A quarter of prescriptions are likely inappropriate for the conditions they were prescribed for, according to a new study published Wednesday in BMJ.
And that estimate is probably conservative.
The study was conducted by looking at the diagnostic codes for just more 19 million privately insured individuals to determine whether conditions reported within the three days prior to an antibiotic prescription justified their use.
Researchers found 23.2% to be inappropriate, 35.5% potentially appropriate and only 12.8% appropriate. The remaining 28.5% had no related diagnosis code on which to base any judgment.
As high as those numbers appear, they’re likely just the tip of the iceberg, according to the study’s lead author, Kao-Ping Chua, M.D., Ph.D., of the University of Michigan Medical School.
They considered prescriptions appropriate when they were associated with conditions that almost always respond well to antibiotics. Potentially appropriate diagnoses covered situations in which the antibiotics could be appropriate, though the researchers had no way to know whether the prescribers actually followed appropriate guidelines.
Prescriptions only received the inappropriate label if they were associated with diagnoses such as acute upper respiratory tract infections or acute bronchitis, which almost never justify the use of antibiotics.
“We were trying to err on the side of being lenient, saying OK, we’re going to assume appropriate use, and even with that leniency, it’s only 13% where you say, this is definitely OK,” Chua said.
As a consequence of that leniency, the real number of inappropriate prescriptions could be higher. The codes that lacked a diagnosis, for example, could cover a range of circumstances from patient visits paid with cash at an urgent care center or retail clinic, to people who receive prescriptions via telehealth services, and therefore never actually visit their provider to receive a diagnosis. Even the potentially appropriate prescriptions could be worth further investigation, Chua said. He points out that the study’s findings that fewer children received unnecessary antibiotics may not be as substantial as it appears.
“If you look really carefully, the percentage of prescriptions written to children that were in that potentially appropriate category was way higher than adults, so what could actually be happening is there could be a bunch of overprescribing in that category for children,” he said.
Given these numbers, it appears that—despite recent efforts at improving antibiotic stewardship—little headway is being made on the problem.
Chua sees fixing the problem as a long, slow, probably multigenerational effort focused on incremental education of patients.
“I think every acute respiratory illness is an opportunity for that type of education, to really frame what are the harms and benefits of antibiotics—and actually on the benefit side I think we overexaggerate how beneficial antibiotics are in many cases,” he said.