Journal of Adolescent Health 56 (2015) 402e407

www.jahonline.org Original article

Socioeconomic Predictors of Human Papillomavirus Vaccination Among Girls in the Danish Childhood Immunization Program Selma Marie Slåttelid Schreiber a, Kirsten Egebjerg Juul, Ph.D. a, Christian Dehlendorff, Ph.D. b, and Susanne Krüger Kjær, M.D., Dr. Med. Sci. a, c, * a

Unit of Virus, Lifestyle, and Genes, Danish Cancer Society Research Center, Copenhagen, Denmark Unit of Statistics, Bioinformatics, and Registries, Danish Cancer Society Research Center, Copenhagen, Denmark c Department of Gynecology, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark b

Article history: Received September 5, 2014; Accepted December 11, 2014 Keywords: Human papillomavirus; Vaccination; Vaccine coverage; Socioeconomic status

A B S T R A C T

Purpose: In 2009, human papillomavirus (HPV) vaccination was introduced in the Danish national childhood immunization program targeting all 12-year-old girls. Previous findings suggest that 10% e13% of girls born in 1996e1997 have not initiated vaccination despite free access. This study aims to identify socioeconomic predictors of initiation and completion of HPV vaccination. Methods: Girls born in 1996e1997 and their guardians were identified through the Danish Civil Registration System. Information on socioeconomic variables and HPV vaccination status was obtained by linkage to Statistics Denmark and the Danish National Health Insurance Service Register. Through logistic regression, we examined associations between socioeconomic variables and HPV vaccine initiation (N ¼ 65,926) and completion (N ¼ 61,162). Results: Girls with immigrant ethnicity (odds ratio [OR] ¼ .49; 95% confidence interval [CI], .42 e.57) had lower HPV vaccine initiation than Danish girls. Girls of mothers with basic education (OR ¼ .75; 95% CI, .69e.82) or low disposable income (OR ¼ .67; 95% CI, .61e.73) had decreased initiation compared with girls of mothers with higher education/income. Girls of unemployed mothers (OR ¼ .75; 95% CI, .69e.82) or mothers being unmarried (OR ¼ .70; 95% CI, .65e.76) had lower initiation than girls of employed or married mothers. Finally, vaccine initiation varied depending on place of residence. The predictors of HPV vaccine completion were similar to those of initiation. Conclusions: We found social inequality in the initiation and completion of HPV vaccination despite free access. As socioeconomic risk factors identified for cervical cancer also are associated with decreased HPV vaccination, social inequalities in cervical cancer have the potential to increase. Ó 2015 Society for Adolescent Health and Medicine. All rights reserved.

Conflicts of Interest: K.E.J. received a travel grant from Merck. S.K.K. received lecture fees, scientific advisory board fees, and institutional research grants from Merck and Sanofi Pasteur MSD and scientific advisory board fee from Roche. S.M. S.S. was funded by a scholarship from the Copenhagen University Hospital, Rigshospitalet. The study sponsoring has no role in the study design, the collection, analysis, and interpretation of data, writing of the report, and the decision to submit the manuscript for publication. The first draft of the manuscript was written by S.M.S.S. and K.E.J. No perceived potential conflict of interest regarding C.D. 1054-139X/Ó 2015 Society for Adolescent Health and Medicine. All rights reserved. http://dx.doi.org/10.1016/j.jadohealth.2014.12.008

IMPLICATIONS AND CONTRIBUTION

Despite an overall successfully implemented, free-ofcharge human papillomavirus (HPV) vaccination, the same socioeconomic risk factors identified for cervical cancer are also associated with lower HPV vaccination. Therefore, social inequalities in cervical cancer have the potential to increase. This article provides individual-level socioeconomic predictors of HPV vaccine initiation and completion.

Disclaimer: The corresponding author (S.K.K.) hereby declares that everyone who contributed significantly to the work is listed on the title page. * Address correspondence to: Susanne Krüger Kjær, M.D., Dr. Med. Sci., Unit of Virus, Lifestyle, and Genes, Danish Cancer Society Research Center, Strandboulevarden 49, 2100 Copenhagen, Denmark. E-mail address: [email protected] (S.K. Kjær).

