Issue: February 2016
February 16, 2016
15 min read

Antibiotic overuse: A multidisciplinary concern

Issue: February 2016
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Since their discovery in the early 20th century, antibiotics have revolutionized the treatment of bacterial infections and are currently used in a wide variety of medical specialties. Over the years, however, unnecessary prescriptions have diminished their efficacy against certain infections, prompting WHO, the CDC and other leading health organizations to call for a renewed focus on overprescribing across all areas of health care.

“What we know is that in pretty much every study that’s been done, there is quite a bit of overprescribing, and that cuts across almost all settings,” Arjun Srinivasan, MD, associate director for the Healthcare Associated Infection Prevention Programs at the CDC, told Infectious Disease News. “If you look at outpatient, inpatient, nursing home, acute care and other settings, you can make the general statement that any group that uses antibiotics is going to have some degree of overuse — and that’s true of all specialties.”

To explore this ongoing concern, Infectious Disease News spoke with several experts in different specialties about the factors driving antibiotic overuse and the interventions designed to curb its impact.

A problem for all specialties

In a recent review of nationwide prescription data, Lauri A. Hicks, DO, medical epidemiologist in the CDC’s Division of Healthcare Quality Promotion, and colleagues estimated that 262.5 million oral antibiotics were prescribed by U.S. health care providers in 2011 at a rate of 842 prescriptions per 1,000 persons. The majority of these were distributed by primary care providers, among which family medicine was most active with 64.1 million prescriptions at a rate of 667 prescriptions per provider. However, high prescription rates also were observed within the dermatology (724 per provider), pediatrics (598 per provider), otolaryngology (430 per provider) and emergency medicine (427 per provider) specialties. Children aged 9 years and younger received antibiotic prescriptions at a much higher rate than the general population, and the most commonly prescribed agents were azithromycin and amoxicillin.

Arjun Srinivasan, MD, from the CDC, stressed the importance of expanding awareness of antibiotic overprescribing to other provider groups.

Photo by Jim Gathany

These estimates come alongside data suggesting frequent misuse of antibiotics. According to the CDC, approximately 20% to 50% of all antibiotics prescribed in U.S. acute care hospitals are given incorrectly, a trend recently corroborated in a study led by Tamar F. Barlam, MD, MSc, director of the antimicrobial stewardship program at Boston Medical Center and associate professor of medicine at Boston University School of Medicine. In reviewing past visits at their medical center for respiratory tract infections, Barlam and colleagues identified highly variable rates of inappropriate prescribing among general internal medicine and, more frequently, family medicine practitioners.

“We looked at the visits over a number of years, and we found that there was still a marked level of prescribing for respiratory tract infections that are generally recognized as viral,” Barlam told Infectious Disease News. “When we broke it down a little bit, it was interesting to see the range of different practices where some physicians were actually pretty good, and they were prescribing in maybe one-third of the visits. But, it went all the way up to three-quarters or higher of the visits for other practitioners.”

Antibiotic overuse has persisted for decades, but recent developments have raised the stakes and increased the scrutiny of provider prescription practices. Due to the increasing prevalence of antibiotic-resistant pathogens, appropriate drug use is now a major goal of numerous health care organizations, as well as a primary component of the White House’s National Action Plan for Combating Antibiotic-Resistant Bacteria.

The issue is further compounded by the increasing frequency of antibiotic shortages described by Farha Quadri, of the department of emergency medicine at George Washington University, and colleagues. Along with identifying a significant escalation in shortage rates from 2007 onward, the researchers noted that many of the 138 drugs affected between 2001 and 2013 were broad-spectrum antibiotics, and for 32 of these drugs, there were multiple shortages that could have had a detrimental impact on patient care.

“Shortages have significant implications in the context of other factors that limit availability of effective antimicrobials, including a shrinking pipeline of new antibiotics and increasing drug resistance,” Quadri and colleagues wrote. “Given that 46% [of affected drugs] are used in the treatment of high-risk pathogens, such as MRSA, [carbapenem-resistant Enterobacteriaceae], and Pseudomonas, it is imperative to develop a comprehensive strategy to mitigate the implications of shortages on clinical practice and patient outcomes.”


Awareness varies across settings

The issue of overprescribing may be well-known within ID, but according to Srinivasan, the majority of infections where antibiotics are prescribed will never reach an ID specialist.

“If you look at the prescribing of antibiotics in hospitals, the top three conditions for which antibiotics are prescribed are respiratory infections, urinary tract infections, and skin and soft tissue infections,” Srinivasan said. “Those three conditions alone comprise more than half of all antibiotics that are prescribed in hospital settings, and those three conditions are overwhelmingly cared for nowadays by hospitalists. I’m an ID consultant, but I rarely see those three types of infections. It’s only in really unusual circumstances that I’m going to get involved.”

