Estimates and determinants of HPV non-vaccination in 14-year-old Canadians: Results from the childhood national immunization coverage survey, 2019

ABSTRACT

Human papillomavirus (HPV) infections, the most common sexually transmitted infections, are associated with various health outcomes including anogenital warts and cancers. Despite significant investments in HPV vaccination programs, ensuring adequate vaccination coverage for adolescents remains a challenge in Canada. This analysis used data collected through the 2019 Childhood National Immunization Coverage Survey (CNICS) to determine national estimates of HPV non-vaccination and investigate determinants of HPV non-vaccination for adolescents aged 14-years old in Canada, both overall and stratified by gender. The primary outcome of interest was HPV vaccination status, categorized as vaccinated with at least one dose or unvaccinated. Simple and multiple logistic regression models were used to investigate determinants of HPV non-vaccination. In 2019, an estimated 19.8% of the 14-year-olds in Canada were unvaccinated for the HPV vaccine, with males having higher non-vaccination rates than females (27.0% compared to 12.9%). In the unstratified analysis, factors associated with HPV non-vaccination for 14-year-olds were gender and region of residence. These factors differed by gender – for males, region of residence and respondent’s age were significant factors, whereas for females, total household income was a significant factor. These results could help public health officials and policymakers develop and implement tailored interventions to enhance the delivery of HPV vaccination programs for male and female adolescents. By targeting populations that are under-vaccinated, vaccine uptake could be better facilitated to help reduce inequalities in access to the HPV vaccine, which could also potentially reduce disparities in HPV-related health outcomes.

KEYWORDS:

• Non-vaccination
• Canada
• determinants
• HPV
• vaccine
• adolescents

Introduction

Human papillomavirus (HPV) infections are the most common sexually transmitted infections, affecting more than 70% of sexually active Canadian males and females.¹ They are associated with various health outcomes, including anogenital warts and cancers (cervical, anal, vaginal, vulvar, penile, and oropharyngeal), which negatively impact quality of life, can lead to morbidity and mortality, and accrue substantial costs to the healthcare system.^2–4 Primary prevention of such HPV-associated health outcomes is available through safe and effective vaccines.³

In Canada, the National Advisory Committee on Immunization (NACI) has recommended HPV vaccines for females and males since 2007 and 2012, respectively.² These vaccines have been rigorously promoted for pre-adolescents and adolescents to provide them with the highest level of immunogenicity and to protect them before the onset of sexual activity.⁵ As such, publicly funded school-based HPV vaccine programs targeting pre-adolescents and adolescents have been implemented in all Canadian provinces and territories for females and males as of 2010 and 2018, respectively.⁶ Depending on the provinces and territories, the target age group for HPV vaccines ranges from grade 4 to 8 (approximately ages 8–13 years).^2,3 Previous coverage estimates demonstrate heterogeneity in HPV vaccination and non-vaccination rates across the country.² For example, an analysis conducted using the 2013 Childhood National Immunization Coverage Survey (CNICS) found that provincial and territorial HPV non-vaccination rates in 2013 for 12 to 14-year-old females ranged from 10.9% in Newfoundland and Labrador to 47.6% in Northwest Territories.²

Using data from the 2017 CNICS, HPV coverage for at least one dose of the HPV vaccine for 14-year-old females was reported to be 83.0%.⁷ Despite the significant investments in HPV vaccination programs, ensuring adequate vaccination coverage and achieving the national vaccination coverage goal of 90% for at least two doses of the HPV vaccine by 17 years of age ultimately remains a challenge in Canada.^8,9 Many factors related to a child’s identity, household, and area of residence have been previously identified in influencing HPV vaccination uptake.^{2–8,10–14} For instance, male or Indigenous children were reported less likely to be vaccinated for HPV when compared to female or non-Indigenous children, respectively.^10,11 Similarly, having a parent/guardian born outside Canada was a risk factor for preadolescents and adolescents not being vaccinated for HPV.² There have, however, been contradicting or ambiguous results within the literature concerning the impact of income, education, and community size. Specifically, it is unclear whether having a high household income, high household education status, or residing in an urban area is associated with an increased or decreased likelihood of receiving the HPV vaccine.^{2–8,12–14} The knowledge and beliefs of parents/guardians concerning vaccines also play an essential role in HPV vaccine uptake, especially considering HPV vaccination is not mandatory, and programs target preadolescents and adolescents.¹⁵

Current literature focuses primarily on determinants affecting HPV non-vaccination in female children and there is a lack of information pertaining to male children. Since HPV vaccine programs and policies have differentially targeted males and females, it is important to examine the varying effect of specific determinants on non-vaccination between males and females.⁶ Overall, a better understanding of the underlying determinants of HPV non-vaccination can help inform vaccination programs and campaigns to increase HPV vaccination.⁹ Given this, the main objective of this study was to use data collected through the 2019 CNICS to determine national estimates of HPV non-vaccination and investigate determinants of HPV non-vaccination for adolescents aged 14-years old in Canada, both overall and stratified by gender.