S.M. Slåttelid Schreiber et al. / Journal of Adolescent Health 56 (2015) 402e407

Human papillomavirus (HPV) is one of the most common sexually transmitted infections [1] and the high-risk HPV types are recognized as necessary for developing cervical cancer and its precursor lesions [2]. Worldwide, cervical cancer is the third most frequent female cancer with about 530,000 new cases per year [3]. Two prophylactic HPV vaccines are now available and have been shown to be generally safe and efficacious in preventing new HPV infections with HPV 16 and 18, which together cause approximately 70% of cervical cancers [4,5]. The quadrivalent HPV vaccine also protects against HPV 6 and 11, which cause up to 90% of genital warts. In January 2009, HPV vaccination was introduced as a routine vaccination in the Danish national childhood immunization program. The quadrivalent HPV vaccine (Gardasil; Merck, Kenilworth, NJ) was selected through a tender process [6] and has since then been offered to all girls 12 years of age until they turn 15 years, that is, girls born in 1996 or later. Two catch-up programs have been added, and by December 2013, all girls and women from birth cohorts 1985e2001 (age, 12e28 years) have been offered the three-dose vaccination schedule [7,8]. At the start of HPV vaccination in the childhood immunization program in 2009, the Danish Health and Medicines Authority initiated the mailing of invitation letters and information about HPV vaccination initially to all girls born in 1996. Subsequently, public health strategies to improve vaccination coverage have been conducted, primarily through the Danish Health and Medicines Authority. Furthermore, a national network communication campaign, initiated by the Danish Cancer Society, has been used to disseminate knowledge about prevention of HPV. The free-of-charge HPV vaccination is provided by the general practitioner (GP) in an office setting. Because the GP is responsible for the preventive children’s medical checkups and vaccinations, the GP also has a central role in the education of the girl’s guardians about HPV vaccination. Although Denmark has achieved a very high participation rate in the national HPV childhood vaccination program [9], previous findings suggest that still 10%e13% of girls born in 1996 and 1997, the two initial birth cohorts to whom HPV vaccination was offered through the childhood immunization program, have not initiated HPV vaccination [10], despite the opportunity of receiving it free of charge. A correlation between lower socioeconomic status and incidence of cervical cancer has been found in both developing and developed countries [11]. This is also observed in Denmark where a nationwide cohort study has shown that women with lower socioeconomic status (lower education and disposable income, no affiliation to working market, and being unmarried/divorced) have both an increased incidence and a decreased relative survival after a diagnosis of cervical cancer [12]. Thus, having a health care system with free and equal access [13] does not necessarily secure social equality. HPV vaccination could potentially diminish the social inequality in cervical cancer incidence if distributed equally across the target group. In this context, it is interesting to investigate whether the socioeconomic risk factors for cervical cancer are also determinants for participation in the HPV vaccination program. A Danish study has previously examined demographic predictors of HPV vaccine initiation and found low initiation rates among girls with mothers younger or older than 25e34 years, with more than five siblings, and among girls born in other European Union/European Free Trade Association countries than Denmark [14]. However, the follow-up period was limited to the first year of the vaccination program and socioeconomic factors, and determinants for completion of the threedose vaccination schedule were not investigated [14].

403

The aim of our study was to identify socioeconomic and demographic predictors of initiation and completion of HPV vaccination among girls from the two birth cohorts (born in 1996 or 1997), which were the first to be included in the national childhood HPV vaccination program in Denmark, covering the whole period where the girls were eligible for the vaccination program. Methods Data sources and study population Every citizen in Denmark has, since 1968, been registered in the Danish Civil Registration System with an assigned unique 10-digit personal identification number (PIN) [15], which is used universally in society. This ensures correct and high-quality linkages between different registers. On the basis of the Civil Registration System, which contains information on gender, date of birth, and dates for immigration, emigration, and death if relevant, we identified our study population, which comprised all girls from the initial two birth cohorts included in the HPV childhood vaccination program (born 1996 or 1997). It was also possible to identify the guardian of each girl, by the accurate linkage of PINs between guardians and children in the Civil Registration System. Using the PINs as key identifiers, the two birth cohorts of girls were linked to the Danish National Health Insurance Service Register [16] to obtain the individual HPV vaccination status of each girl. This register contains information on services held by the primary health care professionals. HPV vaccines given within the vaccination program were identified through service numbers for each dose (the first dose of HPV vaccination 8328, second dose 8329, and third dose 8330). Furthermore, by linkage to Statistics Denmark’s population-based databases [17], which contain individual socioeconomic information about all residents in Denmark, we obtained information on socioeconomic variables. Inclusion in the study population for both the analysis of initiation (first vaccination) and that of completion of the threedose HPV vaccination schedule required that the girls resided in Denmark from the program start (1 January, 2009) until the end of the program (31 December, 2011 and 2012 for birth cohorts 1996 and 1997, respectively). Therefore, girls born in 1996 were included in the program for 3 years and girls born in 1997 had 4 years to receive the HPV vaccination within that program. If girls were vaccinated before the program start, they were excluded. Information on HPV vaccinations, received outside the program was obtained from the National Prescription Registry, which contains information on all pharmacy-purchased HPV vaccines [18]. The study was approved by the Danish Data Protection Agency. According to the Danish law, ethical approval is not relevant in register-based studies. Statistical analysis In the present study, indicators of socioeconomic status included place of residence, ethnicity, highest attained education, disposable income, employment status, and marital status, which are all commonly used in the social epidemiology in quantifying inequalities in society. All variables were defined at program start and refer to the mothers of the girls, except for ethnicity and place of residence, which refer to the girls. If information on the socioeconomic status of the mother was not available, the socioeconomic status of the father was used instead.

