Abstract
Alternative immunization schedules increase the time a child is unvaccinated and require greater resources from providers. Understanding what drives interest in alternative immunization schedules can potentially inform the design of effective, targeted messages that help to reduce time spent counseling and decrease requests for alternative immunization schedules. This study used the Theory of Planned Behavior to explore associations between sources of vaccine information, parental vaccine concerns, peer norms for vaccine behavior and intentions to follow an alternative immunization schedule. We performed logistic regression using medical record data from a private pediatric practice in a large northeastern city. Routine data were recorded in the EMR by the pediatrician during an initial vaccine counseling conversation with the parent(s). Parents who received vaccine information from doctors were less likely to have immunization concerns while parents who got vaccine information from friends and family or from books were more likely to report specific vaccine concerns. Our multivariate analysis shows that number of reported vaccine concerns and concerns about the utility or necessity of vaccines are strongly associated with alternative immunization intentions. We also find a direct relationship between sources of information about vaccines and alternative immunization intentions. Our results suggest that vaccine concerns and non-physician information sources play an important role in alternative immunization intentions while communication from physicians may play an important role in addressing vaccine concerns and promoting adherence to the ACIP immunization schedule.
Keywords: childhood immunization schedule, parents, vaccination intention, vaccine hesitancy, information source
Introduction
Preventing disease through vaccines is one of the greatest public health achievements in the US.1 A key component of this success is the childhood immunization schedule recommended by the Advisory Committee on Immunization Practices (ACIP). This schedule is designed to maximally protect individuals and the population from vaccine-preventable childhood diseases (VPCDs) and to promote adherence to the schedule by linking immunization to routine well-child visits. A large body of evidence points to the effectiveness of this schedule in reducing the burden of VPCDs in the US. The schedule is estimated to save $14 billion in direct medical costs and $76 billion in indirect social costs per US birth cohort.2 However, the importance of adhering to the ACIP schedule is not universally accepted by parents.
While refusing all vaccines remains rare, parents are increasingly opting for alternative vaccination schedules that omit or delay receipt of one or more vaccines.3 In a recent national survey, 1 in 10 parents reported intentionally delaying vaccines.4 Physician reports also indicate that requests for alternative schedules have become a regular occurrence.5,6
This rising prevalence of alternative immunization schedules is cause for concern. Vaccine delay and refusal are costly in many ways. Alternative schedules increase the time a child is unvaccinated, which puts children at greater risk for contracting VPCDs or transmitting diseases to other children.7,8 Clusters of unvaccinated children are associated with a higher incidence of vaccine-preventable diseases such as measles, pertussis, and varicella.8-11 Even fully vaccinated children can be at risk when exposed to disease, given that vaccines are not 100% efficacious.12,13 Delay also makes it less likely an immunization series will be completed.14 In addition, alternative schedules carry financial costs, requiring additional office visits, provider time, and expensive single-antigen vaccines.
The purpose of this study was to examine factors associated with a parent’s intended choice of vaccine schedule using medical record data routinely collected during a well-baby visit at a private pediatric practice. Our analysis was informed by the Theory of Planned Behavior (TPB), a cognitive processing model which posits that behavioral outcomes are linked to behavioral beliefs and attitudes via behavioral intentions.15 This model is widely used in health behavior research and has been shown to predict health behavior.16,17 A simplified presentation of the model is shown in Figure 1. We focused on three components of the model: (1) the linkages between knowledge and behavioral beliefs; (2) the linkages between behavioral beliefs, subjective norms and behavioral intentions; and (3) the mediating role of behavioral beliefs between knowledge and intentions.
Figure 1. Schematic diagram of the Theory of Planned Behavior. Adapted from “The Influence of Attitude on Intentions,” by I. Ajzen and M. Fishbein, 2005, The Handbook of Attitudes, p.194.16
We first hypothesized that physician-derived information (knowledge) would be positively associated with the reporting of no concerns (behavioral beliefs) while information derived from non-physician sources would be positively associated with the reporting of specific vaccine concerns. Next, we expected to find that specific vaccine concerns (behavioral beliefs) and high levels of non-vaccination by peers (subjective norms) would be associated with a higher likelihood of intending to follow an alternative immunization schedule. Lastly, we expected to find that any direct relationship between information and intentions would be mediated through vaccine concerns. Our results illustrate the heterogeneity of parent concerns, demonstrate the association between vaccine concerns and alternative immunization intentions and additionally reveal the major role played by sources of vaccine information. A nuanced understanding of parents’ intended immunization plans—the direct antecedent to actual immunization behavior—is critical to the design of interventions aimed at encouraging adherence to the recommended ACIP schedule among vaccine-hesitant parents.
