2.1

Study design

We conducted a cross-sectional study, embedded in the Generation R Study, situated in the city of Rotterdam, the Netherlands. The Generation R Study is a prospective population-based cohort study and was designed to identify environmental and genetic determinants of growth, development and health. The design of this cohort is described in detail elsewhere. For this purpose, children have been followed from fetal life until adulthood in Rotterdam, the Netherlands . The Medical Ethics Committee of Erasmus Medical Center, Rotterdam, approved this study (MEC-2007-413). All participants of this study gave written informed consent.

2.2

Study population

All pregnant women with a delivery date between April 2002 and January 2006 and who lived in the study area of Rotterdam, the Netherlands, were eligible for participation in the Generation R Study. From all included children, 8305 mothers gave consent to participate in the school aged period (5 years onwards) of the Generation R Study. Ultimately, 6690 children had actually visited the research center. From these, we excluded all participants with incomplete information on dental caries (n = 1367) and all twin participants (n = 134), leaving a total study population of 5189 children.

2.3

Socioeconomic position

Since young children cannot have yet established their own socioeconomic level, they were classified according to their parents’ socioeconomic position. We named this family SEP. We considered the following socioeconomic and sociodemographic factors as indicators for family SEP: parental education level, parental employment status (paid job vs. unpaid job), net household income, parenting, and teenage pregnancy. Educational level was categorized based on the Dutch Standard Classification of Education . We defined four educational levels that could be obtained by a parent: low (no education, primary school, lower vocational training, intermediate general school, or three years or less general secondary school), mid-low (more than three years general secondary school, intermediate vocational training, or first year of higher vocational training), mid-high (higher vocational training), and high (university or PhD degree). Paternal and maternal employment status was defined as paid job or no paid job. Net household income was divided into three categories; <2000 euro/month, 2000–3200 euro/month and >3200 euro/month. Teenage pregnancy was defined as pregnancy in girls aged 19 years or younger and was based on maternal age at enrolment. All information on socioeconomic and sociodemographic factors were obtained by questionnaires .

2.4

Dental caries

We scored the presence of dental caries on intraoral photographs by using the decayed, missing, and filled teeth index (dmft index). We took these pictures with either the Poscam USB intraoral (Digital Leader PointNix) or Sopro 717 (Acteon) autofocus camera. Both cameras had a resolution of 640 × 480 pixels and a minimal scene illumination of 1.4 and 30 lx. The usage of intraoral photographs, compared to ordinary oral examination, in scoring dental caries with a dmft index has been described elsewhere and showed to have high sensitivity and specificity (85.5% and 83.6% respectively) . Furthermore, we evaluated intra-observer reliability (K = 0.98) and inter-observer reliability (K = 0.89), both indicating an almost perfect agreement .

For the analyses, we categorized the children as having no dental caries (dmft = 0), having mild caries (dmft = 1–3), or having severe caries (dmft >3). The cut-off values for mild and severe caries were based on the mean dmft index of five-year-old Dutch children obtained from a recent report by Schuller et al. .

2.5

Covariates

We considered children’s age, sex, ethnicity, and oral health behavior as potential confounders in the relationship between dental caries and SEP. Children’s ethnicity was based on the birth country of both parents . If one of the parents was born in another country than the Netherlands, the child’s ethnicity was defined as the birth country of that parent. Birth country of the mother was conclusive if both parents were born in another country. For this study we categorized children into two different groups; Dutch/Western and non-Western. Mothers had to fill in questionnaires on oral health behavior of their children at age six. Oral health related behavior was measured by age at first dental visit (0–3 years, >3 years, or never), dental visits in the past year (yes or no), and tooth brushing frequency (once a day, twice a day, or more than twice a day).

2.6

District approach

To compare caries prevalence between districts with different levels of SEP, the four-digit postal codes were collected from the mothers when the children were six years of age. The postal code had to be their current living area at the moment of dental caries measurement. We only considered children living in the city of Rotterdam, the Netherlands, for analysis (n = 3942). In general, Rotterdam has fourteen different districts (Centrum, Prins Alexander, Pernis, Overschie, Noord, Kralingen-Crooswijk, IJsselmonde, Hoogvliet, Hoek van Holland, Hillegersberg-Schiebroek, Feijenoord, Delfshaven, Charlois, and Rozenburg), however we excluded Pernis, Hoek van Holland, and Rozenburg due to low sample sizes within these districts (n < 20). We ranked the districts from socioeconomically weakest to strongest district, using individual based indicators of SEP. For this, we calculated the prevalence of low SEP, for each SEP indicator, within a district. Afterwards, we took the sum of these to calculate a total prevalence of low SEP indicators per district. By using these numbers, each district could be ranked from lowest to highest total prevalence. We used this ranking to compare our data to the municipality of Rotterdam, the Netherlands, and to visually depict the districts from weakest to strongest. Moreover, the ranking was used to assess the reference district for regression analysis (socioeconomically strongest district). Caries prevalence was calculated for each district by dividing the total number of children with a dmft >0 by the total number of children living in that district. We used the open-source desktop Geographic Information System (GIS), QGIS 2.8.4-Wien, for Mac, for geo-mapping our ranking of the districts and caries prevalence per district. The difference between the highest and lowest total prevalence of low SEP indicators was divided by four to construct four even quartiles. We applied the same method for caries prevalence.

