OBJECTIVE: To assess physician attitudes regarding school-located adolescent vaccination and influenza vaccination.
METHODS: From July through September 2010, a 20-item survey was mailed to 1337 practicing Colorado family physicians and pediatricians. Standard statistical methods were used to examine unadjusted and adjusted odds ratios of factors associated with physician support for school-located vaccination programs.
RESULTS: Overall, 943 physicians were survey-eligible, and 584 (62%) responded. More than half of physicians supported both school-located influenza and adolescent vaccination. However, fewer physicians supported school-located adolescent vaccination compared with influenza vaccination. More physicians supported school-located vaccination for their publicly insured patients compared with their privately insured patients. Some family physicians (32%) and pediatricians (39%) believed that school-located vaccination would make their patients less likely to attend well-child visits, and half of respondents believed that school-located vaccination would have a negative financial impact on their practice. In multivariate analyses, physicians concerned about the financial impact of school-located vaccination were less likely to support such programs.
CONCLUSIONS: Although a majority of Colorado physicians supported influenza and adolescent vaccination at school, they expressed concerns regarding the implications on their practice. Lesser support for vaccination of their privately insured patients and concerns regarding attendance at well-child visits suggests the perceived financial impact from school-located vaccination is a barrier and merits additional examination.
What’s Known on This Subject:
Implementing expanded vaccination recommendations has challenged primary care providers, who administer the majority of vaccines in the United States. School-located vaccination has been proposed as a means of increasing vaccination rates while reducing the burden on primary care providers.
What This Study Adds:
This study assesses physicians’ support for their patients’ receipt of vaccines in school. Additionally, it compares physicians’ support for adolescent versus influenza vaccination and compares support by insurance status of their patient.
Vaccination is a cornerstone of pediatric preventive medicine and provides substantial public health benefits by reducing childhood morbidity and mortality from vaccine-preventable diseases.1,2 In addition to the direct benefits experienced from vaccination, protection against disease for unvaccinated classmates, families, and the community can result if local vaccination rates are sufficiently high.3–7 Since 2000, the Advisory Committee for Immunization Practices has expanded recommendations for children and adolescents to include vaccines protecting against human papillomavirus, meningococcal disease, and pertussis, as well as annual influenza vaccination for all children >6 months of age.8–11 Implementing expanded vaccination recommendations has challenged primary care physicians, who administer the majority of vaccines to children in the United States.2,12–15 Sole reliance on primary care physicians to provide these vaccines may not achieve high coverage rates among children and adolescents.5,13,14
Supplementary venues to deliver vaccines (eg, schools, emergency departments, pharmacies) have been proposed as a means of increasing vaccination rates while reducing the burden on primary care physicians.2,14,16,17 School-located vaccination, also known as school-based vaccination, is a promising option for vaccine delivery16,18–20 but might not be successful without support from students’ primary care physicians. Recent research has focused largely on feasibility and sustainability of providing vaccines in schools.2,19,21 Studies of physician attitudes regarding school-located vaccination for their patients and the perceived impact on their practice are limited. A physicians’ recommendation is associated with a child’s vaccination status and studies have shown that physicians generally support vaccine delivery at supplementary venues.20,22–25 These studies did not examine whether physician support varies by the insurance status of their patients or by the type of vaccines offered.
We surveyed practicing Colorado physicians to (1) assess their attitudes, practices, and perceived barriers regarding school-located vaccination for their patients and (2) assess whether physician support differed for adolescent versus influenza vaccination at school and differed for their publicly versus privately insured patients. We hypothesized that more Colorado physicians would be supportive of school-located vaccination for their publicly insured patients compared with privately insured patients and that fewer would be supportive of school-located adolescent vaccination compared with school-located influenza vaccination.
A survey of Colorado family physicians and pediatricians was conducted as part of Denver’s In-School Immunization Program, a Centers for Disease Control and Prevention–funded study testing the feasibility of school-located vaccination. Colorado family physicians and pediatricians were surveyed from July through September 2010. The Colorado Multiple Institutional Review Board approved the study and did not require written informed consent.