404

S.M. Slåttelid Schreiber et al. / Journal of Adolescent Health 56 (2015) 402e407

Place of residence was defined by the five governing regions in Denmark. The regions are responsible for the health care sector [13]. Ethnicity was defined as Danish (Danish-born girl and guardian), descendant (Danish-born girl but guardian not born in Denmark), or immigrant (girl not born in Denmark). Mother’s highest attained education was categorized as either basic (primary school), vocational (vocational education, secondary school, or high school), or higher (academy profession degree, professional bachelor degree, or university degree). Mother’s disposable income reflects mother’s income after taxation and interest expenses and was divided into three groups on the basis of tertiles. Mother’s employment status was classified as employed, unemployed, or uncategorized (those not included in the labor classification module, not having an income [e.g., housewives], or having initiated education within the last 3 months). Finally, mother’s marital status was classified as married, widow, divorced, or unmarried. Using logistic regression models, we examined the association between indicators of socioeconomic status and, respectively, HPV vaccine initiation and completion. Initially, we compared girls who had received at least one dose with those who had not received any HPV vaccination. Among the girls who received a minimum of one HPV vaccination dose, we subsequently compared girls having received all three doses (completion) with those who had only received one or two doses. Crude odds ratios (OR) and OR adjusted for potentially confounding factors were estimated with corresponding twosided 95% confidence intervals (CIs). There were relatively few missing values for the different variables and they were included as an “unknown” category of the respective variable. We also evaluated potential effect measure modification by education and ethnicity in the unadjusted analyses by including interaction terms between each socioeconomic factor and education and ethnicity, respectively. The analyses were performed in R 2.15.1 (R Core Team, Vienna, Austria) [19], using a 5% significance level. Results This Danish nationwide study included 66,986 girls from the birth cohorts 1996 and 1997. Overall, 674 girls were vaccinated before vaccination program start and 386 girls either emigrated or died during follow-up and were therefore excluded. This left the total initiation and completion study population with 65,926 and 61,162 girls, respectively. We found that at the end of the program 92.8% of the girls had received at least one HPV vaccine dose and among these, 83.6% received all three doses. Table 1 shows the proportion of girls who initiated vaccination and girls who completed the threedose HPV vaccination schedule within the HPV childhood vaccination program in relation to indicators of socioeconomic status. The proportion of girls who received at least one dose of HPV vaccine differed between the regions in Denmark, however, the absolute differences were not substantial. Regarding ethnicity, a lower proportion of girls initiating HPV vaccination was observed among immigrant girls (84.4%). For the characteristics related to the mother, a lower vaccine initiation was observed with, for example, basic education (89.3%) and being unemployed (88.1%). Regarding vaccine completion, a lower proportion was also observed for immigrant girls (75.3%) and for example, girls with mothers having a basic education or low income.

Table 1 Indicators of socioeconomic status in girls from birth cohorts 1996e1997 in relation to human papillomavirus (HPV) vaccine initiation (N ¼ 65,926) and completion (N ¼ 61,162) in Denmark Variable

Birth cohort 1996 1997 Girl’s place of residence Region of Zealand Region of the Capital Region of Central Jutland Region of Northern Jutland Region of Southern Denmark Unknown Girl’s ethnicitya Danish Descendant Immigrant Unknown Mother’s education Higher Vocational Basic Unknown Mother’s disposable income High Middle Low Unknown Mother’s employment status Employed Unemployed Uncategorized Unknown Mother’s marital status Married Widow Divorced Unmarried Unknown

HPV vaccine initiation

HPV vaccine completion

n

Initiated (%)

n

Completed (%)

32,925 33,001

30,326 (92.1) 30,836 (93.4)

30,326 30,836

25,130 (82.9) 25,986 (84.3)

10,271 17,865 15,407

9,436 (91.9) 16,589 (92.9) 14,676 (95.3)

9,436 16,589 14,676

7,875 (83.5) 13,059 (78.7) 12,555 (85.5)

7,078

6,628 (93.6)

6,628

5,917 (89.3)

15,063

13,827 (91.8)

13,827

11,709 (84.7)

6

1 (16.7)

242

6 (2.5)

59,404 4,758 1,522 242

55,657 4,214 1,285 6

(93.7) (88.6) (84.4) (2.5)

55,657 4,214 1,285 6

46,832 3,315 968 1

(84.1) (78.7) (75.3) (16.7)

22,741 30,076 12,605 504

21,468 28,189 11,260 245

(94.4) (93.7) (89.3) (48.6)

21,468 28,189 11,260 245

18,052 23,849 9,033 182

(84.1) (84.6) (80.2) (74.3)

