Opioid use disorder can lead to clinically significant impairment or distress. Researchers have now used machine learning to develop a prediction model for the early diagnosis of opioid use disorder.
Opioid use disorder
Opioid use disorder is the chronic use of opioids. It causes significant clinical distress or impairment and affects an estimated 16 million people worldwide. In the United States, the diagnosis is based on the American Psychiatric Association DSM‐5 and involves a desire to obtain and take opioids regardless of the consequences. Opioid use disorder causes approximately 12,000 annual deaths worldwide and is more prevalent among men between 40 and 50 years.
Opioid use disorder is defined as opioid consumption at repeated occurrence within 12 months. The signs and symptoms include drug‐seeking behaviour, legal or social consequences, subsequent adverse health outcomes and multiple prescriptions from different clinicians.
Unfortunately, only a subset of those with opioid use disorder have had their disorder recognised by a medical professional. As it can have a significant impact on health and well-being, it is essential to diagnose the condition as early as possible and begin treatment.
Prediction model
In this study, published in Pharmacology Research & Perspectives, researchers tested the utility of applying machine learning to big data to create a prediction model and algorithm for early diagnosis of opioid use disorder. They also wanted to determine the typical delay in diagnosis. The main aim was to identify patients at high risk early in order to be able to offer them early prevention and interventions.
The model was generated from information obtained from commercial claim databases from 2006-2018. This specifically consisted of 10 million medical insurance claims from 550,000 patient records. The team compiled 436 predictors candidates, which were divided into six feature groups. These groups included demographics, chronic conditions, diagnosis and procedures features, medication features, medical costs and episode counts.
The new algorithm offered a mean 14.4 month reduction in time to diagnosis of opioid use disorder. This algorithm has the potential to save in further morbidity, medical costs, addictions and mortality.
Gideon Koren, senior author, stated:
“Opioid use disorder has led a very serious epidemic in the U.S. and many other countries, with devastating rates of morbidity and mortality due to missed and delayed diagnoses. The novel ability of our algorithm to identify affected individuals earlier will likely save lives and health care costs.”
Image credit: By jcomp – www.freepik.com
