Find care now
If you are experiencing a medical emergency, please call 911 or seek care at an emergency room.
New MedStar Health research demonstrates that patient-reported data combined with electronic medical record data can more accurately predict the likelihood of opioid overuse in patients who have undergone hand surgery.
The opioid crisis continues to upend the lives of U.S. patients. Doctors are faced with an ethical imperative to exercise great care in prescribing opioid medicines for pain relief. MedStar Health Research Institute and Curtis National Hand Center are collaborating to break new ground in predicting which patients are most at risk for developing long-term opioid use after surgery in hopes of learning more about how to further prevent misuse and opioid dependence.
Opioids are a class of drugs that includes legal prescription pain relievers such as oxycodone, hydrocodone, codeine, and morphine, as well as illegal drugs like heroin. These drugs interact with receptors in the brain to block pain and are generally safe when taken for a short time under a doctor’s supervision. However, because they produce pain relief and euphoria, opioids can be misused and lead to addiction, overdose, and death.
According to the Center for Disease Control and Prevention’s National Center for Health Statistics, deaths from a drug overdose in the U.S. continue to rise. In 2021, 80,816 people died of an opioid overdose, an increase of nearly 15% from 2020.
Surgery, in general and particularly hand and wrist surgery, involves creating a moderate amount of short-term discomfort for our patients. Although many patients never use any opioids after surgery, and of those that do most only use for a few days, studies have shown that as many as 13 percent of patients who had elective hand surgery were still taking opioids 90 days after surgery. Safely and respectfully identifying patients at risk of addiction before surgery so we can intervene is critical to helping reduce opioid overuse.
Understanding patients means better risk prediction.
Scientists have been working to predict the risk of opioid overuse for some time. Tools like electronic medical records (EMR) provide a wealth of essential data on patient health, and some studies have successfully used EMR data to predict long-term opioid use.
At Curtis National Hand Center, we collect a large volume of patient-reported data through questionnaires, before and after surgery. A patient’s experience with hand surgery is in many ways more important to their surgical outcome than any X-ray or test we can perform. We want to know about our patients’ lives. What stresses them out? How do they feel on a regular day? How is their mental health? What makes their recovery more challenging?
Our research has shown that the answers to questions like these even before surgery strongly predict future pain management challenges. The MedStar Health Research Institute Center of Biostatistics, Informatics, and Data Science (CBIDS) team performs sophisticated statistical analyses that combined aspects of pre-surgical EMR and unique patient-reported data. Our models accurately predicted the risk of long-term opioid use after surgery for many patients. Our analyses indicated that patients who were underweight and had undergone trauma-related surgery had a higher risk of remaining on opioids three months after surgery than other patients. But, most importantly, questionnaire scores were some of the most strongly predictive of this prolonged opioid use – showing that these unique data provide novel and valuable insights into our patients.
Every patient we can keep from a life of addiction and drug overuse is a life saved, so our next step is to develop a treatment protocol for patients at risk.
Integrating data-driven patient care in the clinic.
Working closely with MedStar National Rehabilitation Hospital, we have collaborated with pain psychiatrist Dr. Natasha Durant and a referral network of providers to help patients learn cognitive behavioral therapy techniques that reduce their risk of developing long-term opioid dependency.
We have applied for federal funding from the National Institutes of Health to explore integrating this predictive model into patient care. For patients who have an identified risk for long-term opioid use, we could refer them for immediate pain-supportive therapy before or immediately following surgery. These therapy techniques assist patients in developing a practical approach to coping with emotional distress and the problems of living with pain.
As we refine our questionnaires, we expect to be able to deploy this prediction method for other surgical procedures that require pain management.
MedStar Health Research Institute and Curtis National Hand Center are uniquely positioned to undertake this research because of the quantity and diversity of our patients, and quality of our data collection. Collaboration throughout our network of academic medical organizations enables us to lead the way in integrating leading-edge analytics to improve care for all patients.
