Part of a growing body of evidence putting EHRs at the center of opioid use reduction efforts, a new study has concluded that small changes to EHR presets can lower the number of opioid pills physicians prescribe. This is particularly interesting when thinking back to the Practice Fusion opioid kickbacks we wrote about last week.
After concluding their analysis, researchers at UC San Francisco found that when EHR presets suggesting the prescription of a standard number of opioid pills were lowered, physicians prescribed fewer pills. The study was published this month in JAMA Internal Medicine.
Previous research has already examined the extent to which leveraging existing EHR data could predict post-hospital use. For example, a 2018 study published in the Journal of General Internal Medicine described a model predicting the likelihood of future chronic opioid use based on hospital data.
To conduct the study, researchers attempted to project which patients would progress to chronic opioid therapy after discharge. Researchers developed a statistical model designed to predict which hospitalized patients who had not previously been on COT would end up on COT after discharge.
After taking into account a wide array of variables in their analysis, including medical and mental health diagnoses, substance and tobacco use and chronic or acute pain, researchers found that they could predict COT in 79% of patients.
In this case, meanwhile, the researchers were interested in learning changes in default opioid settings could influence the amount of opioids physicians prescribed. To examine the issue, the investigators looked at opioid prescribing at two emergency departments, UCSF Medical Center and Highland Hospital, gathering data between November 2016 and July 2017. During that period, the research team randomly changed EHR default settings for commonly-prescribed opioids like oxycodone, Percocet and Norco, for four weeks at a time.
At the start of the study, the Highland EHR had a default setting of 12 pills and UCSF’s system a 20-pill default. To see what would happen if those presets were changed, the researchers substituted preset quantities of 5, 10 and 15 pills, along with a blank setting allowing physicians to whatever level they thought best. All told, researchers analyzed 4,320 opioid prescriptions.
During the study period, the researchers found that lower default pill presets were associated with lower volumes of opioids being prescribed, as well as a lower proportion of scripts exceeding prescribing recommendations issued by the CDC.
The conclusions drawn by these researchers feel fairly intuitive. If you change things up and give physicians a lower recommendation for opioid dosing, it’s almost inevitable that they’ll issue smaller opioid orders. It also stands to reason that when the EHR makes no recommendation as to what a standard opioid script should be, it forces physicians to be more creative and to grapple a bit more with what their patients actually need.