Outpatient overprescribing: ‘Cultural shift’ needed to spare antibiotics
In recent years, much emphasis has been placed on the cautious and appropriate use of antibiotics as a means of mitigating the threat of antimicrobial resistance. Yet antibiotic overprescribing remains a concern, particularly in outpatient settings such the physician’s office, urgent care centers, clinics and EDs.
According to the CDC, in 2015 alone, roughly 269 million antibiotic prescriptions were written from outpatient pharmacies in the United States, and at least 30% were not necessary. Additionally, amid a 16.6% decrease in overall antibiotic spending between 2010 and 2015, antibiotic expenditures in outpatient clinics surged by 148%, according to a study published in Clinical Infectious Diseases.
Experts cite multiple reasons for antibiotic overprescribing, including lack of provider education, an overloaded or rushed patient flow, diagnostic uncertainty or pressure from patients to prescribe unnecessary antibiotics.
“One thing that many outpatient providers talk about is that patients expect antibiotics,” Sara C. Keller, MD, MPH, assistant professor of medicine in the division of infectious diseases at Johns Hopkins University School of Medicine, told Infectious Diseases in Children. “If they do not get antibiotics, they’re in this world of patient satisfaction scores and Yelp reviews. Outpatient providers are very nervous that a bad Yelp review is going to negatively impact their revenue. There’s a lot of angst around this, and understandably so.”
The drivers of antibiotic overprescribing are complex and manifold. Infectious Diseases in Children spoke with several experts in the field about the scope of the problem and what can be done to address it.
According to Lauri A. Hicks, DO, director of the Office of Antibiotic Stewardship in the CDC’s Division of Healthcare Quality Promotion, the first step is to change the way antibiotics are perceived.
“We need to see a cultural shift in how we think about antibiotics in our country,” Hicks said in an interview. “We’ve tended to think of antibiotics as the cure-all for decades. We really need to think about antibiotics as a precious resource that should be used when needed, so that they will work when they’re needed.”
Inappropriate prescribing of antibiotics can undermine the quality of both inpatient and outpatient health care and leaves the population increasingly vulnerable to previously treatable conditions.
“We’re already seeing infections for which antibiotics are ineffective,” Hicks said. “As time goes on, my concern is that common, garden-variety infections that usually would be easy to treat with oral antibiotics are going to require a hospital admission to get very broad-spectrum antibiotics. Or they will just be untreatable. What will happen is that the person will have to rely on their own immune defenses to fight off the infection. That’s obviously not a situation we want to be in.”
Inappropriate antibiotic prescribing is also costly. The CDC reported that in 2009, roughly $10.7 billion was spent in the U.S. on antibiotic treatment, with $6.5 billion of that cost incurred in the outpatient setting. The annual cost of antibiotic resistance on the U.S. economy is an estimated $20 billion.
Moreover, the adverse effects often caused by inappropriate antibiotics can lead to further health care visits.
“We know a lot of those antibiotic prescriptions that are written in outpatient settings can eventually lead to patients having unnecessary emergency department visits,” Michael J. Durkin, MD, MPH, assistant professor of medicine at Washington University School of Medicine, said in an interview. “These are visits that could have been prevented.”
Adverse reactions to antibiotics are not uncommon in children, according to Sean T. O’Leary, MD, MPH, associate professor of pediatrics at the University of Colorado Denver Anschutz Medical Campus and a member of the AAP’s Committee on Infectious Diseases.
For example, O’Leary told Infectious Diseases in Children that up to 10% of children prescribed antibiotics for otitis media have an adverse reaction, and that many of those children will have another visit with a physician because of the reaction.
In addition to causing immediate reactions, antibiotics can also affect the patients’ microbiome.
“In some studies, children who were prescribed antibiotics were at higher risk of developing atopic diseases like asthma and allergies,” he said.
Main drivers of overprescribing
The desire to please patients is a driving force in outpatient antibiotic overprescribing. According to Michael S. Calderwood, MD, MPH, regional hospital epidemiologist at Dartmouth-Hitchcock Medical Center and assistant professor of medicine at the Geisel School of Medicine at Dartmouth College, patients presenting with acute illness may have gone to some trouble to get to the physician’s office and expect something tangible for their efforts.
“They have often taken time off from work to come to the visit, or had to be unexpectedly absent from work due to their illness,” Calderwood told Infectious Diseases in Children. “In such situations, the patient is looking for treatment that will get them back on their feet as fast as possible. This often means that an antibiotic is requested, even when that antibiotic may not be needed, and is potentially harmful.”
In some cases, the pressure that physicians feel to satisfy patients may be based as much on their own perceptions as on the patient’s demands, according to Durkin.
