The present analysis assessed efficacy and costs of prophylaxis with palivizumab versus absence of prophylaxis in the prevention of RSV infections in high-risk preterm infants. The (incremental) cost-effectiveness analysis was conducted from the perspective of the National Health System, updating and adapting a pre-existing lifetime decision-making model . We wish to point out that as in the previously published model, RSV infection mortality during the first hospitalization was estimated based on the results of the IMPACT study: mortality rates were noted to be 0.2% and 0.0% in subjects who received palivizumab prophylaxis and subjects who did not receive prophylaxis, respectively . Other probability rates of occurrence of the events investigated in the present study are also the same ones used in the original model , with the exception of the odds of developing recurrent wheezing within 24 months post-enrolment, data deriving from Simoes study .
We wish to briefly describe the changes made to the original model. Costs associated with high-risk preterm patients refer to 24 months after enrollment instead of 14 years so as to uniform the financial evaluation to the new data available in literature . Palivizumab administration scheme was changed, with three 50 mg vials and two 100 mg vials instead of the five 100 mg vials administered in the first model. Finally, all of the health care resources used by the patient were revaluated based on year 2007 costs and fees.
As regards general results, patients receiving palivizumab prophylaxis show better efficacy in terms both of LY (+ 0.088) and QALY (+ 0.159) compared to those who did not receive any prophylaxis. The best efficacy result achieved by palivizumab is confirmed also when differentiating patients per gestational age and presence of BPD.
On the other hand, in terms of the health care resources used, the mean cost per patient receiving prophylaxis (€ 6,244.20) was higher (+28.3%) compared to the estimated cost per patient without prophylaxis (€ 4,867.70). Therefore, due to the greater efficacy and greater costs associated with palivizumab versus the comparator (absence of prophylaxis), it was necessary to determine the incremental cost-effectiveness ratio per LY and QALY.
The calculation of the incremental cost per life year gained was based on an overall mean cost per patient of € 6,244.20 for palivizumab and of € 4,867.70 for absence of prophylaxis, with a mean survival of 29.842 years and 29.754 years, respectively. The incremental cost for prophylaxis (€ 1,376.50) was thus compared to an incremental efficacy of 0.088 years, leading to a cost per life year gained of € 15,568.65. In the case of the incremental cost per QALY the same mean total costs were recorded per patient (€ 6,244.20 and € 4,867.70), with however different outcomes. Palivizumab is characterized by 29.202 QALYs and absence of prophylaxis is associated with 29.043 QALYs. Incremental efficacy in this case is equal to 0.159 QALYs; therefore the incremental cost per QALY is € 8,676.74. Taking into account patient subgroups, the incremental cost effectiveness ratio per LY ranges between € 4,332.29 and € 28,417.08, whereas those per QALY range between € 2,731.81 and € 14,937.32.
All incremental costs calculated in this paper were finally compared against internationally acknowledged threshold values, which reflect the decision-making bodies' willingness to pay in order to achieve additional health units. Several international studies addressed the issue and established threshold values [22–25] and € 50,000 were set as the threshold below which a therapy is acceptable. Incremental costs per life year gained and per QALY calculated for palivizumab versus the absence of prophylaxis were always noted to be below the aforesaid threshold. The incremental cost per LY or per QALY of palivizumab was quite different observing the results of the three infant groups. The 33–35 weeks infant group showed the higher ICER per LY (€ 28,417.08) and per QALY (€ 14,937.32). If we considered a more restricted threshold value (< € 50,000), this third infant group (33–35 weeks) could be included in the prophylaxis programs, provided additional risk factors are present.
The univariate analyses conducted on several clinical and financial parameters taken into consideration in the model confirmed the robustness of the results, producing ICERs that were always below € 50,000. Also the threshold analysis conducted on the probability of an RSV hospitalization occurring for a patient receiving palivizumab prophylaxis estimated that to reach an incremental cost per LY or per QALY exceeding € 50,000 the efficacy of palivizumab would have to drop to non-realistic levels, i.e., down to efficacy levels equal to or even lower than those for patients without prophylaxis.
