ORIGINAL ARTICLE
Urban and Rural Differences in Parental Attitudes About Influenza Vaccination and Vaccine Delivery Models Sean T. O’Leary, MD, MPH;1,2 Juliana Barnard, MA;2 Steven Lockhart, BA;2 Maureen Kolasa, RN, MPH;3 Doron Shmueli, MS;2 L. Miriam Dickinson, PhD;2,4 Deidre Kile, MS;2 Eva Dibert, MPH;2 & Allison Kempe, MD, MPH1,2,4 1 Department of Pediatrics, University of Colorado Anschutz Medical Campus, Aurora, Colorado 2 The Children’s Outcomes Research Program, Children’s Hospital Colorado, Aurora, Colorado 3 Centers for Disease Control and Prevention, National Center for Immunizations and Respiratory Diseases, Atlanta, Georgia 4 Colorado Health Outcomes Research, University of Colorado Anschutz Medical Campus, Aurora, Colorado
Abstract Disclosures: None of the authors has any financial disclosures. The manuscript was reviewed by Centers for Disease Control co-authors and others at the CDC, although the findings and conclusions in this report are those of the authors, and do not necessarily represent the views of the CDC. Funding: This investigation was funded by a grant from the Centers for Disease Control and Prevention (U01IP000320). For further information, contact: Sean T. O’Leary, MD, MPH, University of Colorado, Department of Pediatrics, Mail Stop F443, 13199 E Montview Blvd, Suite 300, Aurora, CO 80045; e-mail: sean.o’[email protected]. doi: 10.1111/jrh.12119
Objectives: To assess and compare among parents of healthy children in urban and rural areas: (1) reported influenza vaccination status; (2) attitudes regarding influenza vaccination; and (3) attitudes about collaborative models for influenza vaccination delivery involving practices and public health departments. Methods: A mail survey to random samples of parents from 2 urban and 2 rural private practices in Colorado from April 2012 to June 2012. Results: The response rate was 58% (288/500). In the prior season, 63% of urban and 41% of rural parents reported their child received influenza vaccination (P < .001). No differences in attitudes about influenza infection or vaccination between urban and rural parents were found, with 75% of urban and 73% of rural parents agreeing their child should receive an influenza vaccine every year (P = .71). High proportions reported willingness to participate in a collaborative clinic in a community setting (59% urban, 70% rural, P = .05) or at their child’s provider (73% urban, 73% rural, P = .99) with public health department assisting. Fewer (36% urban, 53% rural, P < .01) were likely to go to the public health department if referred by their provider. Rural parents were more willing for their child to receive vaccination outside of their provider’s office (70% vs. 55%, P = .01). Conclusions: While attitudes regarding influenza vaccination were similar, rural children were much less likely to have received vaccination. Most parents were amenable to collaborative models of influenza vaccination delivery, but rural parents were more comfortable with influenza vaccination outside their provider’s office, suggesting that other venues for influenza vaccination in rural settings should be promoted. Key words health services research, influenza, rural, utilization of health services, vaccine.
The burden of influenza in the United States is substantial.1 Children have the highest rates of infection and play an important role in initiating and maintaining influenza epidemics in the community.2,3 Influenza is also an important cause of pediatric deaths, with about half of children dying from influenza having no
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 421–430
identifiable risk factor.4,5 Annual influenza vaccination is the most effective way to prevent influenza and its complications and is now recommended for all persons over 6 months of age.1 However, influenza vaccination rates in children are very low compared to other recommended childhood vaccines. The most recently reported rate in
421
O’Leary et al.