“Generally, a lot of providers are trying to satisfy their patients, so if there are patients they think may want antibiotics or might want to feel better quickly, there may be some perceived pressure on the provider’s part that the patient may be trying to pressure them to prescribe antibiotics,” Durkin said. “I think that’s probably more the provider’s perception than what’s actually the case in terms of what the patients want.”
Hicks called the fear of patient dissatisfaction “probably the single most important factor that is contributing to overprescribing.”
“We hear from doctors as well as nurse practitioners and physician assistants that they feel the need to please their patients, and that is one of the reasons this is happening,” she said.
Calderwood said providers are now asked to see more patients in less time — a factor that intensifies the pressure to acquiesce to a patient’s requests for antibiotics.
“Prescribing an antibiotic, as requested by the patient, may be seen as the path of least resistance and the choice that is likely to satisfy the patient,” he said. “The conversation involved in not prescribing an antibiotic often takes longer. This conversation involves educating the patient on the nature of their symptoms, the inability of antibiotics to treat viral infections, the potential harms from antibiotics, and treatment options that do work.”
Diagnostic uncertainty is another cause of unnecessary antibiotic prescriptions. When physicians have doubt about an illness, they may default to an antibiotic just in case, Durkin said.
“We’re never 100% certain, for a lot of conditions, what the underlying etiology is — sometimes we’ll see infections that have a bacterial superinfection on top of a viral infection,” he said. “If providers aren’t 100% confident that this is a viral condition, or they think there could be a bacterial component, these providers will often prescribe antibiotics.”
Unlike pediatricians and family physicians who frequently treat children, providers in urgent care centers and EDs may not be as comfortable sending away a child with a viral infection without antibiotics because they are concerned about the possibility of “missing something,” according to O’Leary.
“So, a lot of clinicians who are not experienced with treating children will overprescribe antibiotics to younger children because of their own discomfort,” he said.
Beyond the various other reasons for antibiotic overprescribing, there is also a certain amount of simple denial at work, according to Calderwood.
“Until recently, there was less concern about antibiotic resistance in the ambulatory setting, with many outpatient providers thinking that drug resistance was really mostly a problem in hospitals,” he said. “However, common first-line agents are no longer working for common clinical infections seen in the ambulatory setting. This is because of overuse/misuse.”
A 2016 study published in JAMA based on an evaluation of 184,032 ambulatory and ED visits from the 2010-2011 National Ambulatory Medical Care Survey and the National Hospital Ambulatory Medical Care Survey found that per 1,000 population, an estimated 506 antibiotic prescriptions were written annually among all ages and conditions combined in 2010-2011. Of these, 30% were deemed unnecessary.
Acute respiratory infection is the most common condition for which unnecessary antibiotic prescriptions are written. An acute respiratory infection is also the most common reason for an outpatient visit, peaking during cold and influenza season, when people are contracting illnesses caused by viruses, Hicks said.
According to the JAMA report, among annual antibiotic prescriptions written for acute respiratory illnesses, only 111 of 221 annual antibiotic prescriptions written for these illnesses per 1,000 people were deemed appropriate. A more recent report in JAMA Internal Medicine showed that 45.7% of patients who visited an urgent care center in 2014 for a respiratory illness that did not require an antibiotic were inappropriately prescribed one. The proportion was lower in EDs (24.6%), medical offices (17%) and retail clinics (14.4%).
“Over four out of every 10 antibiotics prescribed in the ambulatory setting are prescribed for acute respiratory conditions, half of which are viral,” Calderwood said. “Antibiotics are not needed to treat flu, sore throat that is not strep throat, common cold/runny nose and bronchitis.”
Durkin said the scope of the problem is likely even greater than the JAMA study suggests.
“This is probably a conservative estimate — when you factor in inappropriate antibiotic selection and inappropriate treatment duration, the numbers will probably be much higher,” he said. “We previously showed that about 75% of antibiotic prescriptions for UTIs are for a treatment duration that is outside of clinical practice guidelines. Other groups have shown issues with skin and soft tissue infections, community-acquired pneumonia, and other conditions.”
Keith S. Kaye, MD, professor of medicine in the division of infectious diseases at the University of Michigan Medical School, said campaigns led in part by the CDC have targeted parents of children with otitis media and respiratory infections.
“These campaigns have been successful in decreasing antibiotic demand and prescribing in children. I think similar campaigns can be successful in adults as well,” he said.
According to a CDC report, the top prescribers of oral antibiotics are family practice physicians, physician assistants and nurse practitioners, internal medicine physicians, pediatricians and dentists — in that order.
Keller said efforts have been made to increase awareness among some physicians but have been inconsistent in reaching nurses and physician assistants.
“The training is not quite as standardized,” he said. “They’re figuring out how to get the message across.”
