Objectives. To evaluate the costs and benefits of two new agents, respiratory syncytial virus immune globulin (RSVIG) and palivizumab, to prevent respiratory syncytial virus (RSV) infection among premature infants discharged from the neonatal intensive care unit (NICU) before the start of the RSV season.
Method. Decision analysis was used to compare the projected societal cost-effectiveness of three strategies—RSVIG, palivizumab, and no prophylaxis—among a hypothetical cohort of premature infants. Probabilities and costs of hospitalization were derived from a cohort of 1721 premature infants discharged from six Kaiser Permanente–Northern California NICUs. Efficacies of prophylaxis were based on published trials. Costs of prophylaxis were derived from published sources. Mortality among infants hospitalized for RSV was assumed to be 1.2%. Future benefits were discounted at 3%.
Results. Palivizumab was both more effective and less costly than RSVIG. Cost-effectiveness varied widely by subgroup. Palivizumab appeared most cost-effective for infants whose gestational age was ≤32 weeks, who required ≥28 days of oxygen in the NICU, and who were discharged from the NICU from September through November. In this subgroup, palivizumab was predicted to cost $12 000 per hospitalization averted (after taking into account savings from prevention of RSV admissions) or $33 000 per life-year saved, and the number needed to treat to avoid one hospitalization was estimated at 7.4. However, for all other subgroups, ratios ranged from $39 000 to $420 000 per hospitalization averted or $110 000 to $1 200 000 per life-year saved, and the number needed to treat extended from 15 to 152. The results were sensitive to varying assumptions about the cost and efficacy of prophylaxis, as well as the probability of hospitalization, but were less sensitive to the cost of hospitalization.
Conclusion. In our model, the cost of prophylaxis against RSV for most subgroups of preterm infants was high relative to the benefits realized. Lower costs might permit the benefits of prophylaxis to be extended to additional groups of preterm infants.
- respiratory syncytial virus infections
- cost-effectiveness analysis
- neonatal infections
- intensive care
- bronchopulmonary dysplasia
- chronic lung disease
- seasonal variation
- passive immunization
- RSV =
- respiratory syncytial virus •
- CLD =
- chronic lung disease •
- FDA =
- US Food and Drug Administration •
- RSVIG =
- respiratory syncytial virus immune globulin •
- AAP =
- American Academy of Pediatrics •
- NICU =
- neonatal intensive care unit •
- KPMCP–NC =
- Kaiser Permanente Medical Care Plan (Northern California hospitals) •
- NNT =
- number needed to treat •
- CI =
- confidence interval •
- CMIS =
- Cost Management Information System
Respiratory syncytial virus (RSV) is a major cause of morbidity and mortality in preterm infants. Younger infants and those with chronic lung disease (CLD) are at elevated risk of severe RSV disease.1–13 In the past 3 years, the US Food and Drug Administration (FDA) has approved two new agents, RSV immune globulin (RSVIG; Respigam, MedImmune, Inc, Gaithersburg, MD) and palivizumab (Synagis, MedImmune, Inc), for prophylaxis of RSV disease among high-risk infants. The American Academy of Pediatrics (AAP) recently published recommendations for the use of both RSVIG and palivizumab.14 The AAP identified the following as appropriate candidates for prophylaxis: 1) infants younger than age 2 years who currently receive or have recently required medical therapy for CLD, 2) infants born at ≤28 weeks' gestation who are ≤12 months old at the start of the RSV season, and 3) infants born at 29 to 32 weeks who are ≤6 months old at the start of the RSV season.
The high cost of both of these agents remains a concern.14 ,15 Limited cost-effectiveness data are available for RSVIG, and no published study has addressed the cost-effectiveness of palivizumab.16 ,17 Additional cost-effectiveness analyses of both RSVIG and palivizumab are needed to permit more refined recommendations regarding their use.14 ,15 Ideally, clinical policymakers should identify subgroups of preterm infants at the highest risk of severe RSV disease and recommend prophylaxis for those in whom it is most cost-effective.