18,683 23,113 23,971 159

17,775 21,710 21,651 26

(95.1) (93.9) (90.3) (16.4)

17,775 21,710 21,651 26

14,893 18,517 17,689 17

(83.8) (85.3) (81.7) (65.4)

55,486 8,397 1,884 159

52,169 7,401 1,566 26

(94.0) (88.1) (83.1) (16.4)

52,169 7,401 1,566 26

44,054 5,823 1,222 17

(84.4) (78.7) (78.0) (65.4)

46,573 626 9,042 9,439 246

43,619 582 8,352 8,570 39

(93.7) (93.0) (92.4) (90.8) (15.9)

43,619 582 8,352 8,570 39

36,847 470 6,781 6,993 25

(84.5) (80.8) (81.2) (81.6) (64.1)

a Danish ¼ Danish-born girl and guardian; Descendant ¼ Danish-born girl but guardian not born in Denmark; Immigrant ¼ girl not born in Denmark.

Table 2 shows unadjusted and adjusted ORs and 95% CIs for initiation of HPV vaccination in association with indicators of socioeconomic status. We found a significant difference in vaccine initiation in relation to place of residence; for example, girls living in Central Jutland (OR ¼ 1.76; 95% CI, 1.59e1.95), Northern Jutland (OR ¼ 1.30; 95% CI, 1.16e1.47), and the Capital (OR ¼ 1.15; 95% CI, 1.04e1.26) had an increased likelihood of initiating HPV vaccination, when compared with the region of Zealand. A decreased likelihood of vaccine initiation was observed for descendants (OR ¼ .68; 95% CI, .61e.76) and immigrants (OR ¼ .49; 95% CI, .42e.57) compared with ethnic Danish girls. A lower vaccine uptake was also observed among girls of mothers with basic education (OR ¼ .75; 95% CI, .69e.82) or with low disposable income (OR ¼ .67; 95% CI, .61e.73). Additionally, a lower likelihood of HPV vaccine initiation was seen among girls of unemployed mothers (OR ¼ .75; 95% CI, .69e.82) and in girls of divorced (OR ¼ .86; 95% CI, .79e.94) or unmarried mothers (OR ¼ .70; 95% CI, .65e.76).

S.M. Slåttelid Schreiber et al. / Journal of Adolescent Health 56 (2015) 402e407 Table 2 Odds ratios (ORs) and 95% confidence intervals (CIs) for initiation of human papillomavirus vaccination in relation to indicators of socioeconomic status among girls from birth cohorts 1996e1997 in Denmark (N ¼ 65,926) Variable

Girl’s place of residence Region of Zealand Region of the Capital Region of Central Jutland Region of Northern Jutland Region of Southern Denmark Girl’s ethnicityb Danish Descendant Immigrant Mother’s education Higher Vocational Basic Mother’s disposable income High Middle Low Mother’s employment status Employed Unemployed Uncategorized Mother’s marital status Married Widow Divorced Unmarried

Unadjusted

Adjusteda

OR

95% CI

OR

95% CI

1 1.15 1.78 1.30 .99

ref. 1.05e1.26 1.60e1.97 1.16e1.47 .90e1.08

1 1.15 1.76 1.30 1.00

ref. 1.04e1.26 1.59e1.95 1.16e1.47 .91e1.10

1

ref. .47e.57 .32e.42

1

.52 .37

.68 .49

ref. .61e.76 .42e.57

.89 .50

ref. .82e.95 .46e.54

1 1.03 .75

ref. .95e1.11 .69e.82

ref. .73e.86 .44e.52

1

.79 .48

.86 .67

ref. .78e.94 .61e.73

ref. .44e.51 .28e.36

1

.47 .31

.75 .62

ref. .69e.82 .53e.72

ref. .66e1.22 .75e.90 .62e.72

1

.90 .82 .67

.94 .86 .70

ref. .69e1.29 .79e.94 .65e.76

1

1

1

1

a

405

Table 3 Odds ratios (ORs) and 95% confidence intervals (CIs) for completion of human papillomavirus vaccination in relation to indicators of socioeconomic status among girls from birth cohorts 1996e1997 in Denmark (N ¼ 61,162) Variable

Girl’s place of residence Region of Zealand Region of the Capital Region of Central Jutland Region of Northern Jutland Region of Southern Denmark Girl’s ethnicityb Danish Descendant Immigrant Mother’s education Higher Vocational Basic Mother’s disposable income High Middle Low Mother’s employment status Employed Unemployed Uncategorized Mother’s marital status Married Widow Divorced Unmarried

