Human papillomavirus vaccination uptake and factors related to uptake in a traditional desert city in the United Arab Emirates

Abstract

It is important to assess uptake of a new vaccine. The present survey estimated the uptake of the recently introduced human papillomavirus vaccine (HPVV) in the emirate of Abu Dhabi and also explored barriers to uptake. A questionnaire survey was conducted in secondary schools, enrolling 334 girls. The uptake percentage was estimated and factors associated with uptake were assessed with multilevel logistic regression. Crude vaccination uptake was 53%. Factors positively associated with uptake were, among others, having knowledge on human papillomavirus (vaccine) and having United Arab Emirates (UAE) nationality. The HPVV uptake of just over 50% could probably be improved by educating the target group and their parents before offering the HPVV and by making it free of costs for everyone.

Keywords

human papillomavirus HPV vaccination uptake girls Muslim

INTRODUCTION

Human papillomavirus (HPV) is one of the most widespread sexually transmitted infections (STIs).^1,2 Approximately 20 million people are currently infected with genital HPV in the USA,³ and as many as half of these infections are among adolescents and young adults aged 15 through 24 years. After HPV infection was demonstrated to be a necessary cause for invasive cervical cancer,^4,5 prevention of infection by a prophylactic vaccine was a logical step. Vaccination would have a high public health impact, as cervical cancer is the second most common malignant neoplasm (after breast cancer) among women worldwide⁶ and cervical screening programmes, although having caused in some countries reductions in incidence and mortality of up to 70%,⁷ require consistent compliance of the population.⁸ Randomized controlled trials have shown that currently available vaccines are safe and efficacious,^9,10 and are thus suitable for inclusion into national immunization programmes.

In 2006, the US Food and Drug Administration approved Gardasil^®, a vaccine highly effective in preventing infection with HPV types 16 and 18, two HPV strains causing about 70% of all cervical cancers, and types 6 and 11 causing 90% of genital warts. This quadrivalent vaccine is given in three doses.¹¹

The key for success of any (new) vaccination is a high uptake, which can be hampered by the fear of adverse effects. For example, the vaccination coverage of the measles-mumps-rubella (MMR) vaccination decreased significantly, reaching dangerous levels, below 80% in some areas of Scotland,¹² in response to its hypothesized link to autism.¹³ Similarly, the uptake of the recently introduced HPV vaccination (HPVV) in the Dutch National Immunization Program, was unexpectedly low (60% as compared with the 80% hoped for) after reports in the press and popular media such as YouTube spread misconceptions about adverse effects of the vaccination. These included infertility, and cancer in HPV-infected women. Another obstacle against public acceptance of HPVV is the notion common in the USA, that vaccination against HPV may promote promiscuity.¹⁴ These values and beliefs in conservative, traditional and religious groups may hamper acceptance of HPVV, while at the same time risky sexual behaviour may warrant it.¹⁵ This argument may also find resonance in Muslim countries, where it may be argued that – as Islam proscribes sex before marriage – HPV infection is probably rare, and thus vaccination not necessary. However, for the United Arab Emirates (UAE) we do not know what the prevalence of HPV is, and although cervical cancer was only the sixth most common cancer among women in 2007,¹⁶ this relatively low incidence only reflects the HPV infection rate of several decades ago. After a pilot study in 2008 with an uptake of 70%, HPVV was implemented in the cities of Abu Dhabi and Al Ain in April 2009 as a non-compulsory (in fact, consent from both the girls and their parents is required) part of the National Immunization Program, targeting schoolgirls in grade 11 (roughly 17 years of age). Vaccinations are administered by school nurses in three doses and is free for nationals, while non-nationals pay 50 dirham (14 US$). Other adolescent vaccinations include rubella in grade 6 (age 12 years) and hepatitis B in grade 11 (for those born before 1995).

The objective of this study was to determine the uptake of the HPVV, identify reasons for not taking it, and assess knowledge and attitudes regarding HPV infection and vaccination in the target group of female secondary school students in Al Ain, UAE.

METHODS

Study area, population and design

Our study area was Al Ain, the second largest city in the Emirate of Abu Dhabi, the largest of the seven UAE. It is a traditional, but well-developed oasis city, with a population of approximately 400,000, and, of the larger cities in the UAE, has the greatest proportion of Emirati nationals.

