[Skip to Content]
[Skip to Content Landing]
Research Letter
March 2016

Resource Burden During the 2014 Enterovirus D68 Respiratory Disease Outbreak at Children’s Hospital Colorado: An Unexpected Strain

Author Affiliations
  • 1Section of Hospital Medicine, Children’s Hospital Colorado, University of Colorado, Aurora
  • 2Department of Pediatrics, Children’s Hospital Colorado, University of Colorado, Aurora
  • 3Section of Infectious Diseases, Children’s Hospital Colorado, University of Colorado, Aurora
  • 4Section of Pulmonary Medicine, Children’s Hospital Colorado, University of Colorado, Aurora
  • 5Department of Epidemiology, Children’s Hospital Colorado, University of Colorado, Aurora
JAMA Pediatr. 2016;170(3):294-297. doi:10.1001/jamapediatrics.2015.3879

Enterovirus D68 (EV-D68) is a unique enterovirus that shares biological properties with human rhinoviruses.1 It primarily causes respiratory disease, particularly in children with asthma. Although rarely reported from 1970 to 2005, small clusters of EV-D68 respiratory disease have been increasingly reported since 2008.2,3 From August to November 2014, an outbreak of EV-D68 respiratory disease occurred throughout the United States, with 1153 microbiologically confirmed infections in 49 states.4,5 However, owing to marked underascertainment of cases, the true magnitude and impact of this outbreak are difficult to estimate. The objective of this study is to characterize and quantify the resource burden at a tertiary care children’s hospital during the 2014 EV-D68 respiratory disease outbreak.

Methods

Children’s Hospital Colorado (CHCO) Anschutz Medical Campus is a tertiary care children’s hospital in Aurora, Colorado, operating 444 inpatient beds including 32 pediatric intensive care unit beds, with approximately 17 000 inpatient admissions and 70 000 emergency department visits in 2014. Based on epidemiologic data from respiratory virus testing at CHCO, the EV-D68 outbreak period was defined as August 1 through September 30, 2014 (Figure 1).

Figure 1.
Respiratory Virus Testing at Children’s Hospital Colorado in 2014
Respiratory Virus Testing at Children’s Hospital Colorado in 2014

The numbers of respiratory specimens testing positive for viral pathogens on multiplex respiratory pathogen panel polymerase chain reaction testing (Biofire Diagnostics) at Children’s Hospital Colorado by week in 2014 are depicted by colored lines corresponding to the left axis. Of note, this technology detects rhinoviruses and enteroviruses but cannot differentiate the two. For pediatric intensive care unit (PICU) patients, the numbers of rhinovirus/enterovirus-positive respiratory specimens typed as enterovirus D68 by real-time reverse transcription–polymerase chain reaction at the Centers for Disease Control and Prevention Polio and Picornavirus Laboratory by week are depicted by bars corresponding to the right axis. During this period, 564 of 971 children (58%) tested by multiplex respiratory pathogen panel polymerase chain reaction testing at Children’s Hospital Colorado had results positive for rhinoviruses/enteroviruses. Among 156 rhinovirus/enterovirus-positive specimens from PICU patients tested with real-time enterovirus D68 reverse transcription–polymerase chain reaction, 98 (63%) were positive during this period.

aThe defined outbreak period from August 1 to September 30, 2014.

This is an observational retrospective time series study comparing observed resource utilization during the defined 2014 EV-D68 outbreak period vs expected values forecasted from historical data. Monthly resource utilization data were collected through the CHCO infectious disease data warehouse. Respiratory patient volumes include children with International Classification of Diseases, Ninth Revision codes within major diagnostic category 4 (diseases and disorders of the respiratory system). Days of therapy (DOT) reflect documented administration of a medication or treatment on a calendar day in the electronic medical record. Direct patient care services delivered by respiratory therapists were converted into units of service (UOS) using Children’s Hospital Association Pediatric Analysis and Comparison Tool criteria. This study was deemed “not human subjects research” by the Colorado Multiple Institutional Review Board, and informed consent was not required owing to the retrospective nature of aggregate data used in the study.

Observed values during the outbreak period (August 1 to September 30, 2014) were compared with expected values. Seasonal autoregressive integrated moving average (SARIMA) time series analysis of historical data (January 1, 2011, to July 31, 2014) was used to forecast expected values, adjusted for seasonal variation and hospital growth.6 Optimal model parameters were selected using the Akaike information criterion. Statistical analysis was conducted in SAS/ETS version 9.4 statistical software (SAS Institute, Inc).

Results

From August 1 through September 30, 2014, respiratory patient volumes (Figure 2A-C) and asthma medication use (Figure 2D-F) at CHCO significantly increased. Respiratory emergency department visits increased by 1185 (+36%; P < .05), respiratory hospital admissions increased by 387 (+80%; P < .05), and respiratory pediatric intensive care unit admissions increased by 96 (+79%; P < .05) more than expected. Albuterol sulfate use increased by 2473 DOT (+86%; P < .05), depleting supplies to a critical shortage. Increased uses of steroids (+822 DOT; +72%; P < .05) and second-line asthma medications (including terbutaline sulfate, magnesium sulfate, aminophylline; +118 DOT; +101%; P < .05) were also noted. Despite the surge in patient volumes, antibiotic use did not increase (−406 DOT; −7%; P > .05).

