...REPORTS... Childhood Vaccination Against Pneumococcal Otitis Media and Pneumonia: An Analysis of Benefits and Costs

Transcription

1 ...REPORTS... Childhood Vaccination Against Pneumococcal Otitis Media and Pneumonia: An Analysis of Benefits and Costs Derek Weycker, PhD; Erin Richardson, BA; and Gerry Oster, PhD Abstract Objective: To examine the economic benefits and costs of routine vaccination of children younger than 5 years of age against pneumococcal otitis media and pneumonia with the pneumococcal conjugate vaccine. Study Design: A decision-analytic model of the cumulative numbers of cases and costs to age 10 years of acute otitis media (AOM), tympanostomy and related procedures (TRP), and community-acquired pneumonia (CAP) in children who either did or did not receive the pneumococcal conjugate vaccine. Patients and Methods: Seven hypothetical cohorts of 1000 children, stratified by age at initial vaccination, were followed. Outcome measures include costs of vaccination, cumulative numbers of cases of AOM, TRP, and CAP to age 10 years, and related disease costs, including medical treatment and parental work loss. Results: Routine vaccination of 1000 children against pneumococcal infection would cost between $57,000 and $226,000, depending on age (the From Policy Analysis Inc, Brookline, Massachusetts. This study was supported by Wyeth-Ayerst, Saint Davids, Pennsylvania. Address correspondence to: Gerry Oster, PhD, Policy Analysis Inc. (PAI), Four Davis Court, Brookline, MA goster@pai2.com. recommended number of doses varies by age at initial vaccination). Acute otitis media, TRP, and CAP to age 10 years would decline by 139 to 330, 8 to 22, and 15 to 30 cases, respectively; costs of medical treatment and work loss would correspondingly decline by $56,000 to $138,000. Expected net economic benefits (benefits minus costs) of vaccination against pneumococcal otitis media and pneumonia range from -$88,000 to $15,000 for children less than 2 years of age, and from -$1,000 to $31,000 for those aged 2 to 5 years at vaccination. Conclusion: Routine vaccination against pneumococcal otitis media and pneumonia appears to be costincreasing for children less than 2 years of age who require multiple doses, but cost-saving for children aged 2 to 5 years who would require only a single dose of the vaccine. (Am J Manag Care 2000;6(suppl):S526-S535) Streptococcus pneumoniae is a commonly encountered pathogen in several infectious diseases in infants and children in the United States. 1-3 This organism accounts for 30% to 50% of all cases of acute otitis media (AOM), 2 which is the most common reason for physician office visits among children S526 THE AMERICAN JOURNAL OF MANAGED CARE JULY 2000

2 ... ECONOMIC ANALYSIS OF PNEUMOCOCCAL CONJUGATE VACCINE... younger than 5 years of age. 4 S. pneumoniae is also one of the most common causes of community-acquired pneumonia (CAP). 2 Other serious, but less common, bacterial infections caused by S. pneumoniae include bacterial meningitis and bacteremia. 2 While not as frequent as AOM and CAP among young children, these infections can lead to serious morbidity and death. The mortality rate among young children with pneumococcal meningitis ranges from 5% to 10%; among those surviving, 25% to 35% have permanent disability. 5 Despite the increasing prevalence of antibiotic-resistant strains of S. pneumoniae, pneumococcal infections are still generally responsive to antibiotic therapy. 1,2 However, a small but significant proportion of ear infections among young children do not respond to antibiotics, and many experience frequent recurrences of AOM or develop persistent middle-ear effusions that interfere with hearing and language development. 1 Such children are commonly treated with tympanostomy tubes. The limitations of antibiotic treatment and the increasing resistance of S. pneumoniae to common antibiotics highlight the importance of prevention. Polyvalent pneumococcal polysaccharide vaccines have been licensed in the United States since 1977, but are ineffective in children younger than 2 years of age; they also have failed to demonstrate efficacy against otitis media on a consistent basis. 3,6 Recently, a 7-valent pneumococcal conjugate vaccine (PNCRM7) was approved for use in the United States for the prevention of invasive pneumococcal disease in infants and children. In a large randomized, double-blind clinical trial, the PNCRM7 was found to be 100% effective against invasive pneumococcal disease, and to reduce the risk of AOM by 7%, tympanostomy tube placement by 20.3%, and clinical pneumonia by 11% among infants who were vaccinated at 2, 4, 6, and 12 to 15 months of age, compared to controls. 