We conducted a cost-effectiveness analysis to compare RSVIG and palivizumab for prophylaxis of RSV disease among infants born prematurely. The study is unique because 1) it is the first to evaluate the cost-effectiveness of palivizumab, 2) it incorporates newly available data about the varying risks of severe RSV disease among preterm infants, and 3) it is the first to evaluate how the cost-effectiveness of prophylaxis differs among subgroups of high-risk infants.
Decision Analysis Model
We constructed a decision analytic model that compared RSVIG, palivizumab, and no prophylaxis for a hypothetical cohort of premature infants (Fig 1). We adopted the perspective of the clinician or policymaker who, just before the start of the RSV season, must decide which premature infants discharged from the neonatal intensive care unit (NICU) during the previous 12 months should receive immunoprophylaxis. The analysis addresses infants who already are outpatients at the beginning of the RSV season, rather than those who are discharged from the NICU during the RSV season.
Without prophylaxis, an infant's subgroup-specific probability of hospitalization for RSV (P) was assumed equal to that found in a large retrospective cohort study of premature infants at KPMCP–NC.13 With prophylaxis, an infant's probability of hospitalization for RSV was assumed to be equal to (P)× (1−efficacy), where the estimate of efficacy was derived from randomized controlled trials of RSVIG or palivizumab. The analyses were repeated for each of the 8 risk groups identified in the KPMCP–NC cohort (Table 1).
The analysis took the societal perspective, ie, included both medical and work-loss costs. The primary outcome was cost per hospitalization averted (we report net cost—the medical and nonmedical savings associated with reduced hospitalization have already been subtracted from the numerator). Secondary outcomes included 1) cost per discounted year of life saved, and 2) the number needed to treat (NNT) to prevent one hospitalization for RSV. Quality adjustment was not used because of the lack of data showing that prevention of RSV infection alters quality of life beyond the immediate illness period. The NNT was calculated as 1/ARR, where ARR is the absolute risk reduction and is equal to probability (hospitalization among controls)−probability (hospitalization among infants receiving prophylaxis).18 When considerable uncertainty existed about costs or probabilities, we chose base-case assumptions favorable to prophylaxis to bias the analysis against our findings. Costs are reported in 1995 US dollars and are rounded to two significant figures.19 The model was programmed on an Excel spreadsheet (Microsoft Corp, Redmond, WA).
Data and Assumptions
The probabilities of events in the decision tree (Table 1) were derived primarily from 1) subgroup-specific risks of hospitalization in the KPMCP–NC cohort of premature infants, and 2) estimates of the efficacy of prophylaxis in the three published randomized controlled trials of RSV prophylaxis among premature infants.
Probability of Hospitalization for RSV
In the KPMCP–NC cohort of premature infants, 8 risk groups were identified that were defined by combinations of three variables: gestational age (≤32 weeks vs 33–36 weeks), length of oxygen therapy in the NICU (<28 days vs ≥28 days), and month of NICU discharge (December–August vs September–November).13 To calculate the cost-effectiveness of RSVIG and of palivizumab for each of these 8 groups, we used the subgroup-specific risk of hospitalization for RSV within the KPMCP–NC cohort. Groups are labeled A through H for ease of reporting (Table 1).
Efficacies of RSVIG and Palivizumab
Two randomized controlled trials of prophylaxis in high-risk infants with RSVIG have been published, with differing estimates of efficacy at preventing hospitalization for RSV.3 ,8 These two trials were pooled to provide a base-case estimate of efficacy of 48% (Table 1). Only one randomized controlled trial of prophylaxis with palivizumab has been published, and the 55% efficacy observed in that trial was used in the base-case analysis of palivizumab.12 Although subgroup analysis in the palivizumab trial showed that prophylaxis was more effective among infants without CLD (78%) than among those with CLD (39%), the results of the PREVENT trial of RSVIG demonstrated a nonsignificant trend in the opposite direction (20% vs 49%). Thus subgroup-specific efficacies were not used in the base-case scenario. Instead, efficacy was varied in a sensitivity analysis to permit consideration of the impact of differential effectiveness of prophylaxis.
Probability of Death, if Hospitalized for RSV
In calculating cost per discounted year of life saved, we estimated that mortality among high-risk infants who were hospitalized for RSV, with or without prophylaxis, was 1.2%. We derived this figure by pooling placebo and intervention subjects from the three trials.3 ,8 ,12 In this combined group, there were two deaths among 173 infants hospitalized for RSV, for a pooled mortality of 1.2% (95% confidence interval [CI]: 0–2.8%).
