Risk factors for persisting measles susceptibility: a case-control study among unvaccinated orthodox Protestants

Risk factors for persisting measles susceptibility: a case-control study among unvaccinated... Abstract Background Measles is an infectious disease providing lifelong immunity. Epidemics periodically occur among unvaccinated orthodox Protestants in the Netherlands. During the 2013/2014 epidemic, 17% of the reported patients was over 14 years old. Apparently, they did not catch measles during the previous 1999/2000 epidemic and remained susceptible. We wanted to identify risk factors for this so-called persisting measles susceptibility, and thus risk factors for acquiring measles at older age with increased risk of complications. Methods A case-control study was performed among unvaccinated orthodox Protestants born between 1988 and 1998; cases had measles in 2013/2014, controls during or before 1999/2000. Associations between demographic, geographical and religion-related determinants and persisting measles susceptibility were determined using univariate and multivariable logistic regression. Analyses were stratified in two age-groups: infants/toddlers and primary school-aged children during the 1999/2000 measles epidemic. Results In total, 204 cases and 563 controls were included. Risk factors for persisting measles susceptibility for infants/toddlers in 1999/2000 were belonging to a moderately conservative church, absence of older siblings and residency outside low vaccination coverage (LVC)-municipalities. Risk factors for primary school-aged children were residency outside LVC-municipalities and attendance of non-orthodox Protestant primary school. Conclusion Unvaccinated orthodox Protestant adolescents and adults who resided outside the LVC-municipalities, did not attend an orthodox Protestant primary school, had no older siblings and belonged to a moderately conservative church were at risk for persisting measles susceptibility and, thus, for acquiring measles at older age with increased risk of complications. For this subgroup of orthodox Protestants targeted information on vaccination is recommended. Introduction Measles is a highly infectious disease that causes high fever, rash, cough and conjunctivitis.1 Measles provides lifelong immunity; after recovery the patient is not susceptible to measles anymore.2 In Western countries, measles is no longer considered a childhood disease as it affects adolescents and adults as well.1,3 In several European countries a shift towards a higher median age of unvaccinated cases is reported during measles epidemics.4–6 Measles at higher age is associated with an increased risk of complications such as pneumonia, diarrhoea, acute encephalitis, hepatitis and post-infectious encephalomyelitis, resulting in relatively high hospitalization rates and deaths.1,7 Measles can be prevented by vaccination. In the Netherlands measles vaccination has been offered free of charge to all children since 1976. Catch-up vaccination is possible until 18 years and actively offered during epidemics. Before introduction of vaccination, large measles epidemics occurred every other year. Despite a national vaccination coverage of 96% (at the age of two), measles epidemics still occur in the Netherlands.8 These epidemics are largely confined to an orthodox Protestant minority of 250 000 people with religious objections to vaccination.7,9 They are members of various small orthodox Protestant church denominations (OPD), each with varied interpretations of the Bible with respect to vaccination. Vaccination coverage varies between the different OPDs from <15 to >85%.10 Overall, vaccination coverage among orthodox Protestants is ∼60%.10 Historically, Dutch orthodox Protestants live in rural areas stretching from the south-west to the north-east of the Netherlands, commonly referred to as the Bible Belt.11 Nowadays, about 75% of the orthodox Protestants live geographically clustered in this area.11,12 In 2013, 29 municipalities in this area had a vaccination coverage <90% [low vaccination coverage (LVC)-municipalities)].13 Apart from geographical clustering, strong social clustering is common among orthodox Protestants. Orthodox Protestant families are characterized as large, close-knit families with an average of four children per family, the national average being 1.7.14,15 Furthermore, the orthodox Protestant minority has its own political party (SGP), schools, newspaper, magazines, social media platforms and websites. Almost half of the 160 orthodox Protestant primary schools are located in (rural) LVC-municipalities, the others are located in larger towns and cities.16,17 The seven orthodox Protestant secondary schools and two colleges are all centrally located to serve orthodox Protestants from a large area.17 Due to this geographical and social clustering of unvaccinated children, the measles virus is easily transmitted within this minority. During the 2013–14 measles epidemic 2700 measles cases were notified compared to 3292 cases during the 1999/2000 epidemic.7,9 However, underreporting of measles cases is common during outbreaks and epidemics, either because patients do not consult a physician or because physicians do not report all cases to public health authorities.18 Compared to the 1999/2000 epidemic, the 2013/2014 epidemic showed a considerable higher median age of infection and higher incidence in older age groups.7 As infection with measles provides lifelong immunity, measles patients born before 1999 must have been susceptible for measles during the 1999/2000 epidemic as well. Apparently, they escaped infection during the 1999/2000 epidemic and were thus persistently susceptible to measles, until they were infected in 2013/2014. We aimed to identify characteristics of unvaccinated orthodox Protestants associated with persistent measles susceptibility after the 1999/2000 epidemic, and, thus, with increased risk of complications when acquiring measles at older age. Based on these characteristics, targeted information can be developed for unvaccinated orthodox Protestant adolescents and adults who are at increased risk of measles and its complications, in order to make them aware of their susceptibility and consider vaccination. Moreover, knowledge of the characteristics of those with persistent susceptibility can be used for estimating the potential burden of disease, and health care use in future measles epidemics. Methods Study design and participants We performed a retrospective case-control study among unvaccinated orthodox Protestants born between 1988 and 1998. These individuals were 14–26 years of age during the measles epidemic of 2013/2014 and born before the 1999/2000 measles epidemic. Cases were individuals who were notified with measles during the 2013/2014 epidemic. In the Netherlands measles is a mandatory notifiable disease; laboratory confirmed and epidemiologically linked cases have to be reported to the Regional Public Health Service (RPHS) by physicians and laboratories.19 All RPHSs that covered one or more LVC-municipalities (12 of the 25 Dutch RPHS regions) participated in this study. Controls were individuals with self-reported measles (including symptoms of fever, rash, red and watery eyes, rhinitis and/or cough) during or before the 1999/2000 epidemic. Data collection Data collection was performed in 2015 Cases received a personal invitation from their RPHS to participate in our study and to fill out an online questionnaire. A reminder was sent two weeks later. Controls were approached via orthodox Protestant (social) media and orthodox Protestant secondary schools and invited to fill out an online questionnaire. In addition, for controls snowball sampling was used as respondents were requested to invite friends and relatives, and contacts from student societies. Personal invitation of controls was not feasible as registrations containing personal data of the 1999/2000 epidemic were not available anymore at the RPHSs. The questionnaire contained questions