Associated Factors with Vaccination among Girls Aged between (11) and (13) against Papillomaviruses in Koumpentoum Health District (Senegal) ()
1. Introduction
Cervical cancer develops in the anatomical part of the female reproductive system that separates the vagina from the uterus. Almost all cases are due to chronic infection with the human papillomavirus (HPV), making it a preventable and curable cancer [1]. It is the (4th) most common cancer in women, with approximately (604,000) new cases. It is also the (2nd) most common female cancer in less developed regions, which accounts for (85%) of new cases. In (2020), cervical cancer caused (320,000) deaths worldwide [2]. Almost all, or 90% of these deaths, occurred in developing countries, particularly in sub-Saharan Africa, where it is steadily and markedly increasing [3] [4].
In Senegal, in (2020), cervical cancer occupies the (1st) place, with (10%) of all cancers and all sexes combined. It is also the leading female cancer with (26.8%) of cancer cases [4]. The survival rate at (1) year is (32%) and (1%) at (5) years [5]. The World Health Organization (WHO) has approved the use of bivalent vaccines against HPV (16) and (18) or quadrivalent vaccines against types (6), (11), (16) and (18). Vaccination of girls aged between (11) and (13) years against HPV should be the central element of national prevention strategies [6]. After a pilot phase (2014-2016) in the health districts of Dakar West and Mékhé, the quadrivalent HPV vaccine (Gardasil®) has been introduced in the Expanded Program on Immunization (EPI) since November (2018) at a rate of (2) doses spaced (6) months apart for girls aged (9) to (10) years. During the pilot phase, complete vaccination coverage in Dakar West was (90.5%) in (2015) and (96%) in (2016) [7]. These very satisfactory results contrast with those of the scale-up. Indeed, in (2020), for a target of (90%), national HPV vaccination coverage was (32%) for the (1st) dose and (13%) for the (2nd). The majority, (67) out of (79) districts, had not reached the (90%) target [8]. In (2020), according to administrative data, HPV vaccination coverage in the Koumpentoum health district was (45%) for the (1st) dose and (35%) for the (2nd) [8].
Caroline D et al. in their systematic review conducted in sub-Saharan Africa with a majority of articles from South Africa, Uganda and Kenya, found that the causes of non-vaccination against HPV were lack of information of mothers or guardians and fear of infertility associated with the disease [9].
What are the factors associated with HPV vaccination coverage in the Koumpentoum health district? The objective of this research was to determine HPV vaccination coverage in the Koumpentoum health district among girls aged between (11) and (13) years and to identify associated factors in order to make recommendations for improvement.
2. Methodology
2.1. Study Framework
The population of the district was estimated in (2020) at (164,409) inhabitants on a surface area of (7652) km2, or a density of (21.48) inhabitants/km2. The town of Koumpentoum is home to the health centre, which provides reference activities and also offers primary health care (PHC). It polarizes (22) health posts that only offer PHC. The expanded program of immunization (EPI) target is accessible to (51%) in fixed and displaced fixed strategy, (35%) in advanced and (14%) in mobile [10].
2.2. Study Type and Period
This was a cross-sectional, descriptive and analytical study. Data collection period was from (1st) to (30th) September (2020) throughout the district.
2.3. Study Population
The study population consisted of girls aged between (11) and (13) years. However, for ethical and cultural reasons, mothers or guardians were interviewed.
2.4. Inclusion and Non-Inclusion Criteria
Mothers or guardians of girls aged between (11) and (13) years who refused to participate were absent or ill at the time of the survey were not included. Similarly, girls aged (11) to (13) years who refused to participate were absent or ill at the time of the survey were not included.
