Journal of Adolescent Health 56 (2015) S33eS39
www.jahonline.org Original article
Trust and a School-Located Immunization Program Tiana L. Won a, Amy B. Middleman, M.D., M.S.E.D., M.P.H. b, *, Beth A. Auslander, Ph.D. c, and Mary B. Short, Ph.D. d a
Baylor College of Medicine, Houston, Texas Department of Pediatrics, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma Department of Pediatrics, University of Texas Medical Branch, Galveston, Texas d Department of Psychology, University of HoustoneClear Lake, Clear Lake, Texas b c
Article history: Received July 11, 2014; Accepted September 29, 2014 Keywords: School-located immunization programs; Adolescent health services; Trust; Socioeconomic status; Hispanic
A B S T R A C T
Purpose: To determine variables associated with parental trust in a school-located immunization program (SLIP) and the effect of trust-building interventions on trust and participation in SLIPs. Methods: Parents among eight schools randomized to a trust-building intervention or a control condition (four schools each) completed a five-item trust survey before SLIP implementation both inyear 1 (fall 2012) and in year 2 (fall 2013). Mean trust scores were calculated. Associations between baseline demographic and experiential variables and mean trust scores were analyzed. Mean trust scores in intervention and control schools were compared before SLIP in years 1 and 2, and SLIP participation rates were noted. Results: From year 1, 1,608 parent surveys were analyzed. Baseline mean trust score across schools was 3.59 of 5 (5 ¼ highest trust). In a multiple linear regression model, annual household income, survey language version, participation in a previous SLIP, child’s health insurance status, and perceived vaccine importance were significantly associated with parental trust in SLIPs (R2¼ .06, p < .001). There was no difference in mean trust scores between intervention and control schools (p ¼ .8). In year 2, 844 surveys were analyzed, and a modest difference was observed between intervention and control schools (mean trust score ¼ 3.66 and 3.57, respectively, p ¼ .07). SLIP participation rates appeared higher in intervention (7.7%) versus control schools (4.3%) in year 1. Conclusions: Baseline trust in SLIPs among a low-income, largely Hispanic group of parents in Texas was moderately high. Factors associated with trust included demographic and experiential variables, and interventions aimed at increasing parents’ perception of vaccine importance and participation in SLIPs may be effective in increasing parental trust in SLIPs. Ó 2015 Society for Adolescent Health and Medicine. All rights reserved.
Since 2005, vaccine recommendations for adolescents by the Advisory Committee on Immunization Practices have expanded to include the tetanus toxoid, diphtheria toxoid, acellular Conflicts of Interest: The authors declare no conflicts of interest. Disclaimer: Publication of this article was supported by the Society for Adolescent Health and Medicine through a grant from Merck & Co., Inc. The opinions or views expressed in this supplement are those of the authors and do not represent the position of the funder. * Address correspondence to: Amy B. Middleman, M.D., M.S.E.D., M.P.H., Department of Pediatrics, University of Oklahoma Health Sciences Center, 1200 Children’s Avenue, Oklahoma City, OK 73104. E-mail address: [email protected] (A.B. Middleman). 1054-139X/Ó 2015 Society for Adolescent Health and Medicine. All rights reserved. http://dx.doi.org/10.1016/j.jadohealth.2014.09.018
IMPLICATIONS AND CONTRIBUTION
Baseline trust in schoollocated immunization programs (SLIPs) among low-income, primarily Hispanic, middle school parents is moderately high. Variables predicting greater trust in SLIPs include lower income, completing the survey in Spanish, prior SLIP participation, self-pay health insurance status, and greater perceived vaccine importance. Participation rates were relatively higher in schools receiving a trustbuilding intervention versus control schools.
