Abstract

Objective

To describe the drivers associated with HPV vaccination in adolescent girls and their parent's opinion on the vaccine.

Methods

We conducted an observational and cross-sectional study on adolescent girls and their parents in Valencia (Spain), between September 2011 and June 2012. A consultation was made at a random sample of schools of the 14-year-old girls that should have received the vaccine in the free vaccination programme. We ran a personal survey on knowledge and attitudes regarding HPV infection and the vaccine. A binary logistic regression model was performed to determine which factors were most associated with vaccination.

Results

The survey was run on a binomial of 1,278 girls/mothers in 31 schools, to which 833 girls and their mothers responded (64.0%). The factors associated with vaccination were: country of origin of the families (adjusted OR [aOR]: 0.49; 95% confidence interval [95%CI]: 0.24-0.98), civil status of the parents (aOR: 0.33; 95%CI: 0.13-0.81), knowledge/beliefs about the vaccine when the source of information was the nurse (aOR: 1.83; 95%CI: 1.01-3.35), information source about the vaccine (aOR: 2.32; 95%CI: 1.37-3.92), preventive health centre visits (aOR: 2.1; 95%CI: 1.10-4.07), and nurse advice (aOR: 6.6; 95%CI: 3.19-13.56).

Conclusions

The main factor associated with HPV vaccination was the advice of health professionals. Therefore, the most effective interventions to improve vaccination coverage should focus on health professionals.

Keywords:
Human papillomavirus; Vaccines; Drivers; Attitudes; Nurse

Resumen

Objetivo

Describir qué factores se asocian a la vacunación contra el virus del papiloma humano (VPH) en adolescentes y la opinión de sus padres/madres sobre dicha vacuna.

Métodos

Se llevó a cabo un estudio observacional y transversal en chicas adolescentes y sus progenitores en Valencia (España), entre septiembre de 2011 y junio de 2012. Se realizaron consultas a las niñas de 14 años que tendrían que haber sido vacunadas dentro del programa de vacunación gratuita, en una muestra aleatoria de colegios. Se preguntó sobre conocimientos y actitud respecto a la infección por el VPH y la vacuna, mediante entrevista personal. Se llevó a cabo un modelo de regresión logística binaria para determinar qué factores estaban más asociados con la vacunación.

Resultados

Se entrevistó a 1278 binomios chica/madre, con una tasa de respuesta del 64,0% (833). Los factores asociados a la vacunación fueron el país de origen de las familias (odds ratio ajustada [ORa]: 0,49; intervalo de confianza del 95% [IC95%]: 0,24-0,98), el estado civil de los progenitores (ORa: 0,33; IC95%: 0,13-0,81), los conocimientos/creencias sobre la vacuna cuando la principal fuente de información fue la enfermera (ORa: 1,83; IC95%: 1,01-3,35), la fuente de información sobre la vacuna (ORa: 2,32; IC95%: 1,37-3,92), las visitas preventivas al centro de salud (ORa: 2,12; IC95%: 1,10-4,07) y el consejo de la enfermera (ORa: 6,57; IC95%: 3,19-13,56).

Conclusiones

El principal factor asociado a la vacuna del VPH fue el consejo del personal sanitario, por lo que las intervenciones para aumentar la cobertura vacunal deben centrarse en este colectivo.

Palabras clave:
Virus del papiloma humano; Vacuna; Factores asociados; Actitud; Enfermería

Introduction

Human papillomavirus (HPV) is a necessary cause of cervical dysplasia and cancer. Genital warts are one of the most prevalent sexually transmitted diseases in Europe. HPV vaccine coverage is thus an important public health concern and understanding the reasons for accepting or rejecting vaccination is essential for increasing compliance.¹1. Mortensen GL. Drivers and barriers to acceptance of human-papillomavirus vaccination among young women: a qualitative and quantitative study. BMC Public Health. 2010;10:68.