Results
We abstracted data about an initial vaccine counseling session from 237 unique medical records for clinical encounters between December 2009 and April 2011. The data come from routine parent-provider interactions that took place at a private pediatric practice in a large northeastern city. The practice serves a densely populated urban area with a population that is majority white (58.8%) and highly educated (46% with bachelor’s degree or higher).
Table 1 contains descriptive statistics for the predictor variables across the total sample and by the outcome variable. In our sample, 16% of parents intended to follow some kind of alternative vaccination plan. This outcome encompasses a variety of non-ACIP immunization schedule plans where parents intend to space out, delay or decline some or all vaccines. For example, 9% of the sample intended to decline or postpone all immunizations (not shown in Table 1). Vaccine concerns were even more prevalent (32%), with overtaxing the immune system being the most common concern. We also observed heterogeneity in peer norms and information sources. While the majority of parents reported getting vaccine information from their physician, a substantial proportion reported getting information from friends and family (19%) and the Internet (21%). More than one quarter of parents reported multiple information sources, while a small number reported no information source. The mean number of sources reported was 1.2.
Table 1. Sample summary statistics by intended schedule choice (%).
| Immunization intentions | |||
|---|---|---|---|
| Total | ACIP schedule | Alternative schedule | |
| Vaccine concerns | |||
| No concerns | 68 | 81 | 3 |
| Overtax immune system | 10 | 5 | 38 |
| Autism | 7 | 6 | 15 |
| Mercury | 3 | 2 | 10 |
| Aluminum | 4 | 1 | 21 |
| Utility/necessity of vaccines | 6 | 3 | 23 |
| Number of concerns | |||
| 0 | 68 | 81 | 3 |
| 1 | 23 | 16 | 56 |
| 2+ | 9 | 3 | 41 |
| Alternative immunization among peersa | |||
| None | 57 | 62 | 31 |
| Some | 30 | 30 | 33 |
| Most/all | 13 | 8 | 36 |
| Sources of information about vaccination | |||
| Doctors | 55 | 64 | 10 |
| Friends and family | 19 | 15 | 36 |
| Parent professional background | 14 | 12 | 23 |
| Internet | 21 | 18 | 36 |
| Books | 16 | 10 | 46 |
| Total reported number of sources | |||
| 0 | 11 | 13 | 5 |
| 1 | 59 | 62 | 49 |
| 2 | 24 | 21 | 38 |
| 3 | 5 | 5 | 5 |
| 4 | 0 | 0 | 3 |
| N | 237 | 198 | 39 |
|
aParent estimate of percent of friends with kids choosing to follow an alternative immunization schedule. | |||
Our first analysis, shown in Table 2, uses logistic regression to measure the association between sources of information about vaccines and specific vaccine concerns. The outcome is odds of reporting a specific concern. Each column in Table 2 represents a separate model with a single vaccine concern as the outcome. We use the same predictor variables, the set of information source variables, across all models. Parents who reported getting vaccine information from doctors were more likely to have no immunization concerns, controlling for other information sources reported (aOR 7.09, p < 0.001), while parents who reported getting vaccine information from sources other than their doctor showed lower odds of reporting no concerns. Conversely, parents getting vaccine information from non-physician sources were more likely to report specific concerns. Parents who got vaccine information from friends and family had higher odds of being concerned about overtaxing the immune system (aOR 4.23, p < 0.01), aluminum exposure (aOR 5.74, p < 0.05), mercury exposure (aOR 6.0, p < 0.05), autism (aOR 22.1, p < 0.001) and the utility/necessity of vaccines (aOR 3.54, p < 0.05), controlling for other sources of information. Parents getting information from books or from their own professional background were also more likely to report some specific concerns, such as overtaxing the immune system (aOR 5.72, p < 0.01) and mercury exposure (aOR 6.91, p < 0.05), respectively.