2.5. Sampling
The study target of girls aged between (11) and (13) years was (12,955) [11]. We calculated the sample size according to Schwartz’s formula [12]: N = Z2 × p × (1 − p) × d/i2 with p = desired full vaccination coverage estimated at (90%), q = (1) − p = (10%), i = margin of error set at (0. 05), d = (1.5) is the design effect; N = (1.962) × (0.90) × (0.10) × (1.5)/(0.05)2 = (207), add (10%) non-responders and the final size N = (228). The sampling was stratified in two stages. The first stage corresponded to villages or neighborhoods. A simple random selection of one village or neighborhood from each area of responsibility was made. The second stage corresponded to households. To select the compound, the interviewer stood in the sociological centre of the village or neighborhood and determined a random direction by throwing a pen in the air with the tip pointing in the direction of departure, always starting with the first compound. An exhaustive census of the mother or guardians of young girls aged between (11) and (13) years was carried out.
2.6. Study Variables
The dependent variable was the uptake of the first dose of HPV vaccine by girls aged between (11) and (13) years. The independent variables were related to socio-demographics, knowledge about cervical cancer and the HPV vaccine, attitudes towards EPI and HPV vaccination and HPV vaccination practices.
2.7. Data Collection
We used an HPV vaccination data collection form and administered a closed questionnaire directly to the mothers or guardians of young girls aged between (11) and (13) years.
2.8. Pre Test
A pre-test was carried out with (30) mothers or guardians of young girls aged between (11) and (13) years in the health centre area with the administration of the survey forms and the stabilization of the data entry mask. The pre-test data were not included in the analysis.
2.9. Data Management and Analysis
Preparation of the data entry mask in Epi-info version (7.2) made it possible to constitute a database. Data analysis was carried out with the softwares Epi Info (7.25.0), SPSS version (20) and R version (4.05).
In descriptive part, the quantitative variables were described with their extremes, means and standard deviations, while the qualitative variables were described with their frequencies surrounded by their confidence intervals (CI) at (95%). In the bivariate analysis part, the Pearson’s chi2 statistical test and the Fisher’s test under conditions of applicability verified existence of statistically significant relationships.
Multiple logistic regression was used for the multivariate analysis. All independent variables with a p < (0.25) in the bivariate analysis and those for which the literature review revealed a link with the dependent variable. A stepwise top-down approach was used to retain variables associated with vaccination at the p < (5%) threshold. These were then removed one by one with a nested model comparison by the Aikake information criterion (AIC) [13] until no improvement was found by the maximum likelihood test. The Hosmer-Lemoshow test [14] was used to check the fit of the final model. The bivariate OR and multivariate adjusted OR with their (95%) CIs were used to assess the strength of the relationship between the independent and dependent variables.
2.10. Ethical Considerations
The study protocol was submitted to the health authorities for approval. The survey was explained to the participants through an information letter read in the local language. Free consent was collected on a form and signed by mothers or guardians who consented to the survey. Data collected remained anonymous and confidential.
3. Results
3.1. Description of the Study Population
In this study, (228) mothers or guardians were interviewed. The average age was (37.67) years (±8.59). More than two thirds of them (67.11%) were under the age of (40).The notion of living in a union was found in (92.54%) of the cases and only (7.02%) resided in urban areas. Among mothers or guardians, (41.67%) were educated, (61.91%) had young girls in school and (22.81%) were engaged in income generating activities (IGA) (Table 1).
While information on cervical cancer was found in (78.51%) of the cases, only (41.67%) knew the risk factors and (51.14%) the means of prevention. Mothers or guardians were aware of the HPV vaccine in (65.35%) of the cases, (41.67%) knew the target of the vaccine and (27.19%) the HPV vaccination schedule. The acceptability of routine EPI vaccination was (91.67%). The coverage of the (1st) dose was (44.74%) and that of the (2nd) dose was (25%), i.e. a drop-out rate of (44.11%). The main sources of information were community health workers with (34.64%) and health providers with (32.96%). The main reasons for non-vaccination were lack of knowledge (71.43%) and fear of side effects (10.32%) (Table 2).
