Visualizing patterns of health care use as a screen for opioid medication misuse

Background:

Misuse of and addiction to prescription medication is a growing problem. Identifying patients who misuse can be difficult and alternative strategies, such as decision support systems and visualization techniques, may help clinicians better detect misuse.

Methods:

Fifty veterans who received a prescription for an opioid medication from the VA Palo Alto Health Care System in the last year were interviewed. Based on interviews, patients were categorized as having acute or chronic pain, managed or unmanaged pain, and appropriate use or misuse of opioids. Four combinations of these categories were represented in the sample: acute, managed pain with appropriate use; chronic, managed pain with appropriate use; chronic, unmanaged pain with appropriate use; and chronic unmanaged pain with misuse. Patterns of health care utilization from one patient in each group were determined from medical record review and displayed graphically as egonets.

Results

Health care utilization patterns of the opioid misusers were visually distinguishable. Patients with acute or chronic managed pain and patients with chronic, unmanaged pain with appropriate opioid use tended to obtain opioid medications from a single provider. When they showed high utilization of health care, most care was by referral from their primary provider. In contrast, patients with medication misuse showed numerous visits to the emergency department, prescriptions of potentially addictive medications from multiple providers, and a greater number of mental health and substance use disorder treatment visits.

Conclusions:

Visualization of health care utilization patterns using egonets may allow rapid screening of patients for medication misuse. Automated display of health care utilization using egonets may be a useful addition to a decision support system in the electronic medical record and improve public health.