These results must be compared with those generated by other studies. Nuijten et al.  conducted in the United Kingdom an incremental cost effectiveness analysis for palivizumab in the prophylaxis of RSV infections in preterm infants with gestational age < 36 weeks and the possible presence of complications such as BPD or CHD (Congenital Heart Disease). The assessment was conducted from the perspective of the UK NHS (National Health Service), comparing administration of palivizumab to absence of prophylaxis. Applying a discount (3.5%) both to costs and outcome, an incremental cost per QALY equal to £ 16,720 for children with gestational age < 36 weeks and for children with BPD and an ICER per QALY of £ 6,664 for children with CHD. The results of the study conducted by Nuijten et al.  suggest that from the NHS viewpoint, palivizumab is cost effective compared to the therapeutic alternative represented by absence of prophylaxis.
Lazaro et al.  assessed efficacy and costs of palivizumab in the prevention of RSV infections in preterm children with gestational age in the 32–35 week range with two or more risk factors. The study was conducted from the perspective of the Spanish health care system, with absence of prophylaxis as the comparator. The Authors, thanks to a decision-making model, have calculated (with a 3% discount applied to efficacy and costs) an incremental cost effectiveness ratio per QALY of € 13,849. Once again, palivizumab was noted to be cost-effective versus absence of prophylaxis.
The study conducted in the USA by Elhassan et al.  intended to assess both the cost-effectiveness ratio of palivizumab as RSV prophylaxis in preterm children without BPD and the impact of decreasing recurrent wheezing risks in patients receiving prophylaxis on the cost effectiveness ratio. The Authors constructed two decision-making models, one of which took into account also risks of recurrent wheezing after an RSV infection and another one that did not consider this risk. The patients included in the model were preterm children with gestational age ranging from 26 to 32 weeks. Prophylaxis with palivizumab was compared against absence of prophylaxis. The results of the study conducted by Elhassan et al.  estimated that administration of palivizumab was cost effective only when taking into account also the benefits yielded by a decreased risk of recurrent wheezing in RSV infection patients.
The results of the present review present limitations and therefore several considerations must be taken into account when interpreting the data. As the study conducted by Simoes et al. has estimated the probabilities of developing recurrent wheezing episodes during 24 months of follow-up after the enrollment in preterm children with gestational age equal or less than 35 weeks, with or without RSV hospitalization, the present model is based on the assumption that such probability was identical for all children, without differentiating based on gestational age and the presence of BPD. Moreover, since the study performed by Simoes et al. did not report the odds of developing recurrent wheezing in preterm children receiving prophylaxis with RSV hospitalization, for this group of patients the same percentage (17%) calculated for preterm children with RSV hospitalization not receiving prophylaxis was conservatively adopted. To assess the impact of this double assumption, a sensitivity analysis was performed to test the probability of developing recurrent wheezing during the 24 months of follow-up post-enrollment, with a ± 10% change of the base value, assuming the worst case scenario for palivizumab; also in this case, the incremental cost effectiveness ratios per LY and per QALY show values lower than € 50.000.
Another limitation of the study might be that a pre-existing simulation model was adopted for the comparison between the two alternatives, and was adapted with clinical data deriving from international literature. This choice might be justified by the fact that to date no studies are available providing national data suitable for constructing the model population.
The dosage scheme (three 50 mg vials + two 100 mg vials) adopted herein for prophylaxis administration is different from that used in the original model. In support thereof, a dosage of 15 mg of active principle per kg of weight was considered, assuming that average weight of the preterm infant in the first three months of life was less than 3.4 kg, thus such as to justify the administration of a 50 mg vial; moreover the model conservatively included also the non-administered drug. Moreover, by means of the sensitivity analysis, also the five 100 mg vials administration assumption was taken into consideration, and also in this case ICERs per LY and per QALY remained lower than the € 50,000 threshold.
Finally, as already pointed out by the original study, a further limitation might be the use of average costs referring to an adult sample for the appreciation of the annual mean cost for recurrent wheezing of pediatric patients. However, comparing the cost included in the present model (€ 1,226.88) against that of children suffering from asthma (US$ 1,129) within a study conducted in the United Stated in 1999 , we have reason to believe that the data we used are conservative.