BACKGROUND: Oseltamivir prescribing among pediatric inpatients with influenza varied from 2% to 48% prior to the 2009 H1N1 pandemic. After the pandemic, prescribing guidelines were expanded, and studies reported benefits for hospitalized children. Post-pandemic prescribing practices among children are unclear.
OBJECTIVES: To report the rate of oseltamivir use and to identify factors associated with its use among inpatients with confirmed influenza infection from 2010 to 2014 at a tertiary children’s hospital.
METHODS: We conducted a retrospective cohort study of inpatients with polymerase chain reaction–confirmed influenza from December 2010 to April 2014 at Children’s Hospital Colorado. The primary outcome was oseltamivir use. Variables regarding demographics, underlying medical conditions, diagnoses, and hospital course were also explored. Univariate and multivariate logistic regression analyses were performed.
RESULTS: Among 395 inpatients with influenza, 323 (82%) received oseltamivir. In univariate analyses, oseltamivir use was associated with admission within 48 hours of symptom onset (89% vs 77%), ICU admission (88% vs 79%), longer length of stay (90% for >6 days vs 77% for ≤2 days), and influenza A H1N1 infection (P < .05 for all). In multivariate logistic regression analysis, longer length of stay, illness during the 2013–2014 season, and admission within 48 hours of symptom onset were associated with higher odds of oseltamivir use.
CONCLUSIONS: Oseltamivir use for children with influenza in the postpandemic era is increasing at our institution, aligning with official recommendations and reported benefits. We report highest use for patients in the 2013–2014 season, those who present early in their illness, and those requiring a prolonged hospital stay.
Oseltamivir is an oral antiviral medication approved by the US Food and Drug Administration (FDA) for the treatment of influenza infection and the prevention of its complications. The American Academy of Pediatrics (AAP) provides updated guidelines annually regarding the appropriate use of oseltamivir to treat influenza infection in children. Before the 2009 influenza A H1N1 pandemic, oseltamivir was FDA-approved and AAP-endorsed for the treatment of acute influenza infection in patients older than 1 year of age with symptom onset of <48 hours.1,2
During this prepandemic era, several studies estimated oseltamivir prescribing rates among hospitalized children with influenza infection, with a high proportion of these patients being critically ill. One single-institution retrospective study from 2000 to 2005 reported a prescribing rate of 16.1% among 305 eligible inpatients with laboratory-confirmed influenza who met the initial FDA prescribing criteria.3 Another review of laboratory-confirmed influenza hospitalizations among children in 9 states from 2003 to 2004 reported oseltamivir treatment rates from 3% to 34%, depending on geographic location.4 Other studies of inpatients with influenza from 2003 to 2009 suggested that as few as 2% of pediatric inpatients received oseltamivir5 or that prescribing increased to 48% by 2008.6 Thus, there is significant variation in reported prescribing in the prepandemic era.
During the 2009 pandemic year, the FDA authorized emergency use of oseltamivir for the treatment of children <1 year of age7,8 with published pandemic prescribing rates as high as 77% to 90%.9,10 After 2009, the FDA expanded prescribing guidelines to include neonates,11 and a growing body of literature emerged reporting benefits of oseltamivir for hospitalized children.9,12–16 Conversely, literature also emerged contesting the efficacy of oseltamivir, failing to find benefit in comprehensive meta-analyses of published and unpublished work,17 raising concerns of reporting bias in industry-funded oseltamivir trials and potential methodological flaws.18,19 Thus, in the context of evolving prescribing guidelines and conflicting bodies of literature, oseltamivir prescribing patterns have likely changed. Yet, to our knowledge, a long-term study of oseltamivir prescribing among inpatient children with influenza infection in the postpandemic era has not yet been reported.
In addition, exploring patterns of oseltamivir use can provide insight into the extent to which physicians follow new recommendations regarding influenza treatment, contribute to the understanding of oseltamivir use in hospitalized children, and identify whether certain patient-related factors, such as sociodemographic characteristics or risk groups, influence prescribing.