Table 2. Odds ratios from logistic regression models predicting vaccine concerns.
| No concerns | Overtax immune system | Aluminum | Mercury | Autism | Utility/necessity of vaccines | |
|---|---|---|---|---|---|---|
| Doctors | 7.09 | 0.22 | --- a | 0.56 | 2.21 | 0.29 |
| (2.92)*** | (0.13)* | — | -0.51 | -1.34 | -0.2 | |
| Friends and family | 0.07 | 4.23 | 5.74 | 6 | 22.1 | 3.54 |
| (0.03)*** | (2.14)** | (4.23)* | (5.42)* | (14.40)*** | (2.12)* | |
| Parent prof. background | 0.46 | 2.05 | 4.04 | 6.91 | 1.89 | 1.93 |
| -0.24 | -1.39 | -3.76 | (6.23)* | -1.73 | -1.47 | |
| Internet | 0.37 | 1.47 | 1.67 | 1.91 | 1.43 | 2.72 |
| (0.17)* | -0.78 | -1.34 | -1.77 | -0.93 | -1.67 | |
| Books | 0.14 | 5.72 | 6.2 | 1.22 | 4.49 | 0.77 |
| (0.07)*** | (3.05)** | (4.91)* | -1.45 | (3.10)* | -0.64 | |
| Observations | 237 | 237 | 237 | 237 | 237 | 237 |
a Variable not included in the model due to lack of variation on the dependent variable. Exponentiated coefficients; Standard errors in parentheses; * p < 0.05; ** p < 0.01; *** p < 0.001.
Next, we looked for bivariate associations between alternative immunization intentions and various explanatory variables (Table 3, column 1). With the exception of one information source variable—parent’s professional background—all of our explanatory variables show a significant bivariate association with alternative immunization intentions. Reporting no vaccine concerns and getting vaccine information from physicians were each associated with lower unadjusted odds of intending to follow an alternative immunization schedule (OR 0.01, p < 0.001; OR 0.07, p < 0.01). The remaining variables—individual vaccine concerns, number of concerns, alternative vaccination by peers and non-physician sources of information about vaccines—were all associated with increased odds of intending to follow an alternative immunization schedule. For example, parents concerned about aluminum exposure have 50 times higher odds of intending to follow an alternative immunization schedule. Odds of alternative immunization intentions increase the most for parents with 2 or more concerns, relative to parents with no concerns (OR 426.67, p < 0.001).
Table 3. Odd ratios from logistic models predicting intention to pursue an alternative immunization schedule.
| Unadjusted | Adjusted | |||
|---|---|---|---|---|
| Odds ratio | SE | Odds ratio | SE | |
| Vaccine concernsa | ||||
| No concernsb | 0.01 | (0.01)*** | – | – |
| Overtax immune system | 13.12 | (6.22)*** | 3.92 | (3.11) |
| Autism | 3.09 | (1.67)* | 2.25 | (2.25) |
| Mercury | 7.43 | (5.84)* | 1.26 | (1.55) |
| Aluminum | 50.84 | (54.81)*** | 16.11 | (24.05) |
| Utility/necessity of vaccines | 11.58 | (6.85)*** | 8.57 | (8.74)* |
| Number of concerns reported (Ref. = 0) | 1.00 | – | 1.00 | – |
| 1 concern | 110.00 | (114.47)*** | 41.59 | (50.09)** |
| 2 or more concerns | 426.67 | (474.24)*** | 119.67 | (191.88)** |
| Social norms about vaccination | ||||
| Alternative immunization among peersc (Ref. = None (0%)) | 1.00 | – | 1.00 | – |
| Some (25–50%) | 2.26 | (0.97) | 0.31 | (0.23) |
| Most/All (75–100%) | 8.97 | (4.26)*** | 3.84 | (3.60) |
| Sources of information about vaccinationa | ||||
| Doctors | 0.07 | (0.04)*** | 0.19 | (0.16)* |
| Friends and family | 3.14 | (1.22)** | 0.20 | (0.17) |
| Parent professional background | 2.28 | (1.00) | 7.84 | (7.05)* |
| Internet | 2.61 | (1.00)* | 1.20 | (0.83) |
| Books | 7.63 | (3.04)*** | 10.57 | (8.32)** |
| Observations | 237 | 237 | ||
a Individual dummy variables each regressed separately on the outcome variable; bVariable excluded from the full model due to collinearity with number of concerns; cParent estimate of percent of friends with kids choosing to follow an alternative immunization schedule; Exponentiated coefficients; Standard errors in parentheses; * p < 0.05; ** p < 0.01; *** p < 0.001.
