The

n e w e ng l a n d j o u r na l

of

m e dic i n e

original article

The Burden of Respiratory Syncytial Virus Infection in Young Children

Caroline Breese Hall, M.D., Geoffrey A. Weinberg, M.D., Marika K. Iwane, Ph.D., M.P.H., Aaron K. Blumkin, M.S., Kathryn M. Edwards, M.D., Mary A. Staat, M.D., M.P.H., Peggy Auinger, M.S., Marie R. Griffin, M.D., M.P.H., Katherine A. Poehling, M.D., M.P.H., Dean Erdman, Dr.P.H., Carlos G. Grijalva, M.D., M.P.H., Yuwei Zhu, M.D., M.S., and Peter Szilagyi, M.D., M.P.H.

A bs t r ac t
Background
From the Departments of Pediatrics (C.B.H., G.A.W., A.K.B., P.A., P.S.) and Medicine (C.B.H.), the University of Rochester School of Medicine and Dentistry, Rochester, NY; the National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta (M.K.I., D.E.); the Departments of Pediatrics (K.M.E., K.A.P.), Preventive Medicine (M.R.G., C.G.G.), and Biostatistics (Y.Z.) all at the Vanderbilt University Medical Center, Nashville; and the Department of Pediatrics, Cincinnati Childrens Hospital Medical Center, University of Cincinnati College of Medicine, Cincinnati (M.A.S.). Address reprint requests to Dr. Hall at the Division of Infectious Diseases, Departments of Pediatrics and Medicine, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Box 689, Rochester, NY 14642, or at [email protected]. N Engl J Med 2009;360:588-98.
Copyright 2009 Massachusetts Medical Society.

The primary role of respiratory syncytial virus (RSV) in causing infant hospitalizations is well recognized, but the total burden of RSV infection among young children remains poorly defined.
Methods

We conducted prospective, population-based surveillance of acute respiratory infections among children under 5 years of age in three U.S. counties. We enrolled hospitalized children from 2000 through 2004 and children presenting as outpatients in emergency departments and pediatric offices from 2002 through 2004. RSV was detected by culture and reverse-transcriptase polymerase chain reaction. Clinical information was obtained from parents and medical records. We calculated population-based rates of hospitalization associated with RSV infection and estimated the rates of RSV-associated outpatient visits.
Results

Among 5067 children enrolled in the study, 919 (18%) had RSV infections. Overall, RSV was associated with 20% of hospitalizations, 18% of emergency department visits, and 15% of office visits for acute respiratory infections from November through April. Average annual hospitalization rates were 17 per 1000 children under 6 months of age and 3 per 1000 children under 5 years of age. Most of the children had no coexisting illnesses. Only prematurity and a young age were independent risk factors for hospitalization. Estimated rates of RSV-associated office visits among children under 5 years of age were three times those in emergency departments. Outpatients had moderately severe RSV-associated illness, but few of the illnesses (3%) were diagnosed as being caused by RSV.
Conclusions

RSV infection is associated with substantial morbidity in U.S. children in both inpatient and outpatient settings. Most children with RSV infection were previously healthy, suggesting that control strategies targeting only high-risk children will have a limited effect on the total disease burden of RSV infection.

588

n engl j med 360;6

nejm.org

february 5, 2009

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

Respir atory Syncytial Virus Infection in Healthy Children

he primary role of respiratory syncytial virus (RSV) in causing lower respiratory disease among infants has made its control a worldwide priority.1,2 However, in the United States, the total burden of RSV infection during the first 5 years of life remains poorly defined, particularly in the outpatient setting. Previous studies have been limited because they were retrospective, of short duration, lacked laboratory confirmation of RSV infection, or focused only on hospitalizations or bronchiolitis.3-8 The Centers for Disease Control and Prevention (CDC) initiated the New Vaccine Surveillance Network (NVSN), a prospective, population-based inpatient and outpatient surveillance for acute respiratory infections among children under 5 years of age in 2000 in Nashville and Rochester, New York, and in 2003 in Cincinnati. The aim of our study was to determine the populationbased burden of RSV infection among hospitalized children and outpatients in emergency departments and primary care settings. We further sought to describe the effect of potential risk factors on the severity of illness.

Eligible children were under 5 years of age and had received a diagnosis of acute respiratory infection, which was defined as an illness presenting with one or more of the following symptoms: fever, cough, earache, nasal congestion, rhinorrhea, sore throat, vomiting after coughing, wheezing, and labored, rapid, or shallow breathing. Excluded were children who had respiratory symptoms lasting more than 14 days, had neutropenia from chemotherapy, had been hospitalized elsewhere within 4 days, or were newborns who had been hospitalized since birth. Written informed consent was obtained from a parent or guardian of each child. The institutional review board at each site and the CDC approved the study. The authors vouch for the completeness and accuracy of the data and the analyses presented.
Patients

Me thods
Study Design

The design and methods of NVSNs surveillance for acute respiratory infections have been described previously.9-16 Inpatient surveillance was conducted for 4 years, from October 2000 through September 2004, during the winter months in the counties surrounding Nashville and Rochester and during the 20032004 season in Cincinnati. Children were enrolled within 48 hours after admission from Sunday through Thursday. Outpatient surveillance was conducted concurrently from November through April in Nashville and Rochester from 2002 through 2004 and similarly in Cincinnati during the 20032004 season. Outpatients were enrolled at each site from one to four urban and suburban pediatric practices either 1 or 2 days per week and from emergency departments during daytime and nighttime shifts in Nashville and Rochester 3 or 4 days per week and in Cincinnatis emergency department every fourth day. During the study period, surveillance hospitals cared for more than 95% of each countys children; emergency departments cared for 30% of the children in Nashville, 60% in Rochester, and 95% in Cincinnati.

