Automating the Addiction Behaviors Checklist to identify early problem opioid use in pain patients

Claiming that diagnosis codes are not used enough and sometimes just wrong, researchers are proposing that the Addiction Behaviors Checklist can be automated to better identify problematic opioid use. The researchers used a retrospective cohort and evaluated the data used by electronic health records with an automated clinical tool could be used. Data was deidentified. The cohort of 8,063 chronic pain patients was validated with 100 individuals with chronic pain. The researchers found that compared with manual record review and diagnosis, the automated approached was better. Using automated data may make it easier to identify problematic opioid use earlier. The Addiction Behaviors Checklist has never before been automated, according to the researchers. The study is by Angus H. Chatham and colleagues. While clinical text and diagnosis billing codes can help, so can automation, the researchers wrote in “Automating the Addiction Behaviors Checklist for Problematic Opioid Use Identification,” published online April 9 in JAMA Psychiatry . The researchers conceded that using Electronic Health Record data during manual review “might not be as robust as clinical interviews,” they concluded that key word search is simple. However, they noted that their automated Addiction Behaviors Checklist was originally developed to identify opioid use disorder risk among chronic pain patients, which may not extend to others with the condition. Still, they said their automated approach could be expanded.