Finally, to assess which variables independently predict alternative immunization intentions, we estimated a full logistic regression model with all of the explanatory variables except “no concerns”, which had to be excluded due to collinearity with number of concerns (Table 3, column 3). In the full model we find that only one vaccine concern remains significant when controlling for other concerns, number of concerns, peer norms and information sources. Parents’ concerned about the utility or necessity of vaccines have nearly 9 times higher odds of intending to follow an alternative immunization schedule (aOR 8.57, p < 0.05). The largest association we observe is for number of concerns; parents with 1 concern and parents with 2 or more concerns are substantially more likely to intend to follow an alternative immunization schedule relative to parents with no concerns (aOR 41.59, p < 0.01; aOR 119.67, p < 0.01).
The full model also allows us to assess whether the information source variables have a direct association with immunization intentions or whether the relationship is mediated through vaccine concerns. We see that while most of the individual concern variables and peer norms lose significance in the full model, most of the information source variables retain significance after adjusting for concerns and peer norms. Parents who report getting vaccine information from physicians have 80% lower odds of intending to follow an alternative immunization schedule (aOR 0.19, p < 0.05) while parents who get vaccine information from their own professional background have nearly 8 times higher odds (aOR 7.84, p < 0.05) and parents who get vaccine information from books have 10 times higher odds (aOR 10.57, p < 0.01) of intending to follow an alternative immunization schedule.
Discussion
This study finds that 1 in 3 parents voiced vaccine concerns to their pediatrician and > 1 in 10 parents intended to follow an alternative immunization schedule for their child. Consistent with prior research, we find that parents’ vaccine concerns are varied and many center on vaccine safety in general as well as specific vaccine ingredients in particular.14,18-21 The prevalence of alternative immunization intentions in our data also matches previous findings from retrospective reports of actual vaccine behavior.4 Although only a minority of parents requested alternative immunization schedules, these requests matter because they delay immunization of the children involved, leaving them at risk for serious disease. In practices less tolerant of vaccine hesitancy and refusal, requests for alternative schedules also have the potential to lead to dismissal for the patient or cause negative experiences for the physician.5,22
To better understand these requests for alternative immunization schedules, we used the Theory of Planned Behavior to guide our investigation into factors that predict alternative immunization intentions. We first explored the linkages between sources of vaccine information and vaccine concerns and found that receiving vaccine information from doctors is associated with a greater likelihood of reporting no vaccine concerns. This finding is in agreement with several previous studies.20,21,23-25 Non-physician sources of information, on the other hand, were associated with higher odds of reporting certain concerns. Getting information from friends and family and from books, in particular, showed increased likelihood of reporting several vaccine concerns. Though we do not know the specific content of these sources, particularly for friends and family, our findings suggest that non-physician sources of information are more effective at emphasizing the risks rather than the benefits of vaccines and the recommended childhood immunization schedule. This finding illustrates the important role that providers play in promoting vaccine uptake, and it also suggests the need to support physician efforts with additional pro-vaccine information sources that can reach concerned parents.
We next explored associations between vaccine concerns, peer norms, information sources and alternative schedule intentions. As expected, we found parents with vaccine concerns are more likely to intend an alternative immunization schedule. Though our results model intentions, they are consistent with various prior studies showing a link between parental vaccine beliefs and child’s vaccination status.26-29 Our findings add to this body of literature by showing that the association between concerns and alternative immunization intentions increases with number of concerns. Furthermore, we find a significant association between concerns about the utility and necessity of vaccines and alternative immunization intentions. These findings suggest that both parents’ overall level of worry about vaccines and specific beliefs about vaccines play a role in shaping vaccine choices. Finding effective messages and communication strategies to combat these specific vaccine concerns and to put parents at ease could help to reduce the likelihood of parents choosing an alternative schedule.
Contrary to our expectations and the predictions of the Theory of Planned Behavior, we did not find evidence to support the mediating role of vaccine beliefs between sources of vaccine information and immunization intentions. Rather, we observed a direct association between information sources and immunization intentions that remained in the full model. This finding suggests we consider the mediating role of perceived control, the third prong of the TPB, in the process of parents forming and implementing alternative immunization intentions. It could be the case that some information sources—the content of some vaccine-related books or certain types of professional experience, such as in the healthcare or pharmaceutical fields—give parents a greater sense of control over implementing their immunization preferences. For example, we know that the popular vaccine information book, The Vaccine Book: Making the Right Decision for Your Child by Dr Robert Sears, provides parents with recommendations for a modified immunization schedule that they can present to their pediatrician.30 However, we were unable to include measures of perceived control in our present analysis because there were no such measure available in our data. The data used in this analysis were collected for clinical purposes and therefore were not tailored to our investigation. We recommend that future studies look for ways to measure the role of “perceived control” in parents’ immunization decisions. It is also possible that the variables we used to proxy subjective norms and beliefs do not adequately represent these concepts.