Influenza Vaccination in Urban and Rural Children
the 2012-2013 season of 56.6% in children 6 months to 17 years is far below the Healthy People 2020 goal of 70%.6 Barriers to influenza vaccination in children include problems with health care access, missed vaccination opportunities, and parental misperceptions about the need, effectiveness and safety of influenza vaccination.7 The issue of rurality in influenza vaccination is one that is infrequently addressed in the literature, particularly for children. While there have been several studies looking at vaccination rates for other childhood vaccines in rural residents,8-13 with inconsistent findings, there are scant data specific to uptake of influenza vaccine in rural US children, despite the fact that millions of US children live in rural areas. Also, although parental attitudes regarding influenza and influenza vaccine have been examined in previous studies,14-18 there are few data about attitudes of rural parents regarding influenza vaccine, and there have been no direct comparisons of the attitudes of rural and urban parents. A recent Institute of Medicine (IOM) report stressed the importance of collaboration between primary care and public health in the delivery of preventive services.19 The annual effort to deliver influenza vaccine to all children over 6 months may be an ideal situation for such collaboration, as practices may not have the capacity to vaccinate all of their patients every influenza vaccination season.20-23 Collaborations between primary care and public health may be particularly applicable to rural children, as they are known to utilize public health more often than urban children for vaccinations.24,25 This study sought to increase understanding of the role that rurality may play in attitudes and practices regarding influenza and influenza vaccination. In addition, in light of the IOM report, the study examined the potential for public-private collaboration in influenza vaccine delivery and whether attitudes might differ between parents in urban and rural settings. The specific objectives of this study were to assess and compare among parents of healthy urban and rural children: (1) the proportion whose children had received influenza vaccination in the prior season and location of receipt; (2) attitudes and preferences regarding influenza vaccination; (3) attitudes about collaborative models for influenza vaccination delivery involving private practices and public health departments; and (4) characteristics of parents willing to have their child vaccinated outside of his/her usual provider’s practice.
Methods Study Setting and Population From April to June 2012, a mail survey was administered to parents of urban and rural children in 4 counties
422
in Colorado. This study was approved by the Colorado Multiple Institutional Review Board. A sample of 500 patients 2-17 years of age was randomly selected from practice administrative data from 4 private practices (2 urban pediatric and 2 rural family medicine). In rural Colorado, the vast majority of primary care physicians are family physicians,26 while in urban areas such as Denver and Colorado Springs, as elsewhere, the majority of pediatric care is provided by pediatricians,27 which was our reasoning for sampling from the different specialties. Practices were considered urban if they were located in a Metropolitan Statistical Area (MSA) as defined by the White House Office of Management and Budget.28 Both rural practices, in addition to being outside of a MSA, were in counties classified by the Colorado State Office of Rural Health as “rural,” one of which was further classified as “frontier.”29 All practices participated in the Colorado Immunization Information System and the Vaccines for Children (VFC) program. All practices utilized reminder/recall for influenza vaccine, “flu shot clinics,” and evening and weekend hours for influenza vaccine. One child was randomly selected per family. This study focused on urban/rural differences in attitudes and practices regarding influenza vaccination, which we thought might differ between parents with healthy children as opposed to parents of children with high-risk medical conditions. Because there were too few rural high-risk children to provide adequate power for comparisons between healthy and high-risk groups, we chose to exclude parents of high-risk children from this study. High risk was defined as any child seen in the prior 2 years aged 2-17 years as of the start of the prior influenza season (August 1, 2011) who also had 1 or more diagnostic codes within the administrative data within these 2 years from a prespecified list of conditions considered high risk for influenza30 or who was 6-24 months of age at the start of the prior influenza season. Therefore, the study population included: (1) urban healthy children, defined as any patient aged 2-17 years as of August 1, 2011, seen in an urban practice in the prior 2 years without any diagnostic codes for a high-risk condition; and (2) rural healthy children, defined as any patient aged 2-17 years as of August 1, 2011, seen in a rural practice in the prior 2 years without any high-risk codes. There were 250 surveys sent to parents of urban patients and 250 sent to parents of rural patients.
Survey Design and Administration The survey was based on the Health Belief Model and included 16 questions, with either yes/no or 4-point Likert scales. Questions assessed demographics, attitudes about influenza and influenza vaccine, receipt of influenza
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 421–430
O’Leary et al.
vaccine, location of influenza vaccination, attitudes about location of influenza vaccination, the importance of various factors in the decision of where to receive an influenza vaccine, and attitudes about various models of collaborative clinics. The survey was piloted among a convenience sample of parents (n = 5) attending a hospital-based pediatric clinic and a convenience sample of urban and suburban parents not randomly selected to participate in the survey (n = 10). Parents were mailed a preletter followed by the survey 6 days later. Nonrespondents were sent a reminder postcard and up to 2 additional surveys. The first 2 mailings had a gift card incentive with survey return and the third mailing had a cash incentive included in the mailing.