7 Antibody responses to the PNCRM7 in children younger than 2 years of age as well as in older children have been reported to be generally higher than that with older polysaccharide vaccines. 5 The PNCRM7 also has been reported to be safe and well tolerated, 5,7 and the indication for the PNCRM7 is expected to soon be expanded to include the prevention of pneumococcal otitis media and pneumonia. A recent cost-effectiveness analysis of the PNCRM7 in healthy infants and young children found that vaccination would result in substantial reductions in the expected numbers of cases of pneumococcal-related disease and has the potential to be cost effective in children younger than 2 years of age. 8 The PNCRM7 also was found to be cost effective in children 2 to 5 years of age, especially for those in day care settings, who are at increased risk of pneumococcal infections. These results were largely based on the numbers and costs of AOM and pneumonia at a single institution, and the results varied significantly with changes in the assumed incidence and costs of disease in sensitivity analyses. In addition, conservative assumptions underlying the analysis (eg, loss of protective efficacy against pneumococcal disease at age 5 years) were made that most likely biased the analysis against vaccination. In this study, the economic benefits and costs of routine vaccination of children younger than 5 years of age with the PNCRM7 were examined using a model of the risks and costs of AOM, tympanostomy and related procedures (TRP), and CAP to age 10 years. This model was used to answer the following questions: 1) What would be the impact of routine vaccination on the expected numbers of cases of AOM, TRP, and CAP? 2) What would be its impact on the medical-care costs of VOL. 6, NO. 10, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S527

3 ... REPORTS... AOM, TRP, and CAP? 3) What would be the impact of the PNCRM7 on the costs of parental work loss? and 4) Would the economic benefits of vaccination outweigh its costs?... METHODS... Model Overview To examine the economic consequences of routine childhood vaccination against pneumococcal otitis media and pneumonia, a model of the risks and economic costs of AOM, TRP, and CAP in children younger than 5 years of age was used. For convenience, the economic benefits and costs of vaccination were evaluated in 7 hypothetical cohorts of 1000 children each, ages less than 7, 7 to 11, 12 to 17, 18 to 23, 24 to 35, 36 to 47, and 48 to 59 months, respectively, who either did or did not receive the PNCRM7. The expected numbers of cases and associated costs of AOM, TRP, and CAP from current age to age 10 years were then calculated considering these alternative scenarios. The costs of AOM, TRP, and CAP were assumed to include expenses for medical treatment and the value of illness-related work loss by parents. Expenses for medical treatment were assumed to include those for outpatient medical services, prescription medications, and hospitalizations. Illness-related work loss included the value of parental time lost from work to care for children who were sick. Consistent with other analyses of the economic consequences of pediatric vaccine programs, 9-13 attention was focused on the net economic benefits of vaccination, which were defined as the savings from vaccination in medical-care costs and work loss caused by AOM, TRP, and CAP minus the costs of vaccination. All costs were evaluated from a societal perspective, 14 expressed at 1999 average price levels, and all future benefits and costs were discounted at an annual rate of 3%, consistent with the recommendations of the US Public Health Service Task Force on Cost-Effectiveness in Health and Medicine. 15 Data Sources To estimate the safety and efficacy of the PNCRM7, data from the previously described clinical trial were used. 7 The risks and costs of AOM, TRP, and CAP in young children were estimated using the healthcare claims database of a large New England health plan. The plan provides firstdollar coverage (ie, with no deductible amounts) to approximately 1.2 million people. Data for this study consisted of all claims for services provided between July 1, 1997, and June 30, 1998, to plan members younger than 10 years of age as of June 30, 1998, who were continuously enrolled during this period (or from date of birth for those younger than 1 year of age). The final sample consisted of 125,362 children and was approximately uniformly distributed by age. To estimate the value of work loss, published results of interviews with parents whose children had experienced otitis media or pneumonia were used. 8 Medical-care costs and the value of work loss were adjusted to 1999 price levels using the medical care component and all items, respectively, of the Consumer Price Index for All Urban Consumers. 