Persistent oral health inequality in children—repeated cross-sectional studies in 2010 and 2019

doi:10.21203/rs.3.rs-4950240/v1

Download PDF

Research Article

Persistent oral health inequality in children—repeated cross-sectional studies in 2010 and 2019

https://doi.org/10.21203/rs.3.rs-4950240/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background and aim: Children growing up in vulnerable circumstances have a higher risk of caries experience. Tracking the development of caries in relation to socioeconomic variables over time is essential for fair resource distribution to groups with higher caries risk and to even out inequalities in oral health. The aim was therefore to analyse the association between 6-year-olds´ caries prevalence and socioeconomic variables at family and residential area levels in 2010 and 2019 and the trends in these associations.

Methods: The study design is an epidemiological registry-based, repeated cross-sectional study based on caries data (grouped as 0, 1–3 and >3 dmft) for the population of 6-year-olds in 2010 (n =4,408, 95% coverage) and 2019 (n =5,199, 94% coverage) in a Swedish region. Multiple socioeconomic variables for the children’s families and residential areas were retrieved from official registries. Multinomial logistic regression was performed at both levels to produce models for each level and studied year.

Results: The variables that explained most of the association between caries and socioeconomic variables were mainly the same over the years at both levels. At the family level, these were: maternal age when having their first child (explained most of the association in both years, at 30 and 35%, respectively); maternal age when having the child in the study group; parental employment; parental and child’s migration background; maternal educational level; form of housing; and financial assistance (only 2010). At the residential area level, these were: migration background (explained most of the association both years; 82 and 52 %, respectively), educational level and number of persons per household. The association between socioeconomic variables and caries was consistently stronger for severe caries (dmft >3) than moderate (dmft 1–3). Multiple socioeconomic risk variables meant even greater likelihood of caries.

Conclusion: Over the studied years, variables related with socioeconomics vulnerability continued to be associated with caries in young children growing up under socially disadvantaged circumstances. Effective efforts for families living in socially vulnerable contexts are needed to achieve good and equal oral health, as is continued follow-up to evaluate whether the goal is reached.

epidemiology

caries

socioeconomic status

preschool children

family

residential area

For decades it has been acknowledged that good health has become a particular benefit of more well-resourced individuals in society. First described by Black in England in the 1980s, this inequity in health has been established in studies worldwide.(1–3) Health inequity arises from inequity in the social conditions in which people are born, grow, work, live, and age.(4) Social determinants of health refer to an individual's living conditions, which are shaped by socio-economic and political contexts, in interaction with the individual's socio-economic position such as income, education, occupation, and ethnicity, which also affects, for example, material circumstances, behaviours, and biological factors, ultimately affecting health outcomes.(5) In addition, there is a social gradient; health gradually deteriorates with lower social position.(4)

Inequalities in oral health follow the same pattern as inequalities in general health.(6, 7) Worldwide, 3.5 billion people suffer from poor oral health conditions and vulnerable groups are disproportionately affected.(8) Although largely preventable, dental caries remain one of the world's major non-communicable diseases (NCD) and untreated caries was the most common health condition globally in 2019, affecting one third of the world’s population.(9) In total, 514 million children (1–9 years old) in the world are affected, and prevalence ranges from 46% in upper-middle-income countries to 38% to in high-income countries.(8) In its severe form, caries can cause pain and adversely affect sleep, learning, and quality of life.(10) Caries in small children are related to continued caries progression later in life and poorer self-rated general health by midlife.(11, 12)

Caries is preventable and if detected in its´ early stages also reversable.(13) However, preventive actions directed to the general public may increase oral health gaps as they mainly impact resourceful families, since families and children in socially vulnerable areas visit dental care to lesser extent and are less likely to have access to or take part in caries prevention initiatives.(14, 15) Caries preventive actions directed toward at-risk children rather than to the entire population of children are also more cost-effective.(16) Family and area socioeconomic characteristics are suggested to identify families or areas with increased risk of disease to initiate preventive efforts or as a basis for distribution of public resources.(17–19) For example, in regions in Sweden, capitation per child patient is issued to dental care providers based on socioeconomic conditions of the child’s residential area through the Care Need Index (CNI).(20)