Analytic Methods Chi-squared and Wilcoxon rank-sum tests were used for comparisons of characteristics of respondents and nonrespondents and for comparisons between parents of urban and rural children. Bivariate and multivariable analyses were conducted with the primary dependent variable of willingness to receive influenza vaccine outside of their primary provider’s office. The dependent variable was based on the following yes/no question, which was asked after having answered the questions regarding the various scenarios for potential public-private collaboration: “Overall, would you be willing to have your child get a flu shot at a location other than his/her usual provider’s office? Examples are at community locations (such as a fire station or recreation center) or public health clinics (such as at a public health department or a public health immunization clinic).” Parents who had checked a box in an earlier question stating “I do not want my child to get a flu shot anywhere” were excluded from the bivariate and multivariable analyses. Independent variables to include in the model were chosen a priori based on existing literature and clinical relevance. In addition to sociodemographic factors and setting (urban or rural), we examined the importance of factors which may impact parents’ decisions on where to get their child vaccinated, such as cost, convenience, and who was administering the vaccine. These factors were dichotomized as those rated “very important” versus all other responses. Observations with missing values for the chosen covariates were excluded from the analysis. Factors significant at P < .25 in bivariate analyses were tested in multivariable models using a backwards elimination procedure in which the least significant predictor in the model was eliminated sequentially. To avoid problems of collinearity, if a strong correlation was observed between covariates in the model, we selected the covariate with the lower correlation with the other covariates. At each step, estimates were checked to make
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 421–430
Influenza Vaccination in Urban and Rural Children
Table 1 Characteristics of Survey Respondents Regarding Influenza and Influenza Vaccine (n = 288)
Characteristics
Urban (n = 143)
Rural (n = 145)
Age of child (years), mean 8.6 10.3 13% 22% Medical condition for >3 months, child (parental report) Child’s type of health insurance Medicaid/Child Health Plan 12% 30% Plus (CHP-Plus) Private insurance 83% 61% Other/tricare 5% 8% Number of children under 18 who currently live in household 0-2 children 77% 68% 3 or more children 23% 32% Gender (of respondent),% 10% 6% male Age of parent completing 38.7 38.3 survey, mean Parent’s highest grade or year of school completed Less than high school 1% 11% graduate High school graduate 6% 15% Vocational/Some college 26% 26% College graduate/Advanced 67% 48% degree 63% 41% Reported receipt of influenza vaccine by child in prior season Sites of influenza vaccination (n = 142) Provider’s office 93% 79% Public Health Clinic 3% 11% Community Influenza 0% 7% Vaccine Clinic Pharmacy/Retail Outlet 4% 2%
P Value .0042 .05
.0002
.21 .26 .68
Urban and Rural Differences in Parental Attitudes About Influenza Vaccination and Vaccine Delivery Models Sean T. O’Leary, MD, MPH;1,2 Juliana Barnard, MA;2 Steven Lockhart, BA;2 Maureen Kolasa, RN, MPH;3 Doron Shmueli, MS;2 L. Miriam Dickinson, PhD;2,4 Deidre Kile, MS;2 Eva Dibert, MPH;2 & Allison Kempe, MD, MPH1,2,4 1 Department of Pediatrics, University of Colorado Anschutz Medical Campus, Aurora, Colorado 2 The Children’s Outcomes Research Program, Children’s Hospital Colorado, Aurora, Colorado 3 Centers for Disease Control and Prevention, National Center for Immunizations and Respiratory Diseases, Atlanta, Georgia 4 Colorado Health Outcomes Research, University of Colorado Anschutz Medical Campus, Aurora, Colorado
Abstract Disclosures: None of the authors has any financial disclosures. The manuscript was reviewed by Centers for Disease Control co-authors and others at the CDC, although the findings and conclusions in this report are those of the authors, and do not necessarily represent the views of the CDC. Funding: This investigation was funded by a grant from the Centers for Disease Control and Prevention (U01IP000320). For further information, contact: Sean T. O’Leary, MD, MPH, University of Colorado, Department of Pediatrics, Mail Stop F443, 13199 E Montview Blvd, Suite 300, Aurora, CO 80045; e-mail: sean.o’[email protected]. doi: 10.1111/jrh.12119
Objectives: To assess and compare among parents of healthy children in urban and rural areas: (1) reported influenza vaccination status; (2) attitudes regarding influenza vaccination; and (3) attitudes about collaborative models for influenza vaccination delivery involving practices and public health departments. Methods: A mail survey to random samples of parents from 2 urban and 2 rural private practices in Colorado from April 2012 to June 2012. Results: The response rate was 58% (288/500). In the prior season, 63% of urban and 41% of rural parents reported their child received influenza vaccination (P < .001). No differences in attitudes about influenza infection or vaccination between urban and rural parents were found, with 75% of urban and 73% of rural parents agreeing their child should receive an influenza vaccine every year (P = .71). High proportions reported willingness to participate in a collaborative clinic in a community setting (59% urban, 70% rural, P = .05) or at their child’s provider (73% urban, 73% rural, P = .99) with public health department assisting. Fewer (36% urban, 53% rural, P < .01) were likely to go to the public health department if referred by their provider. Rural parents were more willing for their child to receive vaccination outside of their provider’s office (70% vs. 55%, P = .01). Conclusions: While attitudes regarding influenza vaccination were similar, rural children were much less likely to have received vaccination. Most parents were amenable to collaborative models of influenza vaccination delivery, but rural parents were more comfortable with influenza vaccination outside their provider’s office, suggesting that other venues for influenza vaccination in rural settings should be promoted. Key words health services research, influenza, rural, utilization of health services, vaccine.
