Fighting the opioid epidemic
David Sontag (MIT), Dennis Wei (IBM Research), and Kush Varshney (IBM Research)
Opioid drugs are often prescribed for pain management in the U.S.: almost 58 opioid prescriptions were written for every 100 Americans in 2017. Since the 1990s, opioid misuse leading to addiction, overdose, and death has been a growing problem throughout the U.S. In 2016, prescription opioid drugs contributed to 40 percent of all U.S. opioid overdose deaths, and in 2018, more than 115 people died each day from overdoses involving prescription opioids. The duration, dosage, and type of opioid prescribed may all influence a person’s risk of misuse, and public health efforts have focused on developing data, tools, and guidance to improve opioid prescribing practices and patient safety. MIT-IBM Watson AI Lab scientists are bringing the power of AI to tackle this challenge. They are applying machine learning techniques to connect the dots among diagnoses, prescription quantities/durations, medical histories, and behaviors associated with addiction as recorded in healthcare insurance claims. Their goal is to develop a precision medicine model that can infer the causal effect of opioid usage on individuals.