key: cord-0742775-t0so3wuz
authors: Morissette, Alice; Lefebvre, Gabrielle; Bacque-Dion, Claude; Bélanger, Richard; Cazelais-Asselin, Frédérik; Lalonde, Benoît; Dontigny, André; Leatherdale, Scott T.; Haddad, Slim
title: Disparities in high schools' vaccination coverage (COVID-19). A natural experiment in the Province of Quebec
date: 2022-04-20
journal: Prev Med
DOI: 10.1016/j.ypmed.2022.107056
sha: e3324dd4f601c34d1c753db847338be552c69de2
doc_id: 742775
cord_uid: t0so3wuz

Teenagers' vaccination has become crucial to limit the COVID-19 transmission in the population. To increase the vaccination rate of this age group, a school-based vaccination campaign was launched in Québec, Canada from June 7 to 18, 2021. This study aimed to analyze trajectories of vaccination coverage over time among students attending 37 high schools. The study explored whether school-based vaccination campaigns contributed to the progression of the vaccination coverage and attenuated disparities in vaccination coverage across schools. On average, first dose coverage quickly increased from 30.6% to 81.5% between June 6 and 18, 2021, after the launch of the campaign. As of August 13, 2021, first dose coverage had reached 87.9% and 64.9% for the second dose coverage. Public schools with poorer student populations had 6.5 points of percentage lower first dose vaccination rates (95%CI 0.3%; 12.6%) compared to other schools. A higher level of concern related to the pandemic among students was associated with a 4.3 points of percentage increased coverage (95%CI 0.7%; 8.0%). The initial uneven distribution in first dose coverage decreased dramatically by the end of the campaign. Similar trends were observed for the second dose, although between schools' inequality at the end of the period of observation was significantly larger. The school-based vaccination campaign might have initially contributed to a prompt rise in vaccination coverage and helped the disadvantaged schools to reach similar vaccination coverage as seen in other schools. In addition to being an efficient way to achieve rapidly high vaccination coverage, the school-based approach might contribute to increase equity in vaccination distribution.

On May 5, 2021, Health Canada expanded the eligibility for the PFIZER-BioNTech vaccine and approved its use within 12 to 15 year olds. 1 In Québec, teenagers aged [12] [13] [14] [15] [16] [17] have been eligible to receive the PFIZER-BioNTech vaccine for free in vaccination centers since May 25, 2021. A schoolbased vaccination campaign was launched during the weeks of June 7-11 and June [14] [15] [16] [17] [18] 2021 where all students attending a private and public school in Québec had access to a mobile vaccination clinic at their school or transportation to a vaccination center from their school. 2 Initially, 8 weeks between the 2 doses was requested, however since July 15, 2021, teenagers can receive both doses 4 weeks apart in vaccination clinics only. 3 At the beginning of the campaign, the Quebec Minister of Health aimed to ensure that at least 75% 4 of 12-17 year-olds received 2 doses of vaccine before the start of the 2021 school year (September) to consider reducing school-specific pandemic restrictions.

Achieving those objectives relies to some extend on both the ability to reach the target populations and their level of compliance to vaccination. School-based approach have repeatedly been alleged to significantly increase the vaccination coverage among students for vaccines other than COVID-19. [5] [6] [7] [8] [9] It is considered the baseline approach when a high vaccination coverage among children and teenagers must be reached to reduce illness spread in communities. For instance, in 2017, this effective approach was used in 60% of the World Health Organization (WHO) member states for the delivery of routine vaccines, with tetanus and diphtheria vaccines being the most common. 5 The pediatricians and family physicians of Oregon, in the United States, consider vaccination in a school setting to be even more efficient during outbreaks since it allows to reach a large population in a short period of time. 10 Another advantage of school-based vaccine administration (SBA) lies in their possible ability to attenuate social inequalities in vaccination coverage. Previous studies suggest that lower socioeconomic status is linked to unwillingness to get vaccinated. [11] [12] [13] However, a study examining the uptake of a school-based HPV vaccination program in England suggests that with this approach, the level of deprivation does not appear to be associated with vaccination coverage. 14 On the other hand, in a context where immunization is offered on a voluntary basis, promptly immunizing a large segment of youths requires a high level of adherence to immunization among students and parents. Parent's attitude towards vaccination represents a key predictive factor in relation to children vaccination. A meta-analysis on parent's willingness to vaccinate their children against COVID-19 reported that some of the major predictors of vaccine acceptance were higher socio-economic status, higher level of knowledge and higher level of concern J o u r n a l P r e -p r o o f towards COVID-19. 13 Therefore, one might expect that vaccination coverage of a school at any given time would likely be impacted by these contextual factors.

