Education, Immigration and HPV Vaccination: an Informational Randomized Trial

Utbildningsnivå, Invandringsstatus Och Viljan Att Vaccinera Mot HPV Som Funktion av Informationsformulering: En Randomiserad Studie

Counteracting misinformation on childhood vaccines remains a priority for public health in industrialized countries. Previous research showed that misinformation-induced vaccine hesitancy particularly concerns very highly or very lowly educated parents, and, especially in Europe, specific groups of immigrants. Misinformation framing directly targets specific sub-population of parents by exploiting different cognitive biases, and specific concerns based on cultural norms: this project aims at testing the effectiveness of similar framing techniques applied to positive information on the HPV vaccine by conducting a Randomized Controlled Trial in Stockholm, Sweden. It randomizes emotionally and scientifically/statistically framed information addressing the specific concerns reported by previous literature.

In Europe, non-European immigrants display lower vaccination take-up than natives, a higher prevalence of STDs, and lower usage of publicly-funded treatments. Therefore, migrants (and especially women) face worse preventable health issues. Qualitative research suggests that migrants' lower take-up might be due to culturally founded concerns, which are then exploited and targeted by anti-vax information. These are adverse effects on fertility and, in the case of the HPV vaccine, which is administered at a very young age to prevent a sexually-transmissible infection, on reputational concerns related to incentivizing pre-marital sexual activity. Existing literature also investigates parental education as an alternative explanation to vaccines' mistrust, finding mixed evidence which suggests a non-linear relationship between education level and willingness to vaccinate children. In particular, qualitative research highlights that targeted misinformation addresses emotionally charged contents to lowly educated parents, and pseudo-scientific content to highly educated parents. In this study, 7500 mothers of children (girls and boys) who are due to be offered the HPV vaccine in fall 2021 within the Swedish national vaccination program are randomly sampled. Mothers are stratified according to immigration status, and are sampled from the entire relevant population through Swedish population registers. There are 5 strata: (i) Mothers were born in Iran, Iraq, Afghanistan, Somalia, Eritrea (irrespective of education) (N=2500); (ii) Mothers born in Sweden and have attended at most 3 years of high school (N=1600); (iii) Mothers born in Sweden and have attended more than 3 years of high school and have not attended any university (N=1400); (iv) Mothers born in Sweden and have achieved at most undergraduate education (N=1000); (v) Mothers born in Sweden with graduate education after obtaining an undergraduate degree (N=1000). Sample sizes are chosen with the expectation of having 500 actual participants per stratum, based on participation rate estimates from the implementer (Statistics Sweden, SCB). Within each stratum, the investigators randomize 3 types of information sheets on the HPV vaccine: emotionally-framed (T1), scientifically and statistically framed (T2) and uninformative placebo (C). Along the information sheet, all mothers receive a first survey covering their beliefs on vaccines (henceforth defined "baseline survey" for simplicity, despite being administered along the treatment), which does not vary across treatment arms. The information sheet and the baseline survey invitation are sent by ordinary post in June 2021 to the mother's address, and the survey can be compiled both on paper and online. In July and August, mothers who have still not compiled the baseline survey will receive 3 reminders: each reminders contains the information sheet, which depending on the specific reminder can be printed or accessible in digital format when accessing the online survey. Between September and October, all mothers will have to decide whether to authorize HPV vaccination for their child, which takes places directly at school and free of charge. In November, mothers who have responded to the baseline survey receive an invitation to an endline survey. All information sheets and surveys are available in Swedish, English, Arabic and Farsi. Informed consent is obtained in the baseline survey. For mothers who choose to participate, investigators will obtain, from population registers, background variables concerning the mother herself, the other parent and the child who is due to immunization. These cover specific information on parents' education, income, wealth and occupation, marital status, number of total children, gender of the child due to HPV vaccination, and if she/he received the previous vaccine in the national program's schedule (3 to 4 years before our intervention), order of the child among siblings.

Cervical Cancer, Vulvar Cancer, Anus Cancer, Penis Cancer, Mouse Papilloma, Head Cancer, Genital Wart, Neck Cancer

The receiver is a mother born in Sweden whose education is at least 3 years of high school but no more than high school diploma.

The receiver is a mother born in Sweden whose education is at least some college (undergrad) but does not exceed an undergraduate degree

The receiver is a mother born in Sweden whose education is an undergraduate degree plus at least some graduate education

A4 information sheet on HPV vaccine containing a reminder of the upcoming HPV vaccination, and information on: (i) its safety and efficacy in preventing a number of cancerous diseases, affecting both women and men; (ii) what is HPV and how contagion happens; (iii) the consequences of cervical cancer for women, in terms of invasive medical treatments and temporary or permanent loss of fertility. The information is framed in non-scientific, non-statistical terms and is emotionally charged: the consequences of cervical cancers are portrayed by testimonies of cancer survivors from Sweden.