Our study population was grade 11 and grade 12 school girls (the vaccination target group). Questionnaires were distributed in schools among participants. The design of this questionnaire, data collection and data entry were conducted by final year medical students (MS), assisted by supervisors, as part of their Community Medicine rotation of the Faculty of Medicine and Health Sciences of the United Arab Emirates University. The Al Ain Medical District Human Research Ethics Committee approved the study protocols and all subjects gave informed consent.

Sampling and sample size calculation

Our sampling frame comprised all secondary girls schools with grade 11 and 12 in the district of Al Ain, with a total of about 4800 eligible subjects. A two-stage stratified random sample of 336 subjects was drawn. In the first stage, five schools were randomly selected (one public urban, two public rural and two private urban). In the second stage, grade 11 and 12 (Art and Science in a 50:50 distribution) classes were chosen such that the number of students selected was roughly proportionate to the total number of eligible students in that school. A sample size of 336 was sufficient to estimate the percentage uptake with a 95% confidence interval (CI) with a width 10%.

Data collection

Questionnaires were distributed among all girls present in the selected classes on the day the MS visited the school. Participants were given an information sheet on the aim of the study and then were requested to complete a self-administered questionnaire after signing the consent form. The MS were available to answer questions. After distributing the questionnaires, subjects were asked to fill them out and after completion they were collected by the MS.

The questionnaire consisted of demographic questions, questions on knowledge about HPV infection, HPVV, cervical cancer and the relationship between infection and cancer, questions on attitude (towards vaccination) and questions on practice (vaccination received, if no, why not, and wish to receive it in the future). There were also questions on sources of information on HPV and HPVV, the person a subject would consult regarding the vaccine, including possible concerns with respect to the vaccine and the need for more information.

Statistical analysis

All data were entered into Microsoft Access, converted to SPSS 17.0 (SPSS Inc, Chicago, IL, USA) and checked for keystroke errors. From the questions regarding knowledge of HPV infection and vaccination we calculated a knowledge score (range 0–10). The subjects' and the schools' characteristics, data on sources of information on HPVV, the person a subject would consult regarding the vaccine, possible worries with respect to the vaccine, the need for more information and reasons for not taking the vaccine were listed.

We estimated the uptake of HPVV and its 95% CI and similarly of ‘want to receive HPVV in the future’ for those who had not received it. For the HPVV uptake we also calculated an adjusted figure that took account of the fact that not all three groups of schools (private urban, public urban and public rural) were proportionally represented and that we selected different numbers of students from the schools. Therefore, we calculated for each school a weight that was the product of the number of subjects selected from that school and the inverse of the estimated probability for that school of being in our sample. For this so-called ‘inverse probability weighted estimation’¹⁷ we used a logistic model with ‘school being sampled’ as outcome variable and private/public and urban/rural as independent variables. These weights were used to weight the uptakes in each school, thereby obtaining a weighted uptake estimate adjusted for imbalanced selection.

Finally, we assessed which factors were independently related to vaccination uptake by means of multivariable logistic regression using a stepwise method (Backward Wald). All the variables shown in Table 1 (subjects' characteristics) and the variables in Table 2 marked with an asterisk (lack of information and other worries regarding the vaccine, several sources of information regarding the vaccine used in the past, whom to consult on the vaccine, and the knowledge score) were considered as independent variables for this model. The final model was repeated with multilevel analysis where the potential correlation between students within schools was taken into account. A P value of ≤0.05 was considered significant. All analyses were done with SPSS 17.0, only for the multilevel analyses STATA10 (Stata Corp, College Station, TX, USA) was used.

