To describe the prevalence of combination vaccine use and the associated financial barriers faced by pediatric practices, and to identify determinants of adoption of combination vaccines.
Mailed national survey.
Pediatric practices during the period from August through October 2008.
Pediatricians randomly selected from the American Medical Association Masterfile.
Main Outcome Measure
Use of 1 of 2 infant combination vaccines (the diphtheria and tetanus toxoids and acellular pertussis, hepatitis B virus, and inactivated poliovirus [DTaP-HepB-IPV] vaccine or the DTaP, IPV, and Haemophilus influenzae type b [DTaP-IPV/Hib] vaccine).
We received 629 responses (response rate, 67%). Four hundred ninety-two pediatricians (78%) reported using 1 or both of the infant combination vaccines of interest (ie, the DTaP-HepB-IPV or DTaP-IPV/Hib vaccine). More than half of the respondents said their practice did not receive adequate reimbursement for the purchase and administration of vaccines in general. More than one-fifth reported not using 1 or more of the combination vaccines because of inadequate reimbursement for the cost of vaccine doses (23% of respondents) and/or vaccine administration (20% of respondents). The infant combination vaccines studied were less likely to be used by smaller practices, by those with a lower proportion of publicly insured patients, and by those with less inclusive state vaccine financing policies.
One in 5 pediatricians reported that inadequate reimbursement prevented their using 1 or more combination vaccines. Practice size as well as the proportion of children whose vaccinations are paid for by public funds appear to be important determinants of the adoption of combination vaccines.
Combination vaccines, which incorporate antigens against different diseases in 1 injection, are designed to reduce the number of injections needed for routine childhood immunizations. New combination vaccines introduced in the past several years (Table 1) include measles-mumps-rubella and varicella, or MMR-V (ProQuad; Merck & Co, Inc, Whitehouse Station, New Jersey); diphtheria and tetanus toxoids and acellular pertussis, hepatitis B, and inactivated polio vaccine, or DTaP-HepB-IPV (Pediarix; GlaxoSmithKline Biologicals, Rixensart, Belgium); HepB and Haemophilus influenzae type b, or HepB-Hib (COMVAX; Merck & Co, Inc); DTaP and Hib (TriHIBit; sanofi pasteur, Lyon, France); and, most recently, DTaP, IPV, and Hib (Pentacel; sanofi pasteur, Lyon, France) and DTaP and IPV (Kinrix; GlaxoSmithKline Biologicals, Rixensart, Belgium).
The Advisory Committee on Immunization Practices states that combination vaccines are generally preferred over separate injections of the equivalent component vaccines.1 They offer the benefit of reduced pain for children and may reduce the number of missed opportunities for vaccination. Without the use of combination vaccines, under the current immunization schedule, an infant could receive as many as 5 injections (possibly 6 if the influenza vaccine is included) during a single visit.2 Both parents and health care providers may believe this number to be excessive,3 and as a result, they may delay or forgo vaccines.4 Some data suggest that combination vaccines increase immunization coverage5,6 and improve the timeliness of vaccination.5,7 Combination vaccines may also have some drawbacks, including the administration of unneeded extra doses of some antigens. This can occur, for example, when 3 doses of DTaP-HepB-IPV are given in addition to the HepB vaccine dose recommended at birth.8
In addition, combination vaccines may have financial drawbacks for physician practices. Under fee-for-service contracts, which are nationally the most common mechanism of reimbursement for services, physicians typically receive a separate fee for each injection administered to cover vaccine preparation, injection, and documentation. When a combination vaccine replaces 2 or more injections with a single injection, physicians may lose income from vaccine administration fees. Physicians also receive a fee to cover the cost of purchasing each vaccine. Combination vaccines typically cost more than the equivalent component vaccines. For example, the combination vaccine DTaP-IPV/Hib costs about $73 to purchase in the private sector, whereas the DTaP, IPV, and Hib vaccines together cost $66.9 A recent qualitative study found that some pediatricians receive proportionally less compensation for the cost of purchasing these more expensive combination vaccines compared with noncombination vaccines.10
We conducted a national survey of pediatricians to describe the prevalence of combination vaccine use and the associated financial barriers faced by pediatric practices, and to identify determinants of adoption of combination vaccines.