Costs are summarized in Table 2. These estimates, in 1995 US dollars, were derived both from internal KPMCP–NC data and from published sources. We used costs, rather than charges, to reflect more accurately “the true opportunity cost to society of the resources used.”20
RSVIG and Palivizumab
We estimated the cost of a 2.5 g vial of RSVIG at $500, based on a published wholesale pharmaceutical catalog.21Thus, at a monthly dose of 750 mg/kg, the per-dose cost of RSVIG was assumed to be $750 (750 mg/kg/dose × $500/2500 mg × 5 kg). We assumed further that the average infant who received prophylaxis would weigh 5 kg at the start of the RSV season, based on the 4.8 kg mean weight at study entry among infants enrolled on the IMpact–RSV trial of palivizumab.12 Finally, we assumed that no wastage of drug occurred and that because of the 4-month duration of the RSV season found in the KPMCP–NC cohort, each infant would receive 4, rather than 5, doses of prophylaxis. These assumptions biased the result toward improved cost-effectiveness of the intervention.13 Thus, the total direct per-infant cost of RSVIG for a single RSV season was assumed in our base-case analysis to be $3000.
We assumed that palivizumab would cost $901 per 100 mg vial.21 At a monthly dose of 15 mg/kg, again presuming no wastage, the per-dose cost of palivizumab was assumed to be $675 (15 mg/kg × $901/100 mg × 5 kg). Thus, the cost of 4 doses of palivizumab for a 5-kg infant during a single RSV season was assumed in our base-case analysis to be $2700.
We used internal KPMCP data to estimate the cost of administration of prophylaxis. For RSVIG, we assumed that each administration would take 3 hours, including the time required to obtain intravenous access if necessary, and that each visit for RSVIG infusion would cost $210. As a result, the total administration cost of four visits for RSVIG was assumed to be $840. For palivizumab, we assumed that 2 of 4 doses could be given at the same time as regularly scheduled physician visits and therefore would incur no additional administration cost. The other 2 doses would require brief nurse visits at $50 each.
Hospitalization for RSV
For each infant in the KPMCP–NC cohort who required hospitalization for RSV, we obtained an estimate of actual costs incurred from the KPMCP Cost Management Information System (CMIS). CMIS is a purchased cost accounting software package that integrates the regional utilization databases with the health plan's general ledger to provide fully allocated costs by department, medical center, patient, or service. CMIS derives data from utilization systems that record hospitalizations, nursing acuity levels, operating room time, outpatient visits, laboratory tests, and radiologic procedures. Overhead costs associated with administering a medical care program are allocated to unit costs via a step-down method.
Costs were obtained for all patients with RSV hospitalized entirely at KPMCP facilities during or after 1994, when CMIS became operational. Those patients who were hospitalized in part at non-KPMCP facilities, such as regional tertiary care centers, or whose hospitalizations occurred before 1994, did not have complete information available through CMIS. Costs for those patients with missing or incomplete CMIS data were imputed using a linear regression model (R2 = 0.95) that predicted the cost of hospitalization for RSV based on the number of days of mechanical ventilation, the number of ICU days without mechanical ventilation, and the number of general ward days.
We evaluated whether the cost of hospitalization differed by gestational age (23–32 vs 33–36 weeks), length of oxygen therapy in the NICU (<28 vs ≥28 days), or month of NICU discharge (December–August vs September–November). The mean cost for all 55 hospitalized infants was $8502. Infants who required ≥28 days of oxygen in the NICU tended to have costlier hospitalizations than those who required <28 days of oxygen ($13 518 vs $6061; P= .06), and infants discharged from the NICU in September–November tended to have higher costs than those discharged in December–August ($11 679 vs $5207; P = .08). Because neither risk factor achieved statistical significance, we used $8502 in the base-case analysis and evaluated the impact of varying cost assumptions in subsequent sensitivity analyses.
Finally, we assumed that each hospitalized infant would require an emergency department or acute-care clinic visit before admission. This visit was assumed to cost an average of $198.