The primary objective of this study was to report oseltamivir use among pediatric inpatients with laboratory-confirmed influenza infection at a tertiary children’s hospital during the postpandemic period (2010–2014). Our secondary objective was to identify factors associated with oseltamivir use.
Study Setting and Population
The study was performed at Children’s Hospital Colorado (CHCO), an academic tertiary care center serving the Denver metro area (population of ∼2.5 million people), greater Colorado, and surrounding states. In addition to the main campus (∼500 beds), CHCO provides inpatient care at 6 satellite locations, with an additional 135 beds, with ∼18 000 inpatient admissions per year. To create the study cohort, a computer database queried electronic medical records (EMR) hosted by EPIC (Verona, WI) to identify children hospitalized with laboratory-confirmed influenza infection. Patients were included in this study if they were hospitalized at CHCO (Aurora, CO) or an affiliated site with confirmed influenza infection during the 4 postpandemic seasons starting December 1, 2010, and ending April 12, 2014. Repeat hospitalizations were excluded if hospitalization followed a previous stay within 30 days for confirmed influenza infection. This study was approved by the Colorado Multiple Institutional Review Board.
This retrospective cohort study used data regarding demographics, symptoms, high-risk medical conditions,20 admission diagnoses, viral subtype, oseltamivir treatment, and hospital course abstracted from the EMR. Data were managed with REDCap (Research Electronic Data Capture) tools hosted at the University of Colorado.21
Respiratory specimens were tested by either Direct Fluorescent Antibody, influenza A and B polymerase chain reaction (Genexpert, Sunnyvale, CA), xTag Respiratory Virus Panel (Luminex Molecular Diagnostics, Toronto, Canada), or Film Array (Salt Lake City, UT). Influenza subtypes included H1N1, H3, A (not otherwise specified), and B.
Oseltamivir use was the primary outcome and was defined as EMR-recorded receipt of oseltamivir during hospitalization. An underlying medical condition was defined as any condition considered high risk for influenza related complications and included underlying chronic pulmonary conditions (including asthma), hemodynamically significant cardiac disease, immunosuppressive disorders or therapy, chronic renal dysfunction, metabolic disease, neurologic disorders compromising respiratory function, and long-term aspirin therapy.20 Hypoxia was defined by use of supplemental oxygen during hospitalization. Immunosuppression included patients with any primary immunologic dysfunction, disorders of malignancy, transplantation, HIV, or current receipt of immunosuppressive agents noted in the hospital encounter. PICU admission was considered at any time during the hospitalization. Ventilator dependence was defined as intubation during hospitalization.
Descriptive statistics were performed to determine the proportion of patients receiving oseltamivir throughout the study, by influenza season, and by age group. Univariate analyses identified individual factors predictive of oseltamivir use with χ2 analyses or Fisher’s exact test for proportions and Wilcoxon rank-sum tests for non-normally distributed variables. To select independent predictors of oseltamivir use, multiple logistic regression analysis with a backward elimination selection procedure was applied. Parameters trending toward statistical significance (P < .25) by univariate analysis were included, and predictors with the highest P value were removed from the model 1 at a time until achievement of the best model fit. P values were interpreted at a .05 level of significance. Data were analyzed using SPSS 22 (Chicago, IL) and SAS 9.3 (Cary, NC).
From December 1, 2010, to April 12, 2014, there were 389 patients representing 401 hospitalizations, with 395 inpatients meeting inclusion criteria, with a median age of 48 months (interquartile range [IQR] 16–98 months); 323 (82%) received oseltamivir. Patients experienced a median duration of 3 days of preceding symptoms (IQR 1–6 days) before admission. Figure 1 displays the proportion of patients receiving oseltamivir by study year and location. During the 2011–2012 influenza season, a statistically significant difference in prescribing existed between the PICU and wards (76% vs 37%; P = .009); there were no differences for other prescribing years. Figure 2 depicts oseltamivir prescribing by age group during the study period. A trend toward increased prescribing in patients <3 months of age was not statistically significant.