We further acknowledge that the findings we have documented are cross-sectional associations only. What we capture is the relationship between knowledge, beliefs, norms and intentions as expressed by parents to their provider at a critical moment in time when parents are making decisions about their child’s immunization coverage and when children are at highest risk for contracting VPCDs. Nevertheless, this analysis is unable to comment on whether a dynamic relationship exists between these factors. It would be easy to imagine a back-and-forth cycle between information, beliefs and peer norms as parents seek out opinions of peers, turn to additional sources of information to either confirm or refute beliefs, and potentially change their minds. We also cannot rule out the possibility of confirmation bias or reverse causality. Parents may seek out like-minded sources of information to bolster their already-formed concerns,31 or they may report concerns, information sources, and peer norms that are consistent with their preferred schedule as a way to justify their choice. However, this type of rationalization seems less likely in the context of intentions as opposed to retrospective reports of behavior.
This study has additional limitations. The nature of our sample population—drawn from a pediatric practice characterized by willingness to accommodate alternative vaccination plans, may limit our ability to make conclusions about larger populations from these results. There is also the possibility of omission bias in our data collection methods due to the fact that coding open-ended parent responses into categories was performed by the physician in real time. It is possible that the physician’s selection of what to record and how to categorize it was influenced by issues most salient to the physician.
Despite these limitations, using vaccine intentions as the outcome has the advantage of accurately representing the decision-making process. Our data capture parents’ concerns and intentions at the time they are engaged in vaccine decision-making rather than after they have made the choice, which may be skewed by ex-post rationalization. An important next step would be examining how these intentions predict behavioral outcomes.
Materials and Methods
Study sample and data collection
Data were collected from a solo pediatric practice in a large northeastern city between December 2009 and April 2011. The practice receives many requests for alternative vaccination schedules, which are generally accommodated. Due to the frequent requests for alternative schedules, the provider added modules to the EMR to capture data on parents’ vaccine concerns, sources of vaccine information, and intended immunization plan. In consultation with the researchers, the provider also added a question about the estimated amount of non-vaccination within the parent’s peer network.
All information was recorded by the provider during the initial vaccine counseling conversation with the parent, which typically took place during the one-month well child visit unless the child joined the practice at an older age or there was insufficient time at the one-month visit to cover vaccines. The one-month visit was chosen because while no vaccines are typically given at that visit, the infant will be due for several vaccines at the subsequent two-month visit. Parents provided open-ended responses to questions about concerns, sources of information, peer vaccination, and planned schedule; these responses were categorized by the provider and captured in the EMR during the counseling session using categories determined a priori based on the provider’s clinical experience and in consultation with the researcher (AB).
The full data set consisted of 384 individual records, representing clinical encounters at which the initial vaccine counseling session took place. We retained 237 records for the analytic sample based on two criteria: (1) the infant was less than one year of age at the time of vaccine counseling, and (2) there were no missing data on the predictor variables. The age restriction excluded 140 cases from the sample, while 7 other records were missing data. We felt the age restriction was necessary given that the ACIP recommended childhood immunization schedule begins at birth. Parents who discuss vaccines with a pediatrician when the child is older than one year are already pursuing a particular immunization schedule, and thus, we could not accurately classify their chosen plan as an intention. Delayed vaccine counseling could also indicate transferring into the practice because of the practice’s tolerant policies, which could contribute to bias.
Measures
Dependent variables
The first outcome we examined was parental vaccine concerns. As part of the routine vaccine counseling conversation, the provider elicited open-ended concerns from parents, and then captured those responses in a set of 9 binary, non-exclusive indicator variables, which we have condensed to 6 vaccine concern variables. Five of these concern variables were pre-determined by the pediatrician: no concerns; overtaxing the immune system; risk of autism; mercury exposure; and aluminum exposure. The final concern variable in our analysis was a combination of three separate concerns identified by the pediatrician, which all relate to the utility or necessity of vaccines: preference for natural immunity; belief that vaccines are ineffective; and belief that diseases targeted by vaccines are too rare to necessitate immunization. We combined these three concerns into a single variable because they share a common theme and because of small cell sizes. The pediatrician also employed an “Other” field for concerns that did not fit into the preceding concern categories, which we have excluded from this analysis due to their miscellaneous nature.
