economic impact of respiratory syncytial virus-related illness in the us

Pharmacoeconomics 2004; 22 (5): 275-284ORIGINAL RESEARCH ARTICLE 1170-7690/04/0005-0275/$31.00/0

© 2004 Adis Data Information BV. All rights reserved.

Economic Impact of RespiratorySyncytial Virus-Related Illness inthe USAn Analysis of National Databases

L. Clark Paramore,1 Vincent Ciuryla,2 Gabrielle Ciesla1 and Larry Liu2

1 MEDTAP International Inc., Bethesda, Maryland, USA2 Global Health Outcomes Assessment, Wyeth Research, Collegeville, Pennsylvania, USA

Objective: To determine the impact of respiratory syncytial virus (RSV) infectionAbstracton healthcare resource use and costs in the US from the third-party payerperspective.

Design: The study retrospectively analysed cross-sectional medical encounterdata from three federally funded databases that comprise nationally representativesamples of hospital inpatient stays, physician office visits and visits to hospitaloutpatient departments and emergency rooms.

Methods: Identification of RSV infection-related medical encounters was basedon the occurrence of RSV-specific International Classification of Diseases (9thEdition)-Clinical Modification diagnosis codes (079.6, 466.11, 480.1) as principaldischarge diagnoses or the assumption that 10–15% of all otitis media visits weredue to RSV infection. Outpatient drug costs were estimated based on averagewholesale price, and physician fees and test/procedure costs were estimated basedon prevailing national fees. Inpatient costs were estimated from total billedcharges using a cost-to-charge ratio of 0.53.

Results: In 2000, nearly 98% of RSV infection-related hospitalisations occurredin children <5 years old. There were approximately 86 000 hospitalisations, 1.7million office visits, 402 000 emergency room visits and 236 000 hospital outpa-tient visits for children <5 years old that were attributable to RSV infection. Totalannual direct medical costs for all RSV infection-related hospitalisations ($US394million) and other medical encounters ($US258 million) for children <5 years oldwere estimated at $US652 million in 2000. Otitis media was a major cost driverfor physician visits. RSV infection-related hospitalisations increased from 1993 to2000, but average costs per hospitalisation were relatively stable.

Conclusion: Treatment of RSV infection-related illness represents a significanthealthcare burden in the US. The economic impact of ambulatory care for RSVinfection-related illness could be as important as that for RSV infection-relatedhospitalisation.

276 Paramore et al.

Respiratory syncytial virus (RSV) infection is the The costs of treatment of RSV infection in childrenin other care settings have not been evaluated.most important viral cause of lower respiratory tract

The purpose of this study was to estimate theinfection in infants and young children, infectingannual impact of RSV infection in children onalmost all children by the age of 2 years.[1,2] RSVhealthcare resource use and direct medical costs ininfection is associated with 40–90% of bronchiolitisthe hospital inpatient, hospital outpatient, emer-cases in children <5 years of age and with 50% ofgency room and office visit settings in the US.pneumonia cases in the first 2 years of life.[1,2]

Because a large majority of RSV infection-relatedPremature infants, younger infants and children withencounters were for children <5 years old, the studychronic lung and heart diseases are at an increasedfocused on this age group. The results should pro-risk of severe RSV infection.[2-4] Recent studiesvide insight as to the economic burden that could besuggest that a large proportion of otitis media casesreduced by preventing RSV infections or shorteningmay be associated with RSV infection.[5-7]

the natural history of the illness with therapeuticCurrently, no specific antiviral agent is available agents.

for the treatment of RSV infection. The only therapyis a broad-spectrum antiviral agent (ribavirin aero- Methodssol) that is approved only for hospitalised infants.[2]

Furthermore, despite significant progress in recentData Sources

years, no RSV vaccine is available because of thelack of durable immunity and the complexity of the The study retrospectively analysed cross-section-virus. For premature infants and infants with chronic al medical encounter data, from three separate feder-lung disease, prophylaxis with immune globulin ally funded databases that are comprised of national-(RSV-IGIV) or palivizumab (a monoclonal anti- ly representative samples, including hospital inpa-body) is recommended by the American Academy tient stays, physician office visits and visits toof Pediatrics.[8] However, both RSV-IGIV and hospital outpatient departments and emergency

rooms (ER). We focused on the latest 1-year periodpalivizumab are expensive and their effects are notof data available (2000) to estimate annual cost ofoptimal.illness; other years were examined just to provide aThe primary effect of current prophylactic agentstrend comparison.is to reduce the severity of illness; they do not