While Srinivasan stressed the importance of expanding awareness to provider groups treating the most common infections, there is a chance that many of these heavy prescribers are already aware of the problem. Barlam noted that many of the internal and family medicine practitioners in her academic medical center were up-to-date on current data regarding antibiotic prescribing, yet they continued to overprescribe.

Scott A. Flanders, MD, professor of medicine and director of the hospital medicine program at the University of Michigan, described a similar situation among his hospitalists, and said that more effort is needed to bring these issues into everyday care.

Scott A. Flanders

“Problems with antimicrobial resistance and antibiotic overuse are increasingly becoming known to more and more inpatient prescribers,” Flanders told Infectious Disease News. “It has received increased national attention, and there’s been a lot of work at many hospitals to target the problem and put measures in place to make improvements. It is increasingly on the radar of many hospitalists, but I don’t think it is necessarily ever-present in the prefrontal cortex as they’re caring for their sick, hospitalized patients — that’s where we need to do some work.”

Srinivasan said it is also necessary to challenge inappropriate antibiotic use within the organ-based and surgical subspecialties, where the discussion of overprescribing may just be beginning. One such field is dentistry, which, according to data presented at ICAAC 2015 by Katie Suda, PharmD, MS, of the University of Illinois at Chicago, is responsible for one out of every 10 antibiotic prescriptions. Another is ophthalmology, where Andrzej Grzybowski, MD, PhD, professor of ophthalmology at the University of Warmia and Mazury and head of the department of ophthalmology at Poznan City Hospital, Poland, said empiric prescriptions for common ocular pathogens and surgical procedures are the norm.

Andrzej Grzybowski

“The global risk [of resistance] is much higher when we use antibiotics with no critical approach,” Grzybowski told Infectious Disease News. “We give every patient with red eye topical antibiotic. The negative consequences of this practice are much higher than the real problems we’d face if we stopped giving them these antibiotics ... but unfortunately, awareness of this issue is low.”

Grzybowski pointed to recommendations and awareness campaigns from the American Academy of Ophthalmology against routine antibiotic use for certain conditions as evidence of a movement away from these practices, but admitted that many older specialists accustomed to lax prescription guidelines are late to conform.

In a study on empiric ophthalmologic treatments recently presented at ICAAC 2015, Darlene Miller, DHSc, MPH, research associate professor of ophthalmology and scientific director of the Ocular Microbiology Laboratory at the University of Miami, examined the strategy’s effect on patient outcomes. Among the 176 microbial keratitis patients presenting to the ER and administered antimicrobial therapy by ophthalmologists, 52% were given treatment without any microbiological evaluation, and 92% were prescribed broad-spectrum antibiotics. In addition, those given treatment more often experienced slower recovery, higher culture rates and MRSA, which Miller said demonstrates the need to re-evaluate antibiotic prescription practices in this specialty.

“Antibiotic resistance in ophthalmology is a part of the growing problem in global antibiotic resistance, and it is a contributing factor to patient failure in health care,” Miller said in an interview. “We need to bring all settings into the discussion on antibiotic resistance and patient care in the United States and globally.”


Patient pressure, diagnostic uncertainty drive antibiotic use

Multiple factors persist that may influence providers to overprescribe antibiotics despite being aware of the consequences. Among these, the impact of a patient’s expectations has long been of interest to researchers.

In a survey of 310 adults visiting an acute care clinic for upper respiratory infections published in 2003, researchers reported that 39% of patients wanted antibiotics for their condition, and 46% of these received a prescription. More recently, Mark Ashworth, DM, MRCP, FRCGP, a clinical senior lecturer in the department of primary care and public health services at King’s College London GKT School of Medical Education, and colleagues surveyed 7,800 of all general practices in England and found that frequent antibiotic prescribing was a significant predictor of positive doctor and practice satisfaction scores, and could hint at patient attitudes toward prescription. Furthermore, a 2014 phone survey of U.S. primary care providers by CDC researchers found pressure to satisfy patients was one of the most often cited reasons for antibiotic guideline noncompliance.

Although this influence is common throughout the outpatient setting, Jason G. Newland, MD, MEd, associate professor of pediatrics at Washington University and director of antimicrobial stewardship at St. Louis Children’s Hospital, said the pressure to prescribe can be even stronger in pediatric medicine, where viral respiratory infections and worried parents are a frequent occurrence.

Jason G. Newland

“As pediatricians we don’t realize the influence of perceived parental expectation,” Newland told Infectious Disease News. “Those pressures — that prescriber talking to a family — those are real, and those are probably the hardest because — from the medicine side of it — we know what we’re supposed to give but we feel the need to meet what we think the parents want.”