The second outcome of interest was parents’ intended vaccination schedule as described to the provider during the initial vaccine counseling conversation. The medical record contained 7 categories for vaccination plans: the recommended ACIP schedule; spreading out vaccines; no vaccines for now; no hepatitis B vaccine; Dr. Sears’ schedule (characterized by spreading out vaccines specifically to limit exposure to aluminum, but identified only when parents requested this schedule by name); refusal of all vaccines; and unsure, will decide later, or other. Parents unwilling to commit at the time of the visit were recorded as unsure/other; these records were excluded from the analytic sample on the basis of not yet having clear immunization intentions. For this analysis we created a single binary outcome variable for alternative immunization intentions, where parents who had selected the ACIP immunization schedule were marked as 0 and parents who intended to follow any other schedule were marked as 1.
Independent variables
Sources of information about vaccines were our first set of independent variables. We retained five of the six categories used by the pediatrician (excluding only the “Other” category) and created a non-exclusive, binary variable for each one: doctors, friends and family, parent’s professional background, books, and the Internet.
To examine the influence of peer norms on immunization intentions, we constructed a variable for estimated prevalence of non-vaccination in parents’ peer group. Parents were asked by the provider to estimate the proportion of their friends with children who were choosing an alternative vaccine schedule, including choosing not to vaccinate at all. The physician routinely asks this question to elicit vaccine concerns parents may hear from their friends, to ascertain whether counseling needs to include preparing parents for negative vaccine comments from peers, and to assess risk of exposure to VPCDs from unvaccinated children in peer networks. We used this estimate as a proxy measure for normative beliefs about vaccine delay or refusal. The five response categories used by the pediatrician (0, 25, 50, 75, and 100% of children of parents’ peers who would be unvaccinated) were collapsed into three categories for this analysis: none (0%), some (25–50%), and most/all (75–100%).
Vaccine concerns, which was used as the outcome variable in our first set of analysis, was also used as an explanatory variable in our second set of analyses. We additionally created a categorical count variable for parents total number of reported vaccine concerns: 0, 1, and 2 or more.
Analysis
We used logistic regression to assess the cross-sectional associations between sources of vaccine information and immunization concerns, and between concerns, peer norms, sources of vaccine information and immunization intentions. We first estimated adjusted odds ratios for the association between sources of information about vaccines and parents’ reported vaccine concerns. Separate logistic regression models were estimated for each individual concern, including no concerns, regressed on the set of information source variables. Next, we estimated unadjusted odds ratios for the binary associations between individual vaccine concerns, number of concerns, peer norms and information sources with alternative schedule intentions. Finally we estimated adjusted odds ratios for the association between all explanatory variables and alternative schedule intentions in a full model. All data analyses were conducted using Stata 12 (Stata Statistical Software: Release 12. College Station, TX).
Conclusion
Vaccine hesitancy and requests for alternative schedules are on the rise. These trends are concerning because counseling hesitant parents requires more time from providers and alternative schedules increase time unvaccinated for children. More research is needed to identify the most effective counseling strategies and messages that encourage adherence to the ACIP schedule. Designing such strategies requires a thorough understanding of the correlates of vaccine intentions. Our findings add to this evidence base and provide specific insights about the associations between vaccine concerns, immunization information sources and immunization intentions and about the importance of understanding where parents are getting information from and what messages are being promoted by these sources. Our results also affirm the vital role physicians play in promoting adherence to the ACIP schedule. Important next steps include assessing how well the immunization intentions in these data predict children’s actual immunization status. Assessing change from intending an alternative schedule to implementing the ACIP schedule also represents an important opportunity to study how parental hesitancy was successfully overcome. We also recommend an assessment to examine the effectiveness of specific vaccine counseling strategies and messages in responding to specific parental vaccine concerns.
Disclosure of Potential Conflicts of Interest
We have no conflicts of interest to disclose.
Ethics Statement
The study was approved by the Institutional Review Board of the University of Pennsylvania.
Acknowledgments
The authors thank Cecilia Davis-Hayes and Eileen Wang for manuscript assistance and Yelena Baras for assistance with data cleaning. We also thank the pediatric practice that provided the data used in our analysis.
Sources of Support
This work was supported by the Robert Wood Johnson Foundation Health and Society Scholars Program at the University of Pennsylvania and a training grant in comparative effectiveness research at the University of Pennsylvania (KM1CA156715-01).