The Healthcare Cost & Utilisation Project’sprevent RSV infection. Both RSV-IGIV and(HCUP) Nationwide Inpatient Sample (NIS) data-palivizumab are not approved for infants or childrenbase (1993–2000), funded by the US government’swith congenital heart disease. [2,8] Healthy infantsAgency for Healthcare Research and Qualityand children, who represent a large proportion of(AHRQ), contains annual hospital discharge dataRSV infection burden, do not benefit from the cur-from a 20% sample of US hospitals (approximately

rent US prophylactic recommendations.[1] There is900 hospitals from 28 states). The hospital sampling

clearly an unmet medical need in the treatment andframe is defined as hospitals that were open during

prevention of RSV infection.any part of each calendar year and designated as

The economic burden of RSV infection in the US community hospitals in the American Hospital As-has not been well defined. Previous studies have sociation Annual Survey of Hospitals. Each hospitalexamined direct costs associated with either RSV stay record in the HCUP database contains the fol-infection-associated bronchiolitis or pneumonia lowing information: patient age and gender, princi-hospitalisations in children.[9,10] Leader and Kohl- pal and secondary discharge diagnoses, medical orhase have recently concluded that RSV infec- surgical procedures/tests occurring during stay,tion-associated bronchiolitis is the most important length of stay and total billed charges for the stay. Itcause of hospitalisations of infants for any reason.[11] should be noted that the total charges for the stay do

© 2004 Adis Data Information BV. All rights reserved. Pharmacoeconomics 2004; 22 (5)

Economic Impact of Respiratory Syncytial Virus-Related Illness in the US 277

not include the fees associated with physician ser- sumed that approximately 10–20% of all ambulato-vices provided during the stay. ry/outpatient visits for nonsuppurative otitis media

(ICD-9-CM 381.0–381.4) or suppurative otitisThe National Ambulatory Medical Care Surveymedia (ICD-9-CM 382.x) were RSV infection-relat-(NAMCS) database (1997–2000), funded by the USed.[5-7] The midpoint (15%) of this range (10–20%)government’s Centers for Disease Control and Pre-was used in calculating RSV infection-related otitisvention (CDC), contains annual physician officemedia visits. A sensitivity analysis was conductedvisit data from an approximate 1% sample of USbased on an assumed 5% of otitis media visits beingphysician offices (≈2500 offices). Variables in theRSV infection-related.NAMCS database include: patient age and gender,

up to three listed diagnoses, drugs prescribed andmedical or surgical procedures/tests performed Cost Estimationduring the visit.

The National Hospital Ambulatory Medical Care The perspective of the economic evaluation wasSurvey (NHAMCS) database (1997–2000), also that of a third-party payer. For inpatient data we hadfunded by the CDC, contains annual ER and hospital access to billed charges which we converted tooutpatient department visit data from an approxi- costs. For the other settings we used fees for physi-mate 10% sample of short-stay (average length of cians and facilities to proxy for costs.stay of <30 days) hospitals in the US (600 hospitals). The HCUP provides information on the totalThe NHAMCS database contains information simi- billed charges for a given discharge but not thelar to that included in the NAMCS. actual costs of providing services from the payer

perspective. The HCUP also does not provide hospi-Identification of Respiratory Syncytial

tal-specific cost-to-charge ratios that would facili-Virus-Related Medical Encounters

tate the estimation of costs per discharge. Therefore,Identification of RSV infection-related hospital to allow for an estimation of the total annual costs of

stays from the HCUP was based on the designation RSV infection-related care in the hospital setting inof at least one of the following International Classi- the US, an average cost-to-charge ratio of 0.53,fication of Diseases (9th Edition)-Clinical Modifica- estimated based on publicly available Medicaretion (ICD-9-CM) diagnosis codes as the principal Cost Report Data,[12] was applied to the total billeddischarge diagnosis for the hospital stay: 079.6 charges. Cost estimates were updated to year 2002(RSV); 466.11 (acute bronchiolitis due to RSV in- US dollars based on the Consumer Price Index forfection)1; 466.19 (acute bronchiolitis due to organ- hospital inpatient services.[13]

isms other than RSV; and 480.1 (pneumonia due to Neither costs nor charges are reported in theRSV infection). NAMCS and NHAMCS. Therefore, to estimate the