Table 1

Subjects' characteristics

Variable	n (%)	Median (range)
School type
Private	94 (28.1)
Public	240 (71.9)
School area
Urban	291 (87.1)
Rural	43 (12.9)
Section*
Art	167 (50.0)
Science	160 (47.9)
Not applicable	7 (2.1)
Age (years)		17 (15–20)
Nationality
UAE	181 (54.2)
Other	153 (45.8)
Income family (AED, monthly)**
< 5000	13 (3.9)
5000– <10,000	32 (9.6)
10,000– <15,000	15 (4.5)
≥15,000	78 (23.4)
Don't know	196 (58.7)

UAE = United Arab Emirates

*Section means a school programme that is mainly dominated by language and history subjects (‘art’) or by mathematics, physics, etc. (‘science’). In one class this division was not made

**One AED is about 0.3 US$

Table 2

Satisfaction with information, sources of information on human papillomavirus (HPV) and HPV vaccine (HPVV), attitude questions and knowledge score (n = 334)

Variable	n (%)
Having sufficient information on HPV (subject's opinion)	92 (27.5)
Information source on HPV in the past*
No information at all	119 (35.6)
School	133 (39.8)
Media (TV, radio, newspaper)	117 (35.0)
Relatives/friends	100 (29.9)
Health-care providers	27 (8.1)
Other	35 (10.5)
If you would need more information on HPV, how would you like to get it?*
Health-care providers	223 (66.8)
Internet	140 (41.9)
School	90 (26.9)
Other	34 (10.2)
Worried about HPVV because of:*
Lack of information^†	170 (50.9)
Adverse effects^†	212 (63.5)
Halal ^‡ ^†	20 (6.0)
Social stigma^†	13 (3.9)
Costs^†	10 (3.0)
Promotes promiscuity^†	10 (3.0)
Other^†	18 (5.4)
Parents have sufficient information on HPVV (according to subject)^†	34 (10.2)
Information source on HPVV in the past*
No information at all	118 (35.3)
School^†	163 (48.8)
Media (TV, radio, newspaper)^†	77 (23.1)
Relatives/friends^†	73 (21.9)
Health-care providers^†	34 (10.2)
Other^†	20 (6.0)
Whom would you consult on HPVV?*
Health-care providers^†	215 (64.4)
Relatives/friends^†	196 (58.7)
Internet^†	103 (30.8)
Other^†	14 (4.2)
Should HPVV be free for everybody?
Yes	277 (82.9)
Knowledge score ^†
0–3	142 (42.5)
4–6	153 (45.8)
7–10	39 (11.7)

*Percentages add up to more than 100% as subjects could choose more than one option

^†These variables were, together with the variables in Table 1, considered as independent variables in the logistic model with having received HPVV as outcome (see Table 3)

^‡ Halal means that this is permitted by the Holy Quran

Table 3

Factors associated with receiving the HPV vaccination. Results of multivariable logistic regression: adjusted odds ratios (AOR) and 95% confidence intervals (CI)

	HPV-vaccinated n (%)	AOR	95% CI (P value)
School type
Private	41 (43.6)	1
Public	137 (57.1)	2.05	1.16–3.64 (P = 0.01)
School area
Rural	17 (39.5)	1
Urban	161 (55.3)	2.24	1.09–4.61 (P = 0.03)
Age*
15–16	45 (49.5)	1.29	1.01–1.65 (P = 0.04)
17	60 (50.8)
18	57 (58.8)
19–20	15 (57.7)
Nationality
Non-UAE	64 (41.8)	1
UAE	114 (63.0)	2.50	1.52–4.09 (P < 0.0005)
Knowledge score ^*
0–3	61 (43.0)	1.37	1.21–1.56 (P < 0.0005)
4–6	89 (58.2)
7–10	28 (71.8)
Lacking information
No	99 (60.4)	1
Yes	79 (46.5)	0.60	0.37–0.97 (P = 0.039)

HPV = human papillomavirus; UAE = United Arab Emirates

Significant variables were determined with a backward stepwise method (Wald). All odds ratios were adjusted for the other variables in the table

*These variables were used in the model as continuous variables. This means for age that with an increase of one year the odds of being vaccinated is multiplied with a factor 1.29, and for the knowledge score that an increase of one point multiplies the odds of being vaccinated with a factor 1.37

RESULTS

General data

A total of six schools were approached (2 urban governmental, 2 urban private, 1 rural governmental and one rural private). Both rural private schools available refused to participate and therefore we included five schools only. In these five schools we selected 17 classes.