on year of birth, gender, number of older and younger siblings, postal code, OPD, (orthodox Protestant) school attendance, time and symptoms of measles and measles, mumps and rubella (MMR)-vaccination status (at least one MMR/no MMR/unknown). Participation was anonymous. Data-analysis Based on postal codes, respondents were classified as living in or outside a LVC-municipality. A LVC-municipality was defined as a municipality with a vaccination coverage <90% in 2013 and in which >5% of the population voted for the orthodox Protestant political party (SGP). In 2013, 29 out of 408 municipalities in the Netherlands were LVC-municipalities.13 The OPDs were classified into two categories based on level of conservatism and vaccination coverage.10,20 OPDs with a high level of conservatism (Old Reformed Congregations and Reformed Congregations in the Netherlands) have a vaccination coverage of <25%. OPDs with a moderate level of conservatism have respective vaccination coverages of 50–75% (Reformed Congregations and Restored Reformed Church), and >85% in Christian Reformed Churches and Protestant Church in the Netherlands, including the Reformed Bond.10 For respondents who were school-aged in 1999/2000, the school attendance variable was dichotomized in whether or not they attended an orthodox Protestant school (OPS vs. non-OPS). Statistical analysis The main outcome variable was ‘persisting measles susceptibility’, i.e. not being infected with measles during the 1999/2000 epidemic. Descriptive analyses were conducted using mean and percentages in categorical variables and mean and interquartile range in continuous variables. As school attendance is known to be strongly associated with childhood infections, analyses were stratified into two age-groups.21,22 Group 1 consisted of infants and toddlers born in 1996–98 who were too young to attend school; group 2 consisted of school-aged children born in 1988–95. Univariate and multivariable logistic regression (OR, 95%CI) was used to determine the association between the main outcome variable ‘persisting measles susceptibility’ and the determinants: gender, presence of older siblings, residency in LVC-municipalities, OPS attendance and OPD-membership. Determinants that were statistically significantly associated with the outcome variable (P < 0.05) in the univariate analysis were included in the multivariable analysis. Since almost half of the primary OPSs are located in LVC-municipalities, the determinants ‘residing in LVC-municipalities’ and ‘type of school attendance’ were combined for the group of school-aged children, resulting in four categories: (i) residing in a LVC-municipality and attending an OPS, (ii) residing outside a LVC-municipality and attending an OPS, (iii) resident in a LVC-municipality and not attending an OPS and (iv) residing outside a LVC-municipality and not attending an OPS. All statistical analysis were conducted using SPSS version 21. Ethics The study was approved by the research ethics committee of the Radboud University Nijmegen Medical Centre, Nijmegen, the Netherlands; CMO number 2014/1519. Results The total study population consisted of 767 persons; 204 cases and 563 controls. Of the 304 reported measles cases who were invited to participate in the study, 240 responded to the questionnaire (response rate 79%) and 16 questionnaires were incomplete on essential questions—residency and school attendance—and 20 respondents did not meet the inclusion criteria regarding measles, year of birth, vaccination status or church denomination. Accordingly, 204 cases were included (figure 1). In total 1738 controls responded to the online questionnaire, of which 396 questionnaires were incomplete on essential questions, namely residency and school attendance. The inclusion criteria for controls were not met in 779 questionnaires, most frequently because the respondents did not report having contracted measles as a child or reported to have had measles during the 2013/2014 epidemic instead of the 1999/2000 epidemic. Finally, 563 controls were included for analysis (figure 1). Figure 1 View largeDownload slide Recruitment and response flowchart of cases and controls (n =2042) Figure 1 View largeDownload slide Recruitment and response flowchart of cases and controls (n =2042) Characteristics of cases and controls are shown in table 1. Overall, 74% of the respondents included in the study were female. More than half of the controls were school-aged in 1999/2000 (53.1%) compared to one third of the cases (32.3%). Some of the completed questionnaires which were included for data-analysis lacked postal code data due to a technical problem in the online questionnaire. Postal code data of the cases who were included for data-analysis was therefore compared to the postal code data of all reported measles cases who were initially invited by RPHSs. There was no statistically significant difference in residency in a LVC-municipality between invited cases and included cases (resp. 48.2% and 49.3%, P = 0.83). Table 1 Characteristics of unvaccinated orthodox Protestant cases and controls (n = 767)   Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3    Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3  a IQR: interquartile range. b LVC-municipality: municipality with measles vaccination coverage <90%. c OPS: orthodox Protestant school. Table 1 Characteristics of unvaccinated orthodox Protestant cases and controls (n = 767)   Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3    Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3  a IQR: interquartile range. b LVC-municipality: municipality with measles vaccination coverage <90%. c OPS: orthodox Protestant school. Group 1: infants and toddlers during the 1999/2000 epidemic In the group of infants/toddlers (n = 391), there were 134 cases and 257 controls. In both univariate and multivariable logistic regression analysis, risk factors associated with persisting measles susceptibility after the 1999/2000 measles epidemic were: male gender (OR in multivariable analysis 3.75; 95%CI 2.02–6.98; P < 0.001), absence of older siblings (OR 3.36; 95%CI 1.90–5.95; P < 0.001), membership of a moderately conservative OPD (OR 4.22; 95%CI 2.30–7.74; P < 0.001) and no residency in a LVC-municipality during the 1999/2000 measles epidemic (OR 1.91; 95%CI 1.11–3.27; P = 0.019) (table 2). Table 2 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant infants/toddlers during the 1999/2000 measles epidemic (n = 391)   Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001    Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001  a LVC-municipality: municipality with measles vaccination coverage <90%. Table 2 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant infants/toddlers during the 1999/2000 measles epidemic (n = 391)   Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001    Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001  a LVC-municipality: municipality with measles vaccination coverage <90%. Group 2: school-aged children during the 1999/2000 epidemic Of the 376 school-aged children, 70 were cases and 306 were controls. In univariate analysis residency, school attendance and OPD-membership during the 1999/2000 measles epidemic were associated with persisting measles susceptibility (table 3). Table 3 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant school-aged children during the 1999/2000 measles epidemic (n = 376)   Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108    Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108  a VC-municipality: municipality with measles vaccination coverage <90%. b OPS: orthodox Protestant school. Table 3 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant school-aged children during the 1999/2000 measles epidemic (n = 376)   Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108    Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108  a VC-municipality: municipality with measles vaccination coverage <90%. b OPS: orthodox Protestant school. As