This study aims to explore trajectories of vaccination coverage (first and second dose) over time among high school students in the region of Québec City, Québec. It first focusses on the possible contribution of the school-based approach implemented to: (i) the progression of vaccine coverage among teenagers, (ii) reduce the disparities in the vaccination coverage among schools. The ongoing natural experiment in Québec offers an interesting opportunity to compare the vaccination coverage among schools prior to the launch of the school-based campaign in June with the coverage following this campaign. It also permits to compare the vaccination coverage of the first and second dose since students did not have access to a school-based vaccination program for their second dose. The study also tests the above-mentioned hypotheses of the associations between schools' vaccination coverage, and their student: (i) attitudes towards vaccines and COVID19; (ii) socioeconomic level.

This is a prospective cohort study that uses data form a convenience sample of 37 secondary schools located in the Region of Québec City. School-level vaccination coverage for first and second dose was computed at six time-points; prior to the launch of the school campaign (June 6), and post-campaign approximately every 2 weeks until mid-August. Eligible schools were private and public schools offering regular curricula to students that also had participated in the 2021 wave of the COMPASS study in Québec. COMPASS is a prospective cohort study designed to collect student-and school-level health behaviour data annually across Canada. 15 Among participating schools in Québec, 29 public and 8 private schools (n=37) were included in this study. Given that the unit of observation is the school, the student population of the 37 schools (n=28 966) was used to measure the vaccination coverage in these schools whereas the student population that participated to the COMPASS study (n=21 626) was used for the analysis that were based on student responses collected in the COMPASS survey: (i) support to immunization; (ii) level of concern; (iii) level of knowledge.

The unit of observation is the school. The percentage of students immunized at a given time is the dependent variable. Vaccination coverage for each dose was derived from the regional health authority School trajectories tend to be highly heterogeneous with regards to the first dose ( Figure 2 ). By the end of the observation period, the difference in average school-level vaccination coverage between private schools and more-disadvantaged public schools was 12.7 points of percentage (95%CI 8.5%; 17.0%) and

the difference between private schools and less disadvantaged public schools was 6.7 points of percentage (95%CI 3.1%; 10.2%). Vaccination coverage was 6.8 points of percentage (95%CI 3.2%; 10.4%) higher in schools where students were more willing to be vaccinated, 7.6 points of percentage (95%CI 4.2%;

11.1%) higher in schools where students were more knowledgeable about COVID-19, and 6.2 points of percentage (95%CI 2.6%; 9.7%) higher in schools where students reported greater levels of concern about COVID-19. distinct rates of growth. In this case, the random intercept is highly significant, but the variance in random slopes is negligible. This suggests that after adjusting for confounding, initial coverage values are distinct across schools and the rate of progression of coverage is comparable across schools overall. The model also suggests that being a more disadvantaged public school contributes to lower the school-level vaccination coverage of 6.5 points of percentage in average compared to private schools (Estimate= -6.5

points of percentage (95%CI 0.3%; 12.6%)) and 4.9 points of percentage compared to less disadvantaged public schools (Estimate = -4.9 points of percentage (95%CI 0.8%; 9.0%)), although there is not much of a difference between private schools and less disadvantaged public schools. School-level vaccination coverage also tends to increase by 4.3 points of percentage in average when the students reported greater concerns about COVID-19, compared to schools where the students were less concerned (Estimate= +4.3 points of percentage (95%CI 0.7%; 0.8%)). However, the model did not demonstrate significant associations between the students' levels of knowledge of COVID-19 or supportive attitudes towards vaccination.