Administration of Prophylaxis
We assumed an average wage of $11/h.22 We further assumed that each administration of RSVIG required 1 hour of travel time in addition to the 3-hour infusion time, and that one parent would accompany the infant at each visit. Thus, for four administrations of RSVIG, the total indirect cost was $176.
The work-loss cost associated with each visit for administration of palivizumab was considered to be $22 (1 hour of travel time and 1 hour for the visit itself). Because 2 of the four injections of palivizumab were assumed to occur during regularly scheduled pediatric visits and thus to incur no incremental direct or indirect administration costs, the total indirect cost of palivizumab administration incorporated into the model was $44.
Hospitalization for RSV
Among premature infants in the KPMCP–NC cohort who were discharged from the NICU before the start of the RSV season and then rehospitalized for RSV, the mean length of stay was 5.7 days.13 We assumed that one parent would remain with the infant at all times and considered 5 of every 7 days to be workdays. The estimated work-loss cost was therefore ($11/h × 8 h/d × 5.7 d/hospitalization × 5/7) = $358.
We assumed that the life expectancy of a premature infant being considered for anti-RSV prophylaxis was 75.2 years. This was equivalent to the average life expectancy of a 1-year-old infant in the United States in 1996.23 This assumption that a premature infant with or without CLD suffered no loss of expected life-years as a result of his or her high-risk condition biased our model toward improved cost-effectiveness.
Lost Productivity Resulting From Death of a Child
Our model did not incorporate the cost of lost productivity resulting from the death of a child from RSV infection into the base-case analysis of cost per hospitalization averted. However, we considered the potential impact of lost productivity in a sensitivity analysis.
In our base-case model, all future costs and benefits were discounted at a rate of 3% per year before averaging. Discounting did not impact the cost per hospitalization prevented (because all costs and outcomes occurred in the first year of the program), but it did affect the cost per life-year saved.
We evaluated the sensitivity of the model to variations in key assumptions over plausible ranges. Sensitivity analyses were performed separately for RSVIG and palivizumab and for each of the subgroups of premature infants. We varied 1) the efficacy of prophylaxis (30% to 80%), 2) the cost per dose of prophylaxis ($250 to $1250), and 3) the mean cost of hospitalization for RSV ($5000–$25 000). We also varied the probabilities of RSV hospitalization from half to double those assumed in the base case. This range was chosen because it encompassed the 95% CI for all groups (Table 1) and also allowed for the possibility of underdetection of RSV because of bias in the reference cohort. Bias could have occurred in the KPMCP study, for example, because of an unrepresentative sample population, differential loss to follow-up, or missed cases.13 Finally, we considered the effect of lost productivity among infants who died of RSV. This sensitivity analysis assumed a 1.2% mortality among high-risk infants hospitalized for RSV and a lifetime productivity (discounted at 3%) of $984 000.24
For all subgroups, palivizumab both prevented more RSV hospitalizations and cost less than RSVIG, and thus was the dominant strategy. Among the highest-risk subgroup (group A), palivizumab prophylaxis was estimated to cost $12 000 per hospitalization averted (after subtracting savings from prevention of RSV admissions). In comparison, RSVIG prophylaxis in the same subgroup cost $25 000 per hospitalization averted. At the other extreme (group H), palivizumab prophylaxis cost $420 000 per hospitalization averted, whereas RSVIG cost $690 000 per hospitalization averted. As an example of our calculations, Table 3 shows the detailed analysis of costs and benefits, comparing palivizumab, RSVIG, and no prophylaxis, for infants in group A.
Table 4 compares the costs per hospitalization averted of palivizumab and RSVIG prophylaxis, relative to no prophylaxis, for the 8 subgroups of premature infants. In addition, the Table shows the estimated subgroup-specific costs per year of life saved (discounted) of prophylaxis, as well as the NNT to prevent one hospitalization for RSV.
Sensitivity analyses are presented for the use of palivizumab among the 4 subgroups of infants with a gestation ≤32 weeks (groups A through D). Sensitivity analyses were conducted, but are not presented, for infants with 33- to 36-week gestation. For these less premature infants, either the number of infants who fell into the risk category in the reference cohort were too small and CIs around the probability of hospitalization too wide to permit meaningful conclusions (ie, in groups E and F), or costs were uniformly high regardless of assumptions (ie, in groups G and H). Sensitivity analyses also were conducted, but are not shown, for the use of RSVIG.