According to Srinivasan, providers also may prescribe antibiotics when they are unsure of what is causing a patient’s condition. While this can result from the prescriber’s unfamiliarity with guidelines or drug indications, such circumstances are often the result of inadequate or delayed diagnostic information.

“Infectious disease diagnoses are sometimes hard to make,” Srinivasan said. “We don’t have all of the diagnostic tools that we’d like to have to be able to make definitive diagnoses, and so a lot of times antibiotics are started as a ‘just-in-case’ medication.”

Emergency cases are frequently in need of immediate treatment to stave off infection, Flanders said, and many of the diagnostics available in hospitals are unable to promptly deliver comprehensive results. As more and more patients enter care while on chemotherapy, immunosuppressants or any other treatments for comorbities, front-line health care providers are forced to quickly make what they believe to be the best decision for their patients.

“The diagnostic tests that tell you whether it’s a bacteria or not are not perfect, and it’s often unclear what test to use,” Flanders said. “As a result, most inpatient providers and hospitalists err on the side of using more antibiotics, and for longer periods. Sometimes that’s the right decision, but often it’s not.”

Effective interventions

While the empiric use of antibiotics for undiagnosed patients may be a necessary trade-off, many patients continue to receive these antibiotics well after a better treatment plan has been determined.

“We have a patient who’s sick,” Srinivasan said. “We think that they might have a bacterial infection, so we start antibiotics empirically. But what we find is that after a day or two we have more information: We’ve got culture results, we’ve got imaging, we know more about the patient’s illness. A lot of times, antibiotics are continued without a careful reassessment of whether or not they’re needed.”


To prevent this from happening, Srinivasan recommends the implementation of an “antibiotic timeout” — a reassessment of the patient’s condition, treatment and diagnosis 24 to 72 hours after the initial prescription. Flanders argued that this approach is effective and should be implemented into standard hospital procedure.

“We’ve done some pilot work that has shown that up to one-quarter or more of patients can sometimes stop antibiotics at 72 hours when you are doing that pause and that timeout,” Flanders said. “This needs to be a hard-wired process — the type of thing you want to be happening to every patient.”

Along with the timeout, Flanders and Srinivasan both suggested that the implementation of established guidelines and comprehensive documentation procedures — often through the use of electronic health records — could further reduce unnecessary antibiotic use by clarifying drug indications for uninformed providers and giving others the opportunity to double-check prescriptions.

“For example, say a proceduralist or an interventional cardiologist wants to put in an order for an antibiotic,” Flanders said. “If they’re required to write down the documentation as to why it’s being given, that provides an opportunity for hospitals with built-in processes to review these things to say, ‘Boy, that doesn’t seem like an appropriate indication, I should reach out and get additional information from that provider.’ ”

Many of these processes are among the core elements of antimicrobial stewardship programs. Strongly recommended by CMS and the CDC for all acute care hospitals, the programs instituted in some hospitals have been shown to improve patient outcomes, inform prescribers, reduce unnecessary antibiotic use and cut costs for hospital administrations. As these programs are best implemented with the support of a committed expert, Srinivasan called on ID specialists working in hospitals to take an active role in the formation of systematic stewardship.

“We want to see ID doctors out there leading that charge, connecting with those hospitals, going to the hospital leadership to say, ‘Look, you need to have a stewardship program, and I’m the person to lead that program for you,’ ” Srinivasan said. “One of the key things that we think stewardship programs need to do is educate. Who better to lead that educational effort than an ID physician?”

Communicating with colleagues and patients

Srinivasan said antibiotic timeouts and other interventions offer an ideal opportunity for ID specialists to communicate best practices to peers and other providers offering antibiotic treatments to patients. In addition, Barlam noted that the clustered prescription rates observed in her study suggest that many practitioners are also influenced by the antibiotic prescribing patterns of their peers. As such, she said that fostering a genuine relationship with physician leaders could positively influence antibiotic mindsets throughout the rest of the practice.

“[Consultations] are always learning opportunities to really take the time to explain why you’re suggesting certain antibiotics, why you might be suggesting the escalation or stopping of an antibiotic, and that it’s really important to evaluate the data and constantly be re-evaluating if that’s the right regimen,” Barlam said. “It’s a combination of developing those respectful relationships and providing the scientific evidence to the physician.”

Newland discussed cooperation between specialties as a two-way conversation, as ID specialists may not have the specific knowledge to appropriately limit antibiotic use in other fields with high prescription rates.

“[For example], I’m not a dermatologist, but you wonder can we decrease the number of antibiotic treatments?” Newland said. “Acne’s a real issue and has a real impact on people’s lives. Since I’m not an expert in the treatment of acne, that’s where I and other infectious disease physicians have to partner with dermatologists to understand how we might decrease the use of antibiotics while maintaining good outcomes for those with acne, and in the end decrease the number of adverse events from receiving antibiotics.”