The target population was US pediatricians who regularly provided vaccines to children. Although other practitioners do vaccinate children, we chose to survey only pediatricians because they provide the majority of vaccines to children in the United States.11 A random sample of 1045 pediatricians in active practice was obtained from the American Medical Association Masterfile. This Masterfile includes current and historical data for more than 940 000 residents and physicians and includes both members and nonmembers of the American Medical Association.12 Physicians' records are continuously updated through extensive data collection and verification efforts. The mailed, self-administered survey was fielded in 2 waves starting in August 2008. The first mailing contained a package of cookies as a token of appreciation. Respondents were also given the option of completing the survey online.
The study protocol was approved by the institutional review boards of Harvard Pilgrim Health Care, Boston, Massachusetts, and the Centers for Disease Control and Prevention, Atlanta, Georgia.
The content of the 17-item, closed-ended questionnaire was informed by our previous qualitative study, which included semistructured interviews with pediatricians.10 Domains included current and planned use of newer combination vaccines (specifically MMR-V, DTaP-HepB-IPV, HepB-Hib, DTaP-Hib, DTaP-IPV/Hib, and DTaP-IPV), perceived adequacy of reimbursement for combination vaccines, benefits of combination vaccines, and respondents' practice characteristics, including participation in Vaccines for Children (VFC). VFC is a federally funded program that provides vaccines free of charge to health care providers for children through 18 years of age who are eligible for Medicaid, uninsured, American Indian, or Alaska Native, or, under certain circumstances (depending on the state's vaccine financing policies), underinsured. VFC does not reimburse physicians for the administration of vaccines, and these fees are charged separately to the child's insurer.
We also used information on state vaccine financing policies from the Association of Immunization Managers.13 States were classified as having 1 of 5 general vaccine financing policies (from least inclusive to most inclusive): VFC only (state provides vaccines only for VFC-eligible children), VFC enhanced select (state provides all vaccines to VFC-eligible children but only certain vaccines to underinsured children), universal select (state provides all vaccines to VFC-eligible children and certain vaccines to both insured and underinsured children), VFC enhanced (state provides all vaccines for VFC-eligible and underinsured children), and universal purchase (state provides all vaccines for all children regardless of insurance status using federally purchased and state purchased vaccines).
We conducted descriptive analyses for use of combination vaccines and reimbursement as a barrier to combination vaccines. In the analysis of adoption of new infant combination vaccines, we defined the dependent variable as current use of either DTaP-HepB-IPV or DTaP-IPV/Hib. We chose these 2 vaccines because they are mutually exclusive and are administered on a similar schedule (at 2, 4, 6, and 12-15 months of age). Bivariate analyses were conducted for variables that included practice characteristics, state vaccine financing policy, type and adequacy of reimbursement, and perceived benefits of combination vaccines. The multivariate logistic regression model that predicted use of infant combination vaccines was constructed using factors that were associated with the outcome at P < .10 in bivariate analyses. We also opted to retain variables that were policy relevant in our final regression model. We calculated robust standard errors to control for the possibility of intrastate correlation beyond what would be explained by state vaccine financing policy.
Of the 1045 pediatricians to whom we mailed the survey, 629 completed it (607 mailed or faxed it after completion, and 22 completed it online). Of the other 416 pediatricians who did not complete the survey, 45 were ineligible, 34 had indeterminate eligibility (surveys were returned because of incorrect address information), and 337 did not respond. Given the eligibility rate of 93% among respondents, we estimated that 313 of the 337 nonresponders would have been eligible, yielding a response rate of 67%.14
There were no significant differences between eligible respondents and nonrespondents in terms of type of practitioner (doctor of medicine vs doctor of osteopathy), type of practice, or vaccine financing policy in the practice's state.