In Sweden, the prevalence of caries has decreased over the last few decades and in an international comparison, oral health in Swedish children is fairly good.(21) Between 1973 and 2013, the proportion of caries-free pre-school children (5-year-olds) increased from 9–69%.(21) However, in recent years, the positive trend regarding caries in children of preschool age has stagnated and the percentage of caries-free 6-year-olds decreased between the years 2010 and 2019 from 78–72%, and WHOS´s goal—80% of 6-years-old caries-free by 2020—was not achieved.(22, 23) Furthermore, despite the generally good oral health of the population and the fact that dental health services, including preventive measures, are free of charge for children and young people until the age of 23, inequalities in oral health due to unequal living conditions are also found in Sweden.(24, 25) Similar to international studies, a recent study from Sweden showed multiple family socioeconomic status variables (ethnicity, wealth, parental education, and employment) were associated with dental caries experience in children.(26, 27). Moreover, regardless of the family's social situation, their residential area can be related to children's oral health through the transmission of norms and behaviours in the social interaction in the immediate surroundings; the so-called ‘neighbourhood effect’.(28, 29) A study showed that children in Sweden living in the most deprived areas were five times more likely to experience caries than peers living in the most affluent areas.(30)

In order to even out the modifiable differences in oral health and enable a fair prioritisation of resources based on the population's needs, it is important to follow the development of caries and caries risk over time from a socioeconomic perspective to identify oral health inequities and at-risk groups.(31, 32) As the positive development in the proportion of caries-free 6-year-olds has stagnated in Sweden since 2010, it is of value to analyse how the association between family and residential area socioeconomic variables and caries prevalence in 6-year-olds has developed recent years.

The overall aim of the study was to analyse the association between 6-year-olds´ caries prevalence and socioeconomic variables at family and residential area levels in 2010 and 2019, and the trends in these associations between these years.

The study design is an epidemiological registry-based repeated cross-sectional study. It took place in the region of Östergötland, in southeast Sweden, which is the fourth largest region in Sweden with a population of 468,000 inhabitants (in year 2021) living in both rural and urban areas.(33) The dental health development in the region in terms of caries-free children follows national development.(23, 34) Basic oral health information is delivered to all children staying in the region, from pre-birth until 19 years of age, and their parents, with the aim of ensuring good opportunities for the children to maintain or achieve good oral health. Since 2006, resources have also been allocated for extended preventive efforts to be carried out in areas with low socioeconomic status based on area-based socioeconomics to reduce oral health differences within the region. As a part of the regional and national follow-up of oral health, dental health service providers (public and private dental clinics) in the region report epidemiological caries data to Region Östergötland for all examined 3, 6, 12 and 19-year-olds, and, since 2019, all 23-year-olds.

Study group

Study groups consisted of the population of children (born 2004 and 2013) who, as 6-year-olds, (during 2010 and 2019), visited dental care in the region of Östergötland for oral examination. The population of 6-year-olds in the region was 4,630 in 2010 and 5,506 in 2019. (34, 35) Those 6-year-olds who did not have caries data, a link to residential area, or completely lacked socioeconomic variables were excluded.

Data collection

The caries data were retrieved from the epidemiological caries data reported to Region Östergötland by dental health providers in the region labelled with the individual´s social security number (a unique ten-digit code for all Swedish residents). Caries prevalence variable (dependent variable) was measured in dmft, referring to the number of decayed, missing, or filled primary teeth and categorised into three groups: no caries (dmft = 0), moderate caries experience (dmft = 1–3), and severe caries experience (dmft > 3).(36, 37) In this context, caries refers to manifest (dentin) caries; that is, caries that have passed through the enamel and into the dentin.(21) Caries were diagnosed at oral examinations based on clinical and radiographic findings by the child’s ordinary dental caregiver.