Figure 2.
Observed vs Expected Resource Utilization at Children’s Hospital Colorado During the 2014 Enterovirus D68 Outbreak
Observed vs Expected Resource Utilization at Children’s Hospital Colorado During the 2014 Enterovirus D68 Outbreak

Expected monthly resource utilization at Children’s Hospital Colorado, forecasted by seasonal autoregressive integrated moving average (SARIMA) time series analysis with 95% CIs compared with actual observed values from January 2011 to November 2014. The shaded area indicates the defined outbreak period from August 1 to September 30, 2014. A-C, Respiratory patient volumes, including emergency department visits (A), hospital admissions (B), and pediatric intensive care unit admissions (C) with International Classification of Diseases, Ninth Revision codes within major diagnostic category 4 (diseases and disorders of the respiratory system). D-F, Rates of medication use, including albuterol sulfate (D), steroids (E), and secondary asthma medications (parenteral terbutaline sulfate, parenteral magnesium sulfate, and parenteral or oral aminophylline) (F). G-I, Rates of respiratory equipment use, including ventilator days (G), noninvasive positive airway pressure ventilation days (H), and intermittent positive airway pressure therapy administrations (I). J-L, Staffing demands, including inhaled medication administrations (J), respiratory therapist units of service (K), and respiratory therapist units of service per hour worked (L).

During the 2-month outbreak period at CHCO, respiratory equipment use (Figure 2G-I) and respiratory therapist demands (Figure 2J-L) increased significantly. There were 170 excess ventilator days (+28%; P < .05) and 1272 excess intermittent positive airway pressure therapy administrations (+30%; P < .05) compared with expected values. Noninvasive positive airway pressure ventilation days increased (+84 DOT; +11%; P > .05), but this was not statistically significant. Respiratory therapists administered 6714 more inhaled medication doses (+54%; P < .05) than expected. Respiratory therapist UOS reached an institutional all-time high, 17 646 UOS (+73%; P < .05) more than expected. For every hour worked, respiratory therapists delivered 1.27 UOS, a 70% increase above expected productivity (P < .05).

Discussion

During the 2014 outbreak of EV-D68 respiratory disease, our hospital experienced significant resource burden as evidenced by increased patient volumes, staffing demands, and use of medications and respiratory equipment. In the absence of widespread EV-D68–specific testing for surveillance, SARIMA modeling of resource utilization provides a powerful alternative method to indirectly estimate the magnitude of this outbreak. This study demonstrates the profound impact EV-D68 had on children’s hospitals, which was grossly underestimated by the number of confirmed infections, and illustrates the utility of resource utilization monitoring in future unexpected outbreaks of emerging pathogens.

Back to top
Article Information

Corresponding Author: Kevin Messacar, MD, Section of Hospital Medicine and Pediatric Infectious Diseases, Children’s Hospital Colorado, B055, 13123 E 16th Ave, Aurora, CO 80045 (kevin.messacar@childrenscolorado.org).

Published Online: January 19, 2016. doi:10.1001/jamapediatrics.2015.3879.

Author Contributions: Drs Messacar and Parker had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Messacar, Hawkins, Baker, Dominguez, Parker.

Acquisition, analysis, or interpretation of data: Messacar, Hawkins, Pearce, Tong, Parker.

Drafting of the manuscript: Messacar, Baker, Pearce, Dominguez.

Critical revision of the manuscript for important intellectual content: Messacar, Hawkins, Tong, Dominguez, Parker.

Statistical analysis: Messacar, Tong.

Administrative, technical, or material support: Messacar, Hawkins, Baker, Parker.

Study supervision: Messacar, Dominguez, Parker.

Conflict of Interest Disclosures: None reported.

Funding/Support: Specimen testing and data analysis were supported by grant UL1 TR001082 to the Colorado Clinical and Translational Sciences Institute from the National Center for Advancing Translational Sciences, National Institutes of Health.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: Suchitra Rao, MD, Department of Pediatrics, Section of Hospital Medicine and Infectious Diseases, Children’s Hospital Colorado, University of Colorado, Aurora, Rakesh Mistry, MD, Department of Pediatrics, Section of Emergency Medicine, Children’s Hospital Colorado, University of Colorado, Aurora, Gina DeMasellis, MD, Department of Pediatrics, Section of Critical Care Medicine, Children’s Hospital Colorado, University of Colorado, Aurora, and Jerrod Milton, MBA, Children’s Hospital Colorado, Aurora, provided help in critically reviewing the manuscript; they received no compensation. We thank the medical, nursing, and other clinical staff members at Children’s Hospital Colorado for their heroic efforts to provide expert care to the affected children during this outbreak.

References
1.
Oberste  MS, Maher  K, Schnurr  D,  et al.  Enterovirus 68 is associated with respiratory illness and shares biological features with both the enteroviruses and the rhinoviruses.  J Gen Virol. 2004;85(pt 9):2577-2584.PubMedGoogle ScholarCrossref
2.
Khetsuriani  N, Lamonte-Fowlkes  A, Oberst  S, Pallansch  MA; Centers for Disease Control and Prevention.  Enterovirus surveillance: United States, 1970-2005.  MMWR Surveill Summ. 2006;55(8):1-20.PubMedGoogle Scholar
3.
Centers for Disease Control and Prevention.  Clusters of acute respiratory illness associated with human enterovirus 68: Asia, Europe, and United States, 2008-2010.  MMWR Morb Mortal Wkly Rep. 2011;60(38):1301-1304.PubMedGoogle Scholar
4.
Midgley  CM, Jackson  MA, Selvarangan  R,  et al.  Severe respiratory illness associated with enterovirus D68: Missouri and Illinois, 2014.  MMWR Morb Mortal Wkly Rep. 2014;63(36):798-799.PubMedGoogle Scholar
5.
Centers for Disease Control and Prevention. Enterovirus D68. http://www.cdc.gov/non-polio-enterovirus/about/EV-D68.html. Accessed June 1, 2015.
6.
Brocklebank  JC, Dickey  DA.  SAS for Forecasting Time Series. 2nd ed. Cary, NC: SAS Institute Inc; 2003.
×