About half of the respondents were in single specialty practices, and most participated in VFC (Table 2). Approximately half practiced in states with VFC-only vaccine financing policies, and 4% were in states with universal vaccine financing, which is in keeping with the proportion of the population living in such states. The remaining respondents were in states with VFC enhanced, VFC enhanced select, or universal select vaccine financing policies.
The majority (85%) of the respondents reported that they were involved in their practice's decision making about vaccines. Those in smaller practices with 1 to 5 physicians were more likely to be involved in decision making compared with those in larger practices with more than 20 physicians (92% vs 54%).
Most respondents (70%) were using DTaP-HepB-IPV (Table 3). A smaller proportion (29%) reported currently using DTaP-IPV/Hib, and an additional 30% planned to use it in the next 12 months. Four hundred ninety-two pediatricians (78%) were using 1 or both of these new infant vaccines. Only 1% of pediatricians reported currently using MMR-V, which was not on the market at the time of the survey because of production issues.9,15 However, most pediatricians reported that they planned to use MMR-V again when it became available.
Only 7.5% of respondents reported deferring some injections at the 6-month visit to an extra visit to avoid too many injections at one time. However, respondents who did not use DTaP-HepB-IPV or DTaP-IPV/Hib were significantly more likely to report this practice (17% vs 5%; P < .001).
Almost all respondents believed combination vaccines to be safe (99%) and effective (99%). The majority of respondents (84%) felt that combination vaccines increased or would increase immunization coverage for their practice. When asked whether parents in their practice preferred combination vaccines to separate component vaccines, 88% of respondents agreed or strongly agreed.
Less than half of respondents agreed with the statements that their practice received adequate reimbursement for the cost of vaccine doses and for administration of vaccines in general (Table 4). The same number of respondents agreed with the statements that their practice received adequate reimbursement for the cost of combination vaccines and for administration of these vaccines in general. Moreover, about one-fifth of respondents reported that inadequate reimbursement for purchase (23%) and administration (20%) was preventing their use of 1 or more combination vaccines. Of those for whom reimbursement was preventing use, 56% reported that it was due to inadequate reimbursement for both the cost of purchase and the cost of administration of vaccines, 28% for purchase alone, and 16% for administration alone.
Almost all of the respondents (97%) believed that the financial stability of their practice depends on revenue from vaccines. Despite this, about half agreed or strongly agreed that practices should continue to offer combination vaccines, even if it meant losing revenue.
In bivariate analyses, use of infant combination vaccines was significantly associated with larger practice size, being in a multispecialty practice, having a higher proportion of children with public insurance, participating in VFC, and being in a state with a more inclusive vaccine financing policy (Table 5). Respondents who believed that they were adequately reimbursed for combination vaccines were more likely to report using them.
In the primary multivariate model (Table 6), multispecialty practices were significantly more likely to use combination vaccines compared with solo or 2-physician practices. Practices having the highest proportion of patients with public insurance were also associated with increased use, and those in states with more inclusive vaccine financing policies were more likely to be using combination vaccines. Participation in VFC was not included in this model because it was collinear with the percentage of patients with public insurance and explained less of the variance in the outcome. Similarly, larger practice size was collinear with type of practice but also explained less variance.
In secondary models, we added 2 additional predictor variables: (1) whether the respondent reported that reimbursement for the cost of the purchase of combination vaccines was adequate and (2) whether reimbursement for the administration of combination vaccines was adequate. When adequate reimbursement for vaccine administration alone was added to the primary model, it was significantly associated with combination vaccine use (odds ratio [OR], 3.2; 95% confidence interval [CI], 1.9-5.5). Likewise, when adequate reimbursement for vaccine doses alone was added to the primary model, it was significantly associated with vaccine use (OR, 1.9; 95% CI, 1.1-3.6). These 2 variables were collinear, so that when both were added to the model simultaneously, only reimbursement for the administration of combination vaccines was significantly associated with their use (OR, 2.7; 95% CI, 1.7-4.3) (Table 6). Other covariates in this secondary model had similar patterns to the primary multivariate model, although both type of practice and state vaccine financing policy became not statistically significant, likely because of the smaller number of respondents in this model as a result of missing data.