We obtained childrens demographic, medical, and social histories in standardized interviews of parents or guardians.12,13 The choice of clinical management, including hospitalization, was determined by the childs physician. Laboratory and clinical information was obtained from records from hospitals, emergency departments, and outpatient primary care settings. Recorded high-risk medical conditions included prematurity (<36 weeks of gestation); chronic pulmonary, cardiac, kidney, or immunodeficiency disease; cancer; and sickle cell anemia. We obtained nasal and throat swabs for viral detection for research purposes only, and the personnel caring for the patients were unaware of the results.9-16 All inpatient specimens were tested by reverse-transcriptase polymerase chain reaction (RT-PCR) and culture for RSV, influenza A, influenza B, and parainfluenza viruses 1, 2, and 3. Outpatient samples were assayed by RT-PCR for RSV and influenza. Additional viruses, including human metapneumovirus and rhinoviruses, were detected by RT-PCR among inpatient specimens from Nashville and Rochester in 2000 and 2001.9-17 Specimens were defined as positive if RSV was detected by viral isolation or by duplicate RT-PCR assays.9,10
Statistical Analysis

We determined the weighted number of hospitalizations for RSV infection and acute respiratory infections per 1000 children under 5 years of age
february 5, 2009

n engl j med 360;6

nejm.org

589

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

and age-specific demographic characteristics using bootstrap methods, as described previously.12-15 Rates of visits to emergency departments and primary care offices for RSV-associated acute respiratory infections were calculated by multiplying the RSV-attributable portion from NVSN surveillance by rates of acute respiratory infections according to age group estimated from data from the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS) for the respiratory-infection seasons of November through April from 2002 through 2004.13 The confidence intervals for these rates were calculated with the use of the delta method, which accounted for variation in both the national data sets and the data from this study.13,18 Population-based rates for emergency department visits were calculated on the basis of pediatric visits to emergency departments during the winter months from 2002 through 2004 in Rochester and during the 2003 2004 season in Cincinnati, since these sites accounted for the majority of visits to emergency departments in their counties. Chi-square tests were used for associations between categorical variables, with Yates correction for continuity used for analysis with dichotomous variables and Students t-tests for continuous variables. Three multivariable logistic-regression models were constructed for RSV-positive inpatients versus RSV-positive outpatients, for RSVpositive inpatients versus RSV-negative inpatients, and for RSV-positive outpatients versus RSV-negative outpatients. Candidate covariates in each model were sex, age group (0 to 5, 6 to 11, 12 to 23, and 24 to 59 months), the time in day care (4 and >4 hours per week), the presence or absence of household exposure to smoke, breastfeeding (0 to <1, 1 to 6, and 7 months), the presence or absence of high-risk coexisting conditions, the presence or absence of prematurity, an interaction term between high-risk conditions and prematurity, and the presence or absence of other children under 18 years of age residing in the household. Covariates were included in the multivariable model if the P value was under 0.20 when comparing RSV-positive inpatients with RSV-positive outpatients in bivariate analysis. All statistical analyses were performed with the use of Stata software, version 10.0.19

R e sult s
Study Populations

Of 6225 eligible children, 5067 (81% of inpatients and outpatients) were enrolled, and 919 had confirmed RSV infection (Fig. 1). The major reasons for nonenrollment were a lack of parental consent (65%) or unavailability of parents (20%). In only 1% of cases, the childs physician declined participation.
Laboratory Analyses

Of 564 RSV-positive inpatient specimens, 547 (97%) were identified by RT-PCR and 171 (30%) by tissue culture. Of 500 specimens tested by both methods, 329 (66%) were deemed to be positive only by RT-PCR and 17 (3%) by culture.
Epidemiologic Analysis

RSV outbreaks at each site occurred yearly between November and April. The median onset was December 1. RSV was detected for a median of 18 weeks in Nashville and 20 weeks in Rochester and Cincinnati. RSV was identified in 919 of 5067 specimens (18%) and was associated with 20% of annual hospitalizations, 18% of emergency department visits, and 15% of office visits for acute respiratory infections from November through April. In 58 of the samples (6%), coinfecting viruses were identified, including 26 samples with influenza virus, 8 with parainfluenza virus, 11 with rhinovirus, 6 with adenovirus, 5 with cytomegalovirus, and 4 with other viruses. The circulation of strain groups was similar among sites, with RSV A predominating during 3 of 4 years. Of 753 RSV isolates that were typed, 602 (80%) were group A, 137 (18%) were group B, and 14 (2%) were determined to be both A and B by RT-PCR. RSV A strains comprised 81% of inpatient isolates and 78% of outpatient isolates (P = 0.32).
Demographic Characteristics