The reasons for decreased vaccination coverage seemed to be different according to region and circumstances.²2. Ladner J, Besson MH, Rodrigues M, et al. Performance of 21 HPV vaccination programs implemented in low and middle-income countries, 2009-2013. BMC Public Health. 2014;30:670. There were factors such as the negative impact of mass media, misinformation from anti-vaccination movements, as well as the initial “aggressive” campaigns carried out by commercial companies.³3. Chesson HW, Ekwueme DU, Saraiya M, et al. Cost-effectiveness of human papillomavirus vaccination in the United States. Emerg Infect Dis. 2008;2:244-51. 4. Kim JJ, Goldie SJ. Health and economic implications of HPV vaccination in the United States. N Engl J Med. 2008;359:821-32.^-55. Drolet M, Boily MC, Van de Velde N, et al. Vaccinating girls and boys with different human papillomavirus vaccines: can it optimise population-level effectiveness? PLoS One. 2013;8:e67072. In Spain, these campaigns started a polemical debate even before the vaccination program was launched, possibly creating uncertainty regarding the indications for HPV vaccination among some healthcare professionals. Vaccination coverage in the first campaign, in a cohort of girls born in 1994 vaccinated with three doses, was 75.2% of the total target population. However, in the following campaign (1995 cohort), it decreased to 62.5%. And although it increased again slightly to 67.6% in the 1996 cohort, the overall rate did not improve in the region.⁶6. Navarro-Illana P, Díez-Domingo J, Navarro-Illana E, et al. Knowledge and attitudes of Spanish adolescent girls towards human papillomavirus infection: where to intervene to improve vaccination coverage. BMC Public Health. 2014;14:490. Economic studies carried out before and after the vaccination program stated that vaccination coverage of 70%, or higher, was necessary to predict cost-effectiveness of an intervention.²2. Ladner J, Besson MH, Rodrigues M, et al. Performance of 21 HPV vaccination programs implemented in low and middle-income countries, 2009-2013. BMC Public Health. 2014;30:670. 3. Chesson HW, Ekwueme DU, Saraiya M, et al. Cost-effectiveness of human papillomavirus vaccination in the United States. Emerg Infect Dis. 2008;2:244-51. 4. Kim JJ, Goldie SJ. Health and economic implications of HPV vaccination in the United States. N Engl J Med. 2008;359:821-32.^-55. Drolet M, Boily MC, Van de Velde N, et al. Vaccinating girls and boys with different human papillomavirus vaccines: can it optimise population-level effectiveness? PLoS One. 2013;8:e67072. Nevertheless, these rates were relatively low compared to others obtained in England, Scotland or Wales (84.0%), where funding and implementation conditions were similar.⁷7. Bowyera HL, Marlowa LV, Hibbittsb S, et al. Knowledge and awareness of HPV and the HPV vaccine among young women in the first routinely vaccinated cohort in England. Vaccine. 2013;31:1051-6.