Identification of RSV infection-related visits costs associated with RSV infection-related visits,from the NAMCS/NHAMCS was based on the oc- the resource use items (e.g. drugs, procedures) re-currence of at least one of the RSV codes in any of corded in these visit records2 were assigned unitthe three diagnosis fields. In addition, it was as- costs and then summary costs were calculated. Costs

1 The code 466.11 was not available prior to 1996. Estimates based on ICD-9-CM code 466.1 (acute bronchiolitis) foryears 1993–1995, with adjustment to reflect the proportion of hospitalisations in the period 1996–2000 that had a code of466.11 (versus 466.19).2 The documentation for the NAMCS and NHAMCS states that estimates of visit characteristics (e.g. age, gender,types of drugs prescribed) based on a sample size of <30 visit records are not reliable. Given that the number of visitrecords in each database with RSV infection-specific diagnoses (i.e. 079.6, 466.11, 4801.) was <30, RSV infection-related costs were estimated based on resource use obtained from records with a diagnosis of 466.1 (acute bronchiolitis).It was assumed that visits with this diagnosis would have similar resource use patterns compared with the visits thatrecorded RSV-infection specific diagnoses.


278 Paramore et al.

were estimated separately for visits with RSV infec- Statistical Analysistion-specific diagnoses versus visits for otitis media.

Data analysis was performed using SAS Statisti-Each visit record in the NAMCS and NHAMCS

cal Software (SAS System for Windows, Versionincludes up to a maximum of six drugs that were 8.02, SAS Institute, Cary, North Carolina, USA).prescribed during the visit. The drug name is report- In order to account for the complex design ofed, but not the quantity dispensed, days supplied, or the HCUP NIS survey, the SAS PROCdaily dose. Therefore, each drug was assigned an SURVEYMEANS was used to estimate means andestimated ‘days supplied’ and ‘daily dose’ amount standard errors for outcomes of interest (i.e. lengthbased on drug-specific recommendations in the Phy- of stay, total charges).sician’s Desk Reference. Unit drug costs were esti-mated based on the average wholesale price reported Resultsin the 2002 Drug Topics®RED BOOK®.[14] Oralintake (not intravenous) was assumed for all medi-

Hospital Stayscations in estimating drug costs.

Unit values for physician fees associated with the The 2000 HCUP NIS included a total of 17 879ambulatory/outpatient visits, and the tests and pro- hospital stays with an RSV infection-related princi-cedures performed during these visits, were estimat- pal discharge diagnosis (079.6: 343, 466.11: 14 844ed based on national prevailing fees for 2002.[15] For and 480.1: 2692) across all age groups. The applica-physician fees, the lowest level of visit complexity tion of the appropriate sampling weights (each dis-was assumed in each setting to provide the most charge record had a numeric weight value assignedconservative cost estimate. Corresponding facility to it which were then summed) for the NIS resultedfees were estimated based on publicly available in a weighted estimate of 87 105 RSV infection-claims data utilised by the former Health Care Fi- related hospitalisations in the US in 2000. Of these,nancing Administration (HCFA) to develop the Out- 34.7% were estimated to occur in infants <3 monthspatient Prospective Payment System. Costs were old, 75.3% in infants <1 year old, and approximatelyupdated to 2002 using the Consumer Price Index for 98% in children <5 years old (see figure 1). Due tohospital outpatient services.[13] the large majority of RSV infection-related hospital-

91–182days

18 158

0–90days

30 197

3203

1926

1151

19

672

244

0

10 000

20 000

30 000

40 000

50 000

60 000

70 000

80 000

Num

ber

of h

ospi

talis

atio

ns

Inci

denc

e of

hos

pita

lisat

ions

(pe

r 10

0 00

0)

−1000

−500

0

500

1000

1500

2000

2500

3000

3500Number of hospitalisationsIncidence of hospitalisations

10 854

183–273days

6335

274–364days

18 433

1–2 years

1508

3–4 years

1620

>4 years

Age of child

Fig. 1. Diagramatic representation of the estimated incidence rates (per 100 000 population) for respiratory syncytial virus infection-relatedhospitalisations in 2000 according to age group.



