Hostname: page-component-7b7567c8c-kwt9d Total loading time: 0 Render date: 2026-11-14T17:04:51.745Z Has data issue: false hasContentIssue false
  • English
  • Français

A multicenter analysis of inpatient antibiotic use during the 2015–2019 influenza seasons in the United States: Untapped opportunities for antimicrobial stewardship

Published online by Cambridge University Press:  11 August 2022

Kalvin C. YuOpen the ORCID record for Kalvin C. Yu[Opens in a new window],
Vikas GuptaOpen the ORCID record for Vikas Gupta[Opens in a new window],
Heidi Kabler,
Janet A. Watts and
Amine Amiche
Kalvin C. Yu*
Affiliation:
Becton, Dickinson and Company, Franklin Lakes, New Jersey, United States
Vikas Gupta
Affiliation:
Becton, Dickinson and Company, Franklin Lakes, New Jersey, United States
Heidi Kabler
Affiliation:
Sanofi, Swiftwater, Pennsylvania, United States
Janet A. Watts
Affiliation:
Becton, Dickinson and Company, Franklin Lakes, New Jersey, United States
Amine Amiche
Affiliation:
Sanofi, Dubai, United Arab Emirates
*
Author for correspondence: Kalvin C. Yu, MD, FIDSA, VP, Medical & Scientific Affairs, Becton, Dickinson and Company, 1 Becton Drive, Franklin Lakes, NJ 07417. E-mail: Kalvin.Yu@bd.com

Abstract

Outpatient antibiotic use increases during winter months, but information on temporal changes in inpatient antibiotic use in US hospitals is limited. The use of certain inpatient antibiotics, including extended-spectrum cephalosporins, macrolides, and tetracyclines, was strongly associated with influenza activity during the 2015–2019 viral respiratory seasons.

Information

Type
Concise Communication
Information
Antimicrobial Stewardship & Healthcare Epidemiology , Volume 2 , Issue 1 , 2022, e140
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction in any medium, provided the original article is properly cited.
Copyright
© The Author(s), 2022. Published by Cambridge University Press on behalf of The Society for Healthcare Epidemiology of America

Bacterial coinfection is common during influenza infections and is associated with increased disease severity in patients with community-acquired pneumonia (CAP). Reference Klein, Monteforte and Gupta1,Reference Lim, Kweon, Kim, Kim and Lee2 Many patients with influenza-related symptoms, particularly respiratory manifestations, are empirically treated with antibiotics. Reference Suda, Hicks, Roberts, Hunkler and Taylor3 Although antibiotic treatment is appropriate in some patients, a recent study found that antibiotic prescribing was unsupported in 79.5% of hospitalized patients with CAP in the United States. Reference Magill, O’Leary and Ray4

Previous studies have documented increased outpatient antibiotic use during the winter months in which influenza and other respiratory viruses are often prevalent, Reference Suda, Hicks, Roberts, Hunkler and Taylor3 but recent evaluations of inpatient antibiotic use (IAU) during influenza season are lacking. IAU patterns can be valuable in informing antimicrobial stewardship policies and strategies. We evaluated changes in IAU in hospitalized adults in the United States over time and potential relationships between IAU and community influenza rates based on data in the BD Insights Research Database.

Methods

Study design

We conducted a retrospective, ecological analysis to evaluate possible associations between influenza rates and IAU using real-world data collected between July 1, 2015 and December 31, 2019 from patients ≥18 years old at US acute-care facilities in the BD Insights Research Database (Franklin Lakes, NJ), which includes small and large hospitals with geographical representation across the United States. Reference Gupta, Yu, Kabler, Watts and Amiche5 Geographic localization was based on US Department of Health & Human Services (HHS) regions. Outcome studies using this retrospective, deidentified data set were approved and informed consent was waived by the New England Institutional Review Board (Wellesley, MA).

