Researchers at the University of Alberta are developing a machine learning tool to assess the risk of prescription opioids in order to address the ongoing opioid crisis in Canada. Specifically, doctors could use this AI tool to predict the impacts of a prescription opioid on patients and prevent them from unnecessary emergency department visits or even death within 30 days of starting the medication.
In his interview with Global News, Dr. Dean Eurich, program director for the clinical epidemiology program at the University of Alberta, said the system would allow an additional “level of comfort to clinicians, (knowing) there are also other supports they can use to help (in) making sure the patient is getting the right drug at the right time.”
Dr. Eurich’s team has carried out and published a study which analyzed medical data of more than 850,000 Alberta residents anonymously and predicted the best outcomes for the patients using a data set provided by Alberta Health.
The AI system used various health factors including the history of injury, obesity, depression, diabetes, fluid disorder, and psychosis, to determine risks to a patient. Moreover, these were combined with diagnoses from doctors, healthcare visits, and information regarding where the patient lives.
“The idea is not to make physicians stop prescribing opioids, (but) to minimize the risk after the opioid exposure,” said Dr. Fizza Gilani of the College of Physicians and Surgeons of Alberta.
Moreover, Dr. Eurich noted that machine learning builds systematic models including various key factors to find combinations predicting the best outcomes for a patient, adding that the AI model could “predict correctly (for) four out of every five patients.” Accordingly, a patient who was identified to have a higher risk would have a higher chance of being hospitalized within the first 30 days of prescribing opioids, according to the model.
The AI system will soon be tested with real-time data, and it will be assessed whether the system could limit long-term use of high-dose opioids among patients.
However, some advocacy groups, including Moms Stop the Harm, have expressed concern regarding the AI model’s effectiveness in reducing opioid-related deaths in the province. According to the group’s co-founder, Petra Schulz, most of the opioid-related deaths in Alberta result from the use of street drugs and not prescription opioids. “This kind of AI could make the safer alternatives even less available,” she said. “It’s like you’re doing detective work and wanting to figure out what is not going right for the patient instead of developing a trusting doctor-patient relationship, which allows the patient to (speak) openly.”
In turn, Dr. Gilani said that there is an “indirect linkage” between factors entered into the AI model and that the tool could help in reducing those deaths based on the data, adding that a “good portion” of poor outcomes related to opioid use is not driven by street drugs, but by prescription drug use.
Dr. Eurich added that patients continue to get exposed to opioids for pain medication and eventually start using the health system to “doctor shop (and) obtain massive quantities of opioids… also end(ing) up being cut with other substances.” According to Dr. Eurich, AI could provide a “good continuity of care” even when patients change practitioners, decreasing their risk of harm from prescription opioid drugs.