Efficacy of Prophylaxis
The model was sensitive to assumptions about the efficacy of prophylaxis at preventing hospitalization for RSV (Fig 2). Among infants in group B, when the efficacy of palivizumab was increased to 80%, the cost per hospitalization averted decreased to $24 000. Among infants in group C, when efficacy was assumed to be 80% (as seen in the subgroup without CLD in the IMpact trial), $35 000 was required to prevent a hospitalization.12
Cost of Palivizumab
The cost-effectiveness of palivizumab was very sensitive to the acquisition cost (Fig 3). For group A, the cost per hospitalization averted varied from −$610 to $29 000 as the cost per dose was increased from $250 to $1250. Among infants in group B, costs per hospitalization averted varied from $10 000 to $78 000 over the same range of vaccine price; among those in group C, costs extended from $17 000 to $110 000. Changing the weight of the infant at the start of the RSV season had the same effect as changing the cost of the agent: the cost per hospitalization averted varied linearly with the infant's weight.
Probability of Hospitalization
Palivizumab's cost-effectiveness was sensitive to varying assumptions about the likelihood of hospitalization (Fig 4). When the probability of admission was varied from half to double the base-case estimate, the cost per hospitalization averted ranged from $33 000 to $1500 in group A, from $88 000 to $15 000 in group B, and from $120 000 to $23 000 in group C.
Cost of Hospitalization
The cost-effectiveness of palivizumab was less sensitive to the mean cost of hospitalization (Fig 5). When hospitalization costs were varied from $5000 to $25 000, cost-effectiveness ratios ranged from $15 000 to (−$4500) in group A, from $43 000 to $23 000 in group B, and from $59 000 to $39 000 in group C.
Other Sensitivity Analyses
When lost productivity because of death was incorporated into the model, assuming a discounted lifetime productivity of $984 000, the cost of palivizumab per hospitalization averted decreased to $150 in group A, to $27 000 in group B, to $44 000 in group C, and to $150 000 in group D.
As expected, this study found that the cost-effectiveness of palivizumab and RSVIG varied widely among different subgroups of premature infants based on their risk of hospitalization for RSV. Palivizumab was the preferred intervention because it was both more effective and less costly than RSVIG. For the highest-risk subgroup—infants born at ≤32 weeks, who required ≥28 days of oxygen in the NICU, and who were discharged from the NICU within 3 months before the start of RSV season—palivizumab was projected to cost $12 000 per hospitalization averted. However, among all other subgroups, the cost of preventing a hospitalization was estimated to be >$35 000. The secondary outcomes varied greatly as well—the cost of palivizumab per year of life saved ranged from $33 000 to $1 200 000, and the NNT extended from 7.4 to 152.
There are no published norms that allow us to determine the value of preventing a hospitalization for RSV. However, in general the cost of prophylaxis against RSV infection appeared high, relative to the benefits realized. Robbins et al, using estimates of willingness to pay derived from a small sample of health providers, suggested that prevention of an RSV hospitalization in a high-risk infant might be worth a mean of $5787, with a range of $1325 to $8700.18Our analysis predicted considerably greater costs: in our base-case analysis, the only group of infants in whom prophylaxis cost <$20 000 per hospitalization averted was group A. Among all other groups, relatively extreme assumptions about the efficacy of prophylaxis, the direct cost of hospitalization, and the probability of hospitalization were required to reduce the cost per hospitalization averted to <$20 000. The analysis was most sensitive to the cost per dose of prophylaxis, and a reduction in the price of palivizumab therefore was the most direct way to improve cost-effectiveness ratios.