Significantly more inpatients than outpatients with RSV infection were non-Hispanic white, were under the age of 6 months, and had private insurance (Table 1, and Table 1 in the Supplementary Appendix, available with the full text of this article at NEJM.org). Inpatients with RSV infection were significantly more likely than inpatients who did

590

n engl j med 360;6

nejm.org

february 5, 2009

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

Respir atory Syncytial Virus Infection in Healthy Children

6225 Children 059 mo of age were eligible

5067 (81%) Were enrolled

2892 (57%) Were treated in hospital

1014 (20%) Were treated in emergency department

1161 (23%) Were treated in pediatric office

919 (18%) Were infected with RSV

564 (61%) Were hospitalized

355 (39%) Were outpatients 184 (52%) Were treated in emergency department 171 (48%) Were treated in pediatric office

328 (58%) Were 05 mo 97 (17%) Were 611 mo 99 (18%) Were 1223 mo 40 (7%) Were 2459 mo

88 (25%) Were 05 mo 85 (24%) Were 611 mo 75 (21%) Were 1223 mo 107 (30%) Were 2459 mo

Figure 1. Enrollment and Outcomes. The number of children who were hospitalized includes 78 children who were initially enrolled as outpatients (74 from emergency departments and 4 from pediatric practices).

not have RSV infection to be non-Hispanic white 2.4 per 1000 children in Rochester. Among chil1st RETAKE AUTHOR: Hall ICM and under the age of 6 months. dren under the age 2nd 6 months, rates were 33.9 of REG F FIGURE: 1 of 2 3rd per 1000 in Cincinnati, 12.8 per 1000 in Nashville, CASE Revised Hospitalized Patients Line 14.2 per 1000 in Rochester. and 4-C EMail SIZE
Rates of RSV infection

From 2000 through 2004, the average annual Clinical Characteristics AUTHOR, PLEASE NOTE: rate of RSV-associated hospitalizationhas been redrawn Dischargebeen reset. Figure was 3 per and type has diagnoses were based on clinical and Please check carefully. 1000 children under the age of 5 years and 17 per laboratory information; clinicians were unaware 1000 children under the age of 6 months (Table 2). of NVSN ISSUE: 12-18-08 results. Among 564 inpalaboratory JOB: 35925 Hospitalization rates varied yearly and regionally. tients with RSV infection, bronchiolitis was diagDuring the 20032004 season, among children nosed in 397 (70%) and RSV-associated illness in under the age of 5 years, the rates were 5.7 per 252 (45%). Among RSV-positive inpatients under 1000 in Cincinnati, 2.3 per 1000 in Nashville, and the age of 12 months, bronchiolitis was diagnosed

Enon

ARTIST: ts

H/T Combo

H/T

33p9

n engl j med 360;6

nejm.org

february 5, 2009

591

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

Table 1. Characteristics of Children under 5 Years of Age with or without Respiratory Syncytial Virus (RSV) Infection.* P Value for Inpatients vs. Outpatients with RSV Infection P Value 0.58 160 (45) 195 (55) <0.001 313 (55) 165 (29) 59 (10) 27 (5) 328 (58) 97 (17) 99 (18) 40 (7) 295 (52) 225 (40) 39 (7) 5 (1) 1064 (46) 825 (35) 290 (12) 149 (6) <0.001 1042 (45) 306 (13) 441 (19) 539 (23) 0.15 1237 (53) 860 (37) 214 (9) 17 (1) 220 (62) 119 (34) 14 (4) 2 (1) 1147 (63) 597 (33) 70 (4) 6 (<1) 88 (25) 85 (24) 75 (21) 107 (30) 257 (14) 354 (19) 513 (28) 696 (38) 0.95 0.009 127 (36) 140 (39) 57 (16) 31 (9) 702 (39) 718 (39) 235 (13) 165 (9) <0.001 <0.001 856 (47) 964 (53) 0.42 <0.001 0.63

Variable

Hospitalized Patients RSV-Positive (N = 564) RSV-Negative (N = 2328) P Value 0.54 RSV-Positive (N = 355)

Outpatients RSV-Negative (N = 1820)

no. (%) Sex Male Female Race or ethnic group White Black Hispanic Other or unknown Age (mo) 05 611 1223 2459 Type of health insurance Public Private None Unknown 245 (43) 319 (57) 1045 (45) 1283 (55)

no. (%)

* Inpatient surveillance was conducted for 4 years, from October 2000 through September 2004, in the counties surrounding Nashville and Rochester and during the 20032004 season in Cincinnati. Outpatient surveillance was conducted concurrently from November through April in Nashville and Rochester from 2002 through 2004 and similarly in Cincinnati in the 20032004 season. Percentages may not total 100 because of rounding. Race or ethnic group was reported by parents or guardians.

in 85% and RSV-associated illness in 53%, as compared with rates of 31% and 18% in children between the ages of 24 and 59 months, respectively (P<0.001). The major diagnoses among this older age group were pneumonia (51%) and asthma (60%). Among all RSV-positive inpatients, 95% had labored respirations and required supplemental oxygen, 78% had wheezing, 69% were febrile, and 1% had apnea. There were no deaths. Hospitalizations lasted for a median of 2 days and lasted from less than 1 day to 26 days.
Outpatients