In univariate analyses, oseltamivir use was associated with admission within 48 hours of symptom onset (89% vs 77%), ICU admission (88% vs 79%), longer length of stay (90% for > 6 days vs 77% for ≤2 days), and infection with influenza A H1N1 and was dependent on admission location (P < .05 for all; Tables 1 and 2). In multivariate logistic regression analysis, longer length of stay, illness during the 2013–2014 influenza season, and admission within 48 hours of symptom onset were associated with higher odds of oseltamivir use (Table 3). Changes in oseltamivir treatment recommendations from 2004–2014 are summarized in Table 4.
We report high oseltamivir use among pediatric inpatients for the treatment of influenza in 4 postpandemic influenza seasons at our institution. Oseltamivir use increased during the study period, with highest use during the 2013–2014 season. In adjusted analyses, oseltamivir use was associated with longer length of stay, likely representing patients with more severe illness (specific measures of illness severity were not evaluated) and admission within 48 hours of symptom onset. These findings align with newer recommendations after the pandemic year and with published data indicating greater benefit with earlier treatment.11,16
Our data revealed that oseltamivir prescribing increased overall during our study period, reaching a high of 97% during the 2013–2014 season. While we did observe a decrease in prescribing during the 2011–2012 season, this was notably a less severe influenza season, which may have influenced physician practice. Our prescribing trends differ from other data in the literature which report a decrease in use after the 2009 pandemic season.9,10 However, these data only spanned the 1 season after the 2009 pandemic. Oseltamivir prescribing during the 2013–2014 year was even higher than the pandemic year at our institution, which was reported previously to be 87% among patients with influenza A H1N1 infection.8
The reasons for the increase in prescribing observed in our study are likely multifactorial. First, there have been multiple changes in FDA and AAP recommendations for oseltamivir use among children with influenza infection over the past decade, summarized in Table 4.1,2,8,11,22,23 With 1 exception, each update has liberalized oseltamivir use, such that most hospitalized children of any age with confirmed or suspected influenza infection are eligible to receive antiviral therapy, regardless of underlying conditions. This may explain the lack of association between oseltamivir use and underlying medical conditions or admission diagnoses noted in our study.
Secondly, oseltamivir use also may be increasing as more studies emerge describing the benefits for hospitalized children. Early neuraminidase inhibitor use in hospitalized children has been associated with a decreased length of hospital stay12 and decreased likelihood of ICU admission and need for mechanical ventilation.13,14 Other observational studies have suggested that treatment may improve survival.19,15,16 However, one must take into account the limitations of retrospective, nonrandomized data, which accounts for much of the published work in the postpandemic era, as well as potential bias from industry-sponsored studies.18,19 A recent Cochrane review of >100 clinical studies, including previously unpublished data, reported that oseltamivir therapy did not affect hospitalization and did not reduce serious influenza complications like pneumonia, although the authors reported that methodologic differences between the various studies precluded confidence in the results.17 It will be interesting to note whether these findings will affect oseltamivir prescribing over the subsequent years.
Reasons for high postpandemic oseltamivir use at our institution may include higher illness acuity of patients admitted to a tertiary referral center, institutional guidelines recommending use among hospitalized patients, or a culture of higher antiviral prescribing among physicians and other medical staff. Our univariate analysis demonstrated that a lower proportion of patients admitted to satellite hospitals received oseltamivir, but this is confounded by the lower patient acuity generally seen at these sites. In addition to practice site and performance of rapid testing, physician beliefs about morbidity and mortality benefits have been shown to influence prescribing.24 Factors influencing antiviral prescribing among providers in the postpandemic era warrant further study.