A4 information sheet on HPV vaccine containing a reminder of the upcoming HPV vaccination, and information on: (i) its safety and efficacy in preventing a number of cancerous diseases, affecting both women and men; (ii) what is HPV and how contagion happens; (iii) the consequences of cervical cancer for women, in terms of invasive medical treatments and temporary or permanent loss of fertility. The information is framed in scientific terms and incidence and mortality statistics are reported. The information is not emotionally charged but rather aseptic.

A4 information sheet containing a reminder of the upcoming HPV vaccination, and information on the history and characteristics of the Swedish national vaccination program.

Investigators will assess if there are significant differences in the frequency of parents who decide to vaccinate across treatment arms, within each stratum

From treatment administration: 8 to 18 weeks, depending on when the single subject has read the information from the behavioral intervention

The investigators will examine whether across treatment groups, within strata, parents show different degrees of agreement with false statements concerning the risks of HPV, after reading the information sheet corresponding to their treatment arm

The investigators will examine whether across treatment groups, within strata, parents show different degrees of agreement with false statements concerning the risks of HPV, after they have already decided whether to vaccinate their child and vaccination has taken place

Assessed in the endline survey: from treatment, 8 to 18 weeks, depending on when the single subject has read the information from the behavioral intervention

Behavioral mechanisms' analysis (Measurement: questionnaire, likert scale of agreement with statements on the hypothesized beliefs)

Investigators will assess, within strata and across treatment arms, whether individual treatments affected common beliefs about the benefits and risks of the HPV vaccine, which are observed more often among specific group of parents and which are believed to drive vaccine hesitancy. These beliefs are directly addressed by the informational intervention. They concern the perceived benefits/risks of the HPV vaccine relative to fertility, cancer and other illnesses' incidence and sexual behavior, such as whether the HPV vaccine incentivizes sexual activity of teenagers, or if it changes the probability that the teenager develops a cancer in the future. In the second case, the vaccine actually decreases that probability, but anti-vax information would postulate the opposite: we test whether across treatment arms there are significant differences in the degree of agreement with statements on these beliefs.

These beliefs are elicited after treatment and after the vaccination choice has taken place, in the endline survey (from treatment, 8 to 18 weeks, depending on when the single subject has read the information from the behavioral intervention )

Heterogeneity analysis (ANCOVA regressions including baseline covariates' interaction with treatment binary indicators)

Investigators will assess whether the treatment effect has a significant interaction with the following variables which, unless specified otherwise, are extracted from population registers based on administrative data: -Field of education (medical vs non-medical and quantitative vs non-quantitative) -Presence of a medical doctor in the close network of the parent[from baseline survey] and other parent being a health worker -Self-declared percentage of the information sheet they actually read [from baseline survey] -Preferred source of information on vaccines [from baseline survey] - Household and mother's wealth - Previous knowledge of HPV and HPV vaccine [from baseline survey] -Marital status of the mother - Child being the firstborn - Child being female

All indicators are observed at baseline (observed by investigators on the day of treatment: these covariates are either recorded before treatment in the form of administrative data, or elicited with survey questions on the day of treatment)

Investigators will test, within each stratum, whether the self-reported intention to vaccinate [post-treatment baseline survey] and the objective measure of vaccination [from vaccination registers] are statistically different. While the objective vaccination indicator is by definition a binary variable, the subjective indicator is initially reported on a Likert scale by subjects, and then reduced to a binary indicator by investigators for immediate comparison.

All indicators are observed at endline (i.e. 8 to 18 weeks after treatment, depending on when the single subject has read the information from the behavioral intervention)

We are randomly sampling mothers of children who are due for HPV immunization in fall 2021 according to the Swedish national vaccination program. Only for children who do not have a female guardian we invite a male guardian to participate (N=73).

Inclusion Criteria:- Resident in Stockholm County, Sweden Child attending 5th grade in a school located in the Stockholm County, Sweden Mother born in one of the following countries: Sweden, Eritrea, Somalia, Syria, Afghanistan, Iran, Iraq Gender: female (sampling among mothers, though we observe also information on their partner, even if male). No gender requirement on the child Exclusion Criteria: Mother born outside of Sweden and outside of one of the following countries: Eritrea, Somalia, Syria, Afghanistan, Iran, Iraq Mother already sampled for another child (i.e. excluding children born from the same mother)