16 Expected Numbers of Cases of AOM, TRP, and CAP To estimate the expected numbers of cases of AOM, TRP, and CAP from current age to 10 years of age in the absence of vaccination, age-specific estimates of the expected numbers of cases per child were multiplied by the estimated number of cohort members remaining alive at each age. Age-specific estimates of the expected numbers of cases per child were estimated using claims data from the health plan. All professional service and hospital claims with a principal diagnosis of AOM (International S528 THE AMERICAN JOURNAL OF MANAGED CARE JULY 2000

4 ... ECONOMIC ANALYSIS OF PNEUMOCOCCAL CONJUGATE VACCINE... Code of Diseases, Ninth Edition, Clinical Modification [ICD-9-CM] 381, 381.0, 381.4, 382, 382.0, 382.4, 382.9) or pneumonia (ICD-9-CM , 487.0), or a procedure code for myringotomy with tympanostomy (Current Procedural Terminology, Fourth Edition) [CPT-4] codes 69420, 69421, 69433, 69436) or adenoidectomy (CPT-4 codes 42820, 42830, 42835) were selected; claims were then stratified by current age. Claims with listed diagnoses of AOM or pneumonia were deemed to represent unique cases if no other claims for otitis media (ICD-9-CM ) or pneumonia, respectively, or antimicrobial prescriptions were submitted on behalf of the same child within the previous 30 days. For TRP, each claim was assumed to be a distinct procedure unless accompanied by additional claims with a corresponding code for services performed on the same day. The age-specific annual number of cases was then divided by the total number of children in each age stratum. For children younger than 2 years of age, expected numbers of cases were estimated for periods corresponding to each of the age groups of interest (ie, 7 months for those less than 7 months of age, 5 months for those 7 to 11 months of age, and 6 months for those 12 to 17 and 18 to 23 months of age). The number of cohort members remaining alive at each age was estimated by multiplying the assumed initial size of the cohort (ie, 1000) by the proportion surviving to each future age, which was estimated using 1992 US Vital Statistics. 17 Because survivorship data are not reported for 7 months of age, the number of children surviving to 7 months of age was estimated by multiplying the initial size of the cohort by the neonatal (ie, 28-day) survival rate, and then subtracting 6/11 of postneonatal (ie, days ) mortality. Neonatal and postneonatal rates were used because mortality between birth and 12 months of age is highest in the first month of life. The expected survival of 18-month-old children was estimated similarly. Vaccine Efficacy In children 0 to 23 months of age at initial vaccination, vaccination was assumed to reduce the expected numbers of cases of AOM, TRP, and CAP Neonatal and post-neonatal rates were used because mortality between birth and 12 months of age is highest in the first month of life. until age 5 years (ie, period of full protective benefits) by 7%, 20.3%, and 11%, respectively; from ages 5 to 10 years, vaccine efficacy was assumed to decline by half (ie, period of partial protective benefits). In children 24 to 59 months of age at vaccination, efficacy was assumed to be the same as that in the younger cohort for the first 3 years (ie, period of full protective benefits), at which point it was assumed to decline by half and persist at that level to 10 years of age (ie, period of partial protective benefits). Costs of Vaccination The PNCRM7 was assumed to be administered as a separate injection in conjunction with other currently recommended childhood vaccines at the appropriate ages, and therefore would not require an additional office visit. Administration of the vaccine was assumed to vary by age at initial vaccination as follows, consistent with the manufacturer s recommendation: 3 doses plus a booster at 12 to 15 months for children younger than 7 months of age, 2 doses plus a booster at 12 to 15 months for children 7 to 11 months of age, 2 doses for children VOL. 6, NO. 10, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S529