2.7

Statistical analyses

Characteristics of the study population were calculated and presented as absolute numbers with percentages for categorical variables and median values, with 90% range, for continuous variables.

A multinomial regression analysis was performed to analyze the association between all SEP indicators and dental caries, expressed in odds ratio’s (OR’s) with 95% confidence intervals (CI’s). Children were categorized into three groups based on their dental caries experience (dmft = 0 versus dmft = 1–3 or dmft >3). Three different regression models were constructed. The first model, was adjusted for children’s age, sex and ethnicity only. The second model was additionally adjusted for children’s oral health behavior. The third model was additionally adjusted for all other SEP indicators, to investigate their mediating effects on the association between every single SEP indicator and caries. Moreover, we checked whether the presence of Hypomineralized Second Primary Molars (HSPMs) altered the association between SEP and dental caries . This was not the case and therefore the presence of HSPM was not included in the third model.

Logistic regression analysis was used to analyze the association between the social ranking of a district and dental caries. In this analysis, the district with the highest social ranking (Hillegersberg-Schiebroek, Rotterdam, The Netherlands) was used as the reference category. First, a crude model, including only the districts as independent variables was built. Second, this model was adjusted for child’s age, sex, and ethnicity. Third, this model was additionally adjusted for all SEP indicators.

Children with complete data on all SEP indicators were compared with children with missing data on at least one SEP indicator by using a Pearson Chi-Square test (Table S1). Parents of children with missing information on at least one indicator of SEP were lower educated (p < 0.001), were more often unemployed (p < 0.005), had a lower household income (p < 0.001), were more often single (p < 0.001), and had more often a teenage pregnancy (p < 0.001). To handle the missing data, we performed multiple imputation by using the Markov Chain Monte Carlo method . Ten independent datasets were generated by this method after which the pooled effect estimates were calculated . The relationship between all variables, that have been used in this study, served as a predictor in the multiple imputation models. We used the Statistical Package of Social Sciences version 22.0 for Mac (IBM Corp, Armonk, NY, USA) for our statistical analyses. A p-value < 0.05 was considered to be statistically significant.

2.8

Non-response analysis and stratified analysis

We compared the children which were excluded from analysis (incomplete information on caries, n = 1334) with the children that were included in the study (n = 5189) on all indicators of SEP (Table S2). There were no significant differences of all indicators between participants with incomplete information on dental caries and participants with complete information (p > 0.05). Moreover, since ethnicity is related to SEP, we performed a stratified analysis under Dutch children only for the association between SEP and dental caries .

2.1

Study design

We conducted a cross-sectional study, embedded in the Generation R Study, situated in the city of Rotterdam, the Netherlands. The Generation R Study is a prospective population-based cohort study and was designed to identify environmental and genetic determinants of growth, development and health. The design of this cohort is described in detail elsewhere. For this purpose, children have been followed from fetal life until adulthood in Rotterdam, the Netherlands . The Medical Ethics Committee of Erasmus Medical Center, Rotterdam, approved this study (MEC-2007-413). All participants of this study gave written informed consent.

2.2

Study population

All pregnant women with a delivery date between April 2002 and January 2006 and who lived in the study area of Rotterdam, the Netherlands, were eligible for participation in the Generation R Study. From all included children, 8305 mothers gave consent to participate in the school aged period (5 years onwards) of the Generation R Study. Ultimately, 6690 children had actually visited the research center. From these, we excluded all participants with incomplete information on dental caries (n = 1367) and all twin participants (n = 134), leaving a total study population of 5189 children.