pertussis vaccine (Tdap); meningococcal conjugate vaccine (MCV4); human papillomavirus vaccine (HPV); and annual influenza vaccine. Immunization rates for adolescents have steadily increased but remain below rates achieved for younger children. Coverage for Tdap and MCV4 is closer to the Healthy People 2020 target (of 80%) at 85% and 74%, respectively, but the coverage gap between Tdap and HPV (84.6% for Tdap vs. 53.8% for at least one HPV dose among females in 2012) suggests missed opportunities to fully vaccinate adolescents [1]. There are multiple challenges to achieve high rates of vaccination among adolescents, which requires identifying more effective venues for
S34
T.L. Won et al. / Journal of Adolescent Health 56 (2015) S33eS39
immunization [2]. School-located immunization programs (SLIPs) provide immunizations to students and represent a convenient alternative site. Multiple organizations have acknowledged SLIPs as a way to efficiently access the adolescent population for immunization [3,4]. Their efficacy in reaching adolescents has also been recognized in countries such as the United Kingdom, where a school-based program achieved an HPV vaccination rate of 76.4% for the first dose of HPV in 2009e2010 [5]. Implementing SLIPs in the United States presents several challenges beyond the complex reimbursement issues. A review of influenza-only SLIPs conducted between 2003 and 2010 demonstrated a vaccination rate ranging from 7% to 58%, with higher rates among elementary compared with middle and high school students [6]. Another study confirmed this trend and documented mean immunization rates of 21.5% among elementary school children compared with 10.3% and 5.8% among middle and high school students, respectively [7]. Many variables influence whether parents use SLIPs, including the program’s ability to retrieve consent forms from parents, the use of incentives, parental education and follow-up [8]. Not surprisingly, concerns about vaccine safety also affect parents’ decisions to vaccinate their children [9,10]. A recent qualitative study indicated that parental trust may be another factor in parents’ decisions to use an SLIP [11]. Challenges related to parents’/ adolescents’ trust of vaccine providers may also influence adolescent participation in SLIPs. Although the concept of trust relates to factors such as satisfaction, competency, privacy, and communication, previous research has supported its definition as “the optimistic acceptance of a vulnerable situation in which the truster (i.e., patient) believes the trustee (i.e., provider) will care for the truster’s interests” [12]. The importance of trust in any providerepatient relationship is well recognized and has implications for patient care ranging from adherence to efficacy of treatment [13]. However, it is not known the degree to which parents trust SLIPs, what factors are correlated with trust in SLIPs, and whether interventions aimed at increasing trust could increase parental trust in SLIPs. This study evaluated the baseline level of parental trust in SLIPs and participation rates in years 1 and 2 of an SLIP among a lowincome, largely Hispanic population of middle school students. In addition, this study assessed whether basic demographic variables, perceptions of vaccine importance, and history of SLIP experience were associated with middle school parents’ trust in SLIPs before any intervention in year 1, and whether trust in SLIPs was greater in year 2 among parents from schools that had received a simple trust-building intervention in year 1 compared with schools not receiving a trust-building intervention. Methods School selection and school-located immunization program time line Schools in a large, urban school district with >90% participation in the free lunch program were offered the opportunity to participate in a multivisit SLIP that administered Vaccines for Children (VFC) Program vaccinations without cost to all VFC eligible students. Eight middle schools (grades 6e8) with enrollment varying from 600 to 1,500 students (total of approximately 8,750 in year 1) accepted the invitation to participate in the SLIP project initiated in fall 2012. The eight schools were randomized into four control and four intervention schools.