Rates of RSV Infection

by a factor of 3, but the difference ranged from a factor of 2 to a factor of 11, depending on the calendar year and the age group (Table 2). For children under 5 years of age, population-based rates of RSV-associated emergency department visits in Rochester were 19 per 1000 for the 20022003 season and 36 per 1000 for the 20032004 season; in Cincinnati, the rate was 44 per 1000 children during the 20032004 season. These populationbased rates corresponded closely to emergency department rates estimated from national databases (NAMCS and NHAMCS) for 20022003 and 20032004 (22 and 32 per 1000, respectively).
Clinical Characteristics

Overall, the estimated rates of outpatient office visits for acute respiratory infections among children under 5 years of age were higher than the estimated rates of emergency department visits
592

The most frequent clinical findings among 355 RSV-positive outpatients were cough (98%), fever (75%), labored respirations (73%, including 85%
february 5, 2009

n engl j med 360;6

nejm.org

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

Respir atory Syncytial Virus Infection in Healthy Children

Table 2. Rates of Inpatient and Outpatient Treatment for Children under 5 Years of Age with Confirmed Respiratory Syncytial Virus (RSV) Infection per 1000 Children, According to Year.* Treatment Site and Year 05 Inpatient Hospital 20002001 20012002 20022003 20032004 20002004 Outpatient Emergency department 20022003 20032004 20022004 Pediatric practice 20022003 20032004 20022004 108 (33346) 157 (54462) 132 (46383) 194 (77492) 160 (45576) 177 (61511) 53 (13222) 80 (22282) 66 (18245) 31 (9100) 77 (26230) 57 (19167) 61 (24154) 99 (44219) 80 (36179) 39 (12124) 69 (34143) 55 (24126) 45 (13157) 68 (27175) 57 (20161) 24 (787) 38 (15102) 32 (1192) 15 (544) 11 (339) 13 (441) 22 (1049) 32 (1954) 28 (1550) 18.5 (14.422.9) 11.7 (9.114.7) 12.4 (9.415.2) 21.7 (18.824.6) 16.9 (15.318.5) 7.4 (5.19.9) 4.1 (2.45.8) 3.4 (1.95.0) 5.4 (3.87.0) 5.1 (4.65.5) 3.2 (1.94.8) 2.5 (1.53.6) 1.9 (1.12.8) 3.1 (2.33.9) 2.7 (2.32.7) 0.4 (0.20.7) 0.2 (0.00.4) 0.2 (0.00.4) 0.5 (0.30.8) 0.4 (0.30.4) 3.5 (2.94.1) 2.2 (1.82.7) 2.1 (1.72.5) 3.7 (3.3 4.1) 3.0 (2.83.4) 611 Age, in Months 1223 rate/1000 patients (95% CI) 2459 059

* The presence of RSV infection was confirmed by reverse-transcriptase polymerase chain reaction (RT-PCR) and culture for inpatients and by RT-PCR for outpatients. Inpatient surveillance was conducted for 4 years, from October 2000 through September 2004, in the counties surrounding Nashville and Rochester and during the 20032004 season in Cincinnati. Outpatient surveillance was conducted concurrently in Nashville and Rochester from 2002 through 2004 and similarly in Cincinnati during the 20032004 season. Rates of infection are listed for November through April of each annual season. Rates of infection for outpatients were calculated by multiplying the national rate of outpatient visits for acute respiratory infections from the National Ambulatory Medical Care Survey and the National Hospital Ambulatory Medical Care Survey by the proportion of visits for acute respiratory infections in this study attributable to RSV for November through April 20022003 and 20032004.

and 73% of patients in emergency departments and pediatric offices, respectively; P<0.001), and wheezing (65%). Discharge diagnoses included upper respiratory tract infection (32%), bronchiolitis (20%), asthma (13%), and pneumonia (8%). Only 3% of outpatients with RSV infection received the diagnosis of RSV-associated illness, as compared with 45% of inpatients (P<0.001).
Risk Factors for Hospitalization

To assess the effects that environmental and host factors may have had on severe illness, we compared the results for RSV-positive inpatients with those of RSV-positive outpatients. To assess the effects of these factors on the development of an acute respiratory infection specifically from RSV, as compared with other causes, we compared the results of children with respiratory infections who

were RSV-positive with those who were RSV-negative (Table 3). One or more high-risk conditions were present in 189 of 564 of RSV-infected inpatients (34%) and in 96 of 355 outpatients (27%) (Table 3). The most frequent conditions among high-risk RSVpositive children were previous wheezing, which occurred in 104 inpatients (55%) and 70 outpatients (73%) (P = 0.005), and prematurity, which occurred in 91 inpatients (48%) and 30 outpatients (31%) (P = 0.009). Among children with acute respiratory infections that were either RSVpositive or RSV-negative, the presence of any high-risk condition significantly correlated with hospitalization and increasing age (P<0.001). However, more RSV-negative inpatients than RSVpositive inpatients had high-risk conditions (42% vs. 34%, P = 0.001).