Table I. Estimated average length of stay and average charges per respiratory syncytial virus (RSV) infection-related hospital stay byprincipal discharge diagnosis codea for children <5 years old in the US, 2000 ($US; 2002 values)

Diagnosisa No. of Average length of stay (days) Average costs per stay

hospital stays mean (SE) median 25th 75th mean (SE) median 25th 75thpercentile percentile percentile percentile

079.6 1 554 2.8 (0.15) 2.0 2.0 3.0 4472 (480) 2112 1315 3588

466.11 71 531 3.3 (0.07) 3.0 2.0 4.0 4446 (241) 2664 1646 4498

480.1 12 773 4.2 (0.15) 3.0 2.0 5.0 6455 (454) 2928 1858 5357

Overall 85 858 3.4 (0.06) 3.0 2.0 4.0 4584 (232) 2664 1412 4530

a Based specifically on International Classification of Diseases (9th Edition)-Clinical Modification diagnosis codes 079.6 (RSV), 466.11(acute bronchiolitis due to RSV) and 480.1 (pneumonia due to RSV).

SE = standard error.

isations occurring in children <5 years old, the focus Table I reports the average (mean and median)length of stay and average (mean and median) costsof the study was on this age group.for RSV infection-related hospital stays in 2000.The incidence rate of RSV infection-related hos-Hospital stays for RSV infection-related pneumoniapitalisations in 2000 was 1738 per 100 000 infantshad the longest average mean length of stay and<1 years old, and 453 per 100 000 children <5 yearshighest average mean and median billed costs. Theold. Of infants 3–12 months old, nearly 17% of allmost common tests or procedures performed duringhospitalisations in 2000 were due to RSV-infectionthe RSV infection-related hospital stays included:related illness. In 2000, there were also approxi-respiratory medication administered by nebulizermately 116 in-hospital deaths related to RSV infec-(reported in 6.5% of hospital stays), spinal taption (case-fatality ratio of 1.3 per 1000 RSV infec-(3.8%) and oxygen therapy (4.5%).tion-related hospitalisations).

Trends in the estimated number of RSV infec-In 2000, the gender and racial breakdowns of tion-related hospitalisations, and corresponding in-

children <5 years old for the RSV infection-related cidence for children aged <5 years, are provided inhospitalisations were as follows: 57% male, 54.1% figure 2. The estimated incidence of RSV infection-White, 27.3% Hispanic, 13.3% Black and 5.3% related hospitalisations increased substantially fromother. The incidence rate of RSV infection-related 1993 to 2000, ranging from 335 hospitalisations perhospitalisations was approximately 50% higher in 100 000 children aged <5 years in 1993 to 521 perHispanic children (630 per 100 000 children) than in 100 000 in 1999. However, the associated estimatedeither White children (415 per 100 000) or Black total annual costs did not show a similar pattern overchildren (420 per 100 000). the same period. One possible explanation for this

was a noticeable reduction in the average length ofOnly 0.3% of RSV infection-related hospitalRSV infection-related hospital stays over time.stays involved a comorbid diagnosis of prematurity

(ICD-9-CM diagnosis codes 765.x, 641.2, 644.2).The prevalence of bronchopulmonary dysplasia Ambulatory Care(770.7), considered a significant risk factor for thedevelopment of RSV infection,[16] as a comorbid Table II contains the number of ambulatory/out-diagnosis was also very low (0.5%). Otitis media patient visits for RSV infection for the periodwas the most common comorbid condition for hos- 1997–2000. A noticeable decrease over time in thepital stays with a principal discharge diagnosis of number of office visits was observed between 1997acute RSV bronchiolitis (21.4%) or RSV pneumonia and 1999, but there are no discernible trends in the(20.1%). Other common comorbid conditions in- number of RSV infection-related visits for the othercluded dehydration, dyspnoea, asthma and acute care settings. Overall, otitis media accounted forrespiratory distress. about 80–90% of all RSV infection-related outpa-