A similar approach can be taken with patients as well. Citing research exploring communication breakdowns between physicians and patients, Srinivasan described situations where a patient’s efforts to understand their current treatment were misunderstood by the physician as a demand for more antibiotics.

“There’s no question that pressure from any external source is going to be a factor,” Srinivasan said. “But, I think what we are generally finding is that the ways to solve that problem are to look into issues with communication.”

Newland said that more often than not, patients and families are more interested in reassurance than a prescription. While some requests for antibiotics may be genuine, a straightforward conversation about why an antibiotic may not be necessary is the best way to foster an understanding between both parties.

“It’s going to take a complete societal effort to limit antibiotics use,” Newland said. “We know that clinicians misperceive. We also know that people really want antibiotics, so we really need to work on it together. We need the families to say, ‘Hey, I know you’re giving me an antibiotic, but if you don’t really think I need it, I’m OK not receiving it.’ It needs to be a collaborative approach.” – by Dave Muoio

Disclosures: Newland reports relevant relationships with Cubist, Pfizer and RPS Diagnostics. Ashworth, Barlam, Flanders, Grzybowski, Hicks, Miller, Quadri, Sanchez, Srinivasan and Suda report no relevant financial disclosures.



Harms of inappropriate prescription take priority over patient requests

Antibiotics can do harm to the individual taking them, as they could lead to allergic reactions, alteration of the beneficial microflora of the GI tract or predisposition to Clostridium difficile. This is similar to other types of medicines, such as insulin, antihypertensives or chemotherapy, which also can have adverse effects. A physician would not give chemotherapy to someone who did not have cancer or insulin to someone who was not diabetic. Similarly, physicians should not give antibiotics to patients who do not have a bacterial infection.

Barbara E. Murray

There might be situations, however, when the use of antibiotics is considered optional or conditional; for example, a skin infection might be just on the border of “should be treated with antibiotics” vs. just incision and drainage, and the patient cannot return for follow-up. Similarly, antibiotic prophylaxis to prevent traveler’s diarrhea or antibiotics to treat traveler’s diarrhea that is moderate, might or might not be given. Perhaps in this case, patient preference as well as the patient’s situation and general health might be considered.

Moreover, unlike other drugs, whose use in one patient does not potentially compromise their use in another (insulin, antihypertensives), the use of antibiotics in a patient can proliferate antibiotic resistance, both in the infecting bacteria and in the “bystander” bacterial flora. This then poses a potential risk to others if such organisms spread. For the community, as well as the individual patients, I would encourage physicians to not consider, in most instances, patient preference.

Barbara E. Murray, MD, is the director of the infectious diseases division at the University of Texas Medical School, and a former president of the Infectious Diseases Society of America. Disclosure: Murray reports previous relationships with AstraZeneca, Cubist Pharmaceuticals, Forest Laboratories, Merck and Theravance.



Understanding patient preferences for antibiotics is important to providing optimal care

We increasingly recognize that care should be patient-centered. The Institute of Medicine, in its seminal report, Crossing the Quality Chasm, defined patient-centered care as being “respectful of and responsive to individual patient preferences, needs and values.” While this does not mean that a request for antibiotics should drive clinical decision-making, it does emphasize the need to share the rationale behind these decisions and understand what the patient really wants. Some may want reassurance, to know that their condition is not serious and will improve. Others may ask for antibiotics because of the misperception that antibiotics will make them feel better faster, which is an opportunity for patient education.

Jesse T. Jacob

Understanding the root cause of the request can be challenging. In a busy clinic, it takes more time to explain why an antibiotic is not indicated than it does to write an antibiotic prescription. Not providing an antibiotic prescription may also be perceived as causing lower patient satisfaction. Providing reassurance is a skill that must be developed, as it requires several steps to be successful.

Patients often ask for, and receive, antibiotics for syndromes that are typically viral. Practice patterns vary widely. Some providers prescribe antibiotics for more than 95% of patients with acute respiratory tract infections regardless of symptoms, but others in less than 40%. As part of its Get Smart initiative, the CDC has provided education materials to help structure the conversation with patients, including a “prescription” for supportive care for viral illnesses.

Sir William Osler declared, “One of the first duties of the physician is to educate the masses not to take medicine.” We should listen to our patients and then engage them into a conversation that sometimes less is more. The lack of an antibiotic prescription does not equate to lack of care. Indeed, it often reflects better care.

Jesse T. Jacob, MD, is an associate professor of medicine at Emory University School of Medicine. Disclosure: Jacob reports no relevant financial disclosures.