The outcome of interest was IAU, calculated as days of therapy (DOT) per 1,000 patient days present for commonly prescribed oral or IV antibiotics, including extended-spectrum cephalosporins (ESCs), β-lactam inhibitor combinations (BLICs), fluoroquinolones (FQs), carbapenems, specified anti–methicillin-resistant Staphylococcus aureus (MRSA) drugs (glycopeptides, lipoglycopeptides, and oxazolidinones), lipopeptides, macrolides, tetracyclines, and “other” antibiotics (clindamycin, metronidazole, and aminoglycosides) (see Supplementary Table 1 for specific agents).

Influenza polymerase chain reaction (PCR) and antigen laboratory data were used to determine the influenza positivity rate per 100 tests at inpatient and ambulatory healthcare sites associated with the facility testing laboratory. Influenza testing data therefore represented community influenza rates and were not specific to the inpatient setting. Although other respiratory viruses also display seasonal variations, influenza accounts for most adult cases Reference Choe, Smit and Mermel6 and was therefore chosen as the focus of respiratory virus evaluations.

Statistical analysis

Influenza rates and IAU were calculated for each quarter-year (Q3 2015–Q4 2019) to best encapsulate the concept of a “season.” The sensitivity analyses demonstrated that quarter-years accurately capture seasonal fluctuations in influenza rates. The graphical representations were based on monthly levels to best highlight these seasonal trends.

We used generalized estimated equation models to estimate IAU trends over time and seasonally and the association between IAU and influenza rates. The time-series data were viewed as repeated measures and hospital data were modeled as random effects. Multivariable analyses were performed by antibiotic class while controlling for hospital-level variables (region, bed size, and other characteristics). Beta coefficients indicated the association between variables of interest and were used to evaluate increased IAU for each unit increase in the independent variable; the direction of the association is indicated by positive and negative values. All statistical analyses were conducted using R version 4.0.3 software (R Foundation for Statistical Computing, Vienna, Austria) and the R geepack package. P < .05 was considered statistically significant.

Results

Of 239 acute-care facilities contributing study data, ∼40% (40.6%) were in Health and Human Services (HHS) region 4 (South and Southeast) and 71.6% were in urban locations (Supplementary Table 2). From 2015 to 2019, ESCs, specified anti-MRSA drugs, and BLICs had the highest IAU rates (average of 125, 95, and 94 DOT per 1,000 days present, respectively) (Supplementary Table 3). Data on influenza rates (percentage of positive tests) tracked well with influenza data from the Centers for Disease Control and Prevention. Reference Gupta, Yu, Kabler, Watts and Amiche5,7

Trends in IAU

Unadjusted bivariate results showed ascending trends over time (2015–2019) for IAU for ESCs, BLICs, macrolides, and tetracyclines (Supplementary Table 4). In contrast, descending trends were observed in the IAUs of FQs, “other” antibiotics, specified anti-MRSA drugs, carbapenems, and lipopeptides. Seasonal changes in IAU were significant for BLICs, macrolides, tetracyclines, specified anti-MRSA drugs, and lipopeptides.

Association between IAU and influenza seasons

Visually, we observed strong seasonality that matched the influenza season for macrolides, ESCs, FQs, and tetracyclines IAU (Fig. 1). Influenza rates were significantly associated with IAU for macrolides, ESCs, FQs, tetracyclines, and specified anti-MRSA drugs in bivariate analyses (Table 1). In multivariate analyses, associations with influenza rates were retained for all antibiotics with significant results in bivariate analyses, and a significant association between influenza rates and lipopeptide IAU was revealed (Table 1). We detected no associations between influenza rates and IAU for BLIC, carbapenems, or “other” antibiotics.

Fig. 1.

Trends in IAU and influenza rates from 2015 to 2019. P < .05 for all associations in multivariate analyses. IAU is expressed in DOT per 1,000 days present (95% CI) and influenza rate is expressed as positive tests per 100. The IAU scale for tetracyclines and lipopeptides has been reduced relative to the other antibiotics to better display variations in these drugs with lower usage. Note. BLIC, β-lactam inhibitor combinations; CI, confidence interval; DOT, days of therapy; ESC, extended-spectrum cephalosporins; FQ, fluoroquinolones; IAU, inpatient antibiotic use; MRSA, methicillin-resistant Staphylococcus aureus.