interaction was expected, the determinants ‘residency in LVC-municipality’ and ‘school attendance’ were also combined into four categories. Compared to respondents who did live in an LVC-municipality and attended an OPS, an increased risk of persisting measles susceptibility after the 1999/2000 epidemic was found in those who did not live in an LVC-municipality (OR 5.11; 95%CI 1.84–14.19; P = 0.002), or did not attend an OPS (OR 8.61; 95%CI 2.47–29.98; P = 0.001), or both (OR 6.66; 95%CI 1.87–23.70; P = 0.003). In particular, not attending an OPS was associated with the highest risk for persisting measles susceptibility. The association between membership of a moderately conservative OPD and persisting measles susceptibility did not reach statistical significance in multivariable logistic regression analysis (OR 2.27; 95%CI 0.84–6.17; P = 0.108). Discussion We set out to identify characteristics of unvaccinated orthodox Protestants that were associated with persisting measles susceptibility after the 1999/2000 measles epidemic—i.e. not being infected with measles during this epidemic. Since school attendance is known to be strongly associated with childhood infections, risk factors for persisting measles susceptibility were examined in two age-groups: infants/toddlers and school-aged children. Unvaccinated infants and toddlers who belonged to a moderately conservative church, did not have older siblings and did not live in a LVC-municipality during the 1999/2000 measles epidemic were more likely to remain susceptible for measles. Risk factors for persisting measles susceptibility in unvaccinated school-aged children were not attending an OPS and living outside a LVC-municipality. Although male gender was found to be statistically significantly associated with persisting measles susceptibility, this is not interpreted as a relevant risk factor, since both sexes are equally affected during measles epidemics.7,9 The overrepresentation of females in the study is probably due to a greater interest of women in the subject of health and vaccination, which has been found earlier in comparable studies.23,24 Unvaccinated orthodox Protestants with the identified risk factors may be -unknowingly- susceptible during a new epidemic in their adolescence or adulthood. It is to be expected that during the 2013/2014 measles epidemic orthodox Protestants—with similar risk factors—were not infected with measles and will still be susceptible for measles during a future measles epidemic. Woudenberg et al. state that an increasing vaccination coverage within the orthodox Protestant minority may be reflected in a longer inter-epidemic period resulting in cases of older age during the subsequent measles epidemic.7 This emphasizes the importance for these orthodox Protestants to be aware of their susceptibility and of the increased risk of complications when getting infected in adulthood, in order to reconsider vaccination. A study by Spaan et al. shows that vaccination coverage among orthodox Protestants has increased over the generations, especially in moderately conservative denominations.24 Also, positive vaccination intention for their children was higher among these respondents. It is possible that these young parents, who do want to vaccinate their children, are still susceptible for vaccine-preventable diseases, such as measles, because they were not infected in their childhood. There are several opportunities for health care professionals to discuss the persisting susceptibility to childhood diseases and the increased risk of complications in adulthood and to offer catch-up vaccination, for instance, when vaccinations are offered to protect against work-related or travel-related diseases. There may also be an opportunity to discuss possible susceptibility and catch-up vaccinations with parents, in consultations during which vaccinations for their children in the National Immunization Program are discussed and provided. As vaccination is a delicate subject for orthodox Protestants, the approach should be focussed on informed decision making, taking into account the risk of persisting susceptibility and, the risk of measles and its complications at adult age.23,25 Our study has several limitations. First, RPHSs could only select reported cases of the 2013/2014 epidemic for recruitment of measles cases. Even though a large underreporting of clinical measles cases during the epidemic was suspected, the participating cases are expected to be sufficiently representative. Given that the main study determinants—e.g. residency in LVC-municipality and school attendance as child—do not influence visiting a GP when having measles as adolescent or adult, it is not expected that underreporting of cases by GPs has affected the outcome of this study. Second, it was not possible to select a random sample of unvaccinated orthodox Protestant controls, since in the Netherlands religion is not recorded in public registrations. Therefore, data collection methods were used which have been found successful in this population in the past.10,24,25 For the recruitment of controls, school boards of orthodox Protestant secondary schools were involved, which resulted in a high response rate. However, we had to exclude a high number of questionnaires for data-analysis since the school boards invited every student of the birth cohorts 1988–98, including vaccinated students and students who were infected with measles in 2013/2014 (figure 1). We presume that this does not influence the study results. Third, controls self-reported their measles virus infection during childhood and based on their background as unvaccinated orthodox Protestants it is highly probable that they did have measles as child. It is, however, possible that we included controls who did not have measles as a child, which may have diluted the differences between cases and controls. Nevertheless, this misclassification would have resulted in an underestimation rather than overestimation of associations. Furthermore, involving orthodox Protestants secondary schools in the recruitment of controls could have resulted in selection bias concerning the risk factors ‘attendance to an orthodox Protestant primary school’ and ‘residency in LVC-municipalities’. However, orthodox Protestant secondary schools are centrally located and serve both students from LVC- and non-LVC-municipalities and these secondary schools are not connected to orthodox Protestant primary schools. Moreover, most respondents attending these secondary schools were too young to go to primary school during the 1999/2000 epidemic. Conclusion In this study we identified risk factors of persisting measles susceptibility among unvaccinated adolescents and young adults within an orthodox Protestants community following a measles epidemic in their childhood. These risk factors included not attending an OPS and not living in a LVC-municipality during the childhood measles epidemic. For those who were too young to attend school, not having older siblings and belonging to a moderately conservative church denomination were additional risk factors. Since this specific group of moderately conservative orthodox Protestants may also be more open to discuss the subject of vaccination, they should be informed about these risks and offered catch-up vaccination. Acknowledgements We thank the orthodox Protestant secondary schools (Van Lodenstein College, Driestar College and Driestar Educatief) for their assistance with the recruitment of controls among their students. Furthermore, we thank all participants for completing the study’s questionnaire. We would also like to thank Toos Waegemaekers for her contributions to the research proposal. Funding This work was financially supported by the Program for Strengthening Control of Infectious Diseases by Municipal Health Services of the National Institute for Public Health and the Environment (RIVM), the Netherlands. Conflicts of interest: None declared. Key points In the Netherlands measles epidemics still occur periodically among unvaccinated orthodox Protestant children, adolescents and adults. After the 1999/2000 epidemic, some of the unvaccinated children were persistently susceptible to measles and they got measles at an older age with increased risk of complications. Our study shows that unvaccinated orthodox Protestants who belonged to a moderately conservative church denomination, did not attend an OPS, did not reside in a LVC municipality or did not have older siblings, had a higher risk of persisting measles susceptibility after an epidemic. These specific subgroups should be informed about the risks of their possible susceptibility and measles virus infection at older age, and supported in the decision-making process of catching up childhood vaccination. References 1 Perry RT, Halsey NA. The clinical significance of measles: a review. J Infect Dis  2004; 189: S4– 16. Google Scholar CrossRef Search ADS PubMed  2 Griffin DE, Lin WH, Pan CH. Measles virus, immune control, and persistence. FEMS Microbiol Rev  2012; 36: 649– 62. Google Scholar CrossRef Search ADS PubMed  3 Muscat M, Shefer A, Ben Mamou M, et al.   