Methods

Data source

The 2019 Childhood National Immunization Coverage Survey (CNICS) was a cross-sectional survey conducted by Statistics Canada and the Public Health Agency of Canada. It targeted children living in Canada aged two, seven, fourteen, and seventeen years old as of March 1, 2019. Those residing in First Nations reserves, in foster care, and institutionalized were excluded. The sampling frame was built using the list of children whose parent/guardian claimed the Canada Child Benefit (CCB) as of June 2019, which was estimated to include 96% of the children living in Canada in 2018.

Data collection occurred from December 2, 2019 to April 30, 2020 using three data sources. First, a self-completed electronic questionnaire (EQ) or a computer-assisted telephone interview (CATI) was completed where respondents were asked about the child’s vaccination history, their own knowledge, and beliefs relative to vaccination (for respondents of 2-year-old and 14-year-old children), and sociodemographic information. Respondents were not required to have the vaccination booklet or record available for children aged 14 and 17-years old. Second, if consent was given, Statistics Canada contacted healthcare providers who vaccinated children to supplement the vaccination information provided by respondents. The two sources were combined to give a complete vaccination record. Finally, for children residing in Manitoba and Prince Edward Island, if consent was given, data were extracted from provincial, immunization registries in place of respondent or healthcare provider data.

A more detailed description of data collection and processing methods is available from the 2019 CNICS report.¹⁶

Study population

A total of 2598 children’s parents/guardians (the person most knowledgeable or PMK, hereafter referred to as the respondent) agreed to participate in the 2019 CNICS. Of the 2598 children, there were 476 two-year-olds, 344 seven-year-olds, 674 fourteen-year-old males, 514 fourteen-year-old females, and 590 seventeen-year-olds. This analysis examined the data from fourteen-year-old males and females.

Primary outcome

The primary outcome of interest was HPV vaccination status, categorized as either being vaccinated with at least one dose or unvaccinated. The data collected focused solely on determining whether the child had received the HPV vaccine, as quantifying the number of doses was considered less reliable because parental knowledge was used. Parental consent for the entire vaccine series in some provinces and territories was requested only once, making it challenging for parents to specify the exact number of doses their child received. In this context, children who had at least one dose of the HPV vaccine by the age of 14 years old were considered to have an up-to-date vaccination record of HPV. As such, they were considered vaccinated for HPV and classified in the “vaccinated with at least one dose” category of the HPV vaccination status variable.

Other variables

Potential determinants of HPV non-vaccination were selected based on their reported or suspected association with vaccination status within the literature. Total household income was retrieved using linkage to tax data files from all members of the selected households or was imputed using the trend imputation method or nearest-neighbor imputation method. The highest household education variable was derived using information collected on highest education level obtained from both the respondent and their spouse/partner. Similarly, the vaccine-hesitant variable was derived combining information on whether the respondent decided not to immunize the child with a particular vaccine other than HPV, was reluctant to get a vaccination for the child other than HPV, or decided to delay any vaccines.

Data analysis

Data were analyzed using SAS Enterprise Guide 7.1. Unweighted descriptive statistics of the study population, overall and stratified by gender, were reported across all potential determinants of HPV non-vaccination. Survey sampling weights were applied so that the analyses would be representative of 14-year-old children in Canada. National estimates of HPV non-vaccination, overall and stratified by gender, were calculated as the number of respondents who reported the child was not vaccinated for the HPV vaccine over the total number of respondents who reported the child either vaccinated with at least one dose or not vaccinated for the HPV vaccine. Those with missing information pertaining to the HPV vaccination status variable were excluded from the denominator and from all analyses. If necessary, new categories for categorical variables were created based on frequencies where categories were merged as appropriate if there were too few respondents (i.e., five or less). Continuous variables, such as income and age of the respondent, were categorized based on their quartile distributions for all respondents in the 2019 CNICS. The 95% confidence intervals (95% CIs) of HPV non-vaccination estimates were compared to determine statistically significant differences. Simple and multiple logistic regression models, overall and stratified by gender, were used to calculate odds ratios (OR) and their associated 95% CIs. P-values were also reported, and the statistical significance level was set at 0.05. For logistic regression models, reference categories were either normative or end categories and missing data were coded as missing, resulting in complete-case analysis. All pre-selected potential determinants were included in the multiple logistic regression models, except for the vaccine-hesitant variable, which was excluded due to being a potential mediator in the causal pathway between determinants and HPV vaccination status. Additionally, the Indigenous status variable was omitted from all stratified analyses due to low cell counts. To account for the complex survey design, a bootstrap resampling method with 1,000 samples was used to estimate standard errors, coefficients of variation, and confidence intervals.