The burden of influenza in the United States is substantial.1 Children have the highest rates of infection and play an important role in initiating and maintaining influenza epidemics in the community.2,3 Influenza is also an important cause of pediatric deaths, with about half of children dying from influenza having no
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 421–430
identifiable risk factor.4,5 Annual influenza vaccination is the most effective way to prevent influenza and its complications and is now recommended for all persons over 6 months of age.1 However, influenza vaccination rates in children are very low compared to other recommended childhood vaccines. The most recently reported rate in
421
O’Leary et al.
Influenza Vaccination in Urban and Rural Children
the 2012-2013 season of 56.6% in children 6 months to 17 years is far below the Healthy People 2020 goal of 70%.6 Barriers to influenza vaccination in children include problems with health care access, missed vaccination opportunities, and parental misperceptions about the need, effectiveness and safety of influenza vaccination.7 The issue of rurality in influenza vaccination is one that is infrequently addressed in the literature, particularly for children. While there have been several studies looking at vaccination rates for other childhood vaccines in rural residents,8-13 with inconsistent findings, there are scant data specific to uptake of influenza vaccine in rural US children, despite the fact that millions of US children live in rural areas. Also, although parental attitudes regarding influenza and influenza vaccine have been examined in previous studies,14-18 there are few data about attitudes of rural parents regarding influenza vaccine, and there have been no direct comparisons of the attitudes of rural and urban parents. A recent Institute of Medicine (IOM) report stressed the importance of collaboration between primary care and public health in the delivery of preventive services.19 The annual effort to deliver influenza vaccine to all children over 6 months may be an ideal situation for such collaboration, as practices may not have the capacity to vaccinate all of their patients every influenza vaccination season.20-23 Collaborations between primary care and public health may be particularly applicable to rural children, as they are known to utilize public health more often than urban children for vaccinations.24,25 This study sought to increase understanding of the role that rurality may play in attitudes and practices regarding influenza and influenza vaccination. In addition, in light of the IOM report, the study examined the potential for public-private collaboration in influenza vaccine delivery and whether attitudes might differ between parents in urban and rural settings. The specific objectives of this study were to assess and compare among parents of healthy urban and rural children: (1) the proportion whose children had received influenza vaccination in the prior season and location of receipt; (2) attitudes and preferences regarding influenza vaccination; (3) attitudes about collaborative models for influenza vaccination delivery involving private practices and public health departments; and (4) characteristics of parents willing to have their child vaccinated outside of his/her usual provider’s practice.
Methods Study Setting and Population From April to June 2012, a mail survey was administered to parents of urban and rural children in 4 counties
422
in Colorado. This study was approved by the Colorado Multiple Institutional Review Board. A sample of 500 patients 2-17 years of age was randomly selected from practice administrative data from 4 private practices (2 urban pediatric and 2 rural family medicine). In rural Colorado, the vast majority of primary care physicians are family physicians,26 while in urban areas such as Denver and Colorado Springs, as elsewhere, the majority of pediatric care is provided by pediatricians,27 which was our reasoning for sampling from the different specialties. Practices were considered urban if they were located in a Metropolitan Statistical Area (MSA) as defined by the White House Office of Management and Budget.28 Both rural practices, in addition to being outside of a MSA, were in counties classified by the Colorado State Office of Rural Health as “rural,” one of which was further classified as “frontier.”29 All practices participated in the Colorado Immunization Information System and the Vaccines for Children (VFC) program. All practices utilized reminder/recall for influenza vaccine, “flu shot clinics,” and evening and weekend hours for influenza vaccine. One child was randomly selected per family. This study focused on urban/rural differences in attitudes and practices regarding influenza vaccination, which we thought might differ between parents with healthy children as opposed to parents of children with high-risk medical conditions. Because there were too few rural high-risk children to provide adequate power for comparisons between healthy and high-risk groups, we chose to exclude parents of high-risk children from this study. High risk was defined as any child seen in the prior 2 years aged 2-17 years as of the start of the prior influenza season (August 1, 2011) who also had 1 or more diagnostic codes within the administrative data within these 2 years from a prespecified list of conditions considered high risk for influenza30 or who was 6-24 months of age at the start of the prior influenza season. Therefore, the study population included: (1) urban healthy children, defined as any patient aged 2-17 years as of August 1, 2011, seen in an urban practice in the prior 2 years without any diagnostic codes for a high-risk condition; and (2) rural healthy children, defined as any patient aged 2-17 years as of August 1, 2011, seen in a rural practice in the prior 2 years without any high-risk codes. There were 250 surveys sent to parents of urban patients and 250 sent to parents of rural patients.