The state of measles and rubella in the WHO European Region, 2013. Clin Microbiol Infect  2014; 20: 12– 18. Google Scholar CrossRef Search ADS PubMed  4 ECDC. Surveillance Report: Measles and rubella monitoring February 2014. European Centre for Disease Prevention and Control (ECDC), 2012. Contract No.: August 1, 2017. 5 Mayoral CJ, Perez Morilla E, Gallardo Garcia V, et al.   Measles outbreak in Andalusia, Spain, January to August 2011. Euro Surveill  2012; 17: 20300. Google Scholar PubMed  6 Sabella C. Measles: not just a childhood rash. Cleve Clin J Med  2010; 77: 207– 13. Google Scholar CrossRef Search ADS PubMed  7 Woudenberg T, van Binnendijk RS, Sanders EA, et al.   Large measles epidemic in the Netherlands, May 2013 to March 2014: changing epidemiology. Euro Surveill  2017; 22: 30443. Google Scholar CrossRef Search ADS PubMed  8 van Lier EA, Oomen PJ, Conyn-van Spaendonck MAE, et al.   Immunisation coverage National Immunisation Programme in The Netherlands: year of report 2015 . National Institute for Public Health and the Environment, Bilthoven, 2015. Contract No.: 2015-0067. 9 van den Hof S, Conyn-van Spaendonck MA, van Steenbergen JE. Measles epidemic in the Netherlands, 1999-2000. J Infect Dis  2002; 186: 1483– 6. Google Scholar CrossRef Search ADS PubMed  10 Ruijs WL, Hautvast JL, van Ansem WJ, et al.   Measuring vaccination coverage in a hard to reach minority. Eur J Public Health  2012; 22: 359– 64. Google Scholar CrossRef Search ADS PubMed  11 Knippenberg H. De Religieuze Kaart Van Nederland: Omvang En Geografische Spreiding Van De Godsdienstige Gezindten Van De Reformatie Tot Heden [The Religious Map of The Netherlands: Size and Geographical Dispersal of Religious Denominations from the Reformation to the Present] . Assen: Van Gorcum, 1992. 12 Ruijs WL, Hautvast JL, van der Velden K, et al.   Religious subgroups influencing vaccination coverage in the Dutch Bible belt: an ecological study. BMC Public Health  2011; 11: 102. Google Scholar CrossRef Search ADS PubMed  13 LCI. Overzicht gemeenten met lage vaccinatiegraad BMR [Overview of Municipalities with Low Vaccination Coverage MMR]. Bilthoven: Landelijk coördinatiecentrum Infectieziekten (LCI), 2013. Available at: http://www.rivm.nl/dsresource? objectid=cb19275c-90db-42cc-a60d-e60399fbdf64&type=org&disposition=inline. 14 CBS. Geboorte: kerncijfers [Childbirth: key figures] [Internet]. Centraal Bureau voor de Statistiek, 2016 [cited April 1 2016]. Available at: http://www.cbs.nl. 15 Wagemakers JJMF, Karst WA, van Heukelum A, van den Kerkhof JCTM. Vaccinatiegedrag in gezinnen met kinderen op reformatorische scholen in de regio Zuid-Holland Zuid [Behaviour regarding vaccinations in orthodox Protestant families in the Southern part of the province of Zuid-Holland]. Infectieziektenbulletin  2010; 21: 198– 204. 16 Onderwijsdata: open databestanden [Education data: open databases]. [Internet]. Dienst Uitvoering Onderwijs (DUO), 2017[cited August 1, 2017]. 17 DUO. Vestigingen reformatorisch onderwijs 2012 [Locations of orthodox Protestant education 2012]. Dienst Uitvoering Onderwijs (DUO), Groningen, 2012. Available at: https://www.volksgezondheidenzorg.info/kaart/vestigingen-reformatorisch-onderwijs-2012. 18 van Isterdael CE, van Essen GA, Kuyvenhoven MM, et al.   Measles incidence estimations based on the notification by general practitioners were suboptimal. J Clin Epidemiol  2004; 57: 633– 7. Google Scholar CrossRef Search ADS PubMed  19 LCI. LCI-Richtlijn Mazelen (Morbilli) [LCI Guideline for Measles (Morbilli)]. LCI-Richtlijnen Infectieziektebestrijding . Bilthoven: National Institute for Public Health and the Environment, 2017. 20 Hoekstra EG, Ipenburg MH. Handboek Christelijk Nederland [Handbook Christian Denominations in The Netherlands] . Kampen: Kok, 2008. 21 Mossong J, Hens N, Jit M, et al.   Social contacts and mixing patterns relevant to the spread of infectious diseases. PLoS Med  2008; 5: e74. Google Scholar CrossRef Search ADS PubMed  22 Ruijs WL, Hautvast JL, Akkermans RP, et al.   The role of schools in the spread of mumps among unvaccinated children: a retrospective cohort study. BMC Infect Dis  2011; 11: 227. Google Scholar CrossRef Search ADS PubMed  23 Ruijs WL, Hautvast JL, van 't Spijker K, et al.   Information on vaccination: meeting the needs of unvaccinated youngsters in the Netherlands. Eur J Public Health  2011; 21: 344– 6. Google Scholar CrossRef Search ADS PubMed  24 Spaan DH, Ruijs WLM, Hautvast JLA, Tostmann A. Increase in vaccination coverage between subsequent generations of orthodox Protestants in The Netherlands. Eur J Public Health  2017; 27: 582. Google Scholar CrossRef Search ADS PubMed  25 Ruijs WL, Hautvast JL, van Ijzendoorn G, et al.   How orthodox protestant parents decide on the vaccination of their children: a qualitative study. BMC Public Health  2012; 12: 408. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

Risk factors for persisting measles susceptibility: a case-control study among unvaccinated orthodox Protestants

Loading next page...
 
/lp/ou_press/risk-factors-for-persisting-measles-susceptibility-a-case-control-c80hw8Ae1n
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/cky072
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Measles is an infectious disease providing lifelong immunity. Epidemics periodically occur among unvaccinated orthodox Protestants in the Netherlands. During the 2013/2014 epidemic, 17% of the reported patients was over 14 years old. Apparently, they did not catch measles during the previous 1999/2000 epidemic and remained susceptible. We wanted to identify risk factors for this so-called persisting measles susceptibility, and thus risk factors for acquiring measles at older age with increased risk of complications. Methods A case-control study was performed among unvaccinated orthodox Protestants born between 1988 and 1998; cases had measles in 2013/2014, controls during or before 1999/2000. Associations between demographic, geographical and religion-related determinants and persisting measles susceptibility were determined using univariate and multivariable logistic regression. Analyses were stratified in two age-groups: infants/toddlers and primary school-aged children during the 1999/2000 measles epidemic. Results In total, 204 cases and 563 controls were included. Risk factors for persisting measles susceptibility for infants/toddlers in 1999/2000 were belonging to a moderately conservative church, absence of older siblings and residency outside low vaccination coverage (LVC)-municipalities. Risk factors for primary school-aged children were residency outside LVC-municipalities and attendance of non-orthodox Protestant primary school. Conclusion Unvaccinated orthodox Protestant adolescents and adults who resided outside the LVC-municipalities, did not attend an orthodox Protestant primary school, had no older siblings and belonged to a moderately