All models were also run using only those respondents who indicated that they were involved in decision making about vaccines for those practices. There were no significant differences between these models and those that included all respondents.
Our results suggest that the majority of pediatricians believe that they are not reimbursed adequately for combination vaccines. One in 5 pediatricians report that inadequate reimbursement prevents them from using of 1 or more combination vaccines.
A recent survey of pediatricians and family physicians conducted by Freed et al16 also documented the financial strain that immunizations place on practices. Results from the previous survey were in accord with our results, with most respondents saying they were not adequately reimbursed for the purchase (58%) or administration (51%) of vaccines. Approximately half of respondents said they had delayed the purchase of specific vaccines for financial reasons. Moreover, this financial strain is occurring in the context of perceived inadequate reimbursement for other primary care services, such as developmental screening,17 obesity counseling,18 and mental health care.19,20
Our multivariate analyses indicated that multispecialty practice type, greater proportion of patients with public insurance, and more inclusive state vaccine financing policies are associated with combination vaccine use. It is likely that those pediatricians in multispecialty practices are less reliant on vaccine revenue. Such practices may also have more buying power to purchase vaccines because they tend to be larger practices. This explanation is supported by the association of combination vaccine use with practice size in bivariate analyses.
The finding that practices with higher proportions of patients with public insurance had higher odds of combination vaccine use has important policy implications. Such patients are covered by VFC, and practices receive combination vaccines for these children from the government free of charge. Hence, for children covered by VFC, the potential for underreimbursement of practices for the purchase of the vaccines is eliminated. Participation in VFC was also associated with increased combination vaccine use, presumably for the same reasons. Similarly, practices in states with more inclusive vaccine policies (such as universal and universal select vaccine financing) were more likely to use combination vaccines. In fact, in states where all vaccines are provided free of charge, every practice surveyed reported use of combination vaccines. All of these findings suggest that, when practices are provided with vaccines free of charge, the remaining potential barriers to combination vaccine use may be attenuated. Another important factor identified in our multivariate analyses was perceived adequacy of reimbursement, presumably due to loss of administration fees when using combination vaccines.
For practices that purchase most of their vaccines from manufacturers, increasing reimbursement for the administration of combination vaccines may help to reduce some financial barriers. The American Academy of Pediatrics has successfully advocated for new Current Procedural Terminology codes for administration of combination vaccines (L. Walsh, manager, Committee on Coding and Nomenclature, American Academy of Pediatrics, e-mail communication, August 18, 2009). Such codes would better reflect the associated work for combination vaccines that is similar to the associated work for individual vaccines, which includes counseling on each of the vaccines in the combination and updating patient records and immunization registries with vaccine information. Recommendations for the valuation of these new Current Procedural Terminology codes were presented at the 2009 American Medical Association/Specialty Society Relative Value Scale Update Committee meeting, and they will be available for use in 2011.
Encouraging the use of combination vaccines may be particularly important in light of our finding that respondents who do not use combination vaccines were significantly more likely to defer some injections at the 6-month visit to avoid too many injections. The decision to defer vaccines because of a high number of injections may partially explain the finding of other studies5,7 showing that combination vaccines improve the timeliness of vaccination. Overall, however, the proportion of pediatricians deferring vaccinations was quite low at 7.5%.