The study population included physicians belonging to the Colorado chapters of the American Academy of Family Physicians and the American Academy of Pediatrics as of May 1, 2010. The Colorado membership of these organizations includes 2597 physicians: 1927 American Academy of Family Physicians member family physicians and 670 American Academy of Pediatrics member pediatricians. Two Colorado pediatricians recently sampled in a similar national physician survey were excluded.25 All 668 remaining pediatricians and 669 randomly sampled family physicians (35% of the total) received the survey. Resource limitations prohibited the sampling of all 1927 family physicians. The total sample population included 1337 physicians. Survey respondents were excluded if they were in residency training or not providing primary care for children <18 years of age.
Existing school-located vaccination literature was used to develop the survey instrument.13,17,21,22,25–28 A convenience samples of 25 Colorado family physicians and pediatricians pilot-tested the survey. The final 3-page survey included 20 questions. Using 4-point Likert scales, survey questions assessed physician support for school-located vaccination. Four outcomes were measured:
Physician support of school-located influenza vaccination for their privately insured patients
Physician support of school-located adolescent vaccination for their privately insured patients
Physician support of school-located influenza vaccination for their publicly insured patients
Physician support of school-located adolescent vaccination for their publicly insured patients
Additionally, the survey included questions regarding physician demographic and practice characteristics; general vaccination practices (eg, provision of seasonal influenza vaccine during the 2009–2010 season, provision of the pandemic H1N1 influenza A vaccine during the 2009–2010 season, participation in the Colorado state immunization registry, participation in the federal Vaccines for Children program, physician attitudes regarding school-located vaccination).
Physicians were surveyed by US mail using Dillman’s Tailored Design Method for survey administration.29,30 Physicians received an initial letter, followed by a mailed survey that included a $5 incentive.29,30 Nonrespondents received 1 to 3 additional surveys every 2 weeks after the initial mailing. Survey responses were accepted for 10 weeks after the first survey was mailed.
Simple means and frequencies were calculated, and summary statistics for respondents and survey questions were analyzed. Responses were compared and generally analyzed and presented by physician specialty because of documented differences between family physician and pediatrician vaccine-related attitudes and practices.24,31,32
Physician support was analyzed using multivariate logistic regression. We dichotomized physicians as either very supportive or not very supportive (aggregate of somewhat supportive, not very supportive, or not at all supportive) of school-located influenza and adolescent vaccination. This breakdown was used because only a small proportion of physicians were not very or not at all supportive. Factors associated with physician support for school-located adolescent and influenza vaccination for their privately and publicly insured patients were examined with bivariate analyses and multiple logistic regression models.33 Physician support for school-located influenza and adolescent vaccination were analyzed separately because of the theoretical difference in physician attitudes regarding influenza vaccination versus adolescent vaccination efforts.33
Separate multiple logistic regression models assessed physician support for school-located vaccination for their publicly and privately insured patients; factors associated with physician support were found to be similar across these 2 models. Because physicians expressed more concerns regarding school-located vaccination for their privately insured patients, our final analyses focused on privately insured patients.
Models were constructed by a purposeful selection method. Variables assessed by each model included physician demographics, practice characteristics, and physician beliefs regarding vaccination. Variables were included in the final model at the α = .20 level and considered significant at the α = .05 level.33 Models were examined for colinearity and goodness of fit. Respondents who did not answer all questions included in the final model were excluded from this analysis (n = 50). All statistical analyses used SAS 9.1.3 (SAS Institute, Cary, NC).
Among 1337 Colorado physicians in the original sample, 394 physicians did not meet inclusion criteria (did not provide primary care to children aged <18 years or were in residency [n = 361] or were undeliverable [n = 33]) and were excluded. Of 943 eligible physicians, 584 responded and were included in the analysis (overall response rate 62%).
Respondents included 300 family physicians and 284 pediatricians (Table 1). Respondents and nonrespondents did not differ significantly by specialty (P = .86) or by geographic area (assessed by Colorado city; P = .37). Approximately half of respondents were men, and >90% were medical doctors. A majority of respondents were in a private practice in a suburban setting, and had >2000 patients.