conservative church were at risk for persisting measles susceptibility and, thus, for acquiring measles at older age with increased risk of complications. For this subgroup of orthodox Protestants targeted information on vaccination is recommended. Introduction Measles is a highly infectious disease that causes high fever, rash, cough and conjunctivitis.1 Measles provides lifelong immunity; after recovery the patient is not susceptible to measles anymore.2 In Western countries, measles is no longer considered a childhood disease as it affects adolescents and adults as well.1,3 In several European countries a shift towards a higher median age of unvaccinated cases is reported during measles epidemics.4–6 Measles at higher age is associated with an increased risk of complications such as pneumonia, diarrhoea, acute encephalitis, hepatitis and post-infectious encephalomyelitis, resulting in relatively high hospitalization rates and deaths.1,7 Measles can be prevented by vaccination. In the Netherlands measles vaccination has been offered free of charge to all children since 1976. Catch-up vaccination is possible until 18 years and actively offered during epidemics. Before introduction of vaccination, large measles epidemics occurred every other year. Despite a national vaccination coverage of 96% (at the age of two), measles epidemics still occur in the Netherlands.8 These epidemics are largely confined to an orthodox Protestant minority of 250 000 people with religious objections to vaccination.7,9 They are members of various small orthodox Protestant church denominations (OPD), each with varied interpretations of the Bible with respect to vaccination. Vaccination coverage varies between the different OPDs from <15 to >85%.10 Overall, vaccination coverage among orthodox Protestants is ∼60%.10 Historically, Dutch orthodox Protestants live in rural areas stretching from the south-west to the north-east of the Netherlands, commonly referred to as the Bible Belt.11 Nowadays, about 75% of the orthodox Protestants live geographically clustered in this area.11,12 In 2013, 29 municipalities in this area had a vaccination coverage <90% [low vaccination coverage (LVC)-municipalities)].13 Apart from geographical clustering, strong social clustering is common among orthodox Protestants. Orthodox Protestant families are characterized as large, close-knit families with an average of four children per family, the national average being 1.7.14,15 Furthermore, the orthodox Protestant minority has its own political party (SGP), schools, newspaper, magazines, social media platforms and websites. Almost half of the 160 orthodox Protestant primary schools are located in (rural) LVC-municipalities, the others are located in larger towns and cities.16,17 The seven orthodox Protestant secondary schools and two colleges are all centrally located to serve orthodox Protestants from a large area.17 Due to this geographical and social clustering of unvaccinated children, the measles virus is easily transmitted within this minority. During the 2013–14 measles epidemic 2700 measles cases were notified compared to 3292 cases during the 1999/2000 epidemic.7,9 However, underreporting of measles cases is common during outbreaks and epidemics, either because patients do not consult a physician or because physicians do not report all cases to public health authorities.18 Compared to the 1999/2000 epidemic, the 2013/2014 epidemic showed a considerable higher median age of infection and higher incidence in older age groups.7 As infection with measles provides lifelong immunity, measles patients born before 1999 must have been susceptible for measles during the 1999/2000 epidemic as well. Apparently, they escaped infection during the 1999/2000 epidemic and were thus persistently susceptible to measles, until they were infected in 2013/2014. We aimed to identify characteristics of unvaccinated orthodox Protestants associated with persistent measles susceptibility after the 1999/2000 epidemic, and, thus, with increased risk of complications when acquiring measles at older age. Based on these characteristics, targeted information can be developed for unvaccinated orthodox Protestant adolescents and adults who are at increased risk of measles and its complications, in order to make them aware of their susceptibility and consider vaccination. Moreover, knowledge of the characteristics of those with persistent susceptibility can be used for estimating the potential burden of disease, and health care use in future measles epidemics. Methods Study design and participants We performed a retrospective case-control study among unvaccinated orthodox Protestants born between 1988 and 1998. These individuals were 14–26 years of age during the measles epidemic of 2013/2014 and born before the 1999/2000 measles epidemic. Cases were individuals who were notified with measles during the 2013/2014 epidemic. In the Netherlands measles is a mandatory notifiable disease; laboratory confirmed and epidemiologically linked cases have to be reported to the Regional Public Health Service (RPHS) by physicians and laboratories.19 All RPHSs that covered one or more LVC-municipalities (12 of the 25 Dutch RPHS regions) participated in this study. Controls were individuals with self-reported measles (including symptoms of fever, rash, red and watery eyes, rhinitis and/or cough) during or before the 1999/2000 epidemic. Data collection Data collection was performed in 2015 Cases received a personal invitation from their RPHS to participate in our study and to fill out an online questionnaire. A reminder was sent two weeks later. Controls were approached via orthodox Protestant (social) media and orthodox Protestant secondary schools and invited to fill out an online questionnaire. In addition, for controls snowball sampling was used as respondents were requested to invite friends and relatives, and contacts from student societies. Personal invitation of controls was not feasible as registrations containing personal data of the 1999/2000 epidemic were not available anymore at the RPHSs. The questionnaire contained questions on year of birth, gender, number of older and younger siblings, postal code, OPD, (orthodox Protestant) school attendance, time and symptoms of measles and measles, mumps and rubella (MMR)-vaccination status (at least one MMR/no MMR/unknown). Participation was anonymous. Data-analysis Based on postal codes, respondents were classified as living in or outside a LVC-municipality. A LVC-municipality was defined as a municipality with a vaccination coverage <90% in 2013 and in which >5% of the population voted for the orthodox Protestant political party (SGP). In 2013, 29 out of 408 municipalities in the Netherlands were LVC-municipalities.13 The OPDs were classified into two categories based on level of conservatism and vaccination coverage.10,20 OPDs with a high level of conservatism (Old Reformed Congregations and Reformed Congregations in the Netherlands) have a vaccination coverage of <25%. OPDs with a moderate level of conservatism have respective vaccination coverages of 50–75% (Reformed Congregations and Restored Reformed Church), and >85% in Christian Reformed Churches and Protestant Church in the Netherlands, including the Reformed Bond.10 For respondents who were school-aged in 1999/2000, the school attendance variable was dichotomized in whether or not they attended an orthodox Protestant school (OPS vs. non-OPS). Statistical analysis The main outcome variable was ‘persisting measles susceptibility’, i.e. not being infected with measles during the 1999/2000 epidemic. Descriptive analyses were conducted using mean and percentages in categorical variables and mean and interquartile range in continuous variables. As school attendance is known to be strongly associated with childhood infections, analyses were stratified into two age-groups.21,22 Group 1 consisted of infants and toddlers born in 1996–98 who were too young to attend school; group 2 consisted of school-aged