Associated factors were comparable for receiving the first and second dose at the end of the observation period (Table 1) . However, the magnitude of the association between vaccination coverage and school status here is almost double the association estimated for the first dose: on average, the adjusted coverage rate in disadvantaged schools is 8.6 points of percentage lower than in less disadvantaged public schools (Estimate= -8.6 points of percentage (95%CI 3.4%; 13.7%)) and 11.1 points lower than in private schools (Estimate= -11.1 points of percentage (95%CI 3.1%; 19.0%)). Finally, greater concerns for J o u r n a l P r e -p r o o f Journal Pre-proof COVID-19 among the student population was associated with higher level of coverage of 7.1 points of percentage in average (Estimate= +7.1 points of percentage (95%CI 2.3%; 12.0%)).

Lorenz curves provide a graphical representation of the distribution of the vaccination coverage levels across the schools. The left panel of Figure 3 reveals an uneven distribution on June 6, 2021, when the school-based program was launched (the farther the Lorenz curve is from the 45 degrees line, the more unequal the distribution is). The inequality between schools has almost disappeared by the next wave.

This was also confirmed by the sharp drop in the coefficient of Gini (Table 2 ) before and after the schoolbased intervention. The Gini coefficient continued to decrease gradually until the end of the observation period.

As for the second dose, the inequality between schools dropped markedly during the period covering the Overall, our data suggest that following a provincial initiative to increase student vaccination rates have contributed to reach rapidly a high vaccination coverage in the high schools of the Québec City region.

The program worked very well initially to increase student vaccination rates at a school-level, even in disadvantaged schools, which reflects the responsiveness to vaccination in the Québec youth population.

Nonetheless, school status and students' levels of concern towards the pandemic seem to be significantly associated with vaccination coverage unlike students' levels of knowledge and willingness to vaccinate.

Private and less disadvantaged schools as well as schools where students reported having great concerns towards COVID-19 tended to have a higher vaccination coverage.

This study provides support for a more active approach for the vaccine distribution as these findings suggest that this approach may accelerate the vaccine uptake and reduce inequalities in vaccination coverage among high schools. The most significant increase in the vaccination coverage, regardless of school type, was immediately after the inception of the school vaccination campaign. Significant increases were not observed for the second dose where students were required to take the initiative to go to a vaccination clinic. As shown by the Lorenz curves and the Gini coefficients, the reduction of the inequalities in the vaccination coverage among high schools was more significant for the first dose than for the second one. For instance, on August 13, 2021, the Gini coefficient representing the inequalities between the vaccination coverage of private schools and more-disadvantaged public schools was approximately 2.5 times greater for the second dose. The reduction of the inequalities observed with the vaccination in schools for the first dose is consistent with evidence from a study on a school-based HPV vaccination program in England where the level of deprivation did not appear to be markedly associated J o u r n a l P r e -p r o o f with vaccination coverage. 14 School-based vaccination campaigns are well-established in most of the WHO country members since it has shown to be effective at reducing the transmission of many epidemic diseases such as hepatitis, HPV, meningitis, influenza, rubella, and varicella zoster. 5, 6, 19 This approach lowers virus transmission in communities and protects vulnerable groups such as elderlies and people suffering with immunodeficiency by increasing the vaccination coverage among students. 6 Evidence shows that students represent a key vector in virus spread as they are in contact with many other students who have not been exposed to these viruses before. 6 Moreover, physicians suggest that teenagers are often uninformed about the importance of immunization and underestimate the danger of preventable diseases 20 which increases their risk of spreading viruses in their community. Therefore, it is crucial in an outbreak situation such as COVID-19 to facilitate the vaccination of this high-risk population. For instance, in the United States, where the COVID-19 vaccination uptake among 12-17 year-olds is much lower, experts are encouraging local health services and school districts to offer vaccination in schools, considering the effectiveness of school-based approaches to increase vaccination coverage. 21, 22 This study has some limitations. Notably, definitive conclusions about the value of the school-based vaccination campaigns cannot be made from our findings. An additional limitation is the low number of observations which limits the extent of the analysis. Supplementary observations would have allowed consideration of more covariates such as students' gender, source of information on COVID-19, or the number of COVID-19 outbreaks within a school. Another limitation is that "adequately vaccinated" students who contracted COVID-19 and thus received a single dose of vaccine were not considered in the vaccination coverage for the second dose. There is also limitation in the generalizability of the results since the study took place in a favorable setting where the authorities strongly encouraged the vaccination. The population of the province of Québec is usually supportive of vaccines. 23 Future research may consider reproducing this study in a similar or a less favorable context.

