Summer 2024 Issue

CDI: Understanding Antibiotic Prescribing and Improving Antimicrobial Stewardship
By Selena Chavis
For The Record
Vol. 36 No. 3 P. 26

Two studies shed light on how better documentation practices could combat misuse of antibiotics.

Concerns over the growth of antibiotic-resistant organisms have prompted greater scrutiny of prescribing practices in recent years on the national and international stage. According to the CDC, more than 2.8 million antimicrobial-resistant infections occur each year, contributing to the rise of resistant infections that put people at risk.1

When antibiotics are overused, they become less effective for treating infections, leading to growth in superbugs that make infections hard or impossible to treat. Consequently, the national push for hospitals and health systems to prioritize strategies that promote antimicrobial stewardship has strengthened over the past decade.

Yet amid this national prioritization, two studies suggest that there’s still room for significant improvement, much that can be attributed to better documentation practices.

“Unlike other things where we have to put a diagnosis code, [such as ordering a mammogram for primary care, even X-rays in the hospital or anything like that], where I always have to list a reason, that’s not really a thing for antibiotic statistics,” says Joseph Ladines-Lim, MD, PhD, first author of both studies and a resident in the departments of internal medicine and pediatrics at Michigan Medicine, University of Michigan. “There’s no mechanism, at least at the University of Michigan, which uses Epic, forcing you to code anything. There is just a lot of antibiotic prescribing that’s happening without any kind of coding or diagnosis coding from the clinician.”

Consequently, Ladines-Lim adds that lack of coding and documentation around antibiotic prescribing makes it difficult to identify where the inappropriate use is occurring so that health care organizations can implement policies for improving outlook. “For people in charge of stewardship or documentation, or even more broadly, I think it matters a lot, and deserves closer attention,” he suggests, pointing out that if the data health care organizations are relying on to understand antibiotic prescribing lacks context, it limits efficacy of stewardship strategies.

Study 1
Published in the Journal of Internal Medicine in April, the first study,2 “Prevalence of Inappropriate Antibiotic Prescribing With or Without a Plausible Antibiotic Indication Among Safety-Net and Non-Safety Net Populations,” aimed to determine the prevalence of inappropriate antibiotic prescribing in clinics both when there are plausible indicators for the action and when there are not. The study also looked at the differences between insured populations and those with Medicaid or no insurance (safety net populations).

An analysis was conducted of 67,065,108 and 122,731,809 weighted office visits for children and adults, respectively, National Ambulatory Medical Care Survey. Findings point to common inappropriate antibiotic prescribing practices in all populations with or without a documented indicator that would suggest need for such a medication based on antimicrobial stewardship best practices.

Ladines-Lim points out that there were surprises in the findings. “I totally expected, in general, worse outcomes for people who are on Medicaid or uninsured just because they have lower socioeconomic status. There are all kinds of reasons why this intuitively would make a lot of sense,” he notes. “But when you look at it, it’s actually like not so straightforward.”

In the case of inappropriate prescribing where there was no plausible indication for antibiotics, researchers found that inappropriate prescribing was occurring more in privately insured children and adults. Some theories as to the “why” behind this finding include the following:

• clinicians not recording appropriate ICD-10 codes for secondary bacterial infections;

• clinicians avoiding documentation around infection-related diagnoses as it relates to antimicrobial stewardship;

• differences in reimbursement between those with Medicaid and the privately insured; and

• parental pressures coming from families with means and more desired commercial insurance.

In the case of children, Ladines-Lim notes that there are other kinds of studies that support the parental pressure theory. “When parents have means, or some kind of commercial insurance, there might be some competition with community pediatricians,” he explains, noting that prescribing may occur because a physician wants to keep that business.

In terms of the reimbursement theory, Ladines-Lim points out that clinics receive a fixed amount from Medicaid, whereas with privately insured, reimbursement is more often fee-for-service. “With Medicaid, you get a payment to support these services, and then you’re not going to get anything else,” he explains, noting that Medicaid is capitated, thus there’s less incentive to include detailed documentation that might otherwise support antibiotic prescribing.

