2023-08-06 13:22:08
Researchers at McGill University have designed a model to predict the development of chronic pain in patients who suffer from it. A tool that could be used to manage waiting lists in pain management clinics, according to a co-author of the study.
In the article “A prognostic risk score for development and spread of chronic pain”, published in the journal Natude Medecine, at the beginning of July, the researchers looked at chronic pain which overlaps, i.e. say when a person reports having chronic pain in several places.
To do this, the study authors used data from the UK Biobank, which collects health data from 500,000 people.
“The advantage of that is that for the first time, we have access to half a million people who have filled out different questionnaires, who have done batteries of tests, and who have reported the different types of pain that affected them,” explains Dr. Etienne Vachon-Presseau, assistant professor in the Faculty of Dentistry and Oral Health Sciences at McGill University, and co-author of the study.
The data was collected from UK citizens, aged 50 to 70, whom the biobank follows as they age.
“In this cohort, what we observe is that there are many patients who have chronic pain, about 40% of the people who took part in the study. Then, among these, there is a significant proportion, once again, around 40%, who reported pain in several places on their body, ”explains Mr. Vachon-Presseau.
Thanks to this extensive data, “we developed a predictive model, using machine learning, on a range of environmental, psychological, personality factors, sleep disorders, alcohol consumption, smoking, anthropometric measurement such as body mass index, for example,” he explains.
The researchers entered about 100 variables into the model to find out what combinations of factors can be used to predict the number of sites on the body where a person will suffer from chronic pain.
“We were able to train models where we were able to predict over time, for example, if a patient has chronic knee pain, is this patient at risk over time of seeing his pain develop later in life? other sites. For example, we start at the knee, but we end up with four more sites, nine years later, back pain, hip pain, pain in the neck, shoulders”, illustrates Etienne Vachon-Presseau.
Although the model is most effective when analyzing the development of chronic pain in a person who already suffers from it, it can predict, under certain conditions, whether a person who is at risk of having chronic pain, but does not not suffer from it for the moment, will feel it later. However, in this case, “the performance was a little lower,” said Mr. Vachon-Presseau.
A way to manage waiting lists?
Asking 100 questions to predict the course of a patient’s chronic pain is unrealistic, the study authors concluded. The researchers therefore identified the six main factors of their model, in order to allow clinicians to use it on a day-to-day basis.
“So we sacrificed the performance of the model a little to try to simplify it as much as possible,” says Mr. Vachon-Presseau.
These six factors are sleep, neuroticism (do you often feel overwhelmed?), fatigue, recent consultation with a doctor or psychiatrist for mental health issues, life stressors ( death, divorce, financial difficulties, etc.) as well as the body mass index (BMI).
“The model, we would like to implement it in pain clinics, to try to see how it really performs with patients who are in tertiary care units,” says the professor.
“It could be useful, potentially (…) for example, in the evaluation, the “screening” of patients, to know if we should prioritize them because they are more at risk, or in the opposite case, if the maybe no one can wait a little longer on the waiting list,” he added, pointing out that the waiting time can stretch up to a year for some patients.
If the model “performs well,” it could also prompt clinicians to prescribe more aggressive treatments if a person is particularly at risk of their pain spreading.
This model could also be used in research, underlines Mr. Vachon-Presseau.
1691330845
#Predicting #development #chronic #pain #model #developed #McGill