Glossary
Abbreviations:
- EMR
electronic medical record
- ACIP
Advisory Committee on Immunization Practices
- VPCD
vaccine preventable childhood disease
- TPB
theory of planned behavior
- OR
odds ratio
- aOR
adjusted odds ratio
Footnotes
Previously published online: www.landesbioscience.com/journals/vaccines/article/25959
References
- 1.Centers for Disease Control and Prevention (CDC) Ten great public health achievements--United States, 1900-1999. MMWR Morb Mortal Wkly Rep. 1999;48:241–3. [PubMed] [Google Scholar]
- 2.Zhou F. Updated economic evaluation of the routine childhood immunization schedule in the United States. In: Washington, D.C.; 2011. [Google Scholar]
- 3.Smith MJ, Marshall GS. Navigating parental vaccine hesitancy. Pediatr Ann. 2010;39:476–82. doi: 10.3928/00904481-20100726-05. [DOI] [PubMed] [Google Scholar]
- 4.Dempsey AF, Schaffer S, Singer D, Butchart A, Davis M, Freed GL. Alternative vaccination schedule preferences among parents of young children. Pediatrics. 2011;128:848–56. doi: 10.1542/peds.2011-0400. [DOI] [PubMed] [Google Scholar]
- 5.Leib S, Liberatos P, Edwards K. Pediatricians’ experience with and response to parental vaccine safety concerns and vaccine refusals: a survey of Connecticut pediatricians. Public Health Rep. 2011;126(Suppl 2):13–23. doi: 10.1177/00333549111260S203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wightman A, Opel DJ, Marcuse EK, Taylor JA. Washington State pediatricians’ attitudes toward alternative childhood immunization schedules. Pediatrics. 2011;128:1094–9. doi: 10.1542/peds.2011-0666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Guerra FA. Delays in immunization have potentially serious health consequences. Paediatr Drugs. 2007;9:143–8. doi: 10.2165/00148581-200709030-00002. [DOI] [PubMed] [Google Scholar]
- 8.Sugerman DE, Barskey AE, Delea MG, et al. Measles outbreak in a highly vaccinated population, San Diego, 2008: Role of the intentionally undervaccinated. Pediatrics. 2010peds.2009–1653. 10.1542/peds.2009-1653 [DOI] [PubMed]
- 9.Glanz JM, McClure DL, Magid DJ, Daley MF, France EK, Salmon DA, et al. Parental refusal of pertussis vaccination is associated with an increased risk of pertussis infection in children. Pediatrics. 2009;123:1446–51. doi: 10.1542/peds.2008-2150. [DOI] [PubMed] [Google Scholar]
- 10.Glanz JM, McClure DL, Magid DJ, Daley MF, France EK, Hambidge SJ. Parental refusal of varicella vaccination and the associated risk of varicella infection in children. Arch Pediatr Adolesc Med. 2010;164:66–70. doi: 10.1001/archpediatrics.2009.244. [DOI] [PubMed] [Google Scholar]
- 11.Omer SB, Enger KS, Moulton LH, Halsey NA, Stokley S, Salmon DA. Geographic clustering of nonmedical exemptions to school immunization requirements and associations with geographic clustering of pertussis. Am J Epidemiol. 2008;168:1389–96. doi: 10.1093/aje/kwn263. [DOI] [PubMed] [Google Scholar]
- 12.Diekema DS, American Academy of Pediatrics Committee on Bioethics Responding to parental refusals of immunization of children. Pediatrics. 2005;115:1428–31. doi: 10.1542/peds.2005-0316. [DOI] [PubMed] [Google Scholar]
- 13.Salmon DA, Omer SB. Individual freedoms versus collective responsibility: immunization decision-making in the face of occasionally competing values. Emerg Themes Epidemiol. 2006;3:13. doi: 10.1186/1742-7622-3-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Smith PJ, Humiston SG, Parnell T, Vannice KS, Salmon DA. The association between intentional delay of vaccine administration and timely childhood vaccination coverage. Public Health Rep. 2010;125:534–41. doi: 10.1177/003335491012500408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ajzen I. The theory of planned behavior. Organ Behav Hum. 1991;50:179–211. doi: 10.1016/0749-5978(91)90020-T. [DOI] [Google Scholar]