n engl j med 360;6

nejm.org

february 5, 2009

593

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

594
All Patients (N = 5067) Hospitalized (N = 564) P Value no./total no. (%) 0.05 0.001 0.22 0.48 <0.001 0.006 <0.001 0.02 <0.001 0.78 0.43 80/355 (23) 210 (59) 186 (52) 120 (34) 32 (6) 189/347 (54) 48 (14) 197/352 (56) 0.21 0.55 <0.001 0.21 <0.001 0.75 683 (29) 256/1321 (19) 53/1321 (4) 1723 (74) 1171 (50) 1020 (44) 978 (42) 156/608 (26) 1283 (55) 1491 (64) 637 (27) 185 (8) 716/1198 (60) 93 (26) 117 (12) 355 (15) 984 (42) 495 (27) 181 (10) 36 (7) 28 (15) 885 (49) 628/1800 (35) 181/1800 (10) 1224 (67) 760 (42) 797 (44) 723 (40) 445/1820 (24) 998 (55) 1041 (57) 514 (28) 250 (14) 1027/1803 (57) <0.001 <0.001 0.009 0.008 <0.001 <0.001 <0.001 <0.001 <0.001 0.97 0.12 0.59 0.89 <0.001 <0.53 <0.001 0.14 no./total no. (%) 1765 (35) 657 (13) 174 (10) 140 (21) 1868 (37) 1024/3826 (27) 266/3826 (7) 3616 (71) 2373 (47) 2204 (43) 2081 (41) 764/3090 (25) 2811 (55) 3078 (61) 1437 (28) 515 (10) 2129/3700 (58) 360 (64) 166 (29) 320 (57) 83/307 (27) 239 (42) 141 (40) 235 (42) 152 (43) 305 (54) 137 (39) 427 (76) 242 (68) 6/353 (2) 26/352 (7) 55/353 (16) 85/352 (24) 160 (28) 140 (39) 16 (18) 3 (10) 17 (9) 4 (4) 91 (16) 30 (8) 189 (34) 96 (27) Outpatient (N = 355) Hospitalized (N = 2328) Outpatient (N = 1820) P Value Children with RSV Infection (N = 919) Children without RSV Infection (N = 4148)

Table 3. Proportion of Inpatients and Outpatients with and without Respiratory Syncytial Virus (RSV) with Factors Potentially Associated with the Risk of RSV Infection and Severity of Illness.*

Variable

Underlying condition

Any high-risk condition

Prematurity

Prophylaxis with palivizumab

High-risk children (% of all high-risk children)

Premature birth (% of all premature births)

Day-care conditions
The

>4 hrs/wk

Attendees

>12

Age of other children in home

n e w e ng l a n d j o u r na l

n engl j med 360;6

All

<5 yr

5 to <18 yr

nejm.org

Smoking exposure in home

Any

Mother

of

Breast-feeding history

february 5, 2009

Any duration

m e dic i n e

None or <1 mo

16 mo

7 mo

Maternal education of high school or less

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

* Inpatient surveillance was conducted for 4 years, from October 2000 through September 2004, in the winter months in the counties surrounding Nashville and Rochester and during the 20032004 season in Cincinnati. Outpatient surveillance was conducted concurrently from November through April in Nashville and Rochester from 2002 through 2004 and similarly in Cincinnati during the 20032004 season. P<0.001 for the comparison with patients in the same category without RSV. Data were obtained only from 2002 through 2004. P = 0.04 for the comparison with patients in the same category without RSV.

Respir atory Syncytial Virus Infection in Healthy Children

P<0.001

P=0.002

P=0.007

P>0.05

Patients with RSV

Inpatients vs. outpatients

Patients with RSV vs. Patients without RSV

Inpatients Outpatients

Male Age 05 mo Age 611 mo Age 1223 mo Day care >4 hr/wk Household smoke exposure >1 mo breast-feeding High-risk condition (other than prematurity) Prematurity only Any other child <18 yr in home 1 10 20 30 0 1 2 4 6 8 0.5 1.0 1.5 2.0 2.5 3.0

Odds Ratio

Odds Ratio

Odds Ratio

Figure 2. Odds Ratios for Potential Risk Factors in Patients with and Those without Respiratory Syncytial Virus (RSV) Infection, According to Treatment Site. 1st RETAKE AUTHOR: Hall ICM According to multiple logistic-regression analyses, the only risk factors associated with RSV illness requiring hospitalization were an age FIGURE: 2 of 2prematurity. For age groups,2nd reference group is patients between REG and a history of of less than 2 years (especially under 6 months) F the 3rd CASE the ages of 24 months and 59 months. Horizontal lines indicate 95% confidence intervals. Revised
EMail Enon