Distribution of parent surveys with trust measure A packet containing a two-page questionnaire and a cover letter explaining the study was distributed to all students in all eight schools to take home to parents and return to teachers or school nurses before the initiation of the SLIP in fall 2012, after SLIP in the spring 2013, and again before the SLIP in fall 2013. The questionnaire contained a five-item trust survey and several demographic items and was provided to each family in both English and Spanish. Completed questionnaires were retrieved from schools by a member of the research team approximately 3 weeks after distribution. Each survey was labeled with a unique number to identify the school and prevent duplication of data entry. When returning the survey, parents could write their child’s name on the cover letter for the chance to win an MP3 player. Cover letters with identifying information were removed from the surveys on receipt and retained separately to ensure confidentiality of the survey responses. Survey content and measurement Demographic items included age, race, ethnicity, highest level of education achieved, primary language spoken at home, annual household income, child’s health insurance status, and child’s participation in a medical home. History of participation in SLIPs and the perceived importance of vaccination were assessed. In addition, another variable noted was whether the participants completed the survey using the English or Spanish version. Parental trust in SLIPs was evaluated by adapting a five-item scale developed by Dugan et al. [14] shown to be valid for use in assessing patient trust in physicians, health insurers, and the medical profession as a whole (Figure 1). Parents completed the five-item survey using a five-point Likert scale (ranging from 1¼ strongly disagree to 5¼ strongly agree). One of the five statements was negatively worded and was therefore reverse scored for data analyses. Cronbach alpha measuring internal consistency reliability for the present sample was .71 [15]. Parent/student education During full school student assemblies or successive lunch periods, study personnel visited all eight schools in the program and spoke briefly to students about the importance of adolescent immunizations and the availability of the SLIP. All information was provided using a standard script. Description of intervention. The four intervention schools received an additional, simple trust-building intervention to address parental trust. Research staff visited intervention schools during parent open houses or other parent meetings at the schools. These meetings introduced the SLIP and program personnel to parents in English and Spanish. A standard script described the project (vaccinating students using a mobile unit), the importance of vaccines (to keep their children and the community healthy), information regarding when the vaccinations would take place, and where to get more information about the program. School personnel introduced a representative from the research team who introduced the immunizing team (who spoke in both English and Spanish) and notified parents that the team would be available for questions and answers at the end of the parent meeting/open house. At the end of the parent meetings, members of the team were available for face-to-face
T.L. Won et al. / Journal of Adolescent Health 56 (2015) S33eS39
S35
Figure 1. Items assessing parent trust in school-located immunization programs (SLIPs). Trust scale adapted for SLIPs based on original scale published in Dugan et al. [14]. *Negatively worded item was reverse-numbered for trust scale.
introductions and conversations. During this intervention meeting, flyers with pictures/information regarding SLIP team members and their roles in the program were made available for distribution. The flyers directed parents to the project’s Facebook page. The Facebook page contained a description of the project in English and Spanish, a time line of the immunization dates for each school, and the pictures of the SLIP team and information detailing their credentials and roles in the SLIP. Similar information was posted on the intervention schools’ Web sites. The school-located immunization program The SLIP administered vaccinations at all schools in the fall 2012, the spring 2013, and the fall 2013 to allow for completion of multidose vaccines such as HPV. Each vaccination visit was preceded by the distribution of a packet that contained an explanation of the program, a screening instrument for VFC eligibility, a form allowing parents to consent to individual vaccines, vaccine information sheets for all offered vaccines, enrollment forms for the state immunization information system, and a screening form for allergies to any vaccine