Statistics Sweden´s (SCB, responsible for official statistics in Sweden) classification of Sweden's municipalities into smaller geographical areas, so-called ‘key code areas’ (NYKO), was used as a definition for residential area (henceforth referred to as ‘area’). NYKO is available at several sizes, from single properties to larger areas.(38) Smaller areas expose more of the socioeconomic disparities in society compared to larger areas, and with respect to confidentiality, one of the smallest divisions, the NYKO4 division, was used in the present study.(20, 30)

Socioeconomic variables (independent variables) originated from official registries provided by SCB were used. The link between the children´s caries prevalence, socioeconomic information, and area was established by SCB through the social security numbers. Before the data records were delivered to researchers, the social security numbers were replaced with serial numbers. Variables were divided into relevant categories inspired by previous research.(20, 26, 30, 39) A complete description of each variable and associated categories is presented in appendix 1.

Family-level variables were retrieved for either the children in the study group and/or their parents. Background data (gender), ethnicity, migration background, family type, and number of children in the household was retrieved for each child. Number of persons per household, disposable income, financial assistance, housing allowance, form of housing, urban or rural area, was retrieved for each parent with whom the child was registered. Migration background, age when the child in study group was born, age when first child was born, employment, and educational level was retrieved on both the mother and the father respectively. In cases where an adoptive mother or father existed, they replaced the biological parent. Missing values (see appendix 2a) in the variables on family level were replaced using multiple imputation.(40)

At the area level, socioeconomic data for the total population living in each area were retrieved. This included the following variables: background data (age and gender), ethnicity, migration background, age (when the first child was born), employment, education, family type, number of children in household, household size, disposable income and high-income households, financial assistance, housing allowance, form of housing, urban/rural area, and CNI. Data consisted of proportions or averages of habitants for each categorised variable.

Analyses and statistics

To describe 6-year-olds´ caries prevalence and socioeconomic variables at family (non-imputed data) and area levels in 2010 and 2019, descriptive statistics was reported on categorical (frequency and percentage) and continuous variables (mean, standard deviation, and range) for both years. Independent sample t-tests (continuous variables) and chi² tests (categorical variables) were performed to analyse trends between the studied years.

To analyse the association between 6-year-olds´ caries prevalence and socioeconomic variables, multinomial logistic regression models were performed at the family (imputed data) and area level for 2010 and 2019. The association between caries experience and socioeconomic variables was expressed by odds ratio (OR) and 95% confidence intervals (CI). To analyse time trends in the associations between caries prevalence and socioeconomic variables, time interactions for each variable and caries category were tested. For significance testing on imputed data the median, pooling rules were used for p-values.(41) For the area level, NYKO codes were used as random intercepts to address potential correlations between persons living in the same area.

Primarily, univariable logistic regression analysis for each independent variable per studied level and year was performed. Secondly, four adjusted multivariable models were analysed; one for each level and for each studied year (background variables excluded). Multicollinearity was tested using variance inflation factors (VIF), and values over five were considered problematic. If two independent variables were strongly correlated with each other according to VIF, the variable with the strongest association with 6-year-olds´ caries prevalence was included in the model.

At the family level, all variables were included in the models. At the area level, the included variables (and categories) were: migration background (proportion with foreign background), parental age when the first child was born (average age of mothers/fathers at the birth of their first child), employment (proportion without employment), educational level (proportion with highest attained education elementary school), family type (proportion of single mothers/fathers), number of persons/ household (average number of persons/ household), disposable income (median disposable income/consumption unit/household divided by 100,000), financial assistance (proportion of households receiving financial assistance), housing allowance (proportion of households with housing allowance), form of housing (proportion of families renting), urban/rural area (proportion of households in urban areas), and area CNI (> 1). Ethnicity, number of children/ household, proportion of high-income households were excluded due to multicollinearity.

Descriptive figures were created to visualise changes in caries prevalence between the studied years and different groups based on relevant socioeconomic variables and relevant cut-offs for each variable. Absolute (difference in proportion of caries-free children) and relative (unadjusted OR) inequalities were calculated.