Ethics

The 2019 CNICS was conducted by Statistics Canada as part of its mandate to “collect, compile, analyze, abstract and publish statistical information relating to the commercial, industrial, financial, social, economic and general activities and conditions of the people of Canada.” Participation in the survey was voluntary and data were kept confidential. The authors of this study did not have access to any identifiable record.

Results

Demographics

Of the 1188 fourteen-year-olds accounted for in the 2019 CNICS, 139 (11.7%) had missing HPV vaccination status. As such, the study population consisted of 1049 respondents who reported the child’s HPV vaccination status (Table 1). With respect to the region of residence, 39.9% were in provinces other than Ontario and Quebec and settled in urban communities (86.6%). In addition, a wealthier and more educated study population was observed with 28.8% earning $160,000 or greater and 50.6% having a bachelor’s degree or higher. A larger proportion of respondents were between 43 and 47-years old (36.4%), while a smaller proportion were between 14 and 38-years old (11.7%). Nearly all respondents (98.5%) had a parent relationship with the child. Similar trends as the ones noted above were also observed for both males and females, except for a higher proportion of males residing in Ontario (39.8%) as opposed to provinces categorized in the “other” group. Among the 14-year-olds considered, there were 573 males (54.6%), 471 females (44.9%), and 5 cases where the child’s gender information was missing (0.5%), emphasizing that a larger number of males were captured in this sample compared to females.

Table 1. Characteristics of the population, overall and stratified by gender – CNICS, Canada, 2019.

	Study Population		Males		Females
	n*	%^b	n*	%^b	n*	%^b
Overall	1049	100	573	100	471	100
Region of Residence
Ontario	407	38.8	228	39.8	174	36.9
Quebec	223	21.3	123	21.5	100	21.2
Other	419	39.9	222	38.7	197	41.8
Total Household Income
$160,000+	302	28.8	157	27.4	145	30.8
$110,000 to $159,999	253	24.1	146	25.5	107	22.7
$70,000 to $109,999	246	23.5	133	23.2	109	23.1
Less than $70,000	248	23.6	137	23.9	110	23.4
Community Type
Urban	908	86.6	495	86.4	408	86.6
Rural	136	13.0	75	13.1	61	13.0
Respondent’s Place of Birth
Born in Canada	697	66.4	377	65.8	320	67.9
Born outside of Canada	345	32.9	195	34.0	150	31.8
Respondent’s Marital Status
Married/living common law	827	78.8	458	79.9	369	78.3
Not living common law	215	20.5	114	19.9	101	21.4
Respondent’s Age
48+	316	30.1	158	27.6	158	33.5
43–47	382	36.4	214	37.3	168	35.7
39–42	201	19.2	112	19.5	89	18.9
14–38	123	11.7	73	12.7	50	10.6
Indigenous Status of Child
No	998	95.1	548	95.6	450	95.5
Yes	45	4.3	24	4.2	21	4.5
Gender of Child
Female	471	44.9
Male	573	54.6
Number of Children 18 and Under Living at Address
3+	257	24.5	150	26.2	107	22.7
2	480	45.8	265	46.2	215	45.6
1	303	28.9	154	26.9	149	31.6
Vaccine-Hesitant^d
Yes	180	17.2	101	17.6	79	16.8
No	860	82.0	469	81.8	391	83.0
Highest Household Education^e
Bachelor’s degree or higher	531	50.6	295	51.5	236	50.1
Certificate/diploma below the bachelor’s level	368	35.1	200	34.9	168	35.7
Secondary or less	141	13.4	75	13.1	66	14.0
Relationship of Respondent to Child
Parent	1033	98.5	-^c	-^c	-^c	-^c
Not a parent	10	1.0	-^c	-^c	-^c	-^c

Children whose gender was not disclosed by their respondent (n = 5) were excluded only from gender-specific descriptive statistics.

Totals may differ due to missing data.

*n = number of respondents (unweighted).

^bProportions (unweighted).

^cData suppressed for confidentiality reasons.

^dVaccine-hesitant variable derived by combining information on whether respondent decided not to immunize the child with a particular vaccine or was reluctant to get a vaccination for the child other than HPV or decided to delay any vaccines. If respondent did not immunize the child with the HPV vaccine specifically or was reluctant to get it, this information was excluded from deriving vaccine-hesitant.