ARTIST: ts

Line H/T Combo

4-C H/T

SIZE 36p6

Prophylaxis with palivizumab was adminis- AUTHOR, PLEASE NOTE: RSV illness requiring hospily associated with Figure has been redrawn and type has been reset. tered to 174 of 1765 of high-risk children (10%) talization.carefully. Please check In comparison with RSV-negative paand to 140 of 657 children with a history of pre- tients, RSV-positive inpatients and outpatients JOB: of 5 SSUE: 12-18-08 maturity (21%) (Table 3). The proportion 3592high- were significantly Iyounger (P<0.001) but were not risk patients with RSV infection who received more likely to have been premature, indicating palivizumab was smaller than the proportion of that a young age imposed a greater risk for RSVpatients without RSV infection, but the difference positive illness than for RSV-negative illness. was not significant (21 of 285 patients [7%] vs. 153 of 1479 patients [10%], P = 0.15). The rates of Discussion prophylaxis with palivizumab differed significantly between high-risk inpatients and high-risk Our findings from three geographically diverse outpatients only among children with RSV-nega- populations provide a comprehensive view of the tive acute respiratory infections (P = 0.009). RSV burden among children within the first 5 years Factors indicative of a childs environment and of life. In our population, rates of hospitalization care, including day-care attendance, exposure to for RSV-associated illness were three times as tobacco smoke, and maternal educational level, high as those associated with influenza or parawere also examined (Table 3). None of these fac- influenza viruses, and the proportion of children tors correlated with illness requiring hospital- receiving influenza immunization was low (18%).13 ization among children with RSV-positive or The estimated rates of outpatient visits associatRSV-negative acute respiratory infections. Only ed with RSV infection were similar to those assobreast-feeding of less than 1 months duration ciated with influenza for all children under 5 years and the presence of other children under the age of age but were markedly greater for children unof 5 years in the home significantly correlated der the age of 6 months.12-14 However, prospecwith hospitalization. tive population-based information concerning RSV However, multiple logistic-regression analyses infection among older children and outpatients revealed that none of these potential risk factors, is lacking.3-5,8,20-22 By concurrently addressing these including sex, independently correlated with more populations, our study demonstrates a previously severe illness among patients with RSV infection unrecognized size and spectrum of burden of RSV after adjustment for other covariates (Fig. 2). Only infection among all children under 5 years of age. a younger age and prematurity were independentIf we extrapolate from our population-based

n engl j med 360;6

nejm.org

february 5, 2009

595

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

data to the entire U.S. population, an estimated 2.1 million children under 5 years of age with RSV infection would require medical attention each year. Among these children with RSV-related illnesses, approximately 57,527 (3%) are hospitalized, 517,747 (25%) are treated in emergency departments, and 1,534,064 (73%) are treated by pediatric practices; of these outpatients, 1,256,014 (61%) are between 2 and 5 years of age. These rates of visits for RSV-associated acute respiratory infections indicate that a major proportion of RSVs burden results in outpatient visits among children beyond infancy. Furthermore, the estimated outpatient rates are probably close to the actual national rates, since our directly determined population-based emergency department rates in Rochester and Cincinnati were very similar and well within the confidence intervals of the estimated rates. The outpatient burden of RSV on health care resources is probably not fully recognized by health care providers and by public health officials, since only 3% of outpatients with confirmed RSV infection received the specific diagnosis of RSV infection; bronchiolitis was diagnosed in 20% of such children. Significantly more RSV outpatient visits occurred among primary care practices than among emergency departments, and the illnesses of patients from pediatric practices were similar to those of patients in emergency departments and reflected moderately severe disease. This suggests that outpatient visits contribute greatly to RSVs burden by both their frequency and the severity of illness. The rates of RSV infections requiring medical attention are high not only during infancy but throughout the first 5 years of life. This factor underscores the as-yet-unmet need for an effective vaccine.1,23-26 Of the estimated 2 million children under the age of 5 years who require care for RSV infections annually, 78% are over the age of 1 year. Meanwhile, much effort has been appropriately focused on children who are at highest risk for severe RSV disease and who most need prophylaxis and health care resources.24,27-29 However, our study suggests that since RSV infection among previously healthy children imposes a much larger burden than was previously recognized, the need to address this issue may be even greater. Characteristics that were most frequently asso-

ciated with RSV illness requiring hospitalization included male sex, chronic coexisting medical conditions, lower socioeconomic status, smoke exposure, lack of breast-feeding, and contact with other children.3,6,8,29-36 However, none of these factors in our population independently correlated with illness severity, except for young age and prematurity, and the risk of prematurity was not specific for RSV disease. Only young age imposed a significantly greater risk of severe illness among children with RSV infection than among children with acute respiratory infections caused by other organisms. Our findings may differ from those of previous studies partly because our study was designed to evaluate the role of risk factors in acquiring RSV infection and the development of more severe disease specifically associated with RSV. We assessed the first factor by comparing children with acute respiratory infections who were RSVpositive with those who were RSV-negative. We assessed the second factor by comparing children with RSV infection who were hospitalized with those who were treated as outpatients. Greater use of palivizumab prophylaxis also may have affected the rates and characteristics of recently hospitalized high-risk children.37 This explanation is supported by our findings, which suggest that the proportion of high-risk children receiving prophylaxis who were hospitalized with RSV-negative infection was significantly greater than the proportion of those with RSV-positive infection. Most children with RSV infection, both those who were hospitalized and those who were treated as outpatients, had no coexisting medical conditions or characteristics that significantly identified them as being at greater risk for severe RSV disease, except for being under 2 years of age. RSVs major burden, therefore, occurs among previously healthy children whose risk of severe illness cannot be predicted. On the basis of our findings, we estimate that among children under the age of 5 years, RSV infection results in approximately 1 of 334 hospitalizations, 1 of 38 visits to an emergency department, and 1 of 13 visits to a primary care office each year in the United States.
Supported through cooperative agreements with the CDC. Dr. Hall reports receiving grant support and consulting fees from MedImmune; Dr. Weinberg, research support from Astellas and MedImmune and consulting fees from MedImmune; Dr. Staat, consulting fees or fees for serving on paid advisory boards