- 16.Ajzen I, Fishbein M. The influence of attitudes on behavior. In: The Handbook of Attitudes Mahwah, N.J: Lawrence Erlbaum Associates; 2005:173–221. [Google Scholar]
- 17.Fishbein M, Cappella JN. The role of theory in developing effective health communications. J Commun. 2006;56(s1):S1–17. doi: 10.1111/j.1460-2466.2006.00280.x. [DOI] [Google Scholar]
- 18.Fredrickson DD, Davis TC, Arnould CL, Kennen EM, Hurniston SG, Cross JT, et al. Childhood immunization refusal: provider and parent perceptions. Fam Med. 2004;36:431–9. [PubMed] [Google Scholar]
- 19.Freed GL, Clark SJ, Butchart AT, Singer DC, Davis MM. Parental vaccine safety concerns in 2009. Pediatrics. 2010;125:654–9. doi: 10.1542/peds.2009-1962. [DOI] [PubMed] [Google Scholar]
- 20.Gust DA, Woodruff R, Kennedy A, Brown C, Sheedy K, Hibbs B. Parental perceptions surrounding risks and benefits of immunization. Semin Pediatr Infect Dis. 2003;14:207–12. doi: 10.1016/S1045-1870(03)00035-9. [DOI] [PubMed] [Google Scholar]
- 21.Gust DA, Darling N, Kennedy A, Schwartz B. Parents with doubts about vaccines: which vaccines and reasons why. Pediatrics. 2008;122:718–25. doi: 10.1542/peds.2007-0538. [DOI] [PubMed] [Google Scholar]
- 22.Flanagan-Klygis EA, Sharp L, Frader JE. Dismissing the family who refuses vaccines: a study of pediatrician attitudes. Arch Pediatr Adolesc Med. 2005;159:929–34. doi: 10.1001/archpedi.159.10.929. [DOI] [PubMed] [Google Scholar]
- 23.Benin AL, Wisler-Scher DJ, Colson E, Shapiro ED, Holmboe ES. Qualitative analysis of mothers’ decision-making about vaccines for infants: the importance of trust. Pediatrics. 2006;117:1532–41. doi: 10.1542/peds.2005-1728. [DOI] [PubMed] [Google Scholar]
- 24.Freed GL, Clark SJ, Butchart AT, Singer DC, Davis MM. Sources and perceived credibility of vaccine-safety information for parents. Pediatrics. 2011;127(Suppl 1):S107–12. doi: 10.1542/peds.2010-1722P. [DOI] [PubMed] [Google Scholar]
- 25.Taylor JA, Darden PM, Slora E, Hasemeier CM, Asmussen L, Wasserman R. The influence of provider behavior, parental characteristics, and a public policy initiative on the immunization status of children followed by private pediatricians: a study from Pediatric Research in Office Settings. Pediatrics. 1997;99:209–15. [PubMed] [Google Scholar]
- 26.Allred NJ, Shaw KM, Santibanez TA, Rickert DL, Santoli JM. Parental vaccine safety concerns: results from the National Immunization Survey, 2001-2002. Am J Prev Med. 2005;28:221–4. doi: 10.1016/j.amepre.2004.10.014. [DOI] [PubMed] [Google Scholar]
- 27.Gust DA, Strine TW, Maurice E, Smith P, Yusuf H, Wilkinson M, et al. Underimmunization among children: effects of vaccine safety concerns on immunization status. Pediatrics. 2004;114:e16–22. doi: 10.1542/peds.114.1.e16. [DOI] [PubMed] [Google Scholar]
- 28.Bardenheier B, Yusuf H, Schwartz B, Gust D, Barker L, Rodewald L. Are parental vaccine safety concerns associated with receipt of measles-mumps-rubella, diphtheria and tetanus toxoids with acellular pertussis, or hepatitis B vaccines by children? Arch Pediatr Adolesc Med. 2004;158:569–75. doi: 10.1001/archpedi.158.6.569. [DOI] [PubMed] [Google Scholar]
- 29.Brown KF, Kroll JS, Hudson MJ, Ramsay M, Green J, Long SJ, et al. Factors underlying parental decisions about combination childhood vaccinations including MMR: a systematic review. Vaccine. 2010;28:4235–48. doi: 10.1016/j.vaccine.2010.04.052. [DOI] [PubMed] [Google Scholar]
- 30.Offit PA, Moser CA. The problem with Dr Bob’s alternative vaccine schedule. Pediatrics. 2009;123:e164–9. doi: 10.1542/peds.2008-2189. [DOI] [PubMed] [Google Scholar]
- 31.Nickerson RS. Confirmation bias: A ubiquitous phenomenon in many guises. Rev Gen Psychol. 1998;2:175–220. doi: 10.1037/1089-2680.2.2.175. [DOI] [Google Scholar]