In exceptional circumstances such as the COVID-19 pandemic, mass mobilization of vaccines allows populations to reach a high vaccination coverage in a short period of time. In only a few weeks, students involved in this study reached the same level of vaccination coverage as the adult population who began to get vaccinated 5 weeks prior. In addition to being an efficient way to achieve rapidly high vaccination coverage, the school-based approach contributes to equity in health care. It follows the logic of 

Le vaccin contre la COVID-19 pour les enfants | Société canadienne de pédiatrie

Québec donne le coup d'envoi pour la vaccination des 12 à 17 ans | Coronavirus. Radio-Canada.ca

Les 12 à 17 ans peuvent devancer à quatre semaines leur deuxième dose | La Presse

Consultez le taux de vaccination dans chaque école secondaire de Montréal. Le Devoir

School-based delivery of routinely recommended vaccines and opportunities to check vaccination status at school, a global summary

School-based vaccination in NSW. N S W Public Health Bull

School-based vaccination programmes: a systematic review of the evidence on organisation and delivery in high income countries

School-based human papillomavirus vaccination program for increasing vaccine uptake in an underserved area in Texas

Evaluation of a city-wide school-located influenza vaccination program in Oakland, California, with respect to vaccination coverage, school absences, and laboratory-confirmed influenza: A matched cohort study

Physician Attitudes Regarding School-Located Vaccination Clinics

The COVID States Project #49: Vaccinating America's youth

Patterns in COVID-19 Vaccination Coverage, by Social Vulnerability and Urbanicity -United States

Willingness and Influential Factors of Parents to Vaccinate Their Children against the COVID-19: A Systematic Review and Meta

Examining inequalities in the uptake of the school-based HPV vaccination programme in England: a retrospective cohort study

The COMPASS study: a longitudinal hierarchical research platform for evaluating natural experiments related to changes in school-level programs, policies and built environment resources

COMPASS protocol changes and recruitment for online survey implementation during the COVID-19 pandemic. Compass System

Indices of disadvantaged -Open Government Portal

Overview of Canadian School-Based Immunization Programs | CANVax

Barriers to Adolescent Immunization: A Survey of Family Physicians and Pediatricians

Considerations for Planning School-Located Vaccination Clinics | CDC

COVID-19 vaccination coverage among adolescents aged 12-17 years -United States

Enquête québécoise sur la vaccination contre la grippe saisonnière, le pneumocoque, le zona et sur les déterminants de la vaccination

Investigation, Data collection, Data curation, Analysis, Writing, Original draft preparation, Writing, Reviewing & Editing

Investigation, Data collection, Data curation, Analysis, Writing, Reviewing & Editing

Investigation, Data collection, Data curation, Writing, Reviewing & Editing

Funding acquisition, Supervision, Writing, Reviewing & Editing

) ; Investigation, Data collection, Data curation, Funding acquisition, Writing

) ; Funding acquisition

Conceptualization , Methodology, Funding acquisition, Writing, Reviewing & Editing

) Conceptualization, Methodology, Investigation, Data collection, Data curation, Analysis, Funding acquisition, Supervision, Writing, Original draft preparation, Writing, Reviewing & Editing Affiliations: 6. Faculté de médecine

Direction de santé publique du CIUSSSCN

The authors wish to thank their collaborators from the Public Health Department of the CIUSSS de la 

High schools associated with CIUSSSCN n = 54High schools participating in the COMPASS-Quebec project n = 38High schools included in the study n = 37High schools not following a regular academic curriculum n = 13High schools following a regular academic curriculum, but not participating in the COMPASS-Quebec project n = 4High schools participating in the COMPASS-Quebec project, but not associated with CIUSSSCN n = 1 J o u r n a l P r e -p r o o f Journal Pre-proof