596

n engl j med 360;6

nejm.org

february 5, 2009

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

Respir atory Syncytial Virus Infection in Healthy Children

from GlaxoSmithKline, Merck, and MedImmune, lecture fees from Merck, and research support from MedImmune; Dr. Griffin, grant support from MedImmune and Pfizer; and Dr. Grijalva, lecture fees from Wyeth. No other potential conflict of interest relevant to this article was reported. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the CDC. We thank the children and their parents who participated in this study; all the members of the New Vaccine Surveillance Network, including Geraldine Lofthus, Kenneth Schnabel, Andrea Marino, Lynne Shelley, Jennifer Carnahan, Linda Anderson, References
1. The Jordan Report: accelerated devel-

Gladys Lathan, Charlene Freundlich, Christina Albertin, and Rebecca Martinez at the University of Rochester; John Williams, Diane Kent, Ann Clay, Ayesha Khan, Yi Wei Tang, Haijing Li, Jennifer Doersam, Amondrea Blackman, Nayleen Whitehead, Erin Keckley, and Jody Peters at Vanderbilt University; Linda Jamison, David Witte, David Bernstein, Emilie Grube, Vanessa Florian, Pamela Groen, and Joel Mortensen at the Cincinnati Childrens Medical Center; and Ranee Seither, Aaron Curns, Larry Anderson, Carolyn B. Bridges, Benjamin Schwartz, Frances Walker, John Copeland, Alicia Fry, Jennifer Reuer, and John Zhang at the CDC.

opment of vaccines 2002. Bethesda, MD: National Institute of Allergy and Infectious Diseases, 2002. (Accessed January 12, 2009, at http://www3.niaid.nih.gov/ topics/Malaria/PDF/jordan20_2002.pdf.) 2. Hall CB. Respiratory syncytial virus and parainfluenza virus. N Engl J Med 2001;344:1917-28. 3. Boyce TG, Mellen BG, Mitchel EF Jr, Wright PF, Griffin MR. Rates of hospitalization for respiratory syncytial virus infection among children in Medicaid. J Pediatr 2000;137:865-70. 4. Shay DK, Holman RC, Newman RD, Liu LL, Stout JW, Anderson LJ. Bronchiolitis-associated hospitalizations among US children, 1980-1996. JAMA 1999;282: 1440-6. 5. Leader S, Kohlhase K. Recent trends in severe respiratory syncytial virus (RSV) among US infants, 1997 to 2000. J Pediatr 2003;143:Suppl:S127-S132. 6. Holberg CJ, Wright AL, Martinez FD, Ray CG, Taussig LM, Lebowitz MD. Risk factors for respiratory syncytial virusassociated lower respiratory illnesses in the first year of life. Am J Epidemiol 1991;133:1135-51. 7. Thompson WW, Shay DK, Weintraub E, et al. Mortality associated with influenza and respiratory syncytial virus in the United States. JAMA 2003;289:179-86. 8. Peck AJ, Holman RC, Curns AT, et al. Lower respiratory tract infections among American Indian and Alaska Native children and the general population of U.S. children. Pediatr Infect Dis J 2005;24: 342-51. 9. Weinberg GA, Erdman DD, Edwards KM, et al. Superiority of reverse-transcription polymerase chain reaction to conventional viral culture in the diagnosis of acute respiratory tract infections in children. J Infect Dis 2004;189:706-10. 10. Erdman DD, Weinberg AG, Edwards KM, et al. GeneScan reverse transcription-PCR assay for detection of six common respiratory viruses in young children hospitalized with acute respiratory illness. J Clin Microbiol 2003;41:4298303. 11. Mullins JA, Erdman DD, Weinberg