5 ... REPORTS to 23 months of age, and 1 dose for children 24 to 59 months of age. 5 The average cost of the vaccine was assumed to be $52 per dose, and physicians were assumed to charge $5 for handling and administration, consistent with physician fees reported in similar analyses. 8 The former estimate was based on the weighted average of the manufacturer s direct list price ($58 per dose) plus the federal vaccine tax ($0.75 per dose), and the estimated contract price of the vaccine under the National Immunization Program (NIP) ($48 per dose). 5,18 The latter, which is not yet available, was estimated using the percentage discount for the varicella vaccine under this program (18.5%). 19 The NIP was assumed to purchase 60% of vaccine doses, based on unpublished data from the NIP (Bette Pollard, written communication, April 2000). The total cost of vaccination was therefore assumed to be $228 for children younger than 7 months of age at initial vaccination, $171 for children 7 to 11 months of age, $114 for children 12 to 23 months of age, and $57 for those 24 to 59 months of age. Costs of Treating AOM, TRP, and CAP The age-specific average direct medical-care costs for the treatment of AOM, TRP, and CAP from ages 0 to 10 years were estimated using the claims data from the health plan. To estimate the costs of AOM and CAP, professional service and institutional claims with a diagnosis of acute or other otitis media (ICD-9-CM ) or pneumonia (ICD-9-CM , 487.0), respectively, were selected in the 30-day period following the date of each episode, as well as any pharmacy claims for antimicrobial prescriptions; selected claims were then stratified by current age. Professional service and hospital claims with a CPT-4 code for TRP were excluded. Costs of each case were determined by summing the patient copayment and the amount paid by the insurer on all relevant claims. To estimate the costs of TRP, the patient copayment and the amount paid by the insurer were added together for all professional service and hospital claims on the day of the procedure as well as all follow-up care during the subsequent 15 days. Age-specific average direct costs for the treatment of AOM, TRP, and CAP ranged from $86 to $100, $1641 to $2259, and $146 to $503, respectively, depending on the age of the patient. Costs of Parental Work Loss The expected costs of parental work loss were estimated based on published results of interviews with parents whose children had experienced otitis media or pneumonia. 8 In this analysis, time spent away from work for the care of a sick child, valued at the parent s wage rate, was used to measure parental productivity loss associated with episodes of simple and complex otitis media, pneumonia with consolidation on chest radiograph, and tympanostomy tube placement. As work loss because of otitis media was reported separately for simple and complex cases, the weighted average of the 2 estimates was used. The value of work loss attributable to episodes of AOM, TRP, and CAP was estimated to be $182, $461, and $274, respectively. Sensitivity Analyses The sensitivity of results to changes in a number of key model parameters was examined, including the assumed efficacy of the vaccine during the period of full and partial protective benefits; the expected numbers of cases of AOM, TRP, and CAP at each age; the estimated costs of medical care and parental work loss per case of AOM, TRP, S530 THE AMERICAN JOURNAL OF MANAGED CARE JULY 2000

6 ... ECONOMIC ANALYSIS OF PNEUMOCOCCAL CONJUGATE VACCINE... and CAP; and the costs of vaccine administration.... RESULTS... Among 1000 unvaccinated children younger than 5 years of age, the estimated numbers of cases of AOM, TRP, and CAP to 10 years of age were 2059 to 5066, 47 to 124, and 154 to 333, respectively (Table 1). The associated (discounted) cost of medical treatment was estimated to total $292,000 to $727,000; the value (discounted) of parental work loss was estimated to be $411,000 to $970,000 (Table 2). Vaccination against pneumococcal disease would cost $57,000 to $226,000 per 1000 children; the recommended regimen varies, depending on age (Table 3). The expected numbers of cases of AOM, TRP, and CAP to 10 years of age was estimated to decline to 1920 to 4765, 39 to 102, and 139 to 303, respectively. The total (discounted) cost of medical treatment was estimated to decline to Table 1. Numbers of Cases of Acute Otitis Media, Tympanostomy and Related Procedures, and Community-Acquired Pneumonia to Age 10 Years per 1000 Unvaccinated and Vaccinated Children, by Current Age Tympanostomy and Community- Acute Otitis Media Related Procedures Acquired Pneumonia Age (months) Not Vaccinated Vaccinated Not Vaccinated Vaccinated Not Vaccinated Vaccinated < Table 2. Expected Costs of Acute Otitis Media, Tympanostomy and Related Procedures, and Community-Acquired Pneumonia to Age 10 Years per 1000 Unvaccinated and Vaccinated Children, by Current Age Not Vaccinated Vaccinated Benefits of Vaccination Medical Care Work Loss Medical Care Work Loss Medical Care Work Loss Age (months) (a) (b) (c) (d) (a-c) (b-d) < 7 $727,000 $970,000 $656,000 $903,000 $71,000 $67, $710,000 $956,000 $641,000 $890,000 $69,000 $66, $677,000 $906,000 $610,000 $844,000 $67,000 $62, $589,000 $797,000 $533,000 $744,000 $56,000 $53, $493,000 $686,000 $449,000 $642,000 $44,000 $44, $374,000 $530,000 $340,000 $495,000 $34,000 $35, $292,000 $411,000 $264,000 $383,000 $28,000 $28,000 VOL. 6, NO. 10, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S531