components (including egg allergy). All materials were provided in both English and Spanish. The Texas Children’s Mobile Van and its staff joined study personnel at each school to administer all vaccines for which consent was received. In fall 2012, vaccines offered included influenza (mist and injection), HPV, Tdap, MCV4, varicella, hepatitis A, and measles, mumps, and rubella (MMR). In spring 2013, vaccines offered included influenza (mist only), HPV, Tdap, MCV4, varicella and hepatitis A. In fall 2013, the same vaccines were offered as in fall 2012 with the exception of MMR. State immunization information system records, school records, and parent-provided signed documentation of immunization were used to identify needed vaccines. Because nearly all students were eligible for at least one immunization (seasonal flu) in the fall programs, fall data were used to determine SLIP participation rates with school enrollment numbers as denominators. Data analysis Surveys with incomplete trust item responses or those that were determined to have been completed by students (reported
age less than 18 years and school education less than middle school) were excluded from analyses. The 2012 survey data were used to analyze associations between demographic items and trust using two-sample t tests, one-way analysis of variance, and Pearson correlation. Demographic items found to significantly (p .05) associate with mean trust scores in the bivariate analysis were entered into a multiple linear regression model. A mean trust score was calculated each year from parents’ responses to the trust scale items. T tests (using the central limit theorem) were used to examine the difference in mean baseline trust scores between intervention and control schools in year 1 and also in year 2. All analyses were conducted using SPSS version 20. Results Sample In fall 2012, 1,913 surveys were completed from an estimated 8,900 distributed for a response rate of approximately 21%. The total number of surveys included in analyses was 1,608. In fall 2013, 945 surveys were returned from an estimated 8,175 distributed for a response rate of approximately 12%; 844 surveys were included in the analysis. Response rate for the survey distribution in April 2013 was poor (357/8,900¼ 4%); these data were not analyzed. Demographic data for parents/guardians completing the trust assessment before SLIP implementation in both years are described in Table 1. Associations between demographic factors and baseline mean trust in fall 2012 Variables significantly associated with parental trust are presented in Table 2. In bivariate analysis, items significantly associated with mean trust score included annual household income, survey language version, use of Spanish language at home, participation in a previous SLIP, Hispanic ethnicity, parent education level, child’s health insurance status, and perceived vaccine importance. In the multiple linear regression analysis, annual household income, survey language version, participation in a previous SLIP, child’s health insurance status, and perceived vaccine importance remained significant. The
S36
T.L. Won et al. / Journal of Adolescent Health 56 (2015) S33eS39
Table 1 Demographic characteristics and beliefs about vaccine importance among parents completing surveys in 2012 and 2013
Hispanic ethnicity Race Black White Asian/Native American Other Education Middle school/some high school High school degree/GED Some college College or graduate degree Primary language spoken at home English Spanish Spanish and English Annual household income $10,000 $10,001e$20,999 $21,000e$50,999 $51,000 Child’s insurance Medicaid/CHIP Self-pay Private insurance Child has a medical home Child participated in previous SLIP Agreement with vaccine importance Strongly disagree, disagree, neutral Agree Strongly agree
Fall 2012 (N ¼ 1,608); n (%a)
Fall 2013 (N ¼ 844); n (%a)
1,348 (85)
723 (86.6)
175 521 61 422
(14.8) (44.2) (5.1) (35.8)
75 358 46 126
(12.4) (59.2) (7.6) (20.8)
681 504 234 133
(43.5) (32.5) (15.1) (8.6)
294 294 126 84
(36.8) (36.8) (15.8) (10.5)
500 (31.7) 645 (40.3) 431 (26.9)
254 (31.3) 347 (42.8) 196 (24.2)
497 464 492 71
(32.6) (30.4) (32.3) (4.7)
256 164 219 57
(36.8) (23.6) (31.5) (8.1)
1,044 296 191 1,310 231
(68.2) (19.3) (12.5) (81.9) (14.5)
511 97 119 684 175
(64.6) (12.2) (15) (84.3) (21.5)
115 (7.2) 244 (15.4) 1,229 (77.4)
62 (7.7) 105 (12.9) 645 (79.4)
CHIP ¼ Children’s Health Insurance Program; GED ¼ General Education Development (high school equivalence test); SLIP ¼ school-located immunization program. a Valid percent.
multivariable model had an adjusted R2 of .06 with an F value of 10.59 (p < .001), indicating that the nine variables in the regression model account for 6% of the variance in parental trust in SLIPs.