GA, et al. Human metapneumovirus infection among children hospitalized with acute respiratory illness. Emerg Infect Dis 2004;10:700-5. 12. Iwane MK, Edwards KM, Szilagyi PG, et al. Population-based surveillance for hospitalizations associated with respiratory syncytial virus, influenza virus, and parainfluenza viruses among young children. Pediatrics 2004;113:1758-64. 13. Poehling KA, Edwards KM, Weinberg GA, et al. The underrecognized burden of influenza in young children. N Engl J Med 2006;355:31-40. 14. Miller EK, Griffin MR, Edwards KM, et al. Influenza burden for children with asthma. Pediatrics 2008;121:1-8. 15. Miller EK, Lu X, Erdman DD, et al. Rhinovirus-associated hospitalizations in young children. J Infect Dis 2007;195:77381. 16. Griffin MR, Walker FJ, Iwane MK, Weinberg GA, Staat MA, Erdman DD. Epidemiology of respiratory infections in young children: insights from the new vaccine surveillance network. Pediatr Infect Dis J 2004;23:Suppl:S188-S192. 17. Weinberg GA, Hall CB, Iwane MK, et al. Parainfluenza virus infection of young children: estimates of the populationbased burden of hospitalization. J Pediatr (in press). 18. Properties of a random sample. In: Casella G, Berger RL, eds. Statistical inference. 2nd ed. Pacific Grove, CA.: Duxbury/Thomson Learning, 2002:240-5. 19. StataCorp. Stata statistical software: release 10. College Station, TX: StataCorp, 2007. 20. Schanzer DL, Langley JM, Tam TW. Hospitalization attributable to influenza and other viral respiratory illnesses in Canadian children. Pediatr Infect Dis J 2006; 25:795-800. 21. Izurieta HS, Thompson WW, Kramarz P, et al. Influenza and the rates of hospitalization for respiratory disease among infants and young children. N Engl J Med 2000;342:232-9. 22. Nicholson KG, McNally T, Silverman M, Simons P, Stockton JD, Zambon MC. Rates of hospitalisation for influenza, respiratory syncytial virus and human meta-

pneumovirus among infants and young children. Vaccine 2006;24:102-8. 23. Nokes DJ, Okiro EA, Ngama M, et al. Respiratory syncytial virus infection and disease in infants and young children observed from birth in Kilifi District, Kenya. Clin Infect Dis 2008;46:50-7. 24. Collins PL, Murphy BR. New generation live vaccines against human respiratory syncytial virus designed by reverse genetics. Proc Am Thorac Soc 2005;2:16673. 25. Englund J. In search of a vaccine for respiratory syncytial virus: the saga continues. J Infect Dis 2005;191:1036-9. 26. van Drunen Littel-van den Hurk S, Mapletoft JW, Arsic N, Kovacs-Nolan J. Immunopathology of RSV infection: prospects for developing vaccines without this complication. Rev Med Virol 2007;17:534. 27. Welliver RC. Respiratory syncytial virus infection: therapy and prevention. Paediatr Respir Rev 2004;Suppl:S127-S133. 28. Meissner HC, Rennels MB, Pickering LK, Hall CB. Risk of severe respiratory syncytial virus disease, identification of high risk infants and recommendations for prophylaxis with palivizumab. Pediatr Infect Dis J 2004;23:284-5. 29. Simoes EA. Environmental and demographic risk factors for respiratory syncytial virus lower respiratory tract disease. J Pediatr 2003;143:Suppl:S118-S126. 30. Nielsen HE, Siersma V, Andersen S, et al. Respiratory syncytial virus infection risk factors for hospital admission: a case-control study. Acta Paediatr 2003; 92:1314-21. 31. Reeve CA, Whitehall JS, Buettner PG, Norton R, Reeve DM, Francis F. Predicting respiratory syncytial virus hospitalisation in Australian children. J Paediatr Child Health 2006;42:248-52. 32. Bradley JP, Bacharier LB, Bonfiglio J, et al. Severity of respiratory syncytial virus bronchiolitis is affected by cigarette smoke exposure and atopy. Pediatrics 2005;115(1):e7-e14. 33. von Linstow ML, Hgh M, Nordb SA, Eugen-Olsen J, Koch A, Hgh B. A community study of clinical traits and risk factors for human metapneumovirus and

n engl j med 360;6

nejm.org

february 5, 2009

597

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.

Respir atory Syncytial Virus Infection in Healthy Children

respiratory syncytial virus infection during the first year of life. Eur J Pediatr 2008;167:1125-33. 34. Carbonell-Estrany X, Figueras-Aloy J, Law BJ. Identifying risk factors for severe respiratory syncytial virus among infants born after 33 through 35 completed weeks of gestation: different methodologies yield consistent findings. Pediatr Infect Dis J 2004;23:Suppl:S193-S201.
35. Law BJ, Langley JM, Allen U, et al. The

Pediatric Investigators Collaborative Network on Infections in Canada study of predictors of hospitalization for respiratory syncytial virus infection for infants born at 33 through 35 completed weeks of gestation. Pediatr Infect Dis J 2004;23:806-14. 36. Rossi GA, Medici MC, Arcangeletti MC, et al. Risk factors for severe RSVinduced lower respiratory tract infection

over four consecutive epidemics. Eur J Pediatr 2007;166:1267-72. 37. Romero JR. Palivizumab prophylaxis of respiratory syncytial virus disease from 1998 to 2002: results from four years of palivizumab usage. Pediatr Infect Dis J 2003;22:Suppl:S46-S54.
Copyright 2009 Massachusetts Medical Society.

powerpoint slides of journal figures and tables

At the Journals Web site, subscribers can automatically create PowerPoint slides. In a figure or table in the full-text version of any article at NEJM.org, click on Get PowerPoint Slide. A PowerPoint slide containing the image, with its title and reference citation, can then be downloaded and saved.

598

n engl j med 360;6

nejm.org

february 5, 2009

Downloaded from www.nejm.org on April 14, 2009 . Copyright 2009 Massachusetts Medical Society. All rights reserved.