^eHighest household education derived using information collected on highest education from both the respondent and their spouse/partner.

HPV non-vaccination

In 2019, an estimated 19.8% of the 14-year-old adolescents in Canada had not received at least one dose of the HPV vaccine (Table 2). Higher non-vaccination rates were observed in provinces categorized in the other group (23.8%) compared to Quebec (13.8%).

Table 2. National HPV non-vaccination estimates for 14-year-olds, overall and stratified by gender – CNICS, Canada, 2019.

	Study Population		Males		Females
	Non-vaccination*	95% CI^b	Non-vaccination*	95% CI^b	Non-vaccination*	95% CI^b
Overall	19.8%	(17.4, 22.3)	27.0%	(23.5, 30.8)^c	12.9%	(10.0, 16.5)^c
Region of Residence
Ontario	19.4%	(15.8, 23.7)	24.9%	(19.5, 31.2)	14.4%	(9.8, 20.6)
Quebec	13.8%	(9.8, 18.9)^c	20.7%	(14.3, 29.0)	7.0%	(3.4, 13.9)
Other	23.8%	(19.9, 28.1)^c	33.3%	(27.4, 39.8)	15.0%	(10.3, 21.3)
Total Household Income
$160,000+	15.1%	(11.5, 19.6)	25.2%	(19.0, 32.7)	5.7%	(2.9, 10.9)^c
$110,000 to $159,999	16.5%	(12.4, 21.6)	23.6%	(17.5, 31.2)	8.9%	(4.7, 16.3)
$70,000 to $109,999	22.6%	(17.7, 28.4)	29.1%	(21.9, 37.5)	17.1%	(11.0, 25.7)^c
Less than $70,000	23.8%	(18.7, 29.8)	29.4%	(22.2, 37.9)	18.6%	(12.2, 27.4)^c
Community Type
Urban	19.7%	(17.2, 22.5)	27.0%	(23.2, 31.2)	12.9%	(9.8, 16.7)
Rural	20.8%	(14.4, 29.0)	27.9%	(18.7. 39.4)	13.5%	(6.5, 26.0)
Respondent’s Place of Birth
Born in Canada	19.9%	(17.0, 23.1)	27.7%	(23.4, 32.5)	12.6%	(9.2, 17.0)
Born outside of Canada	20.0%	(16.0, 24.6)	25.8%	(19.9, 32.6)	13.8%	(9.0, 20.4)
Respondent’s Marital Status
Married/living common law	18.9%	(16.3, 21.7)	27.3%	(23.4, 31.6)	10.4%	(7.6, 14.0)
Not living common law	22.6%	(17.3, 28.9)	25.5%	(18.3, 34.3)	20.0%	(12.9, 29.7)
Respondent’s Age
48+	20.6%	(16.2, 25.7)	31.0%	(24.2, 38.8)	11.9%	(7.5, 18.6)
43–47	14.5%	(11.3, 18.5)	18.6%	(14.0, 24.4)	10.6%	(6.6, 16.6)
39–42	23.2%	(17.8, 29.6)	32.4%	(24.2, 41.9)	13.5%	(7.6, 22.8)
14–38	25.5%	(18.4, 34.2)	30.3%	(20.7, 41.9)	19.9%	(10.8, 33.6)
Indigenous Status of Child
No	19.8%	(17.4, 22.5)	27.0%	(23.4, 31.0)	12.8%	(9.8, 16.5)
Yes	21.2%	(11.7, 35.4)	-^d	-^d	-^d	-^d
Number of Children 18 and Under Living at Address
3+	20.1%	(15.5, 25.5)	29.2%	(22.2, 37.4)	9.9%	(5.6, 17.0)
2	17.5%	(14.2, 21.3)	24.1%	(19.2, 29.7)	10.8%	(7.0, 16.2)
1	23.3%	(18.9, 28.4)	29.6%	(23.0, 37.3)	18.2%	(12.4, 25.8)
Vaccine-Hesitant^e
Yes	25.5%	(19.6, 32.4)	30.2%	(22.0, 40.0)	20.6%	(12.7, 31.5)
No	18.3%	(15.7, 21.2)	25.8%	(22.0, 30.0)	11.0%	(8.0, 15.0)
Highest Household Education^f
Bachelor’s degree or higher	17.3%	(14.2, 20.8)	24.0%	(19.5, 29.3)	10.2%	(6.8,15.0)
Certificate/diploma below the bachelor’s level	20.3%	(16.5, 24.7)	28.4%	(22.6, 35.1)	12.7%	(8.3, 19.0)
Secondary or less	27.0%	(20.1, 35.2)	33.3%	(23.2, 45.3)	21.4%	(12.7, 34.0)

Children whose gender was not disclosed by their respondent (n = 5) were excluded only from gender-specific non-vaccination estimates.