Study 2
The second study, “Appropriateness of Antibiotic Prescribing in US Emergency Department Visits, 2016–2021,” published in Antimicrobial Stewardship and Healthcare Epidemiology in May, included a national analysis of US emergency department (ED) visits with antibiotic prescribing from 2016–2021.3 Findings suggest that 27.6% of visits resulted in inappropriate antibiotic prescribing. Of those, 14.9% had diagnosis codes plausibly antibiotic-related, such as acute bronchitis, which would suggest actual inappropriate prescribing. Diagnosis codes not plausibly antibiotic-related, such as hypertension, were found in 12.6%, which would point to poor coding quality.

Ladines-Lim notes the research found that the amount of inappropriate antibiotic prescribing was basically the same as what it was reported a decade ago—roughly 30%. While that would suggest little progress has been made on the antimicrobial stewardship front in EDs since that time, the researchers were able to better contextualize the statistics.

“We were introducing this new concept of plausible vs not plausible, and so almost half of the antibiotics that are prescribed in emergency departments don’t have any codes that are related to antibiotics. So that’s kind of interesting,” Ladines-Lim says. “This [study] adds a little more nuance, because it’s based on a methodology that relies on ICD-10 codes. It could be that half of the time we’re not coding very accurately or completely, whereas the other half of the time it might be because we’re giving antibiotics inappropriately. Those are two very different things.”

Researchers reported that among visits with inappropriate antibiotic prescribing and a plausible indication, the most frequent indications were potential signs and symptoms of infection (54%), bronchitis (17.4%) and upper respiratory infection (10.4%). The most frequent potential signs and symptoms of infection were abdominal pain, headache, nausea with vomiting, dyspnea, and fever. Researchers note that, “If these visits represent instances in which clinicians were unsure of the diagnosis but prescribed antibiotics regardless, stewardship initiatives might focus on safely minimizing unnecessary antibiotic prescribing with diagnostic uncertainty. If these visits represent instances in which clinicians had diagnostic certainty but coded nonspecific signs or symptoms, initiatives should focus on improving coding quality.”

Among ED visits with inappropriate antibiotic prescribing, almost half lacked plausible diagnosis codes. The most frequent diagnosis codes were essential hypertension, chest pain, and joint pain in these cases. According to researchers, “These may represent instances in which clinicians prescribed antibiotics appropriately but neglected to code the condition, or alternatively, prescribed antibiotics for antibiotic-inappropriate conditions and deliberately avoided coding these to avert scrutiny.”

Ladines-Lim believes HIM and clinical documentation improvement departments have an important role to play in ensuring that antibiotic prescribing is backed by appropriate documentation. “They could hopefully try to close the gap in terms of clinicians prescribing antibiotics without appropriate ICD-10 coding or documentation, and I’m sure they could do that in any number of ways,” he says. “You know, anytime you have an antibiotic, you could, in theory, introduce some kind of forcing mechanism where you have to document an ICD-10 code.”

— Selena Chavis is a Florida-based health care writer.

 

References
1. Antimicrobial resistance facts and stats. Centers for Disease Control and Prevention website. https://www.cdc.gov/antimicrobial-resistance/data-research/facts-stats/index.html. Updated April 22, 2024.

2. Ladines-Lim JB, Fischer MA, Linder JA, Chua KP. Prevalence of inappropriate antibiotic prescribing with or without a plausible antibiotic indication among safety-net and non-safety net populations [published online April 26, 2024]. J Gen Intern Med. doi: 10.1007/s11606-024-08757-z.

3. Ladines-Lim JB, Fischer MA, Linder JA, Chua KP. Appropriateness of antibiotic prescribing in US emergency department visits, 2016-2021. Antimicrob Steward Healthc Epidemiol. 2024;4(1):e79.