Unadjusted

Adjusteda

OR

95% CI

OR

95% CI

1

ref. .69e.78 1.09e1.26 1.50e1.81 1.02e1.18

1

ref. .69e.78 1.08e1.25 1.49e1.80 1.02e1.18

ref. .64e.75 .51e.65

1

.69 .58

.88 .67

ref. .81e.95 .59e.77

1 1.04 .77

ref. .99e1.09 .72e.81

1 1.05 .88

ref. .99e1.10 .82e.94

1 1.12 .86

ref. 1.06e1.19 .82e.91

1 1.07 .90

ref. 1.01e1.14 .85e.96

1

ref. .64e.72 .58e.74

1

.68 .65

.83 .81

ref. .77e.90 .71e.92

ref. .63e.95 .75e.84 .77e.87

1

.77 .79 .82

.82 .82 .83

ref. .67e1.01 .77e.88 .78e.89

.73 1.17 1.65 1.10 1

1

.73 1.16 1.64 1.09

a

Mutually adjusted. Danish ¼ Danish-born girl and guardian; Descendant ¼ Danish-born girl but guardian not born in Denmark; Immigrant ¼ girl not born in Denmark.

Mutually adjusted. Danish ¼ Danish-born girl and guardian; Descendant ¼ Danish-born girl but guardian not born in Denmark; Immigrant ¼ girl not born in Denmark.

The correlation between indicators of socioeconomic status and completion of HPV vaccination among girls who received at least one dose of the HPV vaccine is displayed in Table 3. In line with the pattern seen for HPV vaccine initiation, we found an association between HPV vaccine completion and ethnicity (descendants, OR ¼ .88; 95% CI, .81e.95; immigrants, OR ¼ .67; 95% CI, .59e.77), mother’s education (basic, OR ¼ .88; 95% CI, .82e.94), employment status (unemployed, OR ¼ .83; 95% CI, .77e.90), and marital status (divorced, OR ¼ .82; 95% CI, .77e.88; unmarried, OR ¼ .83; 95% CI, .78e.89). In contrast to HPV vaccine initiation, girls living in the Capital Region were less likely to receive all three HPV vaccine doses (OR ¼ .73; 95% CI, .69e.78) when compared with the region of Zealand. Girls living in the region of Northern Jutland were most likely to complete the three-dose HPV vaccination schedule (OR ¼ 1.64; 95% CI, 1.49e1.80). Finally, mother’s income was less strongly associated with vaccine completion than with vaccine initiation (middle income, OR ¼ 1.07; 95% CI, 1.01e1.14; low income, OR ¼ .90; 95% CI, .85e.96). We also assessed potential effect measure modification and we found that the effect of mother’s disposable income on vaccine initiation was modified by ethnicity. Among girls with Danish guardians, decreasing mother’s income was associated with decreasing odds of initiating vaccination (OR ¼ .50; 95% CI, .45e.55 for low vs. higher income). In contrast, decreasing mother’s income was related to increasing odds of initiating HPV vaccination in immigrant girls (OR ¼ 1.90; 95% CI, 1.23e2.96 for low vs. higher income). This pattern was also observed regarding mother’s education (data not shown).

Discussion

b

b

In this nationwide population-based cohort study including all girls from the two initial birth cohorts in the national HPV childhood vaccination program in Denmark, we found social inequality in the initiation and completion of HPV vaccination, despite a free-of-charge immunization program with equal access. We observed statistically significantly decreased vaccination coverage (initiation and completion) with ethnicity, place of residence, disposable income, highest attained education, employment status, and marital status. Ethnicity was one of the strongest predictors of both initiation and completion of HPV vaccination. We found that girls who were not born in Denmark (immigrants) and Danish-born girls with guardians not born in Denmark (descendants) had a lower likelihood of having started HPV vaccination. These results are in line with those of a systematic review and meta-analysis [20] including primarily North American studies, where white young women were more likely to initiate vaccination compared with black young women, and also with a previous Danish study [14]. When considering HPV vaccine completion, the same tendency was observed in a large, representative U.S. study with National Immunization Survey-Teen data [21] and in a study from the United Kingdom [22]. However, the latter study was based on populations from only three primary care trusts, potentially limiting the generalizability of the results, and in addition, the study had missing information about ethnicity for 17% of the study population.