*Weighted proportions based on respondent records, respondent recall, and/or provincial/territorial administrative records.

^b95% CI − 95% Confidence Interval (weighted).

^cDenotes a statistically significant difference at the 5% level. 95% CIs were compared to determine statistically

significant differences between groups of a variable.

^dData suppressed for confidentiality reasons.

^eVaccine-hesitant variable derived by combining information on whether respondent decided not to immunize the child with a particular vaccine or was reluctant to get a vaccination for the child other than HPV or decided to delay any vaccines. If respondent did not immunize the child with the HPV vaccine specifically or was reluctant to get it, this information was excluded from deriving vaccine-hesitant.

^fHighest household education derived using information collected on highest education from both the respondent and their spouse/partner.

HPV non-vaccination was higher among males (27.0%) compared to females (12.9%). Notably, gender-specific trends emerged, such as higher non-vaccination rates among females in households with earnings less than $70,000 (18.6%) as well as between $70,000 to $109,999 (17.1%) compared to those with earnings of $160,000 or more (5.7%).

Determinants of HPV non-vaccination

For the entire study population, region of residence and gender of child appeared as significant determinants of HPV non-vaccination (Table 3). Compared to children in Quebec, those in provinces other than Ontario had higher odds of HPV non-vaccination (OR = 2.00, 95% CI: 1.24–3.23, p = .005). Similarly, compared to females, males had higher odds of HPV non-vaccination (OR = 2.59, 95% CI: 1.80–3.74, p < .0001).

Table 3. Association between sociodemographic variables and HPV non-vaccination in 14-year-olds, overall and stratified by gender – CNICS, Canada, 2019.