280 Paramore et al.

Num

ber

of h

ospi

talis

atio

ns

Inci

denc

e of

hos

pita

lisat

ions

(pe

r 10

0 00

0)

Number of hospitalisationsIncidence of hospitalisations

391

453

521

436

335

404410369

0

20 000

40 000

60 000

80 000

100 000

120 000

140 000

65 939

1993

72 787

1994

79 992

1995

77 914

1996

83 313

1997

74 336

1998

98 744

1999

85 485

20000

100

200

300

400

500

600

Year

Fig. 2. Estimated annual incidence of respiratory syncytial virus-related hospitalisations among children <5 years old.

tient visits, by assuming 15% of all visits with otitis tient visits (4%, $US28 million). Based on the sensi-media diagnoses are caused by RSV infection. tivity analysis that assumed 5% (versus base case of

15%) of ambulatory/outpatient otitis media visitsFor visits with RSV diagnoses, the most fre-were RSV-related, the proportion of ambulatory/quently prescribed drugs in the office visit settingoutpatient RSV costs that were due to otitis mediawere salbutamol (albuterol) [prescribed in 60% ofwas reduced from 83% (base case) to 61%.visits] and amoxicillin (11%). In the ER setting, the

most frequently prescribed drugs were albuterol(61%) and amoxicillin (9%). Albuterol (54%) was Discussionalso the most frequently prescribed drug in the hos-pital outpatient setting, followed by amoxicillin This study indicates that RSV infection-related(15%). illness represents a significant medical and eco-

The most frequently performed tests/procedures nomic burden among children in the US;in the office setting were chest x-ray (20% of visits) ≈60 000–100 000 RSV infection-related hospital-and continuous negative pressure ventilation (15%). isations of children <5 years old are estimated toChest x-ray (45%), pulse oximetry (36%) and com- occur annually, at annual rates of ≈3–5/100 childrenplete blood count (27%) were the most common (figure 2) and at an estimated annual medical cost oftests/procedures performed in the ER setting. Skin $US341–449 million (figure 3). In addition, it isexamination (22%), hematocrit (8%) and chest x-ray estimated that each year there are approximately(7%) were the most common tests/procedures per- 1.5–2.2 million office visits, 180 000–250 000 hos-formed in the hospital outpatient setting. pital outpatient department visits, and 350 000–

400 000 ER visits that can be attributed to RSVinfection-related illness (table II). The ambulatoryTotal Medical Costsmedical care cost was conservatively estimated at$US258 million a year.Table III includes an estimate of the total direct

medical costs for RSV infection-related medical The study results also indicated that the estimatedencounters in the year 2000 (updated to year 2002 number of RSV infection-related hospitalisationsUS dollars). Hospitalisations accounted for 60% has trended slightly upward since 1993. A previous-($US394 million) of total costs ($US652 million), ly published study found that among children <1followed by office visits (22%, $US140 million), year old, annual hospitalisation rates for bronchi-ER visits (14%, $US90 million), and hospital outpa- olitis increased 2.4-fold between 1980 and 1996.[17]



The authors mentioned several possible causes for and $US8900 during the period 1998–2000. Thethe rising RSV infection-related hospitalisation rates mean charge for RSV pneumonia was $US14 961including improved survival rates for premature infor the 10-year period 1991–2000. Stevens et al.[19]

fants, changes in the ICD-9-CM diagnosis codes examined RSV infection-related hospitalisations oc-that included more RSV-specific diagnoses, and a curring at a neonatal hospital network in the Roches-lower sensitivity to hospitalise children. It has also ter, New York, USA from 1992 to 1996. Meanbeen suggested that the increase may be related to hospital charges ranged from $US6176 to 10 864,increasing rates of atopic disease and asthma.[18] depending upon week of age for the neonate. Lof-