Table 1.

Association Between Influenza Rate and Inpatient Antibiotic Use (DOT per 1,000 Patient Days Present)a

Note. BLIC, β-lactam inhibitor combinations; CI, confidence interval; DOT, days of therapy; ESCs, extended-spectrum cephalosporins; FQs, fluoroquinolones; MRSA, methicillin-resistant Staphylococcus aureus.

a Adjusted for hospital geographic region, bed size, urban or rural setting, and teaching or nonteaching status.

Discussion

Our findings demonstrate clear seasonality of IAU for common antibiotic classes that follows trends in influenza rates. The antibiotic classes showing this association were those often prescribed and recommended for CAP, including ESCs, macrolides, FQs, and tetracyclines. An association with IAU for specified anti-MRSA drugs was also observed. Although in many cases antibiotic use during influenza season may be directed at secondary bacterial infections, inappropriate antibiotic use in hospitalized patients with CAP is common. Reference Magill, O’Leary and Ray4

We recently reported that antibiotic resistance also demonstrated seasonal trends and was associated with influenza rates. Reference Gupta, Yu, Kabler, Watts and Amiche5 Our observations on seasonal trends in IAU and antibiotic resistance may highlight an opportunity for improved antimicrobial stewardship efforts during the influenza and viral respiratory season and support the expansion of vaccination programs, a key component of the World Health Organization Action framework on leveraging vaccines to reduce antibiotic use and prevent antibiotic resistance. 8

In addition to documenting the seasonality of IAU for common antibiotics, our study provides data on trends over time in antibiotic use that update an earlier (2006–2012) study. Reference Baggs, Fridkin, Pollack, Srinivasan and Jernigan9 The IAU of ESCs, BLICs, macrolides, and tetracyclines continued to increase during the years spanned by our study (2015–2019), and the IAU for FQs decreased. We further observed a decreasing trend in IAU for specified anti-MRSA drugs and carbapenems, a reversal of the increasing trends for these antibiotic classes reported in the earlier study. Reference Baggs, Fridkin, Pollack, Srinivasan and Jernigan9

Notably, infections with other respiratory viruses, including respiratory syncytial virus and adenovirus, also show seasonal differences that tend to peak around the same time as influenza. Reference Choe, Smit and Mermel6 However, these viruses primarily affect children under 5 years of age, Reference Choe, Smit and Mermel6 a population not included in this study.

The focus of this ecological study was to evaluate trends in IAU; it was not designed to explore specific infections being treated, clinical outcomes, or appropriate or adequate antimicrobial use. Our study did not include data after 2019 and therefore does not reflect significant changes in antimicrobial treatment patterns in response to severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2). Reference Puzniak, Finelli and Yu10 It is too soon to know whether SARS-CoV-2 will also demonstrate an influenza-like seasonality, which could further impact IAU trends.

In conclusion, IAU is significantly associated with influenza rates when controlling for geographic region and hospital-level factors. Information on IAU influenced by influenza activity may be useful in guiding infection prevention and control measures, including rapid diagnostic testing and vaccination campaigns for influenza and other respiratory viruses, and in determining policies, strategies, and programs for antimicrobial stewardship efforts to reduce inappropriate IAU during influenza season.

Supplementary material

To view supplementary material for this article, please visit https://doi.org/10.1017/ash.2022.265

Acknowledgments

The study conception, design, and analysis plan were codeveloped by researchers from Sanofi and from Becton Dickinson. We thank Sharon L. Cross, PhD, Fusion MD Medical Science Network (Montreal, Canada), for providing manuscript support with funding from Becton Dickinson.

Financial support

This study was supported by a grant from Sanofi to Becton Dickinson. Medical writing was supported by Becton Dickinson.