	Study Population						Males						Females
	Simple Logistic Regression Models			Multiple Logistic Regression Model			Simple Logistic Regression Models			Multiple Logistic Regression Model			Simple Logistic Regression Models			Multiple Logistic Regression Model
	OR	95% CI*	p	OR	95% CI*	p	OR	95% CI*	p	OR	95% CI*	p	OR	95% CI*	p	OR	95% CI*	p
Region of Residence
Quebec	Reference			Reference			Reference			Reference			Reference			Reference
Ontario	1.51	(0.96, 2.38)	0.074	1.44	(0.86, 2.40)	0.163	1.27	(0.73, 2.21)	0.406	1.19	(0.65, 2.18)	0.583	2.24	(0.82, 6.14)	0.118	1.92	(0.62, 5.99)	0.260
Other	1.96	(1.28, 3.00)	0.002	2.00	(1.24, 3.23)	0.005	1.91	(1.11, 3.27)	0.019	1.95	(1.07, 3.58)	0.031	2.35	(0.88, 6.27)	0.087	1.95	(0.65, 5.86)	0.232
Total Household Income
$160,000+	Reference			Reference			Reference			Reference			Reference			Reference
$110,000 to $159,999	1.11	(0.69, 1.79)	0.657	1.05	(0.63, 1.76)	0.845	0.92	(0.53, 1.59)	0.759	0.89	(0.48, 1.65)	0.714	1.61	(0.52, 5.01)	0.410	1.50	(0.46, 4.90)	0.503
$70,000 to $109,999	1.65	(1.05, 2.58)	0.029	1.63	(0.99, 2.70)	0.055	1.22	(0.71, 2.08)	0.478	1.25	(0.68, 2.30)	0.477	3.40	(1.28, 9.03)	0.014	2.94	(1.01, 8.56)	0.048
Less than $70,000	1.76	(1.14, 2.72)	0.011	1.53	(0.87, 2.72)	0.143	1.24	(0.72, 2.14)	0.445	1.14	(0.55, 2.35)	0.724	3.78	(1.47, 9.72)	0.006	2.47	(0.77, 7.87)	0.126
Community Type
Urban	Reference			Reference			Reference			Reference			Reference			Reference
Rural	1.07	(0.65, 1.77)	0.795	0.80	(0.44, 1.46)	0.466	1.05	(0.58, 1.90)	0.884	0.85	(0.41, 1.76)	0.654	1.05	(0.39, 2.88)	0.918	0.72	(0.21, 2.41)	0.593
Respondent’s Place of Birth
Born in Canada	Reference			Reference			Reference			Reference			Reference			Reference
Born outside of Canada	1.00	(0.72, 1.41)	0.987	0.87	(0.58, 1.32)	0.515	0.91	(0.60, 1.37)	0.637	0.89	(0.54, 1.47)	0.651	1.11	(0.60, 2.05)	0.745	1.01	(0.49, 2.08)	0.984
Respondent’s Marital Status
Married/living common law	Reference			Reference			Reference			Reference			Reference			Reference
Not living common law	1.25	(0.85, 1.84)	0.253	0.93	(0.57, 1.50)	0.757	0.91	(0.56, 1.49)	0.708	0.75	(0.41, 1.38)	0.352	2.16	(1.11, 4.20)	0.023	1.34	(0.60, 3.00)	0.473
Respondent’s Age
48+	Reference			Reference			Reference			Reference			Reference			Reference
43–47	0.66	(0.43, 1.00)	0.049	0.65	(0.42, 1.01)	0.055	0.51	(0.31, 0.84)	0.008	0.51	(0.30, 0.88)	0.015	0.88	(0.40, 1.94)	0.741	0.86	(0.36, 2.10)	0.747
39–42	1.17	(0.75, 1.82)	0.503	1.24	(0.76, 2.03)	0.395	1.07	(0.63, 1.82)	0.810	1.09	(0.58, 2.05)	0.793	1.16	(0.49, 2.75)	0.744	1.43	(0.53, 3.88)	0.480
14–38	1.32	(0.78, 2.25)	0.298	1.17	(0.63, 2.17)	0.623	0.96	(0.51, 1.84)	0.912	0.85	(0.40, 1.79)	0.663	1.83	(0.70, 4.79)	0.218	1.92	(0.60, 6.16)	0.274
Indigenous Status of Child
No	Reference			Reference				-^b			-^b			-^b			-^b
Yes	1.09	(0.49, 2.43)	0.835	0.76	(0.29, 1.95)	0.563		-^b			-^b			-^b			-^b
Gender of Child
Female	Reference			Reference
Male	2.49	(1.77, 3.52)	<.0001	2.59	(1.80, 3.74)	<.0001
Number of Children 18 and Under Living at Address
3+ children	Reference			Reference			Reference			Reference			Reference			Reference
2 children	0.84	(0.56, 1.27)	0.413	0.83	(0.52, 1.33)	0.344	0.77	(0.48 1.24)	0.278	0.73	(0.42, 1.29)	0.279	1.10	(0.47, 2.60)	0.828	1.08	(0.41, 2.81)	0.877
1 child	1.21	(0.79, 1.85)	0.379	1.27	(0.77, 2.09)	0.441	1.02	(0.60, 1.73)	0.940	0.99	(0.54, 1.85)	0.984	2.02	(0.88, 4.62)	0.096	1.91	(0.71, 5.12)	0.200
Vaccine-Hesitant^d
No	Reference				-^c		Reference				-^c		Reference				-^c
Yes	1.53	(1.03, 2.27)	0.036		-^c		1.25	(0.76, 2.06)	0.386		-^c		2.08	(1.02, 4.27)	0.045		-^c
Highest Household Education^e
Bachelor’s or higher	Reference			Reference			Reference			Reference			Reference			Reference
Below the bachelor’s level	1.22	(0.86, 1.73)	0.258	1.05	(0.68, 1.60)	0.840	1.26	(0.83, 1.90)	0.279	1.19	(0.72, 1,97)	0.504	1.28	(0.65, 2.52)	0.469	0.85	(0.36, 2.00)	0.702
Secondary or less	1.77	(1.12, 2.81)	0.015	1.43	(0.80, 2.58)	0.240	1.58	(0.87, 2.88)	0.136	1.28	(0.63, 2.59)	0.493	2.40	(1.05, 5.48)	0.039	1.37	(0.44, 4.29)	0.589

*95% CI − 95% Confidence Interval (weighted).

^bIndigenous status of child excluded from stratified simple and multiple logistic regression models due to low cell counts.

^cVaccine-hesitant excluded from multiple logistic regression models due to being a potential mediator in the causal pathway between determinants and the outcome.

^dVaccine-hesitant variable derived by combining information on whether respondent decided not to immunize the child with a particular vaccine or was reluctant to get a vaccination for the child other than HPV, or decided to delay any vaccines. If respondent did not immunize the child with the HPV vaccine specifically or was reluctant to get it, this information was excluded from deriving vaccine-hesitant.

^eHighest household education derived using information collected on highest education from both the respondent and their spouse/partner.