land et al.[20] estimated mean RSV infection-relatedThe impact of prophylaxis on the national trendhospitalisation costs of $US10 486, for the periodof RSV infection-related hospitalisations is not1996–1997, based on assigning costs from a hospitalclear. Clinical trial data showed that palivizumabcost-accounting system to resource use collectedcould significantly reduce the risk of RSV infection-during a clinical trial. Finally, an economic evalua-related hospitalisation.[8] However, the follow-uption of RSV infection in Canada found that hospital-time in our study may not be long enough to observeisation accounted for 62% of total direct medicalthe impact of this medicine, as it was only approvedcosts.[21] We did not find any published studies thatin June 1998. The low proportion of the comorbidprovided either estimates of RSV infection-relateddiagnosis of prematurity and chronic lung diseasecosts in non-hospital settings or trend data.observed in our study may suggest some impact of

prophylaxis on RSV infection-related hospitalisa- The results of this study indicate that approxi-tions, but it may also be due to the possibility that mately 40% of the total direct medical cost of RSVthe diagnosis (especially prematurity) was not re- infection in children was from outpatient care. Otitiscorded correctly or completely at the time of hos- media was the major driver for the cost of outpatientpitalisation. care. In this study, we estimated from recent publi-

cations that, on average, 15% of all otitis media-The RSV infection-related hospitalisation ratesrelated outpatient visits (including physician office,and cost results estimated in our study compareER and hospital outpatient visits) were associatedfavourably with published estimates that were de-with RSV infection.[5-7] Anderson stated, after re-rived from independent data sources. Leader andviewing recent studies, that “10-20% of cases ofKohlhase[11] estimated that the annual rate of RSVacute otitis media have had RSV detected in middleinfection-related hospitalisation varied from 17.3ear fluids”.[5] Heikkinen suggested that the low pro-per 1000 infants (<1 year old) in 1998 to 26 per 1000portion of otitis media with identifiable virus wasin 1999, based on data from the National Hospitaldue to the well known limitations of viral cultureDischarge Survey. One study examined inpatientand antigen detection techniques.[22] On the otherbilled charges of RSV infection-related hospitalisa-hand, it is likely that RSV infection-specific dis-tions at Vanderbilt University Medical Center ineases were not appropriately recorded byNashville, Tennessee.[18] The mean charge for RSVICD-9-CM diagnosis codes in outpatient settings.bronchiolitis hospitalisations was between $US5400

Table II. Estimated number of ambulatory respiratory syncytial virus (RSV) infection-related visits for children <5 years in the US betweenthe years 1997 and 2000 by outpatient settinga

Setting 1997 1998 1999 2000

no. % of totalb no. % of totalb no. % of totalb no. % of totalb

Doctor’s office 2 200 000 3.8 2 060 000 2.6 1 500 000 2.5 1 720 000 2.2

Emergency room 385 000 4.1 367 200 4.2 358 000 3.9 401 800 4.1

Hospital outpatient 235 200 3.1 187 500 3.0 248 300 3.2 236 400 3.2

a Assumed 15% of all otitis media visits that were due to RSV.

b Proportion of total annual visits due to RSV in given setting.


282 Paramore et al.

Table III. Estimated average mean and total direct medical costs for respiratory syncytial virus (RSV) infection-related medical encountersfor children <5 years old in the US, 2000 ($US; 2002 values)

Inpatienta Office visits Emergency room Hospital outpatient

Average costs 4584

Physician feeb NA 33 71 63

Tests and procedures NA

RSV visitsc 66 308 38

otitis media visits 11 88 36

Drugs NA

RSV visitsc 49 38 29

otitis media visits 35 20 18

Total costs (million) 394 140 90 28

a Average billed charges from the Healthcare Cost & Utilisation Project (HCUP) adjusted to costs based on cost-to-charge ratio of0.53.[11]

b Physician fees were used to proxy for costs. Lowest level of visit complexity assumed in each setting (to provide most conservativecost estimate).

c Based specifically on International Classification of Diseases (9th Edition)-Clinical Modification diagnosis codes 079.6 (RSV), 466.11(acute bronchiolitis due to RSV) and 480.1 (pneumonia due to RSV).

NA = not available from HCUP, only total billed charges per hospital stay provided in data.