children born in 1988–95. Univariate and multivariable logistic regression (OR, 95%CI) was used to determine the association between the main outcome variable ‘persisting measles susceptibility’ and the determinants: gender, presence of older siblings, residency in LVC-municipalities, OPS attendance and OPD-membership. Determinants that were statistically significantly associated with the outcome variable (P < 0.05) in the univariate analysis were included in the multivariable analysis. Since almost half of the primary OPSs are located in LVC-municipalities, the determinants ‘residing in LVC-municipalities’ and ‘type of school attendance’ were combined for the group of school-aged children, resulting in four categories: (i) residing in a LVC-municipality and attending an OPS, (ii) residing outside a LVC-municipality and attending an OPS, (iii) resident in a LVC-municipality and not attending an OPS and (iv) residing outside a LVC-municipality and not attending an OPS. All statistical analysis were conducted using SPSS version 21. Ethics The study was approved by the research ethics committee of the Radboud University Nijmegen Medical Centre, Nijmegen, the Netherlands; CMO number 2014/1519. Results The total study population consisted of 767 persons; 204 cases and 563 controls. Of the 304 reported measles cases who were invited to participate in the study, 240 responded to the questionnaire (response rate 79%) and 16 questionnaires were incomplete on essential questions—residency and school attendance—and 20 respondents did not meet the inclusion criteria regarding measles, year of birth, vaccination status or church denomination. Accordingly, 204 cases were included (figure 1). In total 1738 controls responded to the online questionnaire, of which 396 questionnaires were incomplete on essential questions, namely residency and school attendance. The inclusion criteria for controls were not met in 779 questionnaires, most frequently because the respondents did not report having contracted measles as a child or reported to have had measles during the 2013/2014 epidemic instead of the 1999/2000 epidemic. Finally, 563 controls were included for analysis (figure 1). Figure 1 View largeDownload slide Recruitment and response flowchart of cases and controls (n =2042) Figure 1 View largeDownload slide Recruitment and response flowchart of cases and controls (n =2042) Characteristics of cases and controls are shown in table 1. Overall, 74% of the respondents included in the study were female. More than half of the controls were school-aged in 1999/2000 (53.1%) compared to one third of the cases (32.3%). Some of the completed questionnaires which were included for data-analysis lacked postal code data due to a technical problem in the online questionnaire. Postal code data of the cases who were included for data-analysis was therefore compared to the postal code data of all reported measles cases who were initially invited by RPHSs. There was no statistically significant difference in residency in a LVC-municipality between invited cases and included cases (resp. 48.2% and 49.3%, P = 0.83). Table 1 Characteristics of unvaccinated orthodox Protestant cases and controls (n = 767)   Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3    Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3  a IQR: interquartile range. b LVC-municipality: municipality with measles vaccination coverage <90%. c OPS: orthodox Protestant school. Table 1 Characteristics of unvaccinated orthodox Protestant cases and controls (n = 767)   Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3    Total  Cases (n = 204)   Controls (n = 563)   Year of birth (IQR)a  767  204  1996 (1994–1997)  563  1995 (1991–1997)  Not attending school in 1999/2000 (IQR)  391  134  1997 (1996–1998)  257  1997 (1997–1998)  Attending school in 1999/2000 (IQR)  376  70  1993 (1991–1994)  306  1992 (1990–1994)  Gender, n | %  N  n  %  n  %  Female  564  122  60.4  442  78.6  Male  200  80  39.6  120  21.4  Current family size, n | %  N  n  %  n  %  0–2 siblings  108  29  14.3  79  14.0  3–5 siblings  355  108  53.2  247  43.9  6–8 siblings  210  46  22.7  164  29.1  >8 siblings  93  20  9.9  73  13.0  Older siblings, n | %  N  n  %  n  %  Yes  518  122  59.8  396  70.3  No  249  82  40.2  167  29.7  Residency in LVC-municipalityb (2013/2014)  N  n  %  n  %  Yes  409  94  47.2  315  57.0  No  343  105  52.8  238  43.0  Residency in LVC-municipalityb (1999/2000)  N  n  %  n  %  Yes  387  60  43.8  327  59.8  No  297  77  56.2  220  40.2  Attendance to OPSc (1999/2000), n | %  N  n  %  n  %  Yes  285  39  60.9  246  84.5  No  70  25  39.1  45  15.5  Too young  391  134    257    Orthodox Protestant church denomination  N  n  %  n  %  High level of conservatism  249  39  20.0  210  38.5  Old Reformed Congregations  90  17  8.7  73  13.4  Reformed Congregations in the Netherlands  159  22  11.3  137  25.1  Moderate level of conservatism  492  156  80.0  336  61.5  Reformed Congregations  326  101  51.8  225  41.2  Restored Reformed Church  95  37  19.0  58  10.6  Christian Reformed Churches  33  10  5.1  23  4.2  Protestant Church in the Netherlands, including Reformed Bond  38  8  3.9  30  5.3  a IQR: interquartile range. b LVC-municipality: municipality with measles vaccination coverage <90%. c OPS: orthodox Protestant school. Group 1: infants and toddlers during the 1999/2000 epidemic In the group of infants/toddlers (n = 391), there were 134 cases and 257 controls. In both univariate and multivariable logistic regression analysis, risk factors associated with persisting measles susceptibility after the 1999/2000 measles epidemic were: male gender (OR in multivariable analysis 3.75; 95%CI 2.02–6.98; P < 0.001), absence of older siblings (OR 3.36; 95%CI 1.90–5.95; P < 0.001), membership of a moderately conservative OPD (OR 4.22; 95%CI 2.30–7.74; P < 0.001) and no residency in a LVC-municipality during the 1999/2000 measles epidemic (OR 1.91; 95%CI 1.11–3.27; P = 0.019) (table 2). Table 2 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant infants/toddlers during the 1999/2000 measles epidemic (n = 391)   Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001    Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001  a LVC-municipality: municipality with measles vaccination coverage <90%. Table 2 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant infants/toddlers during the 1999/2000 measles epidemic (n = 391)   Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001    Total (n = 391)  Cases (n = 134)   Controls (n = 257)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  297  79  59.0  218  84.8  reference    reference    Male  94  55  41.0  39  15.2  3.89 (2.40–6.32)  P<0.001  3.75 (2.0–6.98)  P<0.001  Older siblings  N  n  %  n  %          Yes  274  70  52.2  204  79.4  ref    ref    No  117  64  47.8  53  20.6  3.52 (2.24–5.54)  P<0.001  3.36 (1.90–5.95)  P<0.001  Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  204  47  49.5  157  63.3  ref    ref    No  139  48  50.5  91  36.7  1.76 (1.09–2.84)  P = 0.020  1.91 (1.11–3.27)  P = 0.019  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  150  29  22.3  121  48.6  ref    ref    Moderate level of conservatism  229  101  77.7  128  51.4  3.29 (2.03–5.33)  P <0.001  4.22 (2.30–7.74)  P <0.001  a LVC-municipality: municipality with measles vaccination coverage <90%. Group 2: school-aged children during the 1999/2000 epidemic Of the 376 school-aged children, 70 were cases and 306 were controls. In univariate analysis residency, school attendance and OPD-membership during the 1999/2000 measles epidemic were associated with persisting measles susceptibility (table 3). Table 3 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant school-aged children during the 1999/2000 measles epidemic (n = 376)   Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108    Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108  a VC-municipality: municipality with measles vaccination coverage <90%. b OPS: orthodox