Figure 1 presents physician support for school-located vaccination. More than half of physicians supported both school-located adolescent vaccination and school-located influenza vaccination. However, fewer physicians supported school-located adolescent vaccination compared with influenza vaccination. For example, for their privately insured patients, 50% of pediatricians were very supportive of school-located adolescent vaccination compared with 63% very supportive of school-located influenza vaccination (P < .001). More physicians supported school-located adolescent vaccination for their publicly insured patients compared with their privately insured patients (65% vs 55% very supportive, respectively, P < 0.001). A similar pattern was observed for school-located influenza vaccination. Figure 1 also illustrates that more family physicians were supportive of school-located vaccination compared with pediatricians consistently across vaccination type and patient insurance status.
Physician attitudes about school-located vaccination are presented in Table 2. As shown, some attitudes differed significantly between family physicians and pediatricians, whereas other attitudes did not. Most respondents thought school-located vaccination programs would be convenient for their patients and increase vaccination rates among their patient population. A small percentage of physicians believed that they would be relinquishing responsibility for their patients if they supported school-located vaccination. As seen in Table 2, respondents also expressed several concerns about the implications of school-located vaccination on practice finances and practice immunization delivery. For example, 10% of family physicians and 16% of pediatricians agreed that school-located vaccination would have a negative financial impact on their practice. A majority of family physicians and pediatricians agreed that if their patients received vaccines at school, their practice would have difficulty knowing how much vaccine to order. Thirty-two percent of family physicians and 39% of pediatricians agreed that school-located vaccination would make their patients less likely to attend well-child visits. Finally, 28% of family physicians and 29% of pediatricians expressed concern that if their patients were vaccinated at school, the practice would have difficulty keeping their vaccination records accurate.
Table 3 presents multivariate analyses of factors significantly associated with physician support for school-located influenza vaccination for their privately insured patients; Table 4 presents factors associated with physician support for school-located adolescent vaccination for their privately insured patients. Similar factors were found to be significant when modeling physician support for school-located vaccination for publicly insured patients (data not shown). For both school-located adolescent and influenza vaccination models, positively associated factors included the belief that school-located vaccination would be easier and more convenient for their patients and that it would increase vaccination rates in their practice. Factors negatively associated with physician support for school-located vaccination included the belief that they would be relinquishing responsibility for their patients if they supported vaccination at school, that school-located vaccination would have a negative financial impact on their practice, that it would put their patients at risk for overvaccination, that their practice would have difficulty estimating the amount of vaccine to order (adolescent vaccination only), and that school-located vaccination would put their patients at risk for vaccine administration errors (influenza vaccination only).
This study demonstrated that a majority of surveyed physicians supported school-located vaccination for their patients. Study results confirmed the hypothesis that fewer physicians were supportive of school-located adolescent vaccination, compared with influenza vaccination, and that more physicians were supportive of school-located vaccination for their publicly insured patients, compared with their privately insured patients. These observations may be a consequence, at least in part, of concerns about the financial implications of school-located vaccination on their practices. Successfully addressing the perceived barriers to school-located vaccination may increase the likelihood of future physician support but will require understanding these attitudes and beliefs among unsupportive physicians.
This study found modest physician support for school-located adolescent vaccination, which is consistent with other studies that have found that most physicians prefer for their adolescent patients to receive vaccines in the medical home.34,35 Some studies suggest that physicians prefer for their adolescent patients to receive vaccines in the medical home because of a link between the need to receive vaccines and attendance at routine well-child visits; patients receiving vaccines at school may be less likely to seek well-child care.17,34–37 Having expensive vaccines go unused because of an inability to estimate how many adolescent patients will be vaccinated at their physician’s office may be an additional consequence of school-located adolescent vaccination. Physicians were also concerned about keeping accurate records for patients receiving vaccines administered outside of the medical home, referred to as “record scatter.”38 Previous studies have suggested that record scatter is the greatest barrier to physician support for vaccination outside of the medical home.26,38 Findings from this study suggest that physician concerns regarding the financial implication of school-located vaccination of their patients are as or more significant than their concerns regarding vaccination record scatter.