2.3

Socioeconomic position

Since young children cannot have yet established their own socioeconomic level, they were classified according to their parents’ socioeconomic position. We named this family SEP. We considered the following socioeconomic and sociodemographic factors as indicators for family SEP: parental education level, parental employment status (paid job vs. unpaid job), net household income, parenting, and teenage pregnancy. Educational level was categorized based on the Dutch Standard Classification of Education . We defined four educational levels that could be obtained by a parent: low (no education, primary school, lower vocational training, intermediate general school, or three years or less general secondary school), mid-low (more than three years general secondary school, intermediate vocational training, or first year of higher vocational training), mid-high (higher vocational training), and high (university or PhD degree). Paternal and maternal employment status was defined as paid job or no paid job. Net household income was divided into three categories; <2000 euro/month, 2000–3200 euro/month and >3200 euro/month. Teenage pregnancy was defined as pregnancy in girls aged 19 years or younger and was based on maternal age at enrolment. All information on socioeconomic and sociodemographic factors were obtained by questionnaires .

2.4

Dental caries

We scored the presence of dental caries on intraoral photographs by using the decayed, missing, and filled teeth index (dmft index). We took these pictures with either the Poscam USB intraoral (Digital Leader PointNix) or Sopro 717 (Acteon) autofocus camera. Both cameras had a resolution of 640 × 480 pixels and a minimal scene illumination of 1.4 and 30 lx. The usage of intraoral photographs, compared to ordinary oral examination, in scoring dental caries with a dmft index has been described elsewhere and showed to have high sensitivity and specificity (85.5% and 83.6% respectively) . Furthermore, we evaluated intra-observer reliability (K = 0.98) and inter-observer reliability (K = 0.89), both indicating an almost perfect agreement .

For the analyses, we categorized the children as having no dental caries (dmft = 0), having mild caries (dmft = 1–3), or having severe caries (dmft >3). The cut-off values for mild and severe caries were based on the mean dmft index of five-year-old Dutch children obtained from a recent report by Schuller et al. .

2.5

Covariates

We considered children’s age, sex, ethnicity, and oral health behavior as potential confounders in the relationship between dental caries and SEP. Children’s ethnicity was based on the birth country of both parents . If one of the parents was born in another country than the Netherlands, the child’s ethnicity was defined as the birth country of that parent. Birth country of the mother was conclusive if both parents were born in another country. For this study we categorized children into two different groups; Dutch/Western and non-Western. Mothers had to fill in questionnaires on oral health behavior of their children at age six. Oral health related behavior was measured by age at first dental visit (0–3 years, >3 years, or never), dental visits in the past year (yes or no), and tooth brushing frequency (once a day, twice a day, or more than twice a day).

2.6

District approach

To compare caries prevalence between districts with different levels of SEP, the four-digit postal codes were collected from the mothers when the children were six years of age. The postal code had to be their current living area at the moment of dental caries measurement. We only considered children living in the city of Rotterdam, the Netherlands, for analysis (n = 3942). In general, Rotterdam has fourteen different districts (Centrum, Prins Alexander, Pernis, Overschie, Noord, Kralingen-Crooswijk, IJsselmonde, Hoogvliet, Hoek van Holland, Hillegersberg-Schiebroek, Feijenoord, Delfshaven, Charlois, and Rozenburg), however we excluded Pernis, Hoek van Holland, and Rozenburg due to low sample sizes within these districts (n < 20). We ranked the districts from socioeconomically weakest to strongest district, using individual based indicators of SEP. For this, we calculated the prevalence of low SEP, for each SEP indicator, within a district. Afterwards, we took the sum of these to calculate a total prevalence of low SEP indicators per district. By using these numbers, each district could be ranked from lowest to highest total prevalence. We used this ranking to compare our data to the municipality of Rotterdam, the Netherlands, and to visually depict the districts from weakest to strongest. Moreover, the ranking was used to assess the reference district for regression analysis (socioeconomically strongest district). Caries prevalence was calculated for each district by dividing the total number of children with a dmft >0 by the total number of children living in that district. We used the open-source desktop Geographic Information System (GIS), QGIS 2.8.4-Wien, for Mac, for geo-mapping our ranking of the districts and caries prevalence per district. The difference between the highest and lowest total prevalence of low SEP indicators was divided by four to construct four even quartiles. We applied the same method for caries prevalence.

2.7

Statistical analyses

Characteristics of the study population were calculated and presented as absolute numbers with percentages for categorical variables and median values, with 90% range, for continuous variables.

A multinomial regression analysis was performed to analyze the association between all SEP indicators and dental caries, expressed in odds ratio’s (OR’s) with 95% confidence intervals (CI’s). Children were categorized into three groups based on their dental caries experience (dmft = 0 versus dmft = 1–3 or dmft >3). Three different regression models were constructed. The first model, was adjusted for children’s age, sex and ethnicity only. The second model was additionally adjusted for children’s oral health behavior. The third model was additionally adjusted for all other SEP indicators, to investigate their mediating effects on the association between every single SEP indicator and caries. Moreover, we checked whether the presence of Hypomineralized Second Primary Molars (HSPMs) altered the association between SEP and dental caries . This was not the case and therefore the presence of HSPM was not included in the third model.