Protestant school. Table 3 Risk factors for persisting measles susceptibility in unvaccinated orthodox Protestant school-aged children during the 1999/2000 measles epidemic (n = 376)   Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108    Total (n = 376)  Cases (n = 70)   Controls (n = 306)   Univariate OR (95%CIs)  P  Multivariable ORs (95%CIs)  P  Gender  N  n  %  n  %          Female  267  43  63.2  224  73.4  reference        Male  106  25  36.8  81  26.6  1.61 (0.92–2.80)  P = 0.091      Older siblings  N  n  %  n  %          Yes  244  52  74.3  192  62.7  ref        No  132  18  25.7  114  37.7  0.58 (0.33–1.05)  P = 0.068      Residency in LVC-municipalitya (1999/2000)  N  n  %  n  %          Yes  183  13  31.0  170  56.9  ref        No  158  29  69.0  129  43.1  2.94 (1.47–5.88)  P = 0.002      Attendance to OPSb (1999/2000)  N  n  %  n  %          Yes  285  39  60.9  246  84.5  ref        No  70  25  39.1  45  15.5  3.50 (1.93–6.35)  P < 0.001      Residency in LVC-municipality attendance to OPSa,b  N  n  %  n  %          Residing in LVC-municipality, attending OPS  165  5  12.5  140  48.8  ref    ref    Not residing in LVC-municipality, attending OPS  103  21  52.5  101  35.2  5.82 (2.12–15.96)  P = 0.001  5.11 (1.84–14.19)  P = 0.002  Residing in LVC-municipality, not attending OPS  27  7  17.5  21  7.3  9.33 (2.71–32.12)  P < 0.001  8.61 (2.47–29.98)  P = 0.001  Not residing in LVC-municipality, not attending OPS  32  7  17.5  25  8.7  7.84 (2.30–26.66)  P = 0.001  6.66 (1.87–23.70)  P = 0.003  Orthodox Protestant church denomination  N  n  %  n  %          High level of conservatism  99  10  15.4  89  30.0  ref    ref    Moderate level of conservatism  263  55  84.6  208  70.0  2.35 (1.15–4.83)  P = 0.017  2.27 (0.84–6.17)  P = 0.108  a VC-municipality: municipality with measles vaccination coverage <90%. b OPS: orthodox Protestant school. As interaction was expected, the determinants ‘residency in LVC-municipality’ and ‘school attendance’ were also combined into four categories. Compared to respondents who did live in an LVC-municipality and attended an OPS, an increased risk of persisting measles susceptibility after the 1999/2000 epidemic was found in those who did not live in an LVC-municipality (OR 5.11; 95%CI 1.84–14.19; P = 0.002), or did not attend an OPS (OR 8.61; 95%CI 2.47–29.98; P = 0.001), or both (OR 6.66; 95%CI 1.87–23.70; P = 0.003). In particular, not attending an OPS was associated with the highest risk for persisting measles susceptibility. The association between membership of a moderately conservative OPD and persisting measles susceptibility did not reach statistical significance in multivariable logistic regression analysis (OR 2.27; 95%CI 0.84–6.17; P = 0.108). Discussion We set out to identify characteristics of unvaccinated orthodox Protestants that were associated with persisting measles susceptibility after the 1999/2000 measles epidemic—i.e. not being infected with measles during this epidemic. Since school attendance is known to be strongly associated with childhood infections, risk factors for persisting measles susceptibility were examined in two age-groups: infants/toddlers and school-aged children. Unvaccinated infants and toddlers who belonged to a moderately conservative church, did not have older siblings and did not live in a LVC-municipality during the 1999/2000 measles epidemic were more likely to remain susceptible for measles. Risk factors for persisting measles susceptibility in unvaccinated school-aged children were not attending an OPS and living outside a LVC-municipality. Although male gender was found to be statistically significantly associated with persisting measles susceptibility, this is not interpreted as a relevant risk factor, since both sexes are equally affected during measles epidemics.7,9 The overrepresentation of females in the study is probably due to a greater interest of women in the subject of health and vaccination, which has been found earlier in comparable studies.23,24 Unvaccinated orthodox Protestants with the identified risk factors may be -unknowingly- susceptible during a new epidemic in their adolescence or adulthood. It is to be expected that during the 2013/2014 measles epidemic orthodox Protestants—with similar risk factors—were not infected with measles and will still be susceptible for measles during a future measles epidemic. Woudenberg et al. state that an increasing vaccination coverage within the orthodox Protestant minority may be reflected in a longer inter-epidemic period resulting in cases of older age during the subsequent measles epidemic.7 This emphasizes the importance for these orthodox Protestants to be aware of their susceptibility and of the increased risk of complications when getting infected in adulthood, in order to reconsider vaccination. A study by Spaan et al. shows that vaccination coverage among orthodox Protestants has increased over the generations, especially in moderately conservative denominations.24 Also, positive vaccination intention for their children was higher among these respondents. It is possible that these young parents, who do want to vaccinate their children, are still susceptible for vaccine-preventable diseases, such as measles, because they were not infected in their childhood. There are several opportunities for health care professionals to discuss the persisting susceptibility to childhood diseases and the increased risk of complications in adulthood and to offer catch-up vaccination, for instance, when vaccinations are offered to protect against work-related or travel-related diseases. There may also be an opportunity to discuss possible susceptibility and catch-up vaccinations with parents, in consultations during which vaccinations for their children in the National Immunization Program are discussed and provided. As vaccination is a delicate subject for orthodox Protestants, the approach should be focussed on informed decision making, taking into account the risk of persisting susceptibility and, the risk of measles and its complications at adult age.23,25 Our study has several limitations. First, RPHSs could only select reported cases of the 2013/2014 epidemic for recruitment of measles cases. Even though a large underreporting of clinical measles cases during the epidemic was suspected, the participating cases are expected to be sufficiently representative. Given that the main study determinants—e.g. residency in LVC-municipality and school attendance as child—do not influence visiting a GP when having measles as adolescent or adult, it is not expected that underreporting of cases by GPs has affected the outcome of this study. Second, it was not possible to select a random sample of unvaccinated orthodox Protestant controls, since in the Netherlands religion is not recorded in public registrations. Therefore, data collection methods were used which have been found successful in this population in the past.10,24,25 For the recruitment of controls, school boards of orthodox Protestant secondary schools were involved, which resulted in a high response rate. However, we had to exclude a high number of questionnaires for data-analysis since the school boards invited every student of the birth cohorts 1988–98, including vaccinated students and students who were infected with measles in 2013/2014 (figure 1). We presume that this does not influence the study results. Third, controls self-reported their measles virus infection during childhood and based on their background as unvaccinated orthodox Protestants it is highly probable that they did have measles as child. It is, however, possible that we included controls who did not have measles as a child, which may have diluted the differences between cases and controls. Nevertheless, this misclassification would have resulted in an underestimation rather than overestimation of associations. Furthermore, involving orthodox Protestants secondary schools in the recruitment of controls could have resulted in selection bias concerning the