Table 2 Factors associated with parental trust in SLIPs in year 1: results of bivariate and multivariable analysis (N ¼ 1,608) Factor
Na
Mean trust scoreb Ethnicity Hispanic Non-Hispanic Education level Less than high school diploma High school diploma or greater Annual income $50,999 $51,000 Previous SLIP participation Yes No Survey language version English Spanish Language spoken at home English at home Spanish or English/Spanish at home Insurance type Private insurance (reference) Medicaid/CHIP Self-pay Perceived vaccine importance
Multivariablec,
Bivariate p value
b
1,348 238
3.61 3.51
.048
.041
681 871
3.63 3.56
.025
.037
1,453 71
3.60 3.32
www.jahonline.org Original article
Trust and a School-Located Immunization Program Tiana L. Won a, Amy B. Middleman, M.D., M.S.E.D., M.P.H. b, *, Beth A. Auslander, Ph.D. c, and Mary B. Short, Ph.D. d a
Baylor College of Medicine, Houston, Texas Department of Pediatrics, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma Department of Pediatrics, University of Texas Medical Branch, Galveston, Texas d Department of Psychology, University of HoustoneClear Lake, Clear Lake, Texas b c
Article history: Received July 11, 2014; Accepted September 29, 2014 Keywords: School-located immunization programs; Adolescent health services; Trust; Socioeconomic status; Hispanic
A B S T R A C T
Purpose: To determine variables associated with parental trust in a school-located immunization program (SLIP) and the effect of trust-building interventions on trust and participation in SLIPs. Methods: Parents among eight schools randomized to a trust-building intervention or a control condition (four schools each) completed a five-item trust survey before SLIP implementation both inyear 1 (fall 2012) and in year 2 (fall 2013). Mean trust scores were calculated. Associations between baseline demographic and experiential variables and mean trust scores were analyzed. Mean trust scores in intervention and control schools were compared before SLIP in years 1 and 2, and SLIP participation rates were noted. Results: From year 1, 1,608 parent surveys were analyzed. Baseline mean trust score across schools was 3.59 of 5 (5 ¼ highest trust). In a multiple linear regression model, annual household income, survey language version, participation in a previous SLIP, child’s health insurance status, and perceived vaccine importance were significantly associated with parental trust in SLIPs (R2¼ .06, p < .001). There was no difference in mean trust scores between intervention and control schools (p ¼ .8). In year 2, 844 surveys were analyzed, and a modest difference was observed between intervention and control schools (mean trust score ¼ 3.66 and 3.57, respectively, p ¼ .07). SLIP participation rates appeared higher in intervention (7.7%) versus control schools (4.3%) in year 1. Conclusions: Baseline trust in SLIPs among a low-income, largely Hispanic group of parents in Texas was moderately high. Factors associated with trust included demographic and experiential variables, and interventions aimed at increasing parents’ perception of vaccine importance and participation in SLIPs may be effective in increasing parental trust in SLIPs. Ó 2015 Society for Adolescent Health and Medicine. All rights reserved.
Since 2005, vaccine recommendations for adolescents by the Advisory Committee on Immunization Practices have expanded to include the tetanus toxoid, diphtheria toxoid, acellular Conflicts of Interest: The authors declare no conflicts of interest. Disclaimer: Publication of this article was supported by the Society for Adolescent Health and Medicine through a grant from Merck & Co., Inc. The opinions or views expressed in this supplement are those of the authors and do not represent the position of the funder. * Address correspondence to: Amy B. Middleman, M.D., M.S.E.D., M.P.H., Department of Pediatrics, University of Oklahoma Health Sciences Center, 1200 Children’s Avenue, Oklahoma City, OK 73104. E-mail address: [email protected] (A.B. Middleman). 1054-139X/Ó 2015 Society for Adolescent Health and Medicine. All rights reserved. http://dx.doi.org/10.1016/j.jadohealth.2014.09.018
IMPLICATIONS AND CONTRIBUTION
Baseline trust in schoollocated immunization programs (SLIPs) among low-income, primarily Hispanic, middle school parents is moderately high. Variables predicting greater trust in SLIPs include lower income, completing the survey in Spanish, prior SLIP participation, self-pay health insurance status, and greater perceived vaccine importance. Participation rates were relatively higher in schools receiving a trustbuilding intervention versus control schools.