Determinants of HPV non-vaccination differed between males and females, indicating effect modification. For males, region of residence and respondent’s age appeared as significant determinants. Compared to male children in Quebec, those in provinces other than Ontario had higher odds of HPV non-vaccination (OR = 1.95, 95% CI: 1.07–3.58, p = .031). Additionally, compared to male children whose respondents were 48-years old or over, those whose respondents were between 43 and 47-years old had lower odds of HPV non-vaccination (OR = 0.51, 95% CI: 0.30–0.88, p = .015). A non-linear relationship appeared for the respondent’s age variable among males. In contrast, for females, total household income appeared as a significant determinant. Compared to female children in households earning $160,000 or greater, those in households earning between $70,000 to $109,999 had higher odds of HPV non-vaccination (OR = 2.94, 95% CI: 1.01–8.56, p = .048). A non-linear relationship was also observed for the total household income variable among both females and males. Furthermore, compared to female children without vaccine-hesitant caregivers, those with vaccine-hesitant caregivers had higher odds of HPV non-vaccination (OR = 2.08, 95% CI: 1.02–4.27, p = .045). This association was not observed among male children.

Discussion

The national coverage goal for at least two doses of the HPV vaccine by 17 years of age is 90%.¹⁶ Using a large, nationally representative, cross-sectional survey, our study revealed that in 2019, 19.8% of the adolescents aged 14-years old in Canada had not received at least one dose of the HPV vaccine, with males having higher non-vaccination rates than females (27.0% compared to 12.9%). Notably, coverage increased for females from 83.0% in 2017 to 87.1% in 2019, and recent evidence supports that a single dose of the vaccine provides effective protection.^16,17 Despite the availability of publicly funded school-based HPV vaccination programs across Canada, it is evident that under-vaccination remains a pressing public health challenge.^2,8

Identifying determinants influencing vaccination uptake in different subpopulations is, therefore, a public health priority.¹⁸ This analysis found that factors associated with HPV non-vaccination for adolescents aged 14-years old in Canada were gender and region of residence. Compared to females, males had higher odds of non-vaccination, which was consistent with findings from previous literature. For instance, a systematic review conducted by Bird et al. (2017) concluded that Canadian females were 1.22 times more likely to be vaccinated for HPV compared to Canadian males (95% CI: 1.14–1.30).¹⁰ This discrepancy could be due to jurisdictions more recently offering publicly funded school-based HPV vaccination programs for males (between 2013 and 2018) compared to females (between 2007 and 2010).^4,6 Generally, it is not expected that respondents would make different vaccination decisions based on the gender of the child; however, this may be the case for the HPV vaccine.¹⁶ The gradual yet staggered adoption of school-based HPV vaccination programs for males may have contributed to the “feminization of HPV,” thus affecting vaccination uptake for males.^4,19 Publicly funded school-based HPV vaccination programs were still relatively recent when adolescents, particularly males, were scheduled to receive their HPV vaccinations. Additionally, the vaccine’s benefits may be more obvious for females because it primarily prevents cervical cancer, the cancer with the eighth highest incidence in Canadian females in 2022.²⁰ Conversely, for males, the vaccine provides protection against many rarer cancers like anal and penile cancer. This difference may suggest that males and females have distinct motivations for choosing to receive the HPV vaccine. Also, caregivers of male children may be more inclined to vaccinate their male children with the HPV vaccination if they better understand its role in preventing oral-related cancers, which can affect both genders. In contrast, awareness that HPV vaccinations helps in preventing cervical cancer may not be as compelling to caregivers of male children, as cervical cancer only affects females. Furthermore, compared to children in Quebec, those in provinces other than Ontario had higher odds of non-vaccination. Quebec has been reported to have a history of high vaccine uptake in public programs.¹⁴ Gilbert et al. (2016) also found that Canadians, in 2013, aged 12 to 14-years old residing in the Northwest Territories and Nunavut were most likely to be unvaccinated for HPV (OR = 4.62, 95% CI: 3.43–6.23 and OR = 4.68, 95% CI: 3.36–6.53, respectively).² Although our study could not differentiate between the other provinces due to smaller sample sizes, the higher proportion of non-vaccination in this category may be explained partly by the timing of the HPV vaccination program introduction in various jurisdictions.²

