Stephany Stamatis, MFA Candidate, Herron School of Art + Design
Jay Patel, PhD Candidate, IU School of Informatics and Computing
Jo May, DNP, IU School of Nursing
Amy Little, DNP, IU School of Nursing
Patricia Kline, DNP, IU School of Nursing
Shelly Burns, DNP, IU School of Nursing
Indiana University's School of Nursing, School of Informatics and Computing, Center for Interprofessional Health Education + Practice, Health Information + Translational Services
Our design for the Primary Opioid Intervention Tool won the First Place award for the Interprofessional Education Healthcare Revolution Challenge
The design team started development of a prediction model, named Primary Opioid Intervention Tool (POIT). The POIT would be fully integrated into the existing Electronic Health Record (EHR) system of the health care facility. Using Natural Language Processing (NLP), the tool would search a patient’s EHR for any risk factors that could predict a future risk of addiction. These risk factors were researched through an extensive literature review. POIT would flag any of the risk factors present in the patient's health record. Each of the risk factors would be weighted according to their level of impact and correlation with addiction. The tool would then calculate a risk level for the patient based on the weighted score of any flagged risk factors. Based on the patient's risk level for addiction, POIT would recommend appropriate treatment methods, which may include alternative treatments to prescription opioids. The physician and patient could then collaboratively determine the safest and optimal course of treatment.
For the Interprofessional Education course, students were grouped into competitive teams and challenged to design innovative solutions to the prescription opioid epidemic in Indiana.
1. Our team first developed a System Overview Map of the stakeholders and factors involved in the opioid epidemic. In creating this map, we saw that one of the earliest contributing factors to opioid addiction is the point at which opioids are first prescribed to a patient. (Pictured towards the bottom of the page)
2. We then did a literature review. Through our research, what stood out most was that patients are very often prescribed opioids even if it is not the most appropriate course of treatment. What also stood out was the evidence based research on various risk factors in a patient's health history that correlate with a future risk of addiction.
3. Based on analyzing the data from our literature review, the opportunity statement we came up with was, "How might we help clinicians accurately and efficiently predict a patient's risk of becoming addicted to opioids, so they can provide optimal treatment?" After ideating on many possible solutions to this opportunity statement, we chose to go with a Prediction Modeling tool as our design direction. We felt physicians could use this tool to prevent unnecessarily prescribing opioids in the first place to patients who might be at higher risk for addiction. A Concept Map was created that explores how a Prediction Model might be used in a health care facility. (Pictured towards the bottom of the page)
4. Lastly, we created several wireframe prototypes to illustrate how the Prediction Model would work. We called it the Primary Opioid Intervention Tool, as it is intended to be a primary (first stage) intervention measure that physicians could use. We also designed the prototype to be fully integrated into the health care facility's existing EHR system, such as Epic or Cerner software. This way physicians and nurses would not need to be trained how to use another program, allowing for more efficiency in a fast paced clinical environment. (Pictured to the right)