pertussis vaccine (Tdap); meningococcal conjugate vaccine (MCV4); human papillomavirus vaccine (HPV); and annual influenza vaccine. Immunization rates for adolescents have steadily increased but remain below rates achieved for younger children. Coverage for Tdap and MCV4 is closer to the Healthy People 2020 target (of 80%) at 85% and 74%, respectively, but the coverage gap between Tdap and HPV (84.6% for Tdap vs. 53.8% for at least one HPV dose among females in 2012) suggests missed opportunities to fully vaccinate adolescents [1]. There are multiple challenges to achieve high rates of vaccination among adolescents, which requires identifying more effective venues for
S34
T.L. Won et al. / Journal of Adolescent Health 56 (2015) S33eS39
immunization [2]. School-located immunization programs (SLIPs) provide immunizations to students and represent a convenient alternative site. Multiple organizations have acknowledged SLIPs as a way to efficiently access the adolescent population for immunization [3,4]. Their efficacy in reaching adolescents has also been recognized in countries such as the United Kingdom, where a school-based program achieved an HPV vaccination rate of 76.4% for the first dose of HPV in 2009e2010 [5]. Implementing SLIPs in the United States presents several challenges beyond the complex reimbursement issues. A review of influenza-only SLIPs conducted between 2003 and 2010 demonstrated a vaccination rate ranging from 7% to 58%, with higher rates among elementary compared with middle and high school students [6]. Another study confirmed this trend and documented mean immunization rates of 21.5% among elementary school children compared with 10.3% and 5.8% among middle and high school students, respectively [7]. Many variables influence whether parents use SLIPs, including the program’s ability to retrieve consent forms from parents, the use of incentives, parental education and follow-up [8]. Not surprisingly, concerns about vaccine safety also affect parents’ decisions to vaccinate their children [9,10]. A recent qualitative study indicated that parental trust may be another factor in parents’ decisions to use an SLIP [11]. Challenges related to parents’/ adolescents’ trust of vaccine providers may also influence adolescent participation in SLIPs. Although the concept of trust relates to factors such as satisfaction, competency, privacy, and communication, previous research has supported its definition as “the optimistic acceptance of a vulnerable situation in which the truster (i.e., patient) believes the trustee (i.e., provider) will care for the truster’s interests” [12]. The importance of trust in any providerepatient relationship is well recognized and has implications for patient care ranging from adherence to efficacy of treatment [13]. However, it is not known the degree to which parents trust SLIPs, what factors are correlated with trust in SLIPs, and whether interventions aimed at increasing trust could increase parental trust in SLIPs. This study evaluated the baseline level of parental trust in SLIPs and participation rates in years 1 and 2 of an SLIP among a lowincome, largely Hispanic population of middle school students. In addition, this study assessed whether basic demographic variables, perceptions of vaccine importance, and history of SLIP experience were associated with middle school parents’ trust in SLIPs before any intervention in year 1, and whether trust in SLIPs was greater in year 2 among parents from schools that had received a simple trust-building intervention in year 1 compared with schools not receiving a trust-building intervention. Methods School selection and school-located immunization program time line Schools in a large, urban school district with >90% participation in the free lunch program were offered the opportunity to participate in a multivisit SLIP that administered Vaccines for Children (VFC) Program vaccinations without cost to all VFC eligible students. Eight middle schools (grades 6e8) with enrollment varying from 600 to 1,500 students (total of approximately 8,750 in year 1) accepted the invitation to participate in the SLIP project initiated in fall 2012. The eight schools were randomized into four control and four intervention schools.