406

S.M. Slåttelid Schreiber et al. / Journal of Adolescent Health 56 (2015) 402e407

The burden of HPV infection is strongly associated with onset of sexual activity and number of partners and varies between cultures and social environments [23]. Therefore, if some ethnic groups recognize themselves as having a low risk of HPV infection and therefore reject vaccination, this might partly explain our findings. Lower vaccine coverage among ethnic minority girls could also be explained by integration and language barriers [24], which most likely impede health care seeking and communication, or if both HPV knowledge and HPV vaccine acceptance are lower among ethnic minorities, which has previously been suggested [25]. Additionally, barriers of HPV vaccination among minority groups have included safety concerns and logistical challenges to completion of the three-dose vaccine series [23]. In the present study, we found a significant geographical variation in vaccination coverage that could not entirely be explained by education, income, employment, or marital status as it persisted after adjustment for these factors. Girls from Central and Northern Jutland had a higher vaccine initiation and completion than girls from the region of Zealand. Furthermore, vaccine completion was lowest among girls from the Capital Region. Overall, our results are in agreement with those of previous studies where place of residence was characterized by province [14], region [26,27], state [28], or urbanization [14,29]. A possible explanation for the regional variation of HPV vaccine coverage in Denmark could be attributed to the different communication strategies in relation to vaccination across the governing regions. For example, Central and Northern Jutland, which are the regions having the highest rates of HPV vaccine initiation and completion, are the only regions where the HPV invitation letters/postcards are produced in advance, ready for the GPs to be sent out personally to all 12-year-old girls in the region. Furthermore, Central Jutland is the only region where follow-up-lists are sent out to the GPs with reminders regarding those 12- to 14-year-old girls, who have not initiated or completed HPV vaccination yet (personal written communication). We observed a social gradient regarding education, income, and employment status in relation to HPV vaccine initiation and completion, where decreases in vaccine coverage were associated with girls having mothers being more disadvantaged. This social gradient persisted after mutual adjustment was performed for place of residence, ethnicity, mother’s education, disposable income, and marital status. These observations support findings from some previous studies where shorter education [21,28,30], lower income [21,26,29,30], and higher degree of deprivation [31,32] were found to be associated with lower rates of both HPV vaccine initiation and completion. Previous results were not entirely consistent and lower levels of education and income have also been found associated with higher vaccine initiation [28,29], whereas in some other studies, income, education [20], and deprivation [22] were found not to be associated with HPV vaccination. Additionally, we found that daughters of divorced or unmarried mothers had decreased likelihood for both vaccine initiation and completion, which is in line with, for example, a Californian study on the basis of representative health interview survey data among guardians to 12- to 17-year-old girls where an association with marital status was found, however, the association did not reach statistical significance [30]. Finally, we observed that ethnicity was a statistically significant effect modifier, in particular regarding mother’s disposable income. Our results uncover that girls of immigrant guardians with

lower income are more likely to receive vaccination than those of high-income immigrant guardians. We have no immediate explanation of this finding but hypothetically, this could be because of varying trust in authorities among immigrant guardians with lower and higher income, respectively. An important strength of the present study is the nationwide study population where we had virtually no loss to follow-up. A further strength is the availability of information on an individual level about both socioeconomic status and vaccination status from the population-based registers in Denmark, which eliminates the risk of recall bias. Through the individual-level measures attained from these registers, the risk of socioeconomic status misclassification is diminished, which contrasts other studies with aggregated measures such as area-level measures of deprivation. Moreover, we identified predictors of both initiation and completion of the three-dose HPV vaccination schedule, which was made possible because of inclusion of only those birth cohorts who have been included in the HPV childhood vaccination program for the maximum inclusion period (i.e., 3 and 4 years for birth cohorts 1996 and 1997, respectively), allowing each girl having sufficient amount of time to finish the vaccination schedule. A potential study limitation could be the possibility of missing information on vaccinations given. However, as GPs are reimbursed by registering their services to the Danish National Health Insurance Service Register, severe underreporting seems unlikely. Another limitation could be potential misclassification regarding marital status as the categories widowed, divorced, or unmarried may include both guardians with cohabitant and single status. An additional limitation is the missing information about some guardians’ socioeconomic status, however, the number of missing values in this study is limited in absolute terms. Finally, it is possible that some degree of residual confounding is present as all the socioeconomic determinants are strongly correlated. Overall, the HPV vaccination is successfully implemented in Denmark in the already well-established national childhood immunization program with participation rates among the highest in the world [9]; therefore, in absolute terms, nonparticipation may not be considered a major problem. However, the group of nonparticipants is potentially a vulnerable group and inequality in vaccine coverage has also been described for other vaccines in the Danish childhood immunization program [33]. In our study, the same socioeconomic risk factors as identified for cervical cancer also applied to HPV vaccination, and thus, social inequalities in cervical cancer incidence have the potential to increase further. Therefore, it is important that these girls are targeted to facilitate equal vaccine coverage. In relation to this, there should be a focus on improving the communication strategies and targeting the information campaigns concerning freeof-charge vaccination and prevention against cervical cancer toward the more disadvantaged girls and their guardians, especially when it also has been shown that there is inequality in cervical cancer screening in Denmark [34]. Another suggestion could be to increase the follow-up of vaccinations through patient reminder/recall intervention, which has been found to increase immunization rates [35]dalso for other vaccines in the Danish childhood immunization program [36]. Fortunately, initiatives in this direction are being developed in Denmark. The Danish vaccination register has recently been established [37], and using this, the National Institute for Health Data and Disease Control has authorization to send reminders to girls with lacking vaccinations. This will have the potential to not only increase