Recent studies have gone beyond medical specialty to evaluate other characteristics of high-volume antibiotic prescribers. Hicks and colleagues evaluated prescriptions written in 2011 by almost 110,000 family practitioners, pediatricians and internists and found that older male physicians in small Southern practices were the highest volume prescribers. According to the study, the median number of antibiotic prescriptions written per physician in all three medical specialties increased with each decade that passed since medical school graduation.
Hicks acknowledged that these prescribing habits could be attributed to other factors not related to overprescribing, including that these prescribers are seeing the most patients.
“We can’t rule this out as a possibility,” she said. “There is also a possibility that they are seeing an older and more vulnerable patient population.”
However, Hicks said results from other analyses that she and colleagues have completed suggest that physicians in the South are more likely to prescribe for conditions that do not warrant antibiotic use, such as uncomplicated bronchitis.
“Common reasons physicians prescribe when antibiotics aren’t indicated include: They are concerned about patient satisfaction; they have always prescribed for conditions that don’t warrant antibiotic use and changing behavior is difficult; and they are concerned about ‘missing something,’” Hicks said.
Durkin believes antibiotic prescribing habits are closely related to outdated recommendations.
“My personal suspicion is that a lot of providers are prescribing antibiotics based on trends that they learned during their training,” he said. “I think if you were to look at recent graduates vs. individuals who had graduated further out, the recent graduates may be slightly more adherent to guidelines because they had learned that during their training.”
Breaking the cycle
Raising awareness about the consequences of unnecessary antibiotic prescribing is a fundamental step toward controlling the problem, but many health care organizations and clinics might not know where to begin in terms of disseminating this information. According to Keller, it should be conveyed throughout the patient’s health care encounter, starting from the moment they call the physician’s office.
“Whoever is answering that triage phone call about an upper respiratory infection, whether it’s the front desk staff or a nurse, can set up the expectation at that point by saying something like, ‘Oh, that’s horrible — there’s been a really bad virus going around. Here are some things you can try before you even come in to the doctor’s office,’” she said. “You can also tell the patient that when they come in, they will likely be given some more things to help them feel better, rather than telling them to come into the office so they can get an antibiotic. This helps set expectations from the beginning.”
According to Hicks, “talking points” can reassure patients about forgoing inappropriate antibiotics. The main thing patients want, she said, is to feel that their concerns are being heard and taken seriously.
“Again, it gets back to patient satisfaction,” she said. “New studies have shown us that there are specific ways and strategies that we can use to communicate with patients about their illness to make them feel they’ve gotten something out of the visit.”
She cited a scenario involving a patient with sinusitis, or a child with otitis media. The clinical guidelines on antibiotic use for both are very clear, she said, but patients might still request an antibiotic.
“One of the options for a situation where there’s uncertainty and the patient is not very ill is watchful waiting,” she said. “Or there is another option called delayed prescribing, where the provider can say, ‘You know, my thought right now is that this isn’t an infection that requires antibiotics. However, if you aren’t feeling better in 48 to 72 hours, then I’m much more inclined to think this is a bacterial infection that needs an antibiotic.’ You can either have the patient call you back, or you can write a prescription and put a specific date on it, so the patient can fill it if they don’t get better.”
O’Leary said that otitis media is a good example of a condition that may not necessarily require antibiotics. In this case, physicians should tell concerned parents, “I think this is probably going to get better on its own without antibiotics, but just in case, I am going to go ahead and give you a prescription, and you can fill it in 2 to 3 days if it seems like he is not getting better. Please let me know if you end up filling the prescription.”
He added that physicians should put an expiration date on the prescription.
“That’s an approach a lot of clinicians will use,” he said.
O’Leary said pediatric studies have found that although clinicians assume parents want antibiotics, in many cases, that is not true.
“Even if parents expect antibiotics, they are willing to go without them if the clinician can explain why they are not needed and can give them some alternative treatment recommendation,” O’Leary said.
He added that what parents really want is a plan of action for their children’s treatment. Clinicians should provide parents with three things, he said: the reasoning behind why antibiotics are not needed, instructions for what they can do for their children, and a plan for treatment.
Another key resource for physicians is the CDC’s Core Elements of Outpatient Antimicrobial Stewardship, published in 2016. The document outlines the core elements — commitment, action for policy and practice, tracking and reporting, and education and expertise — and describes ways in which these objectives can be achieved.
“It is a cornerstone, it’s a key part in encouraging providers who are taking care of patients to review or enact some of these core elements in an outpatient setting,” Durkin said. “I think that’s challenging. Some of the incentives aren’t quite there yet in terms of trying to get them to follow those outpatient core elements.”
‘Outside the box’ approaches
Durkin cited other, less conventional approaches to improving outpatient antibiotic stewardship, including methods that use physician comparison to motivate improvement.