3.2. Bivariate Analysis
In bivariate analysis, statistically and significantly factors associated with coverage of the first dose of the HPV vaccine were schooling of mothers or guardians (OR = 5.62 (3.16 - 9.99); p < 0.001), schooling of the young girls (OR = 4.1 (2.23 - 7.53); p < 0.001), information on cervical cancer (OR = 18.97(5.68 - 63.28); p < 0.001), knowledge of risks factors (OR = 8.04 (4.41 - 14.63); p < 0.001), information on HPV vaccine (OR = 16 (6.53 - 39.18); p < 0.001), knowledge on vaccine
Table 1. Distribution of mothers or guardians and young girls by socio-demographic characteristics, Koumpentoum, 2021.
Table 2. Distribution of mothers or guardians according to knowledge, attitudes, practices on cervical cancer and HPV vaccination, Koumpentoum, 2021.
target (OR = 17.11 (8.51 - 34.41); p < 0.001), knowledge of schedule vaccine (OR = 12.06 (5.67 - 25.63); p < 0.001) , knowledge of prevention methods (OR = 26.86 (12.22 - 59.05); p < 0.001),and to be favorable in EPI in general (OR = 16.83 ([2.2 - 128.41]; p < 0.001) (Table 3).
3.3. Multi Variate Analysis
In multiple logistic regression, statistically and significantly factors associated with receiving the first dose of HPV vaccine were information about vaccination (ORa = 10.26 (1.69 - 69.23); p = 0.011), knowledge of the HPV vaccination schedule (ORa = 3.67 (1.2 - 11.51); p = 0.022), and support for EPI in general (ORa = 18.71 (1.51 - 22.23); p = 0.02) (Table 4).
3.4. Limitation of the Study
Although the target was girls aged between (11) and (13) of school age, teachers
Table 3. Identification of factors associated with taking the 1st dose of HPV vaccine in bivariate analysis, Koumpentoum, 2021.
and Koranic masters were not interviewed to collect their knowledge, perceptions, attitudes and practices and to question their influences on HPV vaccination.
4. Discussion
Factors statistically and significantly associated with coverage of the first dose of the HPV vaccine were instruction of mothers or guardians (OR = 5.62 (3.16 - 9.99); p < 0.001), schooling of the young girls (OR = 4.1 (2.23 - 7.53); p < 0.001),
Table 4. Identification of factors associated with taking the 1st dose of HPV vaccine in multiple logistic regression, Koumpentoum, 2021.
information on cervical cancer (OR = 18.97 (5.68 - 63.24); p < 0.001), knowledge of risks factors (OR = 8.04 (4.41 - 14.63); p < 0.001), information on HPV vaccine (ORa = 10.26 (1.69 - 62.23); p = 0.011), knowledge on vaccine target (OR = 17.11 (8.51 - 34.41); p < 0.001), knowledge of schedule vaccine (ORa = 3.67 (1.2 - 22.51); p = 0.022), knowledge of prevention methods (OR = 26.86 (12.22 - 59.05); p < 0.001), and to be favorable in EPI in general (ORa = 18.71 (1.51 - 28.23]; p = 0.02).
In our sample, there was no association between the place of residence of mothers or guardians and HPV vaccination of young girls, contrary to the results of North American [15] and Chinese [16] [17] [18] authors. Indeed, the latter found that overall HPV vaccination coverage was higher in developed countries and urban areas than in developing countries and suburban and rural areas. However, Faye et al. in Kaffrine found the opposite with higher HPV vaccination coverage in rural areas [19]. This lack of difference in our sample can be explained by the fact that the department of Koumpentoum is essentially rural and does not have any large cities [10].
The age of the mothers or guardians was not associated with the vaccination of the young girl in accordance with the results found in national series such as Faye et al. [19] and Ndiaye et al. [20]. However, one might expect greater receptivity on the part of young mothers or guardians and greater mobility and attendance at health facilities due to cultural realities.
We did not find an association between mothers or guardian’s marital status and girl’s vaccination, as in the series by Ndiaye et al. [20]. However, the education of mothers or guardians was associated with vaccination of young girls, as in the series of Ndiaye et al. (OR = 1.97; (1.81 - 2.25); p = 0.01) [20] but in contrast to that of Faye et al. [19]. The dissemination of key messages in local languages (Fulani, Mandingo, Serer and Wolof) by community prevention and promotion actors (community relays and aunts), health providers and Niani FM radio improved access to health information.
