Outpatient Antibiotic Use in the United States: Time to “Get Smarter”

(See the article by Hicks et al, on pages 631–639.) Infections with Streptococcus pneumoniae put a significant burden on the US healthcare system with an estimated 22 000 deaths and 44 5000 hospitalizations in the United States in 2004 [1]. Despite a reduction in the incidence of invasive disease over the last decade due to the introduction of the conjugate vaccine, penicillin-non-susceptible (PNSP) and multidrug-resistant strains remain prevalent, most likely as a consequence of continuing antibiotic selection pressure [2]. In this issue of Clinical Infectious Diseases, Laurie Hicks and colleagues [3] analyze the association between 4 classes of antibiotics commonly prescribed for respiratory tract infections in the outpatient setting and the incidence of nonsusceptible, invasive pneumococcal isolates collected in 7 states participating in the Active Bacterial Core surveillance network between 1996 and 2003. The researchers were able to show a solid link between higher antibiotic prescribing rates (for cephalosporins and macrolides in particular) and a higher proportion of penicillin non-susceptible and multidrug non-susceptible invasive pneumococcal isolates. This finding is in line with several European ecologic studies examining the relationship between outpatient antibiotic use and pneumococcal non-susceptibility in the last 10 years [4–12]. Taken together, these studies convincingly demonstrate that countries or communities with higher levels of outpatient antibiotic use have higher rates of resistance in pneumococci, and that use of a specific class of antibiotics may also affect resistance rates to unrelated classes of antibiotics. Indeed, dynamic models predict that ecologic associations between antibiotics and resistance will be particularly strong when population-level mechanisms of selection are paramount, such as is the case for PNSP [13–16]. A recently published cluster-randomized trial of mass azithromycin treatment for trachoma in Ethiopia illustrated the rapid effect that antibiotic use can have on pneumococcal ecology in communities. In children living in communities exposed to mass azithromycin treatment every 3 months, nasopharyngeal carriage of azithromycin-resistant S. pneumoniae increased from 3.6% at baseline to 46.9% at 12 months, whereas only 9.2% of the children in the untreated communities carried azithromycin-resistant S. pneumoniae at 12 months [17]. Given the ample evidence of the negative ecologic impact of outpatient antibiotic use on resistance—not only with regard to S. pneumoniae but also with regard to Gram-negative pathogens—and given the fact that unnecessary antibiotic use in the primary care setting remains common, we have to ask ourselves where the United States stands relative to other regions of the world with regard to outpatient antibiotic use [18–20]. In their analysis, the authors describe impressive reductions in antibiotic prescriptions of the 4 analyzed classes between 1996 and 2003: 37% for children <5 years of age and 42% for individuals ≥5 years of age. Although at first glance this might seem encouraging, there are several caveats. First, the researchers note that within the macrolide drug class, use of azithromycin, an antibiotic that may disproportionately select for resistance, became increasingly dominant [21–23]. Second, for adults it was not clear how much of the decreasing trend in the 4 drug classes studied was counterbalanced by increased use of fluoroquinolones. Use of this antibiotic class markedly increased between 1995 and 2002, until it became the most commonly prescribed antibiotic class for adults in 2002 [24]. Although fluoroquinolone-resistant pneumococci are still the exception in North America, an association between fluoroquinolone usage and fluoroquinolone resistance in S. pneumoniae has been described in Canada and the United States, and fluoroquinolones have, of course, a well-described impact on resistance in Gram-negative pathogens [25–26]. The period assessed in the current study ends in 2003, raising the question of how antibiotic use has evolved in the United States since then. This brings to light a crucial deficiency of antibiotic stewardship in the outpatient setting in the United States: comprehensive data with regard to outpatient antibiotic use are not easily accessible. For their analysis, the researchers exploited pharmacy-dispensing data collected by IMS Health, a “provider of market intelligence to the pharmaceutical and healthcare industries.” This data source, with a coverage of 70% of all prescriptions in the retail setting, is far more comprehensive and detailed than any of the sources typically used to assess outpatient antibiotic prescribing in the United States such as health plan data or data obtained through the Centers for Disease Control and Prevention’s (CDC) yearly National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS) [27, 28]. However, IMS data are not freely available in most cases. The experience from the European Surveillance of Antimicrobial Consumption (ESAC) program demonstrates that it is possible to collect reliable data on antimicrobial use even across countries with different healthcare systems[29]; as recommended in a recent policy paper by Infectious Disease Society of America (IDSA), similar efforts to collect and publish these data need to be undertaken in the United States [30]. Some more recent IMS antibiotic prescribing data were presented at the 2010 meeting of the IDSA; they show that the nationwide antibiotic prescription rate was 0.86 prescriptions per capita in 2009 [31]. Although this is lower than the prescription rate in high-prescribing European countries such as France (approximately 1.12 prescriptions per capita in 2009; Pierre Chahwakilian, personal communication May 2011), it is much higher than the 0.34 prescriptions per capita reported in the Netherlands in 2001; in addition, some states such as West Virginia, Kentucky, and Tennessee even surpass France at 1.2 prescriptions per capita per year [32]. The amount of antibiotics used is also not the only important dimension. Some studies have reported an increase in broad-spectrum antibiotics for certain conditions in the outpatient setting [33, 34]. And although there has been a promising trend among some physicians to follow guidelines recommending “narrow-spectrum” amoxicillin for otitis media and sinusitis in children, broad-spectrum antibiotic use for these conditions, which often do not require any antibiotics at all, remains widespread [18, 35]. These data illustrate that there is both opportunity and need to improve outpatient antibiotic prescribing in the United States. Although there has been increasing interest in antibiotic stewardship in the hospital setting—indeed, no major infectious disease conference goes without several symposia on this topic—antibiotic stewardship in the outpatient setting has been somewhat neglected, despite the fact that the vast majority of all antibiotics for human use are prescribed in the primary care setting and despite evidence that outpatient antibiotic use can influence resistance rates within facilities [36, 37]. The CDC has promoted judicious antibiotic use since 1995 through the “Get Smart: Know When Antibiotics Work” campaign, but good data linking the campaign to decreased antibiotic use are scarce for the reasons mentioned above [38]. In addition, most statewide campaigns were relatively limited and poorly funded compared with mass-media campaigns in other countries [38, 39]. In 2010, for example, not one of the 5 highest prescribing states in 2009 seems to have received funding in the context of the “Get Smart” campaign [31, 40]. Although campaigns are one approach that might work to improve antibiotic prescribing in certain situations, they should not be the only approach [38]. In addition to improving monitoring of antibiotic prescribing and making these data more accessible, we must also better understand the factors that contribute to the small-area variation observed in prescribing patterns—factors that are unlikely to be simple geographic differences in respiratory pathogens or population genetics and probably include cultural, socioeconomic, and healthcare system-related factors such as physician density and physician remuneration methods [6, 41, 42]. We also must create the correct financial incentives to preserve existing drugs and to develop new drugs [30, 43]. Providing free antibiotics, as is done by some pharmacy chains, is probably not the way to go [44]. It has been suggested that we view antibiotics as a non-renewable resource such as oil [45]. Although humanity’s track record regarding non-renewable resources does not inspire much optimism, it is better late than never to “get smart.” We thank Makoto Jones, Angela Huttner, and Marianne Madsen for their thoughtful comments and manuscript editing assistance. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Department of Veterans Affairs or the University of Utah. Financial support. B. H. was supported by a fellowship grant from Geneva University Hospitals, Switzerland. Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.