Therefore, the actual role of otitis media as a major charged during hospitalisations were not included incost driver for RSV infection-related outpatient care national surveys in the HCUP NIS. Third, the RSVmay not be as dominant as was found in this study. infection-related diagnoses utilised in the current

study (079.6, 466.11, 480.1) may not encompass allThe RSV infection-related costs reported in theof the potential medical encounters impacted bycurrent study most likely represent an underestimate

of the total RSV infection-related economic burden RSV infection. For example, Han et al. estimatedfor several reasons. First, the study considered only that approximately 2–9% of all pneumonia hospital-HCUP NIS hospitalisations with a primary diag- isations (ICD-9-CM 480–486) in the elderly are duenosis associated with RSV infection. In doing so, we to RSV infection.[23] They determined that for theexcluded an additional 20 000 hospitalisations in elderly alone the estimated annual cost of RSVwhich one of the RSV diagnoses was listed as a pneumonia hospitalisations in the US wassecondary diagnosis. Second, physician fees $US150–680 million (1996 values). Fourth, RSV

0

100

200

300

400

500

600

700

800

900

375

1993

428

1994

415

1995

366

1996

375

1997

341

1998

449

1999

394

2000

Ann

ual c

osts

(in

$U

S m

illio

ns)

Year

Fig. 3. Estimated trends in respiratory syncytial virus-related hospitalisation costs for children <5 years old, years 1993–2000.



infection may be associated with exacerbations of Acknowledgementschronic conditions such as congestive heart failure

Larry Liu is now employed at Pfizer.and chronic obstructive pulmonary disease.[23] A This work was funded by Wyeth Research. The authorslarge majority of these costs occur in the adult and have no conflicts of interest that are directly relevant to the

content of this study.elderly populations, and these costs have not beenincluded in this study. In addition, convincing evi-dence has indicated that bronchiolitis caused by References

1. Hall CB, McCarthy CA. Respiratory syncytial virus. In: Man-RSV infection in early life is strongly associateddell GL, Bennett JE, Dolin R, editors. Principles and practicewith an increased risk of developing asthma or reac- of infectious diseases. 5th ed. New York: Churchill Living-stone, 2000: 1782-801tive airway disease during the school years.[24,25]

2. Hall CB. Respiratory syncytial virus and parainfluenza virus. NAlthough the exact mechanism for the association isEngl J Med 2001; 344: 1917-28

not clear, it could add significant cost to the overall 3. Holberg CJ, Wright AL, Martinez FD, et al. Risk factors forrespiratory syncytial virus-associated lower respiratory illnessburden of RSV infection.in the first year of life. Am J Epidemiol 1991; 133: 1135-51

There are a number of limitations associated with 4. Boyce TG, Mellen BG, Mitchel Jr EF, et al. Rates of hospitali-zation for respiratory syncytial virus infection among childrenthis analysis. First, neither the HCUP nor NAMCS/in Medicaid. J Pediatr 2000; 137: 865-70NHAMCS databases included the actual costs of 5. Anderson LJ. Respiratory syncytial virus vaccines for otitis

care for reported services. Such information would media. Vaccine 2001; 19: S59-656. Pitkaranta A, Virolainen A, Jero J, et al. Detection of rhi-allow for a better estimation of the potential cost-

novirus, respiratory syncytial virus, and coronavirus infectionsbenefit of future RSV treatments. Second, the trend in acute otitis media by reverse transcriptase polymerase chain

reaction. Pediatrics 1998; 102: 291-5results may have been impacted by changes in the7. Vesa S, Kleemola M, Blomqvist S, et al. Epidemiology ofICD-9-CM diagnosis codes over time. Prior to 1996,

documented viral respiratory infections and acute otitis mediathe code for RSV bronchiolitis (466.11) was not in a cohort of children followed from two to twenty-four

months of age. Pediatr Infect Dis J 2001; 20: 574-81available. Third, the study utilised cross-sectional8. American Academy of Pediatrics, Committee on Infectiousdata that did not allow for a longitudinal tracking of Diseases and Committee on Fetus and Newborn. Prevention of

respiratory syncytial virus infections: indications for the use ofpatients and their full episodes of care for RSVpalivizumab and update on the use of RSV-IGIV. Pediatricsinfection-related illness. Fourth, the study focused1998; 102: 1211-6