7 ... REPORTS... $264,000 to $656,000, and that of work loss to $383,000 to $903,000. On balance, therefore, vaccination against pneumococcal disease would increase total expected medical-care costs by $13,000 to $155,000 per 1000 children. Expected costs of work loss, however, would decline by $28,000 to $67,000. The net economic benefits of vaccination therefore would be -$88,000, -$34,000, $15,000, and -$5000 for children aged less than7, 7 to 11, 12 to 17, and 18 to 23 months, respectively, at initial vaccination; for children aged 24 to 35, 36 to 47, and 48 to 59 months at vaccination, they would be $31,000, $12,000, and -$1000, respectively. Sensitivity Analyses One-way sensitivity analyses were conducted on a number of key model variables (Table 4): vaccine efficacy during the period of full protective benefits, annual numbers of cases of AOM, TRP, and CAP, and the costs of Table 3. Expected Net Economic Benefits of the Pneumococcal Conjugate Vaccine per 1000 Vaccinated Children, by Current Age Benefits of Costs of Net Economic Vaccination Vaccination Benefits* Age (months) (a) (b) (a-b) < 7 $138,000 $226,000 ($88,000) 7-11 $135,000 $169,000 ($34,000) $129,000 $114,000 $15, $109,000 $114,000 ($5,000) $88,000 $57,000 $31, $69,000 $57,000 $12, $56,000 $57,000 ($1000) *Benefits of vaccination minus the costs of vaccination. Table 4. Sensitivity Analyses Net Economic Benefits Per 1000 Vaccinated Children, by Current Age Vaccine Efficacy Vaccine Efficacy* (period of (period of Numbers of Medical Care Work Loss Administration Age full benefits) partial benefits) Cases Costs Costs Costs (months) 50% 150% 0% 100% 50% 150% 50% 150% 50% 150% $13 < 7 ($157,000) ($19,000) ($110,000) ($66,000) ($157,000) ($19,000) ($124,000) ($53,000) ($122,000) ($55,000) ($120,000) 7-11 ($103,000) $33,000 ($57,000) ($13,000) ($103,000) $33,000 ($70,000) $0 ($68,000) ($2000) ($59,000) ($50,000) $78,000 ($8000) $37,000 ($50,000) $78,000 ($19,000) $47,000 ($17,000) $45,000 ($2000) ($59,000) $50,000 ($28,000) $18,000 ($59,000) $50,000 ($33,000) $23,000 ($32,000) $22,000 ($21,000) ($13,000) $75,000 $8000 $54,000 ($13,000) $75,000 $9000 $53,000 $9000 $53,000 $23, ($23,000) $46,000 ($5000) $28,000 ($23,000) $46,000 ($6000) $28,000 ($6000) $29,000 $3, ($29,000) $26,000 ($13,000) $9000 ($29,000) $26,000 ($15,000) $12,000 ($15,000) $12,000 ($10,000) *Vaccine efficacy during the period of partial protective benefits was varied between 0% and 100% of that during the period of full protective benefits. S532 THE AMERICAN JOURNAL OF MANAGED CARE JULY 2000

8 ... ECONOMIC ANALYSIS OF PNEUMOCOCCAL CONJUGATE VACCINE... medical care and work loss were each varied between 50% and 150% of baseline estimates. Vaccine efficacy during the period of partial protective benefits was varied between 0% and 100% of that during the period of full protective benefits; the cost of administration was increased to $13. 8,9 For scenarios biased against vaccination, vaccination increased total medicalcare costs and overall costs in each age group (except for those who were 24 to 47 months of age). For scenarios favoring vaccination, net economic benefits were positive in each age group (except for those who were 0 to 11 months of age).... DISCUSSION... In this study, the economic benefits and costs of routine vaccination of children younger than 5 years of age with the PNCRM7 were examined using a model of the risks and costs of AOM, TRP, and CAP to 10 years of age. The results of this study indicate that vaccination would yield substantial reductions in the expected numbers of cases of AOM, TRP, and CAP, and related costs of medical treatment and work loss. Net economic benefits of vaccination were estimated to range from -$88 to $31 per vaccinee, depending on age. Similar studies have estimated the economic impact of other childhood vaccines, including the varicella and hepatitis B vaccines. In a study of varicella vaccination of preschool children (and 12-year-old children in a catch-up program), total cost savings (direct and indirect) were found to be $5.40 for every dollar spent on vaccination. 