Want to change physicians' use of antibiotics? Try educating patients
By Phyllis Maguire
In trying to change what analysts call the "culture" of inappropriate antibiotic use, experts emphasize one key lesson: You must work with both patients and physicians.
"Patients and clinicians are part of one dynamic," said Ralph Gonzales, ACP-ASIM Member. "You cannot untangle the two."
Dr. Gonzales has spent the last six years studying the impact of educational interventions on antibiotic prescribing, first as a faculty member of the University of Colorado Health Sciences Center in Denver and now at the University of California, San Francisco. He has helped identify factors that bring about change in physician prescribing patterns and patient expectations—and those that do not.
He was principal investigator of a late-1990s study sponsored by the Robert Wood Johnson Foundation that examined antibiotic prescribing at four Kaiser Permanente primary care practices in Denver. The study followed more than 2,000 adult patients diagnosed with uncomplicated acute bronchitis.
In the study, physicians at a full-intervention site received academic details about antibiotic prescribing, site-specific prescribing profiles, and educational brochures and posters for their offices. The site's patients were sent educational materials at home, including kitchen magnets and brochures on appropriate antibiotic use and antibiotic resistance.
At a limited intervention site, on the other hand, only office-based materials were featured. At two control sites, usual care—with no physician or patient education—was provided.
The results? At the full intervention site, the number of antibiotic prescriptions written for uncomplicated bronchitis fell from 74% of all diagnosed patients to 48%. Neither the limited intervention nor the control sites showed any significant decrease.
Physicians at the full intervention site wrote no additional nonantibiotic prescriptions (for inhaled bronchodilators or codeine, for instance). They also reported no increase in the number of return visits or of pneumonia cases compared to the other sites. (The study was published in April 28, 1999, issue of the Journal of the American Medical Association, which is online at http://jama.ama-assn.org/issues/v281n16/rfull/joc81473.html.)
Finally, there was no change in patient satisfaction at the site with reduced antibiotic use. (See "Impact of Reducing Antibiotic Prescribing for Acute Bronchitis on Patient Satisfaction" in the May/June 2001, issue of Effective Clinical Practice at http://www.acponline.org/journals/ecp/mayjun01/gonzales.htm.)
This year, Dr. Gonzales is principal investigator on a similar but greatly expanded Colorado study on reducing antibiotic use. He spoke to ACP-ASIM Observer about the new study and about the lessons learned from researching antibiotic prescribing patterns and patient demand.
Q: You concluded from your Kaiser study that both patient and clinician interventions are needed to reduce antibiotic use. Why is a combination so important?
A: If you don't educate patients, physicians have to bear the full weight of delivering the message of antibiotic resistance to patients. That's when physicians see themselves becoming barriers to something patients want. It's a position they don't want to be in.
When doctors know that patients are getting materials at home and then seeing similar materials in the office, the physician becomes more of a facilitator, reinforcing a message that the patient has already seen multiple times. When you educate patients, you're less likely to antagonize them when you don't give them a prescription for antibiotics.
Q: Were you surprised that patient satisfaction at the full intervention site was just as high as at the control sites?
A: We weren't, but the doctors were. Part of that reaction comes from physician perceptions: The one patient who leaves your office angry because you didn't prescribe antibiotics can overshadow the hundreds of patients who are much more comfortable not getting drugs they don't need and that may compromise their care down the line.
The study showed—as many others have—that clinician characteristics and the manner in which patients are treated are the biggest indicators of patient satisfaction, not the type of treatment patients receive.
Q: How important was staff education?
A: The message has to be uniform across all provider levels. We provided training to the physicians and nurses at the intervention sites, but initially we missed one key portal: the triage nurse.
Kaiser has a call center where patients talk to an advice nurse before they get an appointment. We discovered that the call-center script told patients with bronchitis to see a doctor for antibiotics. Patients end up getting angry when they get different messages from their health care system, and often impart this frustration to their physician.
Q: So educating physicians alone isn't enough?
A: No, and that point is being borne out by a study of 700 physicians being conducted by the Colorado Medical Society in partnership with us. As part of the study, physicians received individual prescribing profiles for bronchitis and pharyngitis that typically showed high levels of antibiotic use. While prescribing rates dipped at first in response, they returned to baseline within a year, so the reductions weren't sustained. We learned, as others have, that clinician education alone is not sufficient.
In that study, we also learned that it is possible to get all the major managed care organizations at the table. In Colorado, they agreed to give us data that we aggregate to create a single profile, so the profiles actually reflect the majority of a physician's patients.
When doctors see patients from multiple health plans, they tend not to believe one plan's data because they represent just a fraction of their patients. But with aggregate profiles—and chart reviews to validate the data—they are more willing to trust the information.
Q: How is the new study you're working on different from your previous efforts?
A: First, it's much bigger. We will provide patient and office education to 10 different practices, and most primary care physicians in Colorado will also get clinician educational materials.
We're also looking at many different populations instead of just those in a group-model HMO. We're seeing other kinds of insured patients as well as underserved populations to help us find ways to deal with language and literacy barriers in patient education. And through a parallel grant from the Centers for Medicare and Medicaid Services, we'll look at how to educate seniors.
We'll also experiment with different types of patient education. In the first year, we'll send information to patients' houses, like we did with the Kaiser study. But next year, we'll use a mass media campaign on reducing antibiotic use, and see what results we get from that.
Q: Why did you start studying ways to change prescribing patterns?
A: After dealing with the punitive nature of managed care, I wanted to develop innovative ways to improve physician behavior. There are many ways to improve physician practice patterns that don't involve tying doctors' hands, such as system changes that improve shared decision-making and patient education. I've found that every time we go to a practice and say 'We'd like to educate your patients,' we're welcomed with open arms.
Internist Archives Quick Links
What will you learn from your Annals Virtual Patient?
Annals Virtual Patients is a unique patient care simulator that mirrors real patient care decisions and consequences. CME Credit and MOC Points are available. Start off with a FREE sample case. Start your journey now.
Internal Medicine Meeting 2015 Live Simulcast!
Unable to attend the meeting this year? On Saturday, May 2, seven sessions will be streamed live from the meeting. Register for the simulcast and earn CME credit after watching each session. Watch it live or download for later viewing.