solely on the impact of RSV on direct medical costs. 9. Stang P. The economic burden of respiratory syncytial virus-associated bronchiolitis hospitalizations. Arch Pediatr AdolescThere is doubtless also a significant impact on theMed 2001; 155: 95-6productivity (i.e. work loss, reduced leisure time) of 10. Howard TS, Hoffman LH, Stang PE, et al. Respiratory syncyti-

the caregivers of RSV patients. al virus pneumonia in the hospital setting: length of stay,charges, and mortality. J Pediatr 2000; 137: 227-32Cost-of-illness studies can provide useful infor- 11. Leader S, Kohlhase K. Respiratory syncytial virus-coded pedi-

mation on how healthcare resources are currently atric hospitalizations, 1997 to 1999. Pediatr Infect Dis J 2002;21 (7): 629-32being used for a given medical condition, and how

12. Center for Medicare and Medicaid Services (CMS). Data onthey might be better directed in the future. The hospitals’ cost-to-charge ratios. CMS public use files [online].

Available from URL: http: //cms.hhs.gov/providers/results of this study will be valuable in strategypufdownload/default.asp. [Accessed 2002 Sep 23]planning for the development of new therapies and 13. US Bureau of Labor Statistics. Consumer price index [database

interventions for RSV infection. online]. All urban consumers, US city average, medical care.Available from URL: http: //146.142.4.24/cgi-bin/surveymost?.cu. [Accessed 2002 Jul 31]

14. 2002 Drug topics® RED BOOK®. Montvale (NJ): MedicalConclusionEconomics, 2002

15. Physicians fee & coding guide (a comprehensive fee & codingreference). Augusta (GA): HealthCare Consultants of AmericaTreatment of RSV infection-related illness repre-Inc., 2002

sents a significant healthcare burden in the US. The 16. Kim HW, Arrobio JO, Brandt CD, et al. Epidemiology ofrespiratory syncytial virus infection in Washington, DC: I.economic impact of ambulatory care for RSV infec-importance of the virus in different respiratory tract diseasetion-related illness could be as important as that for syndromes and temporal distribution of infection. Am J

RSV infection-related hospitalisation. Epidemiol 1973; 98: 216-25


284 Paramore et al.

17. Shay DK, Holman RC, Newman RD, et al. Bronchiolitis- 23. Han LL, Alexander JP, Anderson LJ. Respiratory syncytialassociated hospitalizations among US children, 1980-1996. virus pneumonia among the elderly: an assessment of diseaseJAMA 1999; 282: 1440-6

burden. J Infect Dis 1999; 179: 25-3018. Bredenberg HK, Graham BS. Hospitalization costs of respira-

24. Martinez FD. Respiratory syncytial virus bronchiolitis and thetory syncytial virus infection. Pediatr Infect Dis J 2001; 20:1100-1 pathogenesis of childhood asthma. Pediatr Infect Dis J 2003;

19. Stevens TP, Sinkin RA, Hall CB, et al. Respiratory syncytial 22 (2 Suppl.): S76-82virus and premature infants born at 32 weeks’ gestation or

25. Sigurs N. Epidemiologic and clinical evidence of a respiratoryearlier: hospitalization and economic implications of prophy-laxis. Arch Pediatr Adolesc Med 2000; 154: 55-61 syncytial virus-reactive airway disease link. Am J Respir Crit

20. Lofland JH, Touch SM, O’Connor JP, et al. Palivizumab for Care Med 2001; 163: S2-6respiratory syncytial virus prophylaxis in high-risk infants: acost-effectiveness analysis. Clin Ther 2000; 22 (11): 1357-69

21. Langley JM, Wang EEL, Law BJ, et al. Economic evaluation ofCorrespondence and offprints: L. Clark Paramore, MEDTAPrespiratory syncytial virus infection in Canadian children: a

Pediatric Investigators Collaborative Network on Infections in International Inc., 7101 Wisconsin Avenue, Suite 600,Canada (PICNIC) study. J Pediatr 1997; 131: 113-7

Bethesda, MD 20814, USA.22. Heikkinen T. The role of respiratory viruses in otitis media.Vaccine 2001; 19: S51-5 E-mail: [email protected]


economic impact of respiratory syncytial virus-related illness in the us

Documents