risk factors ‘attendance to an orthodox Protestant primary school’ and ‘residency in LVC-municipalities’. However, orthodox Protestant secondary schools are centrally located and serve both students from LVC- and non-LVC-municipalities and these secondary schools are not connected to orthodox Protestant primary schools. Moreover, most respondents attending these secondary schools were too young to go to primary school during the 1999/2000 epidemic. Conclusion In this study we identified risk factors of persisting measles susceptibility among unvaccinated adolescents and young adults within an orthodox Protestants community following a measles epidemic in their childhood. These risk factors included not attending an OPS and not living in a LVC-municipality during the childhood measles epidemic. For those who were too young to attend school, not having older siblings and belonging to a moderately conservative church denomination were additional risk factors. Since this specific group of moderately conservative orthodox Protestants may also be more open to discuss the subject of vaccination, they should be informed about these risks and offered catch-up vaccination. Acknowledgements We thank the orthodox Protestant secondary schools (Van Lodenstein College, Driestar College and Driestar Educatief) for their assistance with the recruitment of controls among their students. Furthermore, we thank all participants for completing the study’s questionnaire. We would also like to thank Toos Waegemaekers for her contributions to the research proposal. Funding This work was financially supported by the Program for Strengthening Control of Infectious Diseases by Municipal Health Services of the National Institute for Public Health and the Environment (RIVM), the Netherlands. Conflicts of interest: None declared. Key points In the Netherlands measles epidemics still occur periodically among unvaccinated orthodox Protestant children, adolescents and adults. After the 1999/2000 epidemic, some of the unvaccinated children were persistently susceptible to measles and they got measles at an older age with increased risk of complications. Our study shows that unvaccinated orthodox Protestants who belonged to a moderately conservative church denomination, did not attend an OPS, did not reside in a LVC municipality or did not have older siblings, had a higher risk of persisting measles susceptibility after an epidemic. These specific subgroups should be informed about the risks of their possible susceptibility and measles virus infection at older age, and supported in the decision-making process of catching up childhood vaccination. References 1 Perry RT, Halsey NA. The clinical significance of measles: a review. J Infect Dis  2004; 189: S4– 16. Google Scholar CrossRef Search ADS PubMed  2 Griffin DE, Lin WH, Pan CH. Measles virus, immune control, and persistence. FEMS Microbiol Rev  2012; 36: 649– 62. Google Scholar CrossRef Search ADS PubMed  3 Muscat M, Shefer A, Ben Mamou M, et al.   The state of measles and rubella in the WHO European Region, 2013. Clin Microbiol Infect  2014; 20: 12– 18. Google Scholar CrossRef Search ADS PubMed  4 ECDC. Surveillance Report: Measles and rubella monitoring February 2014. European Centre for Disease Prevention and Control (ECDC), 2012. Contract No.: August 1, 2017. 5 Mayoral CJ, Perez Morilla E, Gallardo Garcia V, et al.   Measles outbreak in Andalusia, Spain, January to August 2011. Euro Surveill  2012; 17: 20300. Google Scholar PubMed  6 Sabella C. Measles: not just a childhood rash. Cleve Clin J Med  2010; 77: 207– 13. Google Scholar CrossRef Search ADS PubMed  7 Woudenberg T, van Binnendijk RS, Sanders EA, et al.   Large measles epidemic in the Netherlands, May 2013 to March 2014: changing epidemiology. Euro Surveill  2017; 22: 30443. Google Scholar CrossRef Search ADS PubMed  8 van Lier EA, Oomen PJ, Conyn-van Spaendonck MAE, et al.   Immunisation coverage National Immunisation Programme in The Netherlands: year of report 2015 . National Institute for Public Health and the Environment, Bilthoven, 2015. Contract No.: 2015-0067. 9 van den Hof S, Conyn-van Spaendonck MA, van Steenbergen JE. Measles epidemic in the Netherlands, 1999-2000. J Infect Dis  2002; 186: 1483– 6. Google Scholar CrossRef Search ADS PubMed  10 Ruijs WL, Hautvast JL, van Ansem WJ, et al.   Measuring vaccination coverage in a hard to reach minority. Eur J Public Health  2012; 22: 359– 64. Google Scholar CrossRef Search ADS PubMed  11 Knippenberg H. De Religieuze Kaart Van Nederland: Omvang En Geografische Spreiding Van De Godsdienstige Gezindten Van De Reformatie Tot Heden [The Religious Map of The Netherlands: Size and Geographical Dispersal of Religious Denominations from the Reformation to the Present] . Assen: Van Gorcum, 1992. 12 Ruijs WL, Hautvast JL, van der Velden K, et al.   Religious subgroups influencing vaccination coverage in the Dutch Bible belt: an ecological study. BMC Public Health  2011; 11: 102. Google Scholar CrossRef Search ADS PubMed  13 LCI. Overzicht gemeenten met lage vaccinatiegraad BMR [Overview of Municipalities with Low Vaccination Coverage MMR]. Bilthoven: Landelijk coördinatiecentrum Infectieziekten (LCI), 2013. Available at: http://www.rivm.nl/dsresource? objectid=cb19275c-90db-42cc-a60d-e60399fbdf64&type=org&disposition=inline. 14 CBS. Geboorte: kerncijfers [Childbirth: key figures] [Internet]. Centraal Bureau voor de Statistiek, 2016 [cited April 1 2016]. Available at: http://www.cbs.nl. 15 Wagemakers JJMF, Karst WA, van Heukelum A, van den Kerkhof JCTM. Vaccinatiegedrag in gezinnen met kinderen op reformatorische scholen in de regio Zuid-Holland Zuid [Behaviour regarding vaccinations in orthodox Protestant families in the Southern part of the province of Zuid-Holland]. Infectieziektenbulletin  2010; 21: 198– 204. 16 Onderwijsdata: open databestanden [Education data: open databases]. [Internet]. Dienst Uitvoering Onderwijs (DUO), 2017[cited August 1, 2017]. 17 DUO. Vestigingen reformatorisch onderwijs 2012 [Locations of orthodox Protestant education 2012]. Dienst Uitvoering Onderwijs (DUO), Groningen, 2012. Available at: https://www.volksgezondheidenzorg.info/kaart/vestigingen-reformatorisch-onderwijs-2012. 18 van Isterdael CE, van Essen GA, Kuyvenhoven MM, et al.   Measles incidence estimations based on the notification by general practitioners were suboptimal. J Clin Epidemiol  2004; 57: 633– 7. Google Scholar CrossRef Search ADS PubMed  19 LCI. LCI-Richtlijn Mazelen (Morbilli) [LCI Guideline for Measles (Morbilli)]. LCI-Richtlijnen Infectieziektebestrijding . Bilthoven: National Institute for Public Health and the Environment, 2017. 20 Hoekstra EG, Ipenburg MH. Handboek Christelijk Nederland [Handbook Christian Denominations in The Netherlands] . Kampen: Kok, 2008. 21 Mossong J, Hens N, Jit M, et al.   Social contacts and mixing patterns relevant to the spread of infectious diseases. PLoS Med  2008; 5: e74. Google Scholar CrossRef Search ADS PubMed  22 Ruijs WL, Hautvast JL, Akkermans RP, et al.   The role of schools in the spread of mumps among unvaccinated children: a retrospective cohort study. BMC Infect Dis  2011; 11: 227. Google Scholar CrossRef Search ADS PubMed  23 Ruijs WL, Hautvast JL, van 't Spijker K, et al.   Information on vaccination: meeting the needs of unvaccinated youngsters in the Netherlands. Eur J Public Health  2011; 21: 344– 6. Google Scholar CrossRef Search ADS PubMed  24 Spaan DH, Ruijs WLM, Hautvast JLA, Tostmann A. Increase in vaccination coverage between subsequent generations of orthodox Protestants in The Netherlands. Eur J Public Health  2017; 27: 582. Google Scholar CrossRef Search ADS PubMed  25 Ruijs WL, Hautvast JL, van Ijzendoorn G, et al.   How orthodox protestant parents decide on the vaccination of their children: a qualitative study. BMC Public Health  2012; 12: 408. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The European Journal of Public HealthOxford University Press

Published: Apr 30, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off