Our finding that most pediatricians currently feel they receive inadequate reimbursement for combination vaccines differs from that of a physician survey conducted by Freed et al21 in 2005, in which only 12% of physicians surveyed reported a significant or moderate decrease in practice revenue as a result of using DTaP-HepB-IPV. There are a number of possible reasons for this discrepancy. At the time of the previous physician survey, 3 of the newest combination vaccines (ie, DTaP-IPV/Hib, DTaP-IPV, and MMR-V) had not yet been introduced. Other vaccines, such as human papillomavirus, also have been introduced since that time. These newer vaccines have tended to be more expensive and may be placing practices under additional financial strain if reimbursement is considered inadequate to cover their purchase.22,23
The primary limitation of our study is that we were not able to quantify the adequacy of reimbursement. For some pediatricians, inadequate reimbursement might mean losing revenue, whereas for others it might signify only a decreased profit margin. It is possible that pediatricians set a relatively low bar for adequacy of reimbursement, since approximately half said they felt that practices should offer combination vaccines even if it meant losing revenue. Regardless, about 20% of respondents reported that their perceived inadequate reimbursement was actually preventing the use of at least 1 combination vaccine.
There is also the possibility of response bias, in that pediatricians who felt strongly about poor reimbursement may have been more likely to respond to our survey. We did not find any significant differences between responders and nonresponders in the relevant characteristics available for analysis (type of practice, type of practitioner, and state vaccine financing policy). Moreover, even if all the nonresponders stated that they were adequately reimbursed, those reporting inadequate reimbursement would still be about 40% of all health care providers.
One additional issue is that our study was conducted during a national shortage of Hib vaccine. In an effort to optimize Hib vaccination, all immunization programs received a mix of monovalent Hib and DTaP-IPV/Hib vaccines starting in July 2008. Practices were encouraged to use DTaP-IPV/Hib when possible,24 and this may have influenced its use. Therefore, estimates for use of DTaP-HepB-IPV and DTaP-IPV/Hib should be interpreted with caution and in this context because use of DTaP-IPV/Hib may have increased during this time. However, the Hib vaccine shortage should not have affected perceived barriers to combination vaccines.
Inadequate reimbursement for the administration and purchase of vaccines is a barrier to the use of combination vaccines for some pediatric practices. However, these barriers may be attenuated by providing vaccines free of charge through state and federal programs. Reducing the remaining financial barriers to combination vaccine use by pediatric practices will most likely require collaboration between physician groups and health insurers.
Correspondence: Courtney A. Gidengil, MD, MPH, RAND Corporation, 20 Park Plaza, Ste 720, Boston, MA 02116 (firstname.lastname@example.org).
Accepted for Publication: April 28, 2010.
Author Contributions: Dr Gidengil had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Gidengil, Messonnier, and Lieu. Acquisition of data: Gidengil, Dutta-Linn, and Rusinak. Analysis and interpretation of data: Gidengil, Messonnier, and Lieu. Drafting of the manuscript: Gidengil and Lieu. Critical revision of the manuscript for important intellectual content: Gidengil, Dutta-Linn, Messonnier, Rusniak, and Lieu. Statistical analysis: Gidengil. Obtained funding: Messonnier and Lieu. Administrative, technical, and material support: Dutta-Linn, Messonnier, and Rusinak. Study supervision: Lieu.
Financial Disclosure: None reported.
Funding/Support: This study was funded by cooperative agreement U01 IP000143-01 from the Centers for Disease Control and Prevention, National Center for Immunization and Respiratory Diseases. Dr Gidengil was supported by grant T32 HS000063-13 from the Agency for Healthcare Research and Quality through the Harvard Pediatric Health Services Research Fellowship Program.
Role of the Sponsor: The Centers for Disease Control and Prevention was involved in the design of the study, interpretation of the data, and review and approval of the manuscript.
Disclaimer: The views in this article are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention, US Department of Health and Human Services.
Additional Contributions: We thank all of the pediatricians who participated in this study. In addition, we thank Ken Kleinman, ScD, for statistical advice and Grace Lee, MD, MPH, for assistance with the design of this study.
Courtney A. Gidengil, M. Maya Dutta-Linn, Mark L. Messonnier, Donna Rusinak, Tracy A. Lieu. Financial Barriers to the Adoption of Combination Vaccines by Pediatricians. Arch Pediatr Adolesc Med. 2010;164(12):1138–1143. doi:10.1001/archpediatrics.2010.222