Goodness of fit (pseudo-R2) for multivariable models were assessed with McFadden and Nagelkerke. Statistical significance level was set to < 0.05. All analyses were performed using the Statistical Package for Social Sciences (SPSS vn.29) and R v.4.2.3.

Population characteristics

In 2010 4,431 6-year-olds (95.7 %) had an oral examination of which 4,408 (95%) was included in the study as they have met the inclusion criteria with eligible data for caries, area and socioeconomic variables. Corresponding figures in 2019 were 5,206 examined 6-year-olds (94.5%) of which 5,199 (94%) could be included in the study. The population was distributed within 622 areas in 2010 and 651 in 2019.

On the family level, there was a difference (p =<0.001) in caries experience between the studied years. In 2010, 78 of the children in the study group were caries-free (dmft =0); for 2019, the corresponding proportion was 74% and the proportion with dmft >3 increased from 9.3% to 11.5% (p<0.001). The percentile with the most caries had dmft of 3 or above in 2010 and dmft 4 or above in 2019; for the other percentiles the dmft remained unchanged. Mean dmft among 6-year-olds per area was 0.74 in 2010 and 0.85 in 2019 (p =0.148).

Descriptive data for caries prevalence and socioeconomic variables

Table 1a–b presents the main results for descriptive data on caries prevalence and the distribution of socioeconomic variables at family and area levels for 2010 and 2019 and whether there were significant differences in proportions between the studied years. For the full table, see appendix 2a–b. At the family level all variables but four displayed significant differences in proportions between 2010 and 2019 (table 1a), e.g., the proportions of children born in Sweden had decreased from 94% to 93%, and the proportion of children born outside Europe grew by 1.3 percentage points (pp). The proportions of children and parents with foreign migration backgrounds had increased among children from 17% to 23%, and for parents from 22% to 30%. For maternal age when child in study group, and for when first child was born, there was an increase in proportions of parents >35 years of age from 19% to 22%, from 6.3% to 7.5%, respectively; and a decrease of proportions of mothers aged 25–34y from 67% to 63% from 62% to 61%. The proportion of mothers with the highest attained education being high school had decreased from 43% to 38%, while the proportion with only elementary school (low educational level) had increased from 10% to 14%.

Table 1a. Main results for descriptive data on caries and socioeconomic variables 2010 and 2019 (family level)