“They’ve done things like benchmarking, where they’ll compare provider A to provider B and say, ‘You’re in the top 15% of antibiotic prescribing,” or, ‘You’re not in the top 15% for this,’” he said. “There have also been efforts involving commitment posters, where they will make a poster for the provider and write down that they’ve committed to optimizing antibiotic prescribing. That’s shown some improvement in outpatient settings.”
Another approach proposed by some policymakers is the imposition of financial penalties on organizations and physicians that regularly recommend antibiotics for uncomplicated upper respiratory infections. However, a 2017 study found that only 31% of physicians surveyed would support fines for excessive antibiotic prescribing.
Calderwood expressed concern about the use of financial penalties for excessive antibiotic prescribing, citing a 2014 study that revealed the limited role of “pay-for-performance” initiatives in primary care.
“In general, the impact of pay-for performance in primary care is short-lived, with the greatest impact seen on process-of-care measures rather than patient health outcomes,” he said.
He noted a 2015 editorial by Woodhandler and colleagues, which said that improving patient care should be a sincere and collaborative effort by providers, rather than a reward/punishment-motivated effort.
“What concerns me is the ability to skew the data such that one looks good on a metric without changing prescribing behavior in a way that truly impacts the patients,” Calderwood said. “The example that I would give is the linking of antibiotic prescribing to specific diagnosis codes. If a provider is ‘fined’ for prescribing an antibiotic for a diagnosis where an antibiotic is inappropriate, the provider will just stop using that diagnosis code, at least when an antibiotic is prescribed.”
Calderwood supports electronic order sets suggesting alternative treatments to antibiotics, as well as accountable justification and peer comparison.
“I feel that each of these interventions, alone or in combination, is more likely to have a long-term impact than ‘fines’ for excessive antibiotic prescribing,” he said.
Changes to electronic medical records also have been used to give clinicians electronic notifications or instructions concerning antibiotic prescribing, but implementing these programs on a large scale will be a significant undertaking, Durkin said.
“We’ve still got a long way to go between translating a few studies that have shown benefits and showing sustainable implementation in a nonacademic setting,” he said. “We’re trying to change antibiotic prescribing on a national scale. How do we disseminate these interventions broadly? If we show that an intervention is successful in a cohort of clinics, how are we going to translate that to clinics throughout the United States? That is the challenge.”
Addressing inappropriate antibiotic prescribing is daunting, but progress is steadily being made, according to Calderwood.
“This message is getting out there to providers — we now need to work to help providers in communicating this message to patients,” he said. “Toolkits and resources are available to help educate patients, provide alternative and effective treatment for viral infections, and engage providers in rational prescribing choices. After all, antibiotic stewardship is everyone’s responsibility.” – by Jennifer Byrne and Bruce Thiel
- Allen T, et al. Risk Manag Healthc Policy. 2014;doi:10.2147/RMHP.S46423.
- CDC. Antibiotic Use in the United States, 2017: Progress and Opportunities. https://www.cdc.gov/antibiotic-use/stewardship-report/outpatient.html.
- CDC. Core Elements of Outpatient Antibiotic Stewardship. 2016. https://www.cdc.gov/mmwr/volumes/65/rr/rr6506a1.htm?s_cid=rr6506a1_e.
- CDC Grand Rounds: Getting Smart About Antibiotics. Available at; https://www.cdc.gov/mmwr/preview/mmwrhtml/mm6432a3.htm
- Fleming-Dutra KE, et al. Am Fam Physician. 2016;Aug. 1;94(3):200-202.
- Fleming-Dutra KE, et al. JAMA. 2016; doi:10.1001/jama.2016.4151.
- Fleming-Dutra KE, et al. Open Forum Infect Dis. 2018;doi:10.1093/ofid/ofx279.
- Liao JM, et al. Ann Intern Med. 2017;doi:10.7326/L17-0102.
- Meeker D, et al. JAMA. 2016; doi:10.1001/jama.2016.0275.
- Palms DL, et al. JAMA Intern Med. 2018;doi:10.1001/jamainternmed.2018.1632.
- Suda KA, et al. Clin Infect Dis. 2017;doi:10.1093/cid/cix773.
- Woodhandler S, et al. Ann Intern Med. 2015; doi:10.7326/M15-1393.
- For more information:
- Michael S. Calderwood, MD, MPH, can be reached at firstname.lastname@example.org.
- Michael J. Durkin, MD, MPH, can be reached at email@example.com.
- Lauri A. Hicks, DO, can be reached at firstname.lastname@example.org.
- Keith S. Kaye, MD, MPH, can be reached at email@example.com.
- Sara C. Keller, MD, MPH, can be reached at firstname.lastname@example.org.
- Sean T. O’Leary, MD, MPH, can be reached at email@example.com.
Disclosures: Durkin reports receiving funding for Medscape CME materials through an unrestricted Merck educational grant. Calderwood, Hicks, Kaye, Keller and O’Leary report no relevant financial disclosures.