Furthermore, factors associated with HPV non-vaccination differed between males and females, indicating effect modification. For starters, the region of residence was a significant determinant of non-vaccination in males. This may have been driven by differences in initiation time of publicly funded school-based HPV vaccination programs for males across jurisdictions in conjunction with variations in target ages and dosing schedules between provinces and territories.^2,3,21 Quebec, for instance, introduced publicly funded school-based HPV vaccination programs for males as early as grade four in 2016, whereas jurisdictions such as British Columbia and Nunavut had slight later introductions in 2017 and 2018, respectively.⁴ Another significant determinant of non-vaccination for males was age of respondent, which suggested that respondents aged 48-years old or over as compared to respondents aged 43 to 47-years old may have been more hesitant to get the vaccine for the child. This may be due to a lack of familiarity with the HPV vaccine in older respondents, especially given its more recent introduction for males. However, as a general trend, younger age groups are more likely to be vaccine-hesitant.²² Additionally, the respondent’s age at which the decision to vaccinate the child was made may not necessarily correlate with the respondent’s age at the time of the survey. This discrepancy arises due to variations in vaccination age requirements across different jurisdictions. Specifically, a younger age at vaccination implies a longer time period has elapsed since the decision was originally made when the survey was conducted. Total household income was a significant determinant of non-vaccination in females as those in households earning between $70,000 to $109,999 had higher odds of HPV non-vaccination compared to the wealthiest households. There have been contradicting and ambiguous results within the literature regarding the impact of total household income on vaccination for females, showing both positive and negative associations.^{2–8,12–14} Some studies reported greater support for HPV vaccination from high-income households, whereas others reported lower likeliness of acceptance.^23,24 The lack of association between the two variables has even been observed, suggesting that the availability of publicly funded school-based HPV vaccination programs limited financial constraints to the receipt of the vaccine.^8,24 Regardless, it may be important to target females from low-income households for cervical cancer screening, given their increased vulnerability to HPV-related diseases.¹³ The non-linear relationships identified in the respondent’s age variable among males and the total household income variable among females pose challenges in interpretation. This challenge is further exacerbated by the wide confidence intervals associated with these relationships. Consequently, definitive conclusions cannot be drawn from this analysis and further research is needed to confirm or infirm these findings. Additionally, having a vaccine-hesitant caregiver was a significant determinant of non-vaccination in females but not for males. This suggests that educating caregivers is an important barrier to address in reducing vaccine hesitancy. However, vaccine hesitancy could also be influenced by multiple factors beyond education, making it one of the aspects that needs to be considered.

Our study did not find a significant association between the highest household education and HPV non-vaccination. Educational attainment is recognized to influence income, and while our study observed an association between highest household education and HPV non-vaccination in the simple logistic regression model, this association did not emerge as significant in the multiple logistic regression model, where the effect of income was more prominent. However, some studies have found respondents with higher educational attainment were more likely to refuse HPV vaccination for the child.^2,22 One study in British Columbia in 2008/2009, for example, reported that females whose respondents had a post-graduate degree had lower odds of HPV vaccination when compared to those whose respondents had a high-school education (OR = 0.6; 95% CI: 0.4–0.9).¹² Yet, the relationship between the highest household education and HPV vaccine acceptance remains unclear, with other recent studies also showing positive associations.²

Strengths and limitations

Notably, our study was the first to explore determinants of HPV non-vaccination for male adolescents in Canada. Despite differences in schedules and program delivery between jurisdictions, non-vaccination was estimated consistently.¹⁶ A representative sample was also collected nationally from a comprehensive sampling frame, the Canada Child Benefit, which included 96% of the children in Canada.¹⁶ There are, however, limitations to consider. For instance, data were collected from respondent-held vaccination records, in which some information may have been incomplete, erroneous, or missing altogether.¹⁶ It is important to consider that some adolescents may have received the vaccine without their respondents’ knowledge if healthcare professionals at school deemed them mature. This could have led to misclassification of the child’s vaccination status since the data were reported by respondents rather than the children themselves. Additionally, a significant proportion, approximately 47% of the respondents of 14-year-olds provided vaccination information based on recall. As a result, there was possible misclassification of the child’s vaccination status and overestimation of non-vaccination.¹⁵ Moreover, by excluding children whose respondents were not fluent in either official language, English or French, we may have failed to capture the differences in non-vaccination and access or utilization of healthcare services this subpopulation of children may have.¹⁶ Total household income was examined without adjusting for household size due to the absence of information on household size. Finally, the stratified analyses had smaller sample sizes that may have affected the precision of the measure of association and may have resulted in reduced power to detect associations.

Conclusion

Estimates of HPV non-vaccination remain high with males having much higher non-vaccination rates than females. Additionally, determinants of non-vaccination differed by gender. The results of this study could help inform future HPV vaccination programs for adolescents by emphasizing the need to target populations that are under-vaccinated.^9,25 Furthermore, the results of this study will provide an important baseline to measure the impact of the COVID-19 pandemic on HPV non-vaccination for adolescents.

Author contributions

All authors developed the analytical plan for this analysis, and AS performed the analysis. All authors contributed to the interpretation of the results. AS drafted the manuscript, to which all authors provided input. All authors approved the final manuscript.

Acknowledgments

The authors are grateful to all survey participants and the staff from Statistics Canada involved in the 2019 CNICS.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The 2019 CNICS was funded by the Public Health Agency of Canada.