We distributed the questionnaire among 336 female students in the age group 16–19 years of whom 334 (99 %) completed and returned it. The median age in this group was 17 years (range 15–20 years) and over 50% were UAE nationals. Twenty-eight percent attended a private school and 87% were students in an urban area. The students were about equally distributed over science and art classes. Almost 60% indicated that they did not know the family monthly income (Table 1).

HPV(V) knowledge, sources of information, attitude and worries regarding the HPVV

Only 92 (27.5%) of the subjects indicated that they had sufficient information on HPV while 170 (50.9%) reported that they did not have enough information on HPVV, and 118 (35%) that they had no information on HPVV at all. For those who had at least some information, the school was reportedly the main source (see Table 2); however, the majority said they would prefer to consult a health-care provider (64%) and/or relatives or friends (59%).

Knowledge on HPV, HPVV and its relation with cervical cancer was tested with 10 questions (yielding a 10-point score): only 12% of subjects scored 7 or higher. It was thus not surprising that 170 (51%) reported lack of information as one of their concerns regarding HPVV. The most prevalent concern was ‘adverse effects’, which was reported by 212 subjects (64%). Other concerns, for example that HPVV would promote promiscuity or that it would stigmatize vaccinated girls, were expressed by few participants. A vast majority (277 [83%]) felt that the HPVV should be free for everybody.

HPVV uptake and reasons for not receiving it

One hundred and seventy-eight subjects (53.3%, 95% CI: 47.8–58.7%) reported to have received the HPVV. The weighted figure for the uptake was very similar (51.5%).

‘Presumed side-effects’ was given as the reason for not receiving the HPVV by 52 subjects (33%), ‘family refusal’ by 32 (21%) and ‘costs’ by only eight subjects (5%, all non-nationals). Sixty-four subjects (41%) gave other reasons. Of those subjects giving ‘family refusal’ or other reasons for not receiving HPVV, 48 (31% of total unvaccinated) specified that this occurred because of a lack of information, 14 (9% of unvaccinated) by fear of unwanted health effects, 12 (8%) by being absent or loss of consent, three (2%) by stating that HPVV was not necessary and two (1%) by being afraid of injections. Seventy-eight subjects in the unvaccinated group (50% of those who did not receive the HPVV) indicated that they would like to receive the HPVV in the future. This intention was not significantly related to knowledge.

Factors associated with receiving the HPVV

Logistic regression, with receipt of the HPVV as outcome, showed, using the Backward Wald method, that governmental school students were significantly more likely than private school students to be HPV-vaccinated (adjusted odds ratio [AOR] = 2.05, 95% CI = 1.16–3.64), that students in an urban area were significantly more often vaccinated compared with their rural peers (AOR = 2.24, 95% CI = 1.09–4.61), that UAE nationals were significantly more often vaccinated than non-nationals (AOR = 2.82, 95% CI = 1.75–4.55), that subjects who were concerned about the lack of information were significantly less often vaccinated (AOR = 0.62, 95% CI = 0.39–0.99), that vaccination coverage increased with increasing age (AOR for one-year increase in age = 1.29, 95% CI = 1.01–1.63) and that knowledge was significantly positively associated with being vaccinated (AOR for 1-point increase in knowledge score = 1.35, 95% CI = 1.19–1.54). Multilevel analysis yielded results that were almost identical to those produced by simple logistic regression.

DISCUSSION

Main findings

This cross-sectional study is to our knowledge the first to assess the knowledge, attitude and practice towards HPVV in a traditional Muslim city in the Arab world. We found that the vaccination uptake of the HPVV in Al Ain in the UAE was just over 50% and that lack of information and knowledge on the HPVV was significantly associated with not having received the HPVV.

Comparison with other studies

The HPVV uptake of 50% in this study is high compared with the Centers for Disease Control and Prevention estimate of 37% for 13–17 year olds in 2008,¹⁸ but lower than the almost 80% found in a school-based study in the UK where different implementation plans were compared,¹⁹ and where it was concluded that uptake can be improved by achieving a high initial consent and by accepting late consenters.