Prior to the vaccination program, literature suggested that ethnicity, socio-demographic and psychosocial factors could play a decisive role in vaccination.⁸8. Reiter PL, Brewer NT, Gottlieb SL, et al. Parents' health beliefs and HPV vaccination of their adolescent daughters. Soc Sci Med. 2009;3:475-80. Research after the vaccination programs were launched⁶6. Navarro-Illana P, Díez-Domingo J, Navarro-Illana E, et al. Knowledge and attitudes of Spanish adolescent girls towards human papillomavirus infection: where to intervene to improve vaccination coverage. BMC Public Health. 2014;14:490. ⁸8. Reiter PL, Brewer NT, Gottlieb SL, et al. Parents' health beliefs and HPV vaccination of their adolescent daughters. Soc Sci Med. 2009;3:475-80. 9. Gottlieb SL, Brewer NT, Sternberg MR, et al. Human papillomavirus vaccine initiation in an area with elevated rates of cervical cancer. J Adolesc Health. 2009;45:430-7. 10. Conroy K, Rosenthal SL, Zimet GD, et al. Human papillomavirus vaccine uptake, predictors of vaccination, and self-reported barriers to vaccination. J Womens Health (Larchmt). 2009;10:1679-86. 11. Jain N, Euler GL, Shefer A, et al. Human papillomavirus (HPV) awareness and vaccination initiation among women in the United States, National Immunization Survey-Adult 2007. Prev Med. 2009;5:426-31. 12. Rosenthal SL, Weiss TW, Zimet GD, et al. Predictors of HPV vaccine uptake among women aged 19-26: importance of a physician's recommendation. Vaccine. 2011;29:890-5. 13. Dempsey AF, Abraham LM, Dalton V, et al. Understanding the reasons why mothers do or do not have their adolescent daughters vaccinated against human papillomavirus. Ann Epidemiol. 2009;8:531-8.^-1414. Caskey R, Lindau ST, Alexander GC. Knowledge and early adoption of the HPV vaccine among girls and young women: results of a national survey. J Adolesc Health. 2009;5:453-62. confirmed that age, perceived access to vaccination, cultural norms, religion, knowledge about the severity of the disease and risk perception, directly influenced the acceptance of the vaccine.¹1. Mortensen GL. Drivers and barriers to acceptance of human-papillomavirus vaccination among young women: a qualitative and quantitative study. BMC Public Health. 2010;10:68. ¹⁵15. Dempsey AF, Patel DA. HPV vaccine acceptance, utilization and expected impacts in the U.S. Where are we now? Human Vaccines. 2010;6:715-20. These factors, together with the usual acceptance difficulties that any vaccine may have, are a challenge to achieve better vaccination coverage rates.

This study was designed to determine the drivers associated with HPV vaccination in the region of Valencia (Spain), and to ascertain whether there was a major determining factor in order to carry out more effective interventions to improve vaccination rates.

Subjects, material and methods

Design

We conducted an observational, and cross-sectional study in adolescent girls and their parents of Valencia (Spain) between September 2011 and June 2012.

Sample size and population

A randomized and stratified selection of schools in the region of Valencia was made, according to funding (public, private or charter school) and geographic location (rural, semirural or urban). This cohort was in the post-vaccination school year and should have received the vaccine during the previous year. We included their mothers (otherwise, fathers or legal guardians) since, in most cases, they had the last word in the decision to vaccinate.

We estimated the prevalence of 70% population adolescent girls vaccinated. We assumed a population of 9,506, with a Type I error (alpha risk) of 0.05, an accuracy of 5% and a dropout rate of 50%. We therefore calculated a sample size of 1,230 girls and respective mothers for our study. Sample size was undertaken using Epidat 3.1.

Measuring instruments and study variables

An envelope with a questionnaire, an information sheet and a parental consent form was given to each girl, which was later returned to the schools. After checking the consent form, we interviewed the girls at the school to guarantee the correct development and understanding of each part of the questionnaire. Before data collection, the research team had carried out a pilot study in the school of a low-income area to assess comprehension difficulties of the questionnaire, as well as the correct data collection procedure.

All families that expressed their refuse to participate were excluded of the study.

The variables collected comprise:

We considered those cases valid where mothers and daughters replied and had consent forms signed.

To avoid response bias and determine the variable ‘vaccination status’ with reliability, we checked the mother and daughter's answers against the Nominal Vaccine Register (NVR) of the Public Health Department, where all the doses administered in the Regional Health System are registered (date, batch, injection site, name of healthcare professional and primary health centre). We found no discrepancy in the ‘Vaccination Status’.

We considered a ‘vaccination status’ to be positive when three doses were administered and registered in the NVR, even if mothers and daughters’ answers did not always coincide. In contrast, a ‘vaccination status’ was negative when the NVR and the mothers, or when the NVR and the girls, both stated that the three doses of the vaccine had not been administered; or when the mothers, daughters and the NVR coincided in not having the three doses.

An uncertain ‘vaccination status’ was that which did not respond to any of the above and those cases were excluded from the multivariate study.