20 In another study of varicella vaccination, the estimated savings in total costs were only $2.40, but this second study used a somewhat higher vaccination cost and did not include costs for acyclovir treatment of chickenpox. 9 For hepatitis B vaccination of 12-year-old school children, study results indicated savings of $1.70 per $1 spent on vaccination. 21 Another study reported savings of $1.42 for every dollar spent on infant hepatitis B vaccination. 22 In comparison, findings of this study translate into savings of $0.61 to $1.54 in medical-care costs and work loss for every dollar spent on vaccination, depending on age. The estimates of the net economic benefits of vaccination against pneumococcal disease are not substantially different from those reported in a previous study, despite several important differences between the 2 analyses. 8 This study used higher estimates of the expected numbers of cases of pneumonia and tympanostomy tube placement for infants and children, while the previous study used higher estimates of the costs per case of otitis media and pneumonia. The higher rate of pneumonia among infants and children in the analysis presented here is likely related to the broad set of diagnosis codes used to define cases. Differences may also be the result of regional variation in pediatric diagnostic patterns. While higher rates of pneumonia favor vaccination, the definition of pneumonia employed in this analysis is consistent with that used in the previously described clinical trial. The model in the previous study included estimates of other medical-care and nonmedical-care costs (eg, nonprescription medications and productivity losses resulting from premature death or disability). The analysis presented here did not account for any effect of the PNCRM7 on the risk of invasive disease, while that of the previous study did. Also, in the analysis presented here, estimates of the costs of vaccination reflect the discounted price of the vaccine available under the NIP. Finally, the assumption was made in this study that protective efficacy would persist to 10 years of age for children who were 0 to 23 VOL. 6, NO. 10, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S533

9 ... REPORTS... months of age at initial vaccination, albeit at half the initial rate after the age of 5 years, while the previous study assumed that vaccine efficacy would persist only to age 5 years. Available information on the persistence of efficacy for pediatric bacterial vaccines is limited. A recent study of antibody levels and antigen responsiveness after immunization with Haemophilus influenzae type b (Hib) conjugate vaccines reported that the level of protection remained high in healthy children who were 4 to 5 years of age. 23 The actual rate of decline in the efficacy of conjugate vaccines, either pneumococcal or Hib, is thereafter unknown but is generally believed to be gradual. Because of the lack of data on vaccine efficacy at older ages and because of the decreased incidence of pneumococcal disease after 5 years of age, the previous study assumed that vaccine efficacy would persist to 5 years of age for infants and for a duration of 3 years for those 2 to 5 years of age at vaccination. In the study presented here, the assumption was made that vaccine efficacy from ages 5 to 10 years would be half that at younger ages for children aged 0 to 23 months at initial vaccination and would decline by half after 3 years to age 10 years for those aged 24 to 59 months at vaccination. The effect of this assumption was somewhat modest, as only about 30% of cases of AOM, TRP, and CAP occur between the ages of 5 and 10 years. Several limitations of this study should be noted. As previously mentioned, the economic impact of vaccination against invasive pneumococcal disease was not examined; while invasive disease is far less frequent than AOM and CAP, it is much more costly. Because the PNCRM7 has been found to be 100% effective against invasive pneumococcal disease among infants and young children, the analysis almost certainly understates the true economic benefits of vaccination. Also, a recent study found that the economic benefits of vaccination against invasive pneumococcal disease alone were substantial. 