Dr Boyce’s group criticizes our study as biased on “confounding by indication” and states that therefore results should be dismissed. Because 41.5% of the otherwise healthy infants born at 29- to 32-weeks’ gestation in our population received ≥1 doses of palivizumab, a very large proportion of this population is represented. To decrease risk of confounding by indication, we carefully excluded infants who had claims suggesting a significant chronic illness that could affect risk for severe RSV disease. The full list of exclusion criteria was published in the Supplemental Materials. To reduce variation by socioeconomic status, our data set was restricted to a Medicaid-insured population. Variation by age group at start of RSV season was accounted for in our multivariate analyses. Furthermore, in the 29- to 32-week age group this variation was small, with only a 17-day difference in mean age between those who received ≥1 palivizumab doses and those who did not receive any.1 Variation in physician practice is the likely explanation for the most of the variation in rates of palivizumab prescribing that we observed.
Boyce et al claim that our results conflict with the results of the randomized controlled clinical trials on palivizumab. This assertion is incorrect. As stated in our discussion, our results are consistent with those of IMPACT-RSV trial.2 Among the subgroup of infants born at 29- to 32-weeks’ gestation who received ≥80% of recommended doses of palivizumab, we saw a statistically significant decrease in hospitalizations with an RSV diagnosis. The absolute magnitude of the difference was small because of the low rate of RSV hospitalization in this population. Those data are reported in Table 2 of our manuscript. Boyce et al cite a study performed in the Netherlands (the MAKI trial) to support the efficacy of palivizumab for infants born at 33- to 35-weeks’ gestation. However, the randomization procedures for that trial did not adequately balance the treatment groups for important variables known to influence outcomes; specifically, their control group had greater rates of maternal smoking, lower rates of breast feeding, and greater day care attendance.3
What raises substantial concern from our study results, and was ignored by Boyce et al, is the increase in hospitalizations for bronchiolitis without an RSV diagnosis associated with palivizumab administration. Dismissing that finding as “confounding by indication” is irresponsible. The most likely explanations are respiratory viral infections picked up in the pediatrician’s office (because palivizumab administration requires multiple trips to the pediatrician’s office) or false negative RSV tests (because viral load would probably be decreased by the palivizumab), both of which have evidence supporting them, as cited in the manuscript. Why was this finding not reported in the randomized controlled clinical trials? There are 2 plausible explanations. Perhaps it was not looked for, because the specific aim of the trials was to examine the impact on hospitalizations for RSV disease. The other plausible (and most likely) explanation is that infection control procedures were much better in the clinical trial than in many physicians’ offices, leading to more respiratory viral infections associated with visits to receive palivizumab in the real world than in the clinical trial setting.
As our data were used to validate the 2014 American Academy of Pediatrics (AAP) Guidance to not administer palivizumab to otherwise healthy infants born at ≥29 weeks’ gestation, the main criteria used by the AAP (that RSV hospitalization rates in the population of otherwise healthy infants born at 29 to 36 weeks’ gestation are low and that rates in infants born at 29 to 32 weeks’ gestation are not substantially different from those of infants born at 33 to 36 weeks’ gestation)4 are validated by our results. The finding of increased hospitalizations for bronchiolitis without an RSV diagnosis associated with palivizumab administration is concerning and lends support to the 2014 AAP guidance statement.
Dr Boyce’s group asserts that support from managed care organizations represents a conflict of interest to “justify cost reductions associated with palivizumab approvals.” It should be noted that 7 of the 9 participating managed care organizations are provider-sponsored, not-for-profit organizations. Managed care organizations are interested in delivering value to members. Cost–benefit analysis is very different from cost reduction, because cost–benefit analysis focuses on value delivered. Determining cost–benefit ratios is a legitimate and important interest of managed care organizations and provides important benefits to society.
FINANCIAL DISCLOSURE: Each of the authors receives salary support for service in the medical affairs sections of their respective health plans.
- Farber HJ,
- Buckwold FJ,
- Lachman B, et al
- ↵The IMpact-RSV Study Group. Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high-risk infants. The IMpact-RSV Study Group. Pediatrics. 1998;102(3 pt 1):531–537
- ↵American Academy of Pediatrics Committee on Infectious Diseases; American Academy of Pediatrics Bronchiolitis Guidelines Committee. Updated guidance for palivizumab prophylaxis among infants and young children at increased risk of hospitalization for respiratory syncytial virus infection. Pediatrics. 2014;134(2). Available at: www.pediatrics.org/cgi/content/full/134/2/e620
- Copyright © 2017 by the American Academy of Pediatrics