There was no relations between the practice of an IGA and the vaccination of young girls, contrary to the series of Ndiaye et al. [20] (ORa = 1.21 (1.13 - 1.85); p < 0.001), and Alene T. et al. [21] (ORa = 3.44 (1.97 - 6.01); p < 0.05), where the income of the head of the household was predictive of the vaccination of young girls. Indeed, even if HPV vaccination is free, there may be direct and indirect costs associated with mothers or guardians and young girls travelling to service points. These costs can be a barrier to vaccination, hence the need to strengthen advanced and mobile strategies.
Young girls whose mothers or guardians were informed about cervical cancer (OR = 18.97 (5.68 - 63.24); p < 0.001), knew its risk factors (OR = 8.04 (4.41 - 14.63); p < 0.001) and knew how to prevent it (OR = 26.86 (12.22 - 59.05); p < 0.001) were more likely to be vaccinated. Similar associations between vaccination of young girls and knowledge of cervical cancer and the possibility of cure were found by Alene T. et al. in Gondar, Ethiopia (ORa = 5.42 (2.69 - 11.52); p < 0.05) [21], Ndiaye et al. in Dakar [18] (ORa = 3.05 (2.75 - 4.53); p < 0.001) [20] and by Thiam in Kédougou (ORa = 2.98 (1.19, 7.47); p = 0.025) [22].
Knowledge of HPV vaccination (ORa = 10.26 (1.69 - 69.23); p = 0.011), knowledge of the target to be vaccinated (OR = 17.11 (8.51 - 34.51); p < 0.001) and knowledge of the HPV vaccination schedule (ORa = 3.67 (1.2 - 11.51); p = 0.022) were predictive of better vaccination of young girls. This association between good knowledge of HPV vaccination and vaccine uptake was also found in the series by Rabiu A. et al. in Nigeria (OR = 6.11 (1.37 - 27.34); p = 0.018) [23], Faye et al. in Kaffrine (OR = 10.92 (2.93 - 40.64); p < 0.001) [19] and Thiam in Kédougou (ORa = 2.98 [1.19 - 7.47]; p = 0.025) [22].
The main reasons for non-vaccination were lack of information in (71.43%) of cases and fear of side effects in (10.32%). This indicates the need to increase awareness among mothers or guardians, by diversifying communication channels, about cervical cancer, its causes, means of prevention of the disease, HPV vaccination and its safety.
Young girls whose mothers or guardians were supportive of the EPI were more likely to be vaccinated against HPV (ORa = 18.71 (1.51 - 22.23); p = 0.02). This indicates the importance of comprehensive interpersonal communication about cervical cancer and its prevention and the fact that the HPV vaccine has become a component of the EPI.
Mothers or guardians who have school education was more likely to vaccine their daughters (OR = 5.62 (3.16 - 9.99); p < 0.001) according to Ndiaye et al. [20] (ORa = 1.97; (1.81 - 2.25); p < 0.001). This link may be related to better understanding and adherence to health policies and a higher standard of living for educated mothers or guardians.
Also, young schooled girls were more likely to be vaccinated (OR = 4.1 (2.23 - 7.53); p < 0.001). This finding was also found during the pilot phase in 2016 by Sy K. et al. in West Dakar [7]. Indeed, the target young girls aged between (9) and (14) are mainly in primary and secondary school. Consequently, awareness, involvement and collaboration with the national education sector are imperative in order to reach this school target and the vaccination coverage objective of (90%).
5. Conclusion
The information provided to mothers or guardians is crucial to achieving the vaccination coverage target. Factors associated with low HPV vaccination coverage in the Koumpentoum health district are related to mothers’ or guardians’ lack of knowledge, which is a consequence of insufficient and inadequate communication. HPV vaccination coverage could be significantly improved through an acceleration plan focusing on interpersonal communication and involving national education sector.
Acknowledgements
We would like to thank the health providers and community health workers in the Koumpentoum health district and the families who participated in the survey.