S.M. Slåttelid Schreiber et al. / Journal of Adolescent Health 56 (2015) 402e407

vaccination coverage but also contribute to limit inequalities in HPV vaccination. Denmark has implemented HPV vaccination in a clinic-based setting and has obtained a high coverage. However, studies have found that school-based vaccination versus community-based public health clinics increases vaccination coverage among girls living in socioeconomically deprived neighborhoods [38,39]. Nevertheless, inequalities are still found despite school-based vaccination [22]. In conclusion, in this nationwide population-based cohort study, we found that HPV vaccination initiation and completion are both significantly influenced by socioeconomic status despite a free-of-charge immunization program. Vulnerable groups should have special attention in future campaigns to increase HPV vaccination coverage and subsequently decrease the prevalence of HPV infection and the incidence of cervical cancer. Future research should aim to uncover more specific reasons for nonvaccination in a context of free-of-charge immunization with equal access to health care. Acknowledgments For educational training purposes results have previously been presented by means of a poster and platform presentation at the Congress of Medical Student Research 2014 and at the PhD day 2014 held at the University of Copenhagen, respectively. References [1] Baseman JG, Koutsky LA. The epidemiology of human papillomavirus infections. J Clin Virol 2005;32(Suppl 1):S16e24. [2] Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189: 12e9. [3] Forman D, de Martel C, Lacey CJ, et al. Global burden of human papillomavirus and related diseases. Vaccine 2012;30(Suppl 5):F12e23. [4] Smith JS, Lindsay L, Hoots B, et al. Human papillomavirus type distribution in invasive cervical cancer and high-grade cervical lesions: A meta-analysis update. Int J Cancer 2007;121:621e32. [5] Schiller JT, Castellsague X, Garland SM. A review of clinical trials of human papillomavirus prophylactic vaccines. Vaccine 2012;30(Suppl 5):F123e38. [6] Sundhedsstyrrelsen. Indstilling vedr. indførelse af humant papillomavirus (HPV)-vaccination i det danske børnevaccinationsprogram. [Danish Health and Medicines Authority. Recommendation regarding introduction of human papillomavirus (HPV) vaccination in the Danish Childhood Vaccination Program]. Available at: http://sundhedsstyrelsen.dk/w/media/ 401A7568F15945B193EC0B0346B58CA5.ashx; Accessed February 2, 2014. [7] Finansministeriet. Aftaler om Finansloven for 2008. Indførelse af vaccination mod livmoderhalskræft og pneumokokker. [Ministry of Finance. Agreements on the 2008 Finance Act. The introduction of vaccination against cervical cancer and pneumococcus]. Copenhagen, 2008: 62e65. Available at: http://www.fm.dk/ publikationer/2008/aftaler-om-finansloven-for-2008/download/w/media/ Files/Publikationer/2008/Download/aftaler_om_finansloven_for_2008_web. ashx; Accessed January 29, 2014. [8] Ministeriet for Sundhed og Forebyggelse. HPV-vaccination catch-up program. [Ministry of Health and Prevention. HPV vaccination catch-up program]. Available at: http://www.sum.dk/Aktuelt/Nyheder/Forebyggelse/ 2012/Juli/HPV-vaccination.aspx; Accessed January 29, 2014. [9] Markowitz LE, Tsu V, Deeks SL, et al. Human papillomavirus vaccine introductioneThe first five years. Vaccine 2012;30(Suppl 5):F139e48. [10] Valentiner-Branth P, Knudsen LK, Andersen PH. HPV vaccinationecoverage 2012. EPI-news:20. Available at, http://www.ssi.dk/English/News/EPINEWS/2013/No%2020%20-%202013.aspx; 2013. Accessed May 14, 2014. [11] Parikh S, Brennan P, Boffetta P. Meta-analysis of social inequality and the risk of cervical cancer. Int J Cancer 2003;105:687e91. [12] Jensen KE, Hannibal CG, Nielsen A, et al. Social inequality and incidence of and survival from cancer of the female genital organs in a populationbased study in Denmark, 1994e2003. Eur J Cancer 2008;44:2003e17.