Distribution of parent surveys with trust measure A packet containing a two-page questionnaire and a cover letter explaining the study was distributed to all students in all eight schools to take home to parents and return to teachers or school nurses before the initiation of the SLIP in fall 2012, after SLIP in the spring 2013, and again before the SLIP in fall 2013. The questionnaire contained a five-item trust survey and several demographic items and was provided to each family in both English and Spanish. Completed questionnaires were retrieved from schools by a member of the research team approximately 3 weeks after distribution. Each survey was labeled with a unique number to identify the school and prevent duplication of data entry. When returning the survey, parents could write their child’s name on the cover letter for the chance to win an MP3 player. Cover letters with identifying information were removed from the surveys on receipt and retained separately to ensure confidentiality of the survey responses. Survey content and measurement Demographic items included age, race, ethnicity, highest level of education achieved, primary language spoken at home, annual household income, child’s health insurance status, and child’s participation in a medical home. History of participation in SLIPs and the perceived importance of vaccination were assessed. In addition, another variable noted was whether the participants completed the survey using the English or Spanish version. Parental trust in SLIPs was evaluated by adapting a five-item scale developed by Dugan et al. [14] shown to be valid for use in assessing patient trust in physicians, health insurers, and the medical profession as a whole (Figure 1). Parents completed the five-item survey using a five-point Likert scale (ranging from 1¼ strongly disagree to 5¼ strongly agree). One of the five statements was negatively worded and was therefore reverse scored for data analyses. Cronbach alpha measuring internal consistency reliability for the present sample was .71 [15]. Parent/student education During full school student assemblies or successive lunch periods, study personnel visited all eight schools in the program and spoke briefly to students about the importance of adolescent immunizations and the availability of the SLIP. All information was provided using a standard script. Description of intervention. The four intervention schools received an additional, simple trust-building intervention to address parental trust. Research staff visited intervention schools during parent open houses or other parent meetings at the schools. These meetings introduced the SLIP and program personnel to parents in English and Spanish. A standard script described the project (vaccinating students using a mobile unit), the importance of vaccines (to keep their children and the community healthy), information regarding when the vaccinations would take place, and where to get more information about the program. School personnel introduced a representative from the research team who introduced the immunizing team (who spoke in both English and Spanish) and notified parents that the team would be available for questions and answers at the end of the parent meeting/open house. At the end of the parent meetings, members of the team were available for face-to-face
T.L. Won et al. / Journal of Adolescent Health 56 (2015) S33eS39
S35
Figure 1. Items assessing parent trust in school-located immunization programs (SLIPs). Trust scale adapted for SLIPs based on original scale published in Dugan et al. [14]. *Negatively worded item was reverse-numbered for trust scale.
introductions and conversations. During this intervention meeting, flyers with pictures/information regarding SLIP team members and their roles in the program were made available for distribution. The flyers directed parents to the project’s Facebook page. The Facebook page contained a description of the project in English and Spanish, a time line of the immunization dates for each school, and the pictures of the SLIP team and information detailing their credentials and roles in the SLIP. Similar information was posted on the intervention schools’ Web sites. The school-located immunization program The SLIP administered vaccinations at all schools in the fall 2012, the spring 2013, and the fall 2013 to allow for completion of multidose vaccines such as HPV. Each vaccination visit was preceded by the distribution of a packet that contained an explanation of the program, a screening instrument for VFC eligibility, a form allowing parents to consent to individual vaccines, vaccine information sheets for all offered vaccines, enrollment forms for the state immunization information system, and a screening form for allergies to any vaccine components (including egg allergy). All materials were provided in both English and Spanish. The Texas Children’s Mobile Van and its staff joined study personnel at each school to administer all vaccines for which consent was received. In fall 2012, vaccines offered included influenza (mist and injection), HPV, Tdap, MCV4, varicella, hepatitis A, and measles, mumps, and rubella (MMR). In spring 2013, vaccines offered included influenza (mist only), HPV, Tdap, MCV4, varicella and hepatitis A. In fall 2013, the same vaccines were offered as in fall 2012 with the exception of MMR. State immunization information system records, school records, and parent-provided signed documentation of immunization were used to identify needed vaccines. Because nearly all students were eligible for at least one immunization (seasonal flu) in the fall programs, fall data were used to determine SLIP participation rates with school enrollment numbers as denominators. Data analysis Surveys with incomplete trust item responses or those that were determined to have been completed by students (reported