We found that more physicians supported school-located influenza than school-located adolescent vaccination. The use of schools to deliver influenza vaccine has been proven effective and is now more widespread and familiar to physicians than ever before.16,20,25 The majority of respondents in our study believed that vaccine delivery at school was effective and safe and agreed that providing vaccines at school would increase vaccination rates among their patients. Potential contributors to the relatively robust support for school-located influenza vaccination include heightened awareness of influenza vaccination after the H1N1 influenza outbreak, minimal cost of influenza vaccine, and the overwhelming number of primary care visits required to fully implement universal annual influenza vaccination.13,15,28,39 Challenges to implementing universal influenza vaccination exclusively by primary care physicians have driven expanded use of school-located influenza vaccination as a viable delivery method.20,21,28 Future school-located influenza vaccination programs might encounter less physician resistance if this delivery method continues to expand toward widespread use.
Although previous studies have shown that physicians generally support school-located vaccination, they have not examined the impact of patient insurance type on that support. We found that more physicians were supportive of school-located vaccination for their publicly insured patients compared with their privately insured patients. Because a majority of physicians vaccinate both privately and publicly insured patients, this finding may result in physicians supporting the delivery of vaccines at school for some of their patients (publicly insured patients) and not others (privately insured patients).40 Most private insurance plans reimburse physicians for recommended vaccines for children and adolescents.41 Although the costs of vaccination to a physician often exceed the reimbursement by private payers,12,40 practices may be able to generate additional revenue through the charges associated with well-child, acute, or chronic care. Despite programs such as Vaccines for Children that purchase vaccine for uninsured patients or those with public insurance (ie, Medicaid), practices are challenged to collect revenue to cover vaccine administration costs. For Medicaid patients, reimbursement rates vary significantly by state and by type of coverage,42 and a recent study assessing reimbursement reported that physicians vaccinating a greater proportion of Medicaid patients showed increased financial losses.40 Financial pressure from vaccine costs and inadequate reimbursement rates have challenged physicians’ ability to provide vaccines to their patients,12 with certain physicians considering not providing vaccines or not vaccinating uninsured or underinsured patients.27,32 Given the cost of vaccination in a primary care setting and the complexities of vaccine financing, it appears that the insurance status of a patient and the prospect of reimbursement are key factors influencing a physician’s support for school-located vaccination programs.
Physicians often cite accurate vaccination records as an important concern about school-located vaccination programs.26 Without accurate records, physicians and supplementary venues risk overvaccination. However, keeping accurate records is challenging because it requires the participation and careful attention of all entities administering vaccines. The majority of physicians responding to our survey agreed that school-located vaccination programs might make maintaining accurate records difficult. Despite concerns regarding keeping accurate vaccination records, >30% of physicians in our study did not participate in the Colorado state immunization registry or were unsure of their participation status. Increasing widespread use of the state immunization registry by physicians and school-located vaccination program to maintain a single comprehensive and accurate vaccination record may mitigate physician concerns and increase the likelihood of support for school-located vaccination in the future.
There are limitations to this study and the generalizability of its findings. The responses from a survey of Colorado physicians may not reflect that of all physicians throughout the United States. The response rate of this study is average for a physician survey of this size.43 As is well-documented with mailed surveys, potential response bias is a limitation of this study. Additionally, physician support is a theoretical concept, the meaning of which is subjective to the individual physician. Because the study sample did not include nurses, nurse practitioners, physician assistants, or administrators who assist or provide care for children, drawing definitive conclusions about their attitudes would be speculative. A majority of our survey respondents (93%) were medical doctors in private practice (55%), which is consistent with expected demographics of Colorado child primary care providers.