Logistic regression analysis was used to analyze the association between the social ranking of a district and dental caries. In this analysis, the district with the highest social ranking (Hillegersberg-Schiebroek, Rotterdam, The Netherlands) was used as the reference category. First, a crude model, including only the districts as independent variables was built. Second, this model was adjusted for child’s age, sex, and ethnicity. Third, this model was additionally adjusted for all SEP indicators.

Children with complete data on all SEP indicators were compared with children with missing data on at least one SEP indicator by using a Pearson Chi-Square test (Table S1). Parents of children with missing information on at least one indicator of SEP were lower educated (p < 0.001), were more often unemployed (p < 0.005), had a lower household income (p < 0.001), were more often single (p < 0.001), and had more often a teenage pregnancy (p < 0.001). To handle the missing data, we performed multiple imputation by using the Markov Chain Monte Carlo method . Ten independent datasets were generated by this method after which the pooled effect estimates were calculated . The relationship between all variables, that have been used in this study, served as a predictor in the multiple imputation models. We used the Statistical Package of Social Sciences version 22.0 for Mac (IBM Corp, Armonk, NY, USA) for our statistical analyses. A p-value < 0.05 was considered to be statistically significant.

2.8

Non-response analysis and stratified analysis

We compared the children which were excluded from analysis (incomplete information on caries, n = 1334) with the children that were included in the study (n = 5189) on all indicators of SEP (Table S2). There were no significant differences of all indicators between participants with incomplete information on dental caries and participants with complete information (p > 0.05). Moreover, since ethnicity is related to SEP, we performed a stratified analysis under Dutch children only for the association between SEP and dental caries .

3.1

Population characteristics

In Table 1 the characteristics of the study population are presented. The children had a median age (90% range) of 6.03 years (5.68–7.83). The prevalence of caries in our study population was 31.7%. About 28.6% of the mothers and 33.7% of the fathers had a high level of education. The majority of mothers and fathers had a paid job (75.4% and 93.9%, respectively) and almost half of the parents reported a household income of more than 3200 euros per month (49.7%). The household was run by a single parent in 21.0% of the families.

Table 1

Characteristics of the study population (n = 5189).

		Total	Missing
		n (%)	n (%)
Maternal educational level	High	1260 (28.6)	784 (15.1)
	Mid-high	1215 (27.6)
	Mid-low	1386 (31.5)
	Low	544 (12.3)
Paternal educational level	High	1348 (33.7)	1192 (23.0)
	Mid-high	913 (22.8)
	Mid-low	1085 (27.1)
	Low	651 (16.3)
Maternal employment status	Paid job	3143 (75.3)	1015 (19.6)
	No paid job	1031 (24.7)
Paternal employment status	Paid job	3644 (93.8)	1305 (25.1)
	No paid job	240 (6.20)
Household income	>€3200	2075 (49.7)	1016 (19.6)
	€2000-<€3200	1081 (25.9)
	<€2000	1017 (24.4)
Single parenting	No	3518 (79.0)	738 (14.2)
	Yes	933 (21.0)
Teenage pregnancy	No	5044 (97.2)	–
	Yes	145 (2.80)
Ethnic background	Dutch/Western	3446 (68.0)	125 (2.41)
	Non-Western	1618 (32.0)
Child’s sex	Boy	2601 (50.1)	–
	Girl	2588 (49.9)
Child’s age	Median (90% range)	6.03 (5.68–7.83)	–
Tooth brushing	Once a day	884 (20.7)	920 (17.7)
	Twice or more a day	3385 (79.3)
Age first dental visit	0–3 years	2389 (54.1)	770 (14.8)
	Older than 3 years	1863 (42.2)
	Never been	167 (3.70)
Dental visit in past year	Yes	3967 (92.4)	897 (17.3)
	No	325 (7.60)
Index dmft	0	3524 (68.3)	–
	1–3	995 (19.2)
	>3	652 (12.6)

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Influence of adhesive strategy on clinical parameters in cervical restorations: A systematic review and meta-analysis
2. No-waiting dentine self-etch concept—Merit or hype
3. Adhesive interfacial characteristics and the related bonding performance of four self-etching adhesives with different functional monomers applied to dentin
4. Boron nitride nanotubes as novel fillers for improving the properties of dental adhesives
5. The development of a novel shade selection program for fixed shade translucent dental materials
6. Mechanical properties of polymer-infiltrated-ceramic (sodium aluminum silicate) composites for dental restoration

Stay updated, free dental videos. Join our Telegram channel

Join