age less than 18 years and school education less than middle school) were excluded from analyses. The 2012 survey data were used to analyze associations between demographic items and trust using two-sample t tests, one-way analysis of variance, and Pearson correlation. Demographic items found to significantly (p .05) associate with mean trust scores in the bivariate analysis were entered into a multiple linear regression model. A mean trust score was calculated each year from parents’ responses to the trust scale items. T tests (using the central limit theorem) were used to examine the difference in mean baseline trust scores between intervention and control schools in year 1 and also in year 2. All analyses were conducted using SPSS version 20. Results Sample In fall 2012, 1,913 surveys were completed from an estimated 8,900 distributed for a response rate of approximately 21%. The total number of surveys included in analyses was 1,608. In fall 2013, 945 surveys were returned from an estimated 8,175 distributed for a response rate of approximately 12%; 844 surveys were included in the analysis. Response rate for the survey distribution in April 2013 was poor (357/8,900¼ 4%); these data were not analyzed. Demographic data for parents/guardians completing the trust assessment before SLIP implementation in both years are described in Table 1. Associations between demographic factors and baseline mean trust in fall 2012 Variables significantly associated with parental trust are presented in Table 2. In bivariate analysis, items significantly associated with mean trust score included annual household income, survey language version, use of Spanish language at home, participation in a previous SLIP, Hispanic ethnicity, parent education level, child’s health insurance status, and perceived vaccine importance. In the multiple linear regression analysis, annual household income, survey language version, participation in a previous SLIP, child’s health insurance status, and perceived vaccine importance remained significant. The
S36
T.L. Won et al. / Journal of Adolescent Health 56 (2015) S33eS39
Table 1 Demographic characteristics and beliefs about vaccine importance among parents completing surveys in 2012 and 2013
Hispanic ethnicity Race Black White Asian/Native American Other Education Middle school/some high school High school degree/GED Some college College or graduate degree Primary language spoken at home English Spanish Spanish and English Annual household income $10,000 $10,001e$20,999 $21,000e$50,999 $51,000 Child’s insurance Medicaid/CHIP Self-pay Private insurance Child has a medical home Child participated in previous SLIP Agreement with vaccine importance Strongly disagree, disagree, neutral Agree Strongly agree
Fall 2012 (N ¼ 1,608); n (%a)
Fall 2013 (N ¼ 844); n (%a)
1,348 (85)
723 (86.6)
175 521 61 422
(14.8) (44.2) (5.1) (35.8)
75 358 46 126
(12.4) (59.2) (7.6) (20.8)
681 504 234 133
(43.5) (32.5) (15.1) (8.6)
294 294 126 84
(36.8) (36.8) (15.8) (10.5)
500 (31.7) 645 (40.3) 431 (26.9)
254 (31.3) 347 (42.8) 196 (24.2)
497 464 492 71
(32.6) (30.4) (32.3) (4.7)
256 164 219 57
(36.8) (23.6) (31.5) (8.1)
1,044 296 191 1,310 231
(68.2) (19.3) (12.5) (81.9) (14.5)
511 97 119 684 175
(64.6) (12.2) (15) (84.3) (21.5)
115 (7.2) 244 (15.4) 1,229 (77.4)
62 (7.7) 105 (12.9) 645 (79.4)
CHIP ¼ Children’s Health Insurance Program; GED ¼ General Education Development (high school equivalence test); SLIP ¼ school-located immunization program. a Valid percent.
multivariable model had an adjusted R2 of .06 with an F value of 10.59 (p < .001), indicating that the nine variables in the regression model account for 6% of the variance in parental trust in SLIPs.
Table 2 Factors associated with parental trust in SLIPs in year 1: results of bivariate and multivariable analysis (N ¼ 1,608) Factor
Na
Mean trust scoreb Ethnicity Hispanic Non-Hispanic Education level Less than high school diploma High school diploma or greater Annual income $50,999 $51,000 Previous SLIP participation Yes No Survey language version English Spanish Language spoken at home English at home Spanish or English/Spanish at home Insurance type Private insurance (reference) Medicaid/CHIP Self-pay Perceived vaccine importance
Multivariablec,
Bivariate p value
b
1,348 238
3.61 3.51
.048
.041
681 871
3.63 3.56
.025
.037
1,453 71
3.60 3.32