Three previous studies have evaluated the costs and benefits of RSVIG; none have considered palivizumab. Hay et al found that the use of RSVIG cost $24 000 per life-year saved.16 In contrast, our model showed that RSVIG cost $70 000 per year of life saved for the highest-risk infants, and substantially more for all other subgroups. There are several reasons for the difference. First, our assumptions about efficacy included data from both randomized trials of RSVIG, whereas the previous study considered only the more favorable National Institutes of Allergy and Infectious Disease trial. Second, Hay et al assumed a uniform probability of severe RSV disease among all preterm infants, whereas we have shown that the likelihood of hospitalization may differ 20-fold between those infants at highest risk and those at lowest risk.13 Third, we assumed a lower mortality in our base case than did Hay et al. Finally, the previous analysis used the unconventional assumption that benefits of RSV prophylaxis would include life-years saved among the potential children, grandchildren, etc, of those infants who would have died of RSV disease, thereby decreasing the cost-effectiveness ratio by almost half. In contrast, we used the standard approach of counting only life-years saved for those patients who received prophylaxis.
Robbins et al showed in an analysis of the NNT that 12 infants with CLD would need prophylaxis with RSVIG to prevent one hospitalization for RSV.18 Among premature infants ≤6 months of age without CLD, the NNT in their study was 63. Although our use of 8 subgroups rather than 2 makes direct comparison difficult, a review ofTable 4 shows that the results of the previous work generally are compatible with our own. Finally, O'Shea and colleagues found that the use of RSVIG prophylaxis resulted in estimated net costs to the health care system of $1689 to $5415 per recipient, but did not express their findings in cost-effectiveness terms.17
To weight the analysis against our ultimate conclusions, we made several assumptions in the base-case model that intentionally biased our findings toward improved cost-effectiveness of prophylaxis. For example, we ignored the possibility that the life expectancy of a former preterm infant may be shorter than that of the average 1-year-old child, we disregarded the likelihood of wastage of drug, and we did not account for toxicities related to the administration of prophylaxis. Most significant, because of the 4-month duration of the RSV season in northern California, we assumed that prophylaxis for a season would consist of 4 doses of prophylaxis rather than the 5 doses used in the clinical trials. This had the effect of reducing the cost of prophylaxis by ∼20%, a saving that might not be realized in areas with a longer RSV season. The substantial costs per hospitalization averted, despite these biases, reinforce our conclusion that for most premature infants, the cost of prophylaxis against RSV is high, relative to the benefits obtained.
Our findings have five important limitations. First, we could not reliably determine which infants in the KPMCP–NC cohort had CLD. As a result, our analysis could not address directly the cost-effectiveness of RSV prophylaxis among infants with CLD. We used oxygen therapy for ≥28 days in the NICU as a surrogate for CLD, but not all infants who required ≥28 days of oxygen will have clinically significant CLD.25 It is likely that infants with active CLD may make up a particularly high-risk subgroup among those infants with an oxygen requirement for ≥28 days. Our results therefore do not contradict the AAP recommendation that “… prophylaxis should be considered for infants and children younger than 2 years of age with CLD who have required medical therapy for their CLD within 6 months before the anticipated RSV season.”14
Second, our analysis was based on the subgroup-specific probabilities of RSV hospitalization found in our retrospective study of high-risk infants enrolled in KPMCP–NC. As we noted previously, it is conceivable that the risks observed in that study were falsely low, because of underdiagnosis of RSV and/or because of biases in the sample population.13 However, sensitivity analysis shows that the magnitude of these effects would have to be quite large to alter substantially our conclusions about cost-effectiveness (Fig 4).
Third, this cost-effectiveness analysis did not address directly infants whose discharge from the NICU occurred during the RSV season. In the KPMCP–NC cohort, 3.5% of 769 infants discharged during the RSV season required readmission for RSV during the same season, and the risk of readmission was unassociated with either gestational age or length of perinatal oxygen therapy.13 Our findings therefore cannot be used to guide decisions about prophylaxis for infants who are discharged from the NICU during an RSV season.
Fourth, it is likely that hospitalizations for RSV among some subgroups of infants are, on average, more costly than those among other subgroups. Unfortunately, the sample of RSV hospitalizations from which we derived our estimates of hospital costs did not have adequate power to detect this difference, and therefore we used mean costs in our analysis. Thus, it is probable that prophylaxis among higher-risk infants may be somewhat more cost-effective than our analysis suggests, whereas among lower-risk infants, prophylaxis may be even less cost-effective. Review of Fig. 5, however, shows that, except among infants in group A, the use of higher hospitalization costs in our model would not have altered our conclusions substantially.