Survey Design and Administration The survey was based on the Health Belief Model and included 16 questions, with either yes/no or 4-point Likert scales. Questions assessed demographics, attitudes about influenza and influenza vaccine, receipt of influenza
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 421–430
O’Leary et al.
vaccine, location of influenza vaccination, attitudes about location of influenza vaccination, the importance of various factors in the decision of where to receive an influenza vaccine, and attitudes about various models of collaborative clinics. The survey was piloted among a convenience sample of parents (n = 5) attending a hospital-based pediatric clinic and a convenience sample of urban and suburban parents not randomly selected to participate in the survey (n = 10). Parents were mailed a preletter followed by the survey 6 days later. Nonrespondents were sent a reminder postcard and up to 2 additional surveys. The first 2 mailings had a gift card incentive with survey return and the third mailing had a cash incentive included in the mailing.
Analytic Methods Chi-squared and Wilcoxon rank-sum tests were used for comparisons of characteristics of respondents and nonrespondents and for comparisons between parents of urban and rural children. Bivariate and multivariable analyses were conducted with the primary dependent variable of willingness to receive influenza vaccine outside of their primary provider’s office. The dependent variable was based on the following yes/no question, which was asked after having answered the questions regarding the various scenarios for potential public-private collaboration: “Overall, would you be willing to have your child get a flu shot at a location other than his/her usual provider’s office? Examples are at community locations (such as a fire station or recreation center) or public health clinics (such as at a public health department or a public health immunization clinic).” Parents who had checked a box in an earlier question stating “I do not want my child to get a flu shot anywhere” were excluded from the bivariate and multivariable analyses. Independent variables to include in the model were chosen a priori based on existing literature and clinical relevance. In addition to sociodemographic factors and setting (urban or rural), we examined the importance of factors which may impact parents’ decisions on where to get their child vaccinated, such as cost, convenience, and who was administering the vaccine. These factors were dichotomized as those rated “very important” versus all other responses. Observations with missing values for the chosen covariates were excluded from the analysis. Factors significant at P < .25 in bivariate analyses were tested in multivariable models using a backwards elimination procedure in which the least significant predictor in the model was eliminated sequentially. To avoid problems of collinearity, if a strong correlation was observed between covariates in the model, we selected the covariate with the lower correlation with the other covariates. At each step, estimates were checked to make
c 2015 National Rural Health Association The Journal of Rural Health 31 (2015) 421–430
Influenza Vaccination in Urban and Rural Children
Table 1 Characteristics of Survey Respondents Regarding Influenza and Influenza Vaccine (n = 288)
Characteristics
Urban (n = 143)
Rural (n = 145)
Age of child (years), mean 8.6 10.3 13% 22% Medical condition for >3 months, child (parental report) Child’s type of health insurance Medicaid/Child Health Plan 12% 30% Plus (CHP-Plus) Private insurance 83% 61% Other/tricare 5% 8% Number of children under 18 who currently live in household 0-2 children 77% 68% 3 or more children 23% 32% Gender (of respondent),% 10% 6% male Age of parent completing 38.7 38.3 survey, mean Parent’s highest grade or year of school completed Less than high school 1% 11% graduate High school graduate 6% 15% Vocational/Some college 26% 26% College graduate/Advanced 67% 48% degree 63% 41% Reported receipt of influenza vaccine by child in prior season Sites of influenza vaccination (n = 142) Provider’s office 93% 79% Public Health Clinic 3% 11% Community Influenza 0% 7% Vaccine Clinic Pharmacy/Retail Outlet 4% 2%
P Value .0042 .05
.0002
.21 .26 .68