To our knowledge, this is the first study to document physicians’ attitudes regarding school-located vaccination that compared support by patient insurance type and contrasted support for school-located adolescent versus influenza vaccination. Fewer physicians supported school-located adolescent vaccination compared with influenza vaccination and more physician supported school-located vaccination for their publicly insured patients compared with privately insured patients. Although the majority of physicians surveyed were supportive of school-located vaccination, concerns regarding the impact of school-located vaccination on practice finances, well-child care attendance, and the accuracy of their vaccination records were identified as possible barriers to a physician’s support of this delivery method. Efforts to address these physician attitudes directly might increase physician support and collaboration with future school-located vaccination programs and subsequently increase vaccination rates.
- Accepted June 14, 2012.
- Address correspondence to Emily V. McCormick, MPH, Public Health Prevention Service Fellow, Denver Public Health Department, 605 Bannock, 3rd floor, Denver, CO 80204. E-mail:
Ms McCormick was the primary author on this manuscript; she developed and administered the survey described and developed the analytic plan; she also participated in the revision of the manuscript and provided final approval. Mr Durfee conducted the multivariate analysis for this study and provided input on the limitations of different analytic tools. Dr Vogt provided subject matter expertise and participated in revision of the survey instrument and manuscript. Dr Daley provided subject matter expertise and participated in revision of the survey instrument and manuscript. Dr Hambidge provided subject matter expertise and participated in revision of the survey instrument and manuscript. Dr Shlay provided subject matter expertise and participated in revision of the survey instrument and manuscript.
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
FINANCIAL DISCLOSURE: Dr Shlay has spoken on vaccination at a presentation sponsored by Pfizer Inc; the other authors have indicated they have no conflicts of interest relevant to this article to disclose.
FUNDING: This work was supported by the Centers for Disease Control and Prevention.
- ↵Centers for Disease Control and Prevention. Ten great public health achievements—United States, 1900–1999. MMWR Morb Mortal Wkly Rep. 1999;48(12)241–243.
- Markowitz LE,
- Dunne EF,
- Saraiya M,
- Lawson HW,
- Chesson H,
- Unger ER
- Broder KR,
- Cortese MM,
- Iskander JK
- ↵Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep. 2010;59(35):1147
- Glazner JE,
- Beaty B,
- Berman S
- Szilagyi PG,
- Iwane MK,
- Schaffer SJ,
- et al
- Yoo BK,
- Szilagyi PG,
- Schaffer SJ,
- et al
- Schaffer SJ,
- Fontanesi J,
- Rickert D,
- et al.,
- Working Group on Complementary Settings
- Lindley MC,
- Boyer-Chu L,
- Fishbein DB,
- et al.,
- Working Group on the Role of Schools in Delivery of Adolescent Vaccinations
- King JC Jr,
- Cummings GE,
- Stoddard J,
- et al.,
- SchoolMist Study Group
- Taylor JA,
- Darden PM,
- Slora E,
- Hasemeier CM,
- Asmussen L,
- Wasserman R
- Oster NV,
- McPhillips-Tangum CA,
- Averhoff F,
- Howell K
- Davis MM,
- Ndiaye SM,
- Freed GL,
- Kim CS,
- Clark SJ
- Hull HF
- ↵Dillman D. Mail and Internet Surveys: The Tailored Design Method. 2nd ed. Hoboken, NJ: John Wiley & Sons; 2007
- Schulte JM,
- Bown GR,
- Zetzman MR,
- et al
- Freed GL,
- Cowan AE,
- Clark SJ
- Hosmer DW,
- Lemeshow S
- Broder KR,
- Cohn AC,
- Schwartz B,
- et al.,
- Working Group on Adolescent Prevention Priorities
- Kaplan DW
- Coleman MS,
- Lindley MC,
- Ekong J,
- Rodewald LE
- Lindley MC,
- Shen AK,
- Orenstein WA,
- Rodewald LE,
- Birkhead GS
- ↵The Department of Health Care Policy and Financing, Colorado Department of Public Health and Environment. 2012. Available at: http://www.colorado.gov/hcpf. Accessed Februrary 1, 2010
- Copyright © 2012 by the American Academy of Pediatrics