaviours to pain that can be misattributed to psychiatric issues and lead to incorrect diagnosis and treatment. While there are methods of assessing pain through the systematic observation of nonverbal behaviours, oftentimes, staff in long-term care facilities don’t have adequate time or resources to do so frequently. Hadjistavropoulos’s and Tran’s research could change that.
In the study, Tran evaluated the readiness of each facility to incorporate the necessary technology, and trained nurses in how to use and respond to the system. A total of 30 participants from four long-term care facilities were included in the study, with residents with dementia and conditions that result in pain being observed.
When the new automated system is fully developed, it will monitor patients in their daily routines and notify nurses when pain is detected. As Hadjistavropoulos states, “that creates a tantalizing possibility of having patients assessed for pain 24 hours a day – something that no human assessor can do. We have demonstrated that when pain is assessed frequently, quality of life improves, pain levels reduce, the quality of interventions directed towards the patients improves and perhaps unexpectedly, nursing staff members feel less stressed in their jobs … nurse stress, over time, reduces when they assess pain frequently because it has tremendous implications for the quality of interactions between staff and patients.”
Caregiver/Patient Partner Mary Brachaniec has helped optimize the team’s impact on knowledge mobilization. She worked on two major team projects to share information on pain assessment in dementia care with key stakeholder groups, including health professionals, policy makers, patients and caregivers. Brachaniec also serves as a caregiver representative on the Health Standards Organization Technical Committee that published the 2023 Canadian Standard for Long-Term Care Services. She was able to share information on pain assessment in dementia with the technical committee. As Hadjistavropoulos indicates, Brachaniec’s connections with various patient, caregiver and dementia groups have contributed to the dissemination of information well beyond the team’s expectations.
Currently, the team is experiencing an 85% correct pain detection rate in lab, although the system is less accurate in clinical settings. They are working on increasing accuracy by utilizing artificial intelligence and other methods.
Says Tran, “Once this technology has been fully implemented, I think it will really help the nurses and other care staff in being able to detect pain earlier, so it is more manageable for both the staff and residents. It’s about improving the overall quality of life for residents in long term care and ensuring all residents have equal opportunity to care, regardless of their cognitive abilities.”
To learn more about this project, watch the video below or click here.