407

[13] Ministry of Health and Prevention. Health care in Denmark. Copenhagen:6e13. Available at: http://www.sum.dk/Aktuelt/Publikationer/ Publikationer/w/media/Filer%20-%20Publikationer_i_pdf/2008/UK_Health care_in_dk/pdf.ashx; 2008. Accessed January 29, 2014. [14] Widgren K, Simonsen J, Valentiner-Branth P, Molbak K. Uptake of the human papillomavirus-vaccination within the free-of-charge childhood vaccination programme in Denmark. Vaccine 2011;29:9663e7. [15] Pedersen CB. The Danish Civil Registration System. Scand J Public Health 2011;39(Suppl 7):22e5. [16] Andersen JS, Olivarius NF, Krasnik A. The Danish National Health Service Register. Scand J Public Health 2011;39(Suppl 7):34e7. [17] Statistics Denmark. Available at: http://www.dst.dk/en; Accessed February 17, 2014. [18] Kildemoes HW, Sorensen HT, Hallas J. The Danish National Prescription Registry. Scand J Public Health 2011;39(Suppl 7):38e41. [19] R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Available at: http://www.r-project.org/; 2013. Accessed April 24, 2014. [20] Fisher H, Trotter CL, Audrey S, et al. Inequalities in the uptake of human papillomavirus vaccination: A systematic review and meta-analysis. Int J Epidemiol 2013;42:896e908. [21] Polonijo AN, Carpiano RM. Social inequalities in adolescent human papillomavirus (HPV) vaccination: A test of fundamental cause theory. Soc Sci Med 2013;82:115e25. [22] Fisher H, Audrey S, Mytton JA, et al. Examining inequalities in the uptake of the school-based HPV vaccination programme in England: A retrospective cohort study. J Public Health (Oxf) 2014;36:36e45. [23] Jeudin P, Liveright E, Del Carmen MG, Perkins RB. Race, ethnicity, and income factors impacting human papillomavirus vaccination rates. Clin Ther 2014;36:24e37. [24] Gerend MA, Zapata C, Reyes E. Predictors of human papillomavirus vaccination among daughters of low-income Latina mothers: The role of acculturation. J Adolesc Health 2013;53:623e9. [25] Marlow LA. HPV vaccination among ethnic minorities in the UK: Knowledge, acceptability and attitudes. Br J Cancer 2011;105:486e92. [26] Lefevere E, Hens N, de Smet F, Van Damme P. Dynamics of HPV vaccination initiation in Flanders (Belgium) 2007e2009: A Cox regression model. BMC Public Health 2011;11:470. [27] Rondy M, van Lier A, van de Kassteele J, et al. Determinants for HPV vaccine uptake in the Netherlands: A multilevel study. Vaccine 2010;28:2070e5. [28] Pruitt SL, Schootman M. Geographic disparity, area poverty, and human papillomavirus vaccination. Am J Prev Med 2010;38:525e33. [29] Smith LM, Brassard P, Kwong JC, et al. Factors associated with initiation and completion of the quadrivalent human papillomavirus vaccine series in an Ontario cohort of grade 8 girls. BMC Public Health 2011;11:645. [30] Tiro JA, Tsui J, Bauer HM, et al. Human papillomavirus vaccine use among adolescent girls and young adult women: An analysis of the 2007 California Health Interview Survey. J Womens Health (Larchmt) 2012;21:656e65. [31] Roberts SA, Brabin L, Stretch R, et al. Human papillomavirus vaccination and social inequality: Results from a prospective cohort study. Epidemiol Infect 2011;139:400e5. [32] Sinka K, Kavanagh K, Gordon R, et al. Achieving high and equitable coverage of adolescent HPV vaccine in Scotland. J Epidemiol Community Health 2014;68:57e63. [33] Michelsen SI, Kastanje M, Flachs EM, et al. Evaluering af de forebyggende børneundersøgelser i almen praksis [Evaluation of the preventive medical check-ups for children in general practice]. Copenhagen: Danish Health and Medicines Authority, National Institute of Public Health, University of Southern Denmark. Available at: http://www.si-folkesundhed.dk/upload/ 2698_-_evaluering_af_de_forebyggende_b%C3%B8renunders%C3%B8gelser_ i_almen_praksis_ii.pdf; 2007. Accessed November 26, 2014. [34] Kristensson JH, Sander BB, von Euler-Chelpin M, Lynge E. Predictors of nonparticipation in cervical screening in Denmark. Cancer Epidemiol 2014;38: 174e80. [35] Szilagyi PG, Bordley C, Vann JC, et al. Effect of patient reminder/recall interventions on immunization rates: A review. JAMA 2000;284:1820e7. [36] Flachs L, Melgaard K. Tilslutningen til børneundersøgelser og vaccinationer. [The use of children’s medical check-ups and vaccinations]. Manedsskr Prakt Laegegern 1999;77:249e56. [37] Krause TG, Jakobsen S, Haarh M, Mølbak K. The Danish vaccination register. Euro Surveill 2012;17. [38] Musto R, Siever JE, Johnston JC, et al. Social equity in human papillomavirus vaccination: A natural experiment in Calgary Canada. BMC Public Health 2013;13:640. [39] Gold R, Naleway AL, Jenkins LL, et al. Completion and timing of the threedose human papillomavirus vaccine series among adolescents attending school-based health centers in Oregon. Prev Med 2011;52:456e8.

Socioeconomic predictors of human papillomavirus vaccination among girls in the Danish childhood immunization program.

In 2009, human papillomavirus (HPV) vaccination was introduced in the Danish national childhood immunization program targeting all 12-year-old girls. ...
325KB Sizes 0 Downloads 9 Views