Conflicts of interest

K.C.Y., V.G., and J.A.W. are employees of Becton, Dickinson & Company (BD), which was contracted by Sanofi to conduct the study. K.C.Y. and V.G. also own stock in BD. H.K. and A.A. are employees of Sanofi and may hold shares and/or stock options in the company.

Footnotes

PREVIOUS PRESENTATION: Part of this research was presented at IDWeek 2021 on September 29, 2021, held virtually.

References

Klein, EY, Monteforte, B, Gupta, A, et al. The frequency of influenza and bacterial coinfection: a systematic review and meta-analysis. Influenza Other Respir Viruses 2016;10:394–403.CrossRefGoogle ScholarPubMed
Lim, YK, Kweon, OJ, Kim, HR, Kim, TH, Lee, MK. Impact of bacterial and viral coinfection in community-acquired pneumonia in adults. Diagn Microbiol Infect Dis 2019;94:50–54.CrossRefGoogle ScholarPubMed
Suda, KJ, Hicks, LA, Roberts, RM, Hunkler, RJ, Taylor, TH. Trends and seasonal variation in outpatient antibiotic prescription rates in the United States, 2006 to 2010. Antimicrob Agents Chemother 2014;58:2763–2766.CrossRefGoogle ScholarPubMed
Magill, SS, O’Leary, E, Ray, SM, et al. Assessment of the appropriateness of antimicrobial use in US hospitals. JAMA Netw Open 2021;4:e212007.CrossRefGoogle ScholarPubMed
Gupta, V, Yu, KC, Kabler, H, Watts, JA, Amiche, A. Antibiotic resistance patterns and association with influenza season in the United States: a multicenter evaluation reveals surprising associations between influenza season and resistance in gram-negative pathogens. Open Forum Infect Dis 2022;9:ofac039.CrossRefGoogle ScholarPubMed
Choe, YJ, Smit, MA, Mermel, LA. Comparison of common respiratory virus peak incidence among varying age groups in Rhode Island, 2012–2016. JAMA Netw Open 2020;3:e207041.CrossRefGoogle ScholarPubMed
Weekly US influenza surveillance report. Centers for Disease Control and Prevention website. https://www.cdc.gov/flu/weekly/index.htm. Accessed February 25, 2022.Google Scholar
Leveraging vaccines to reduce antibiotic use and prevent antimicrobial resistance. An action framework and annexe to immunization agenda 2030. World Health Organization website. https://www.who.int/publications/m/item/leveraging-vaccines-to-reduce-antibiotic-use-and-prevent-antimicrobial-resistance. Published 2021. Accessed October 5, 2021.Google Scholar
Baggs, J, Fridkin, S, Pollack, LA, Srinivasan, A, Jernigan, JA. Estimating national trends in inpatient antibiotic use among US hospitals from 2006 to 2012. JAMA Intern Med 2016;176:1639–1648.CrossRefGoogle ScholarPubMed
Puzniak, L, Finelli, L, Yu, KC, et al. A multicenter analysis of the clinical microbiology and antimicrobial usage in hospitalized patients in the US with or without COVID-19. BMC Infect Dis 2021;21:227.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1.Trends in IAU and influenza rates from 2015 to 2019. P < .05 for all associations in multivariate analyses. IAU is expressed in DOT per 1,000 days present (95% CI) and influenza rate is expressed as positive tests per 100. The IAU scale for tetracyclines and lipopeptides has been reduced relative to the other antibiotics to better display variations in these drugs with lower usage. Note. BLIC, β-lactam inhibitor combinations; CI, confidence interval; DOT, days of therapy; ESC, extended-spectrum cephalosporins; FQ, fluoroquinolones; IAU, inpatient antibiotic use; MRSA, methicillin-resistant Staphylococcus aureus.

Figure 1

Table 1.Association Between Influenza Rate and Inpatient Antibiotic Use (DOT per 1,000 Patient Days Present)a

Supplementary material: File

Yu et al. supplementary material

Tables S1-S4

Download Yu et al. supplementary material(File)
File 23.3 KB