Specific findings

Concerns regarding HPVV

The only commonly reported concerns were ‘adverse effects’ and ‘lacking information’. This perceived lack of information was also significantly related to not receiving the HPVV, suggesting that providing adequate information before offering HPVV may increase uptake. This recommendation is supported by the striking discrepancy between the small proportion that actually received information from health-care providers and the large group that would prefer to get information from these professionals. Obviously, perceptions of HPVV promoting promiscuity or not being halal (i.e. permitted according to the Holy Quran) were not big issues in this population of schoolgirls, although it might be among their parents whom we did not study. This attitude differs from the one expressed in a multiethnic group of young Malaysian women,²⁰ where these two perceptions were prevalent, and from USA religious groups that argue that reducing the risk of cancer would encourage sexual activity and that abstinence is the best way to prevent HPV.^14,15,21

Reasons for not receiving the HPVV

Fear of side-effects was the most frequently reported reason for not taking HPVV (a total of 42% reported this), followed by lack of information (31%). This suggests again that a more extensive information campaign should precede the introduction of the vaccine.

Eight percent reported to have been absent or to have lost the consent paper as the reasons for not having received HPVV. Therefore, a change in the programme so that it accepts late consenters might further increase uptake.

Costs were reported to be prohibitive by only 5% of unvaccinated girls; nevertheless a freely available vaccine will undoubtedly increase HPVV coverage.

Only 2% of unvaccinated girls stated that HPVV was not necessary for them. This small group might represent those who presume that in a traditional culture sexually transmitted infections are non-existent.

Knowledge score and perceived lack of knowledge

The knowledge score was positively related to receiving the vaccination, indicating that providing information on HPVV would increase uptake. This positive association was also observed in studies in the USA and China,^22,23 but not in Malaysia.²⁰ Remarkable was that both being concerned about ‘lacking information’ and the knowledge score itself were independently significantly related to acceptance of HPVV in the multivariable model. Perhaps our knowledge score did not measure essential components of knowledge such as knowledge on adverse effects, about which no questions were included in the score. Alternatively, unvaccinated subjects are more likely to tick the option ‘lacking information’ as an excuse for not being vaccinated. Around 14% of participants thought that HPVV would prevent breast cancer, most likely because the Arabic name for HPV is suggestive of breast pathology.

Knowledge of the parents

Knowledge of the parents, as perceived by subjects, was not related to receiving the vaccine. This lack of an association was also observed in a study that measured knowledge on HPVV in mothers of adolescent girls in Italy,²⁴ suggesting that the main reason for parental refusal is not lack of specific knowledge on HPVV, but concerns regarding the total number of vaccinations received.

Variability between schools

The large variability in vaccination uptake between schools was for almost 100% explained by the variables in the logistic model. Therefore, the results of multilevel logistic regression were almost identical to those of the logistic model that ignored the multilevel structure of the data.

Limitations

No school in the category private/rural could be enrolled because both schools in this category refused to participate. Therefore, our attempt to adjust for imbalanced selection of schools from the four categories (private urban, private rural, governmental urban and governmental rural) could not take any data from the private/rural schools into account. One school (providing 30 subjects, about 10% of our sample) conducted a lecture on HPV vaccine and HPV infection before the MS distributed the questionnaires and after schools were asked to participate, obviously with the intention to improve the students' performance. We repeated the main analyses without the girls attending this school and as expected the average knowledge was lower, but the logistic regression determining predictors of being vaccinated yielded almost identical results.

Recommendations

An educational programme for both parents and girls on HPVV should be an integral part of the programme. Education should be provided shortly before the vaccine is offered. Also concerns regarding side-effects and the large number of vaccinations in general in the National Vaccination Program should be addressed;

The HPVV should be free of cost, although only eight subjects (all non-nationals and 5% of the HPV-unvaccinated subjects) admitted that the cost of HPVV was a reason for not accepting it;

More research on knowledge and attitudes in parents should be conducted;

The programme should be more flexible with respect to late consenters;

The Arabic name for HPV should be changed as this may reduce confusion with breast cancer.

CONCLUSION

The HPVV uptake in the city of Al Ain was just over 50%. Actual and perceived knowledge was poor in this population and uptake may possibly be improved by educating the target group and their parents before offering the HPVV and by making it free of costs for everyone. Culturally sensitive issues in terms of promiscuity do not seem to play an important role in this population of schoolgirls.