Finally, it is possible that the efficacy of palivizumab may differ across subgroups of preterm infants. Indeed, in the IMpact–RSV trial, efficacy was 78% among preterm infants without CLD and 39% among those with CLD, a difference that achieved statistical significance.12 We chose not to model this difference because the results of the PREVENT trial of RSVIG did not show a similar effect.8 If such a difference were confirmed, it would tend to improve the cost-effectiveness of palivizumab among infants who required <28 days of oxygen, while decreasing cost-effectiveness among those who required ≥28 days of oxygen. As shown in Fig 2, however, greater efficacy among infants without CLD would be unlikely to alter the conclusions of this analysis fundamentally.
The relative cost-ineffectiveness of prophylaxis among most subgroups of infants argues for more restrictive recommendations for the use of both RSVIG and palivizumab than are currently in place. Our findings suggest that RSV prophylaxis is most appropriately reserved for infants with active CLD and for premature infants without CLD who have multiple risk factors for RSV hospitalization. Among other premature infants, it would be reasonable not to administer prophylaxis against RSV disease on the grounds of limited cost-effectiveness. However, lower costs might enable the benefits of these interventions to be extended to additional groups of premature infants.
We wish to thank Norris Lieu, PharmD, for his assistance with current pricing information for RSVIG and palivizumab.
- Received December 28, 1998.
- Accepted April 26, 1999.
Reprint requests to (T.L.) Department of Ambulatory Care and Prevention, Harvard Pilgrim Health Care and Harvard Medical School 126 Brookline Ave, Suite 200, Boston, MA 02215. E-mail: tracy–.
- Groothuis JR,
- Gutierrez KM,
- Lauer BA
- Groothuis JR,
- Simoes EAF,
- Hemming VG,
- et al.
- Cunningham CK,
- McMillan JA,
- Gross SJ
- Law BJ,
- De Carvalho V,
- and the Pediatric Investigators Collaborative Network on Infections in Canada
- The PREVENT Study Group
- Wang EEL,
- Law BJ,
- Stephens D,
- et al.
- ↵Wang EEL, Law BJ, Robinson JL, et al. PICNIC (Pediatric Investigators Collaborative Network on Infections in Canada) study of the role of age and respiratory syncytial virus neutralizing antibody on respiratory syncytial virus illness in patients with underlying heart or lung disease. Pediatrics. 1997;99(3). URL: http://www.pediatrics.org/cgi/content/full/99/3/e9
- ↵Nachman S, Navaie-Waliser M, Qureshi M. Rehospitalization with respiratory syncytial virus after neonatal intensive care unit discharge: a 3-year follow-up. Pediatrics. 1997;100(6). URL: http://www.pediatrics.org/cgi/content/full/100/6/e8
- The IMpact–RSV Study Group
- ↵Joffe S, Escobar GJ, Armstrong MA, Black SB, Lieu TA. Rehospitalization for respiratory syncytial virus among premature infants. In press
- Committee on Infectious Diseases and Committee on Fetus and Newborn
- Storch GA
- Hay JW,
- Ernst RL,
- Meissner HC
- O'Shea TM,
- Sevick MA,
- Givner LB
- ↵Siegel JE, Weinstein MC, Torrance GW. Reporting cost-effectiveness studies and results. In: Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996:176–213
- ↵Luce BR, Manning WG, Siegel JE, Lipscomb J. Estimating costs in cost-effectiveness analysis. In: Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996:176–213
- ↵Bindley-Western Wholesale Catalogue. Indianapolis, IN: November 5, 1998
- ↵US Bureau of Labor Statistics. Bulletin 2445. Employment and Earnings. March and June issues. http://www.census.gov/prod/1/gen/95statab/labor.pdf
- ↵Kranczer S. Mixed life expectancy changes. Stat Bull. 1996;77:29–36
- ↵Haddix AC, Teutsch SM, Shaffer PA, Dunet DO. Prevention Effectiveness: A Guide to Decision Analysis and Economic Evaluation. New York, NY: Oxford University Press; 1996
- Copyright © 1999 American Academy of Pediatrics