24 Possible effects of vaccination on mortality among vaccinees were not considered, nor was the value of lifeyears saved as a result of vaccination. In addition, possible benefits of herd immunity were not considered for unvaccinated children or adults, nor was any possible effect of vaccination on the severity of cases of AOM or CAP that occur among vaccinees considered. In the absence of data on the efficacy of the vaccine in reducing the risk of acute sinusitis, it was not included in the analysis presented here even though the vaccine may well be effective for this indication. Finally, any possible effect of vaccination on the emergence of multi drug-resistant bacterial strains was not considered despite the fact that vaccination may ultimately reduce the frequency of use of antibiotics in young children as well as the need for more expensive antibiotics. In conclusion, the analysis presented here suggests that routine childhood vaccination against pneumococcal disease in children younger than 5 years of age would result in substantial reductions in the expected numbers of cases of AOM, TRP, and CAP as well as related costs of medical treatment and work loss. Routine childhood vaccination against pneumococcal otitis media and pneumonia appears to be cost increasing for children less than 2 years of age who require multiple doses, but cost saving for those aged 2 to 5 years, who would require only a single dose of the vaccine. Acknowledgments We thank John Edelsberg, MD, MPH, and Vincent Ciuryla, PhD, for their contributions to this study. S534 THE AMERICAN JOURNAL OF MANAGED CARE JULY 2000

10 ... ECONOMIC ANALYSIS OF PNEUMOCOCCAL CONJUGATE VACCINE REFERENCES Berman S. Otitis media in children. N Engl J Med 1995;332: Centers for Disease Control and Prevention. Prevention of pneumococcal disease: Recommendations of the advisory committee on immunization practices (ACIP). MMWR Morb Mortal Wkly Rep 1997;46(RR-08): Shann F. Pneumococcus and influenza. Lancet 1990;335: Schappert SM. Office visits for otitis media: United States, Advance data from vital and health statistics. Hyattsville, Maryland: National Center for Health Statistics; Publication A pneumococcal conjugate vaccine for infants and children. Med Lett Drugs Ther 2000;42: Pneumococcal vaccine. Med Lett Drugs Ther 1999;41: Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Pediatr Infect Dis J 2000;19: Lieu TA, Ray GT, Black SB, et al. Projected cost-effectiveness of pneumococcal conjugate vaccination of healthy infants and young children. JAMA 2000;283: Huse DM, Meissner HC, Lacey MJ, Oster G. Childhood vaccination against chickenpox: An analysis of benefits and costs. J Pediatr 1994;124: Cochi SL, Broome CV, Hightower AW. Immunization of U.S. children with Hemophilus influenzae type b polysaccharide vaccine. JAMA 1985;253: Koplan JP, Preblud SR. A benefit-cost analysis of mumps vaccine. Am J Dis Child 1982;136: Koplan JP, Schoenbaum SC, Weinstein MC, Fraser DW. Pertussis vaccine: An analysis of benefits, risks and costs. N Engl J Med 1979;301: Schoenbaum SC, Hyde JN, Bartoshesky L, Crampton R. Benefit-cost analysis of rubella vaccination policy. N Engl J Med 1976;294: Eisenberg JM. Clinical economics: A guide to the economic analysis of clinical practices. JAMA 1989;262: Weinstein MC, Siegel JE, Gold MR, et al. Recommendations of the panel on cost-effectiveness in health and medicine. JAMA 1996;276: Bureau of Labor Statistics. Consumer Price Index (CPI), Medical care services. Available at: Accessed March 2, National Center for Health Statistics. Vital Statistics of the United States, 1992, Vol II, Mortality, Part B. Washington, DC: Public Health Service; Health Resources and Services Administration. (1999). National vaccine injury compensation program. Available at: htm. Accessed April 11, Centers for Disease Control and Prevention. CDC vaccine price list. Available at: pricelist.pdf. Accessed April 25, Lieu TA, Cochi SL, Black SB, et al. Costeffectiveness of a routine varicella vaccination program for US children. JAMA 1994;271: Krahn M, Guasparini R, Sherman M, Detsky AS. Costs and cost-effectiveness of a universal, school-based hepatitis b vaccination program. Am J Public Health 1998;88: Margolis HS, Coleman PJ, Brown RE, Mast EE, Sheingold SH, Arevalo JA. Prevention of hepatitis b virus transmission by immunization. JAMA 1995;274: Scheifele DW, Halperin SA, Guasparini R, Meekison W, Pim C, Barreto L. Extended follow-up of antibody levels and antigen responsiveness after 2 Haemophilus influenzae type b conjugate vaccines. J Pediatr 1999;135: Hueston WJ, Mainous AG, Brauer N. Predicting cost-benefits before programs are started: Looking at conjugate vaccine for invasive pneumococcal infections. J Community Health 2000;25: VOL. 6, NO. 10, SUP. THE AMERICAN JOURNAL OF MANAGED CARE S535