References

Dillner

, Meijer

, von Krogh

, Horenblas

. Epidemiology of human papillomavirus infection. Scand J Urol Nephrol 2000;205 Suppl:194–200

Koutsky

. Epidemiology of genital human papillomavirus infection. Am J Med 1997;102:3

Centers for Disease Control and Prevention (CDC). Sexually transmitted diseases treatment guidelines (2006). See http://www.cdc.gov/STD/hpv/STDFact-HPV-vaccine-hcp.htm#ref1 (last accessed 1 June 2011)

Walboomers

, Jacobs

, Manos

, Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189:12–9

Bosch

, Manos

, Munoz

, Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. International Biological Study on Cervical Cancer (IBSCC) study group. J Natl Cancer Inst 1995;87:796–802

Ferlay

, Bray

, Pisani

, Parkin

. GLOBOCAN 2002 cancer incidence, mortality and prevalence worldwide. In: IARC Cancer Base No.5 Version 2.0. Lyon, France: IARC Press, 2004

International Agency for Research on Cancer (IARC). Handbooks of Cancer Prevention: Cervix Cancer Screening. Lyon, France: IARC Press, Vol. 10, 2005

Bosch

, Castellsagué

, de Sanjosé

. HPV and cervical cancer: screening or vaccination? Br J Cancer 2008;98:15–21

Garland

, Hernandez-Avila

, Wheeler

, Quadrivalent vaccine against human papillomavirus to prevent anogenital diseases. N Engl J Med 2007;356:1928–43

10.

Paavonen

, Jenkins

, Bosch

, Efficacy of a prophylactic adjuvanted bivalent L1 virus-like-particle vaccine against infection with human papillomavirus types 16 and 18 in young women: an interim analysis of a Phase III double-blind, randomised controlled trial. Lancet 2007;369:2161–70

11.

Centers for Disease Control and Prevention (CDC). HPV Vaccine information for clinicians. 2008. See http://www.asccp.org/PracticeManagement/HPV/tabid/5822/Default.aspx (last accessed 1 June 2011)

12.

Friederichs

, Cameron

, Robertson

. Impact of adverse publicity on MMR vaccine uptake: a population based analysis of vaccine uptake records for one million children, born 1987–2004. Arch Dis Child 2006;91:465–8

13.

Wakefield

, Murch

, Anthony

, Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet 1998;351:637–41

14.

Thomas

. The new human papillomavirus (HPV) vaccine: pros and cons for pediatric and adolescent health. Pediatr Nurs 2008;34:429–31

15.

Katz

, Reiter

, Kluhsman

, Human papillomavirus (HPV) vaccine availability, recommendations, cost, and policies among health departments in seven Appalachian states. Vaccine 2009;27:3195–200

16.

Health Authority Abu Dhabi. Health Statistics. Abu Dhabi: Health Authority Abu Dhabi, 2008

17.

Wooldridge

. Inverse probability weighted estimation for general missing data problems. J Econometrics 2007;141:1281–301

18.

Centers for Disease Control and Prevention (CDC). National, state, and local area vaccination coverage among adolescents aged 13–17 years—United States, 2008. MMWR Morb Mortal Wkly Rep. 2009;58:997–1001

19.

Stretch

. Implementing a school-based HPV vaccination programme. Nurs Times 2008;104:30–3

20.

Wong

. Young multiethnic women's attitudes toward the HPV vaccine and HPV vaccination. Int J Gynaecol Obstet 2008;103:131–5

21.

Gilman

, Gilman

, Johns

3rd . Human papillomavirus, abstinence, and the other risks. Lancet 2009;373:1420–1

22.

Kahn

, Rosenthal

, Hamann

, Bernstein

. Attitudes about human papillomavirus vaccine in young women. Int J STD AIDS 2003;14:300–6

23.

Chan

, Yan Ng

, Lo

, Cheung

, Hung Chung

. Adolescent girls' attitudes on human papillomavirus vaccination. J Pediatr Adolesc Gynecol 2009;22:85–90

24.

Tozzi

, Ravà

, Stat

, Pandolfi

, Marino

, Ugazio

. Attitudes towards HPV immunization of Italian mothers of adolescent girls and potential role of health professionals in the immunization program. Vaccine 2009;27:2625–9