Our multivariate analysis explored patient-related factors that could influence oseltamivir prescribing and showed that patient age was not associated with oseltamivir use in this study. Although there was a trend to increased use over time for children <3 months, these findings were not statistically significant. This may have been due to a small sample size among this group. This increased trend coincides with the FDA expansion of prescribing guidelines to include children as young as 2 weeks of age in December 2012, and the 2013–2014 AAP recommendations for control and treatment of influenza note that oseltamivir can be used to treat influenza in both term and preterm infants from birth.22 Given the impact of influenza on young children,25 and lack of protection available from immunization for children <6 months of age, the age expansion has important implications regarding the potential to decrease morbidity and mortality in this age group.
There are limitations to this study. The retrospective nature of the study may have introduced misclassification bias as the investigators reviewed other providers’ charting to abstract clinically relevant information. In addition to a small sample size, the higher proportion of patients receiving oseltamivir compared with those who did not receive this agent may have introduced further bias. This study did not assess clinical decision-making processes regarding treatment with oseltamivir. Given the variation in oseltamivir prescribing noted by geographic area in previous work,4 it is difficult to assess the generalizability of our findings to other institutions. Finally, dosing, timing, and duration of therapy were not measured once the patient was discharged from hospital, or before hospitalization.
In conclusion, our study demonstrated that oseltamivir use is increasing in a postpandemic era at our institution, with >80% of hospitalized patients receiving therapy in a 4-year period. These increasing prescribing trends align with pediatric governing body recommendations and are of interest in the setting of conflicting evidence in the literature regarding its morbidity and mortality benefits.
Ultimately, our findings provide insight into oseltamivir prescribing practices for hospitalized children in both the ICU and non-ICU settings, as well as the impact of national guidelines in influencing physician practice changes. Our findings suggest that among a multitude of factors that influence prescribing, illness severity and duration of symptoms are leading considerations. Additional studies are needed to confirm these findings in a broader context, to explore factors influencing individual provider prescribing practices, and to further corroborate reported benefits of oseltamivir therapy in hospitalized children, particularly in the non-ICU setting.
We thank Kelly Pearce, BA, from the Department of Epidemiology for the generation of the initial data set.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: This work was supported in part by a grant from the National Institutes of Health/National Center for Research Resources (Colorado Clinical and Translational Sciences Institute grant UL1 TR001082). Funded by the National Institutes of Health (NIH).
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- ↵Tamiflu (oseltamivir phosphate) capsules and for oral suspension. NDA 21-087/S-016.î US Food and Drug Administration. December, 2007. Available at: http://www.accessdata.fda.gov/drugsatfda_docs/nda/2004/021087_S016_021246s010TamifluTOC.cfm. (Accessed December 12 2014)
- ↵American Academy of Pediatrics Committee on Infectious Diseases. Antiviral therapy and prophylaxis for influenza in children. Pediatrics. 2007;119(4):852–860
- Poehling KA,
- Edwards KM,
- Griffin MR,
- et al
- Louie JK,
- Yang S,
- Samuel MC,
- Uyeki TM,
- Schechter R
- Garg S,
- Chaves SS,
- Pérez A,
- et al
- Pickering LK,
- Baker CJ,
- Kimberlin DW,
- Long SS
- ↵American Academy of Pediatrics, Committee on infectious diseases. Recommendations for prevention and control of influenza in children, 2013–2014. Pediatrics. 2013;132(4). Available at: www.pediatrics.org/cgi/content/full/132/4/e1089
- ↵American Academy of Pediatrics, Committee on Infectious Diseases. Policy statement—recommendations for prevention and control of influenza in children, 2010–2011. Pediatrics. 2010;126(4):816–826
- Ebell MH
- Ebell MH,
- Call M,
- Shinholser J
- Rothberg MB,
- Bonner AB,
- Rajab MH,
- Kim HS,
- Stechenberg BW,
- Rose DN
- Copyright © 2016 by the American Academy of Pediatrics