Statistical analysis

Descriptive statistics data are presented as means and standard deviation for continuous variables. The multivariate analysis, first analysing each variable with the vaccination status individually, and later using a binary logistic regression to find the association of each variable with the vaccination status. For the selection of variables of the predictive model we used significance, suitability and opportunity criteria of each of them, following a logistic regression, adjusting the baseline parameters and without forcing the order of the selected variables. Confidence intervals (CI) were calculated for a confidence level of 95%, and p <0.05 was considered statistically significant. Statistical analysis was undertaken using SPSS (version 22; SPSS, Inc. an IBM Company, Chicago, Illinois, USA).

Ethics

The study was conducted according to the principles included in the Helsinki Declaration. The study was approved by the Committee on Research - Ethics Committee of the Conselleria d’Educació (Generalitat Valenciana) in May 2011. All participants were provided full information to consider on the nature of the study, voluntary participation, and confidentiality.

Results

A questionnaire was distributed among 1,278 girls and their mothers, and we obtained 833 adequate answers (64.5%; 95%CI: 61.9-67.1). We chose a sample of 37 schools but only 31 participated (83.8%; 95%CI: 69.9-97.7), either because they were not given authorization by the parent school governing body (n=4) or due to school program issues (n=2).

From the descriptive analysis of the girls’ socio-demographic and family profile (Table 1), we found that 86.1% (95%CI: 83.7-84.8) of the girls were born in Spain, 79% (95%CI: 76.2-81.8) lived with both parents and 85.5% (95%CI: 83.2-88.0) had siblings. Regarding religion, 76.5% mothers and 66.4% daughters were believers, while 46.5% and 30%, respectively, were practicing individuals.

Regarding the multivariate analysis (Table 2), from 833 cases, 73 (8.7%; 95%CI: 6.8-10.6) were excluded due to uncertain vaccination status. A total of 566 cases (74.5%; 95%CI: 71.4-77.6) were positive vaccinations and 194 (25.5%; 95%CI: 22.4-28.6) were negative vaccinations.

Sixty-five variables were selected from a total of 155 to be included in the logistic regression model (40 variables from the girls’ questionnaire and 25 from the mothers’). The final logistic regression model included 84.5% of the cases, with 0.4 Cox & Snell and, 0.6 Nagelkerke coefficients. The success rate in the model prediction in vaccinated cases was 92.1% and 74.6% in non-vaccinated cases (global 87.9%).

After controlling for the covariates included, the multivariate model showed that the main factors associated with the HPV vaccination were the origin of the families (adjusted odds ratio [aOR]: 0.49; p=0.04), the civil status of the parents (aOR: 0.33; p=0.01), knowledge/beliefs about the vaccine when the source of information was the nurse (aOR: 1.83; p=0.048), the information source about the vaccine (aOR: 2.32; p=0.002), preventive healthcare centre visits (aOR: 2.1; p=0.02) and nurse advice (aOR: 6.6; p <0.001).

Discussion

This was one of the most comprehensive randomized study carried out in Spain. It had a large sample size and a data collection period of eighteen consecutive months. It focused on the drivers associated with HPV vaccination in order to carry out more effective interventions to improve vaccination rates.

We observed that 74.5% of girls were vaccinated, with a vaccination rate slightly higher than that published by the Public Health Department that year (66.5%), probably because the profile of the analysed cases which did not answer the questionnaire or the profile of the excluded cases, responded to a slightly less vaccinated group than that included in the analysis. However, when comparing the resulting sample with the socio-demographic, cultural and family profile of the general population⁶6. Navarro-Illana P, Díez-Domingo J, Navarro-Illana E, et al. Knowledge and attitudes of Spanish adolescent girls towards human papillomavirus infection: where to intervene to improve vaccination coverage. BMC Public Health. 2014;14:490. we can observe similar profiles. The sampling bias is thus compensated and it is considered representative.

The main drivers associated with vaccination were:

Of all the factors associated with vaccination, advice from the nurse together with their physician's advice was the main determining driver. Therefore, it is essential that healthcare professionals are adequately trained to inform patients about HPV infection and vaccination. Interventions aimed at increasing vaccine coverage should be focused on raising nurses’ and doctors’ HPV awareness in order to improve the health of adolescent girls.

References

Publication Dates

History