		2010	2019
Variable	Category	n (%)	n (%)	p-value
6y		4,408	5,199
Caries (dmft)	0	3,426 (77.7)	3,838 (73.8)
	1-3	574 (13.0)	764 (14.7)
	>3	408 (9.3)	597 (11.5)	<0.001
Childs gender	Male	2,260 (51.3)	2,691 (51.8)
Childs gender	Female	2,148 (48.7)	2,508 (48.2)	0.632
Childs´ ethnicity	Sweden	4,156 (94.3)	4,813 (92.6)
	Europe except Sweden	54(1.2)	84 (1.6)
	Outside of Europe	198 (4.5)	301 (5.8)	<0.01
Childs´ migration background	Native¹	3,678 (83.4)	3,998 (76.9)
Childs´ migration background	Foreign²	730 (16.6)	1201 (23.1)	<0.001
Maternal migration background	Native	3,438 (78.2)	3,617 (69.7)
Maternal migration background	Foreign	959 (21.8)	1,573 (30.3)	<0.001
Paternal migration background	Native	3,394 (78.3)	3,541 (69.7)
Paternal migration background	Foreign	942 (21.7)	1,537 (30.3)	<0.001
Maternal age when child in the study group was born	<20y	76 (1.7)	64 (1.2)
	20-24y	530 (12.0)	686 (13.2)
	25-34y	2,970 (67.4)	3,291 (63.4)
	>35y	829 (18.8)	1,152 (22.2)	<0.001
Paternal age when child in the study group was born	<20y	29 (0.7)	19 (0.4)
	20-24y	208 (4.8)	326 (6.4)
	25-34y	2,604 (59.6)	2,770 (54.4)
	>35y	1,528 (35.0)	1,977 (38.8)	<0.001
Maternal age when first child was born	<20y	295 (6.7)	300 (5.8)
	20-24y	1,100 (25.0)	1,356 (26.1)
	25-34y	2,726 (62.0)	3,145 (60.6)
	>35y	276 (6.3)	389 (7.5)	<0.01
Paternal age when their first child was born	<20y	70 (1.6)	85 (1.7)
	20-24y	630 (14.5)	749 (14.7)
	25-34y	2,953 (68.1)	3,359 (66.1)	0.144
Maternal employment status	Employed	3,475 (79.0)	4,221 (81.4)
Maternal employment status	Unemployed	922 (21.0)	965 (18.6)	<0.01
Maternal educational level	Elementary school	446 (10.3)	601 (13.7)
	Highschool	1,889 (43.4)	1,676 (38.2)
	Higher education	2,015 (46.3)	2,110 (48.1)	<0.001
Paternal educational level	Elementary school	460 (10.7)	637 (14.2)
	Highschool	2,186 (50.7)	2,116 (47.3)
	Higher education	1,669 (38.7)	1,722 (38.5)	<0.001
Family type	Single parent	738 (16.7)	968 (18.6)
	Not single	3,668 (83.3)	4,226 (81.4)	0.016
	Else	2	5
Number of children in the household	1	465 (10.6)	609 (11.7)
	2-3	3,499 (79.4)	3,938 (75.8)
	≥4	442 (10.0)	647 (12.5)	<0.001
Number of persons/ household	1-5	3,997 (91.1)	3,999 (98.7)
Number of persons/ household	>5	392 (8.9)	52 (1.3)	<0.001
Familys´ disposable income	Median	506,770	572,050	<0.001
Financial assistance	Yes	385 (8.8)	176 (5.9)
Financial assistance	No	4,004 (91.2)	5,002 (96.6)	<0.001
Housing allowance	No	3,656 (82.9)	4,246 (82.0)
Housing allowance	Yes	733 (16.7)	932 (18.0)	0.095
Form of housing	Renting	1,186(26.6)	965(23.8)
	Owning appartment	159(3.6)	186(4.6)
	Owning house	3,022(68.9)	2,840(70.2)	<0.01
	Else	40	56
Family living in in urban or rural	Urban	3,707 (84.1)	4,481 (86.2)
Family living in in urban or rural	Rural	701 (15.9)	718 (13.8)	0.004

¹ native= Born in Sweden with one or two native Swedish parents, ² foreign=Born abroad or born in Sweden with two foreign born parents

Results for descriptive data on caries prevalence and socioeconomic variables at family level, for year 2010 and 2019 and significant differences in proportions between the years.

In Table 1b, the main results for caries prevalence and socioeconomic variables at the area level for 2010 and 2019 are presented. Background data (age, gender), ethnicity, migration background, maternal age (when the first child was born), employment, education, family type, number of children per household, and disposable income displayed significant differences in proportions between studied years; E.g., the mean proportion born in Sweden had decreased from 91 to 88% and the mean proportion with foreign migration background had increased from 11% to 15%. The mean average maternal age when the first child was born had increased from 27.6 years to 28.4 years. The mean proportion of parents with higher education had increased from 34% to 39%, while the mean proportion with elementary or high-school education had decreased from 15% to 11% and from 50% to 48%, respectively. The mean average number of persons per household was 2.4 in 2010 and 2.3 in 2019 (p =0.066).

Table 1b. Main results on descriptive data for caries and socioeconomic variables 2010 and 2019 (area level)

		2010	2019
Variable	Category	n (%)	n (%)	p-value
NYKO		622	651
N 6y /NYKO	Mean (SD)	7.1 (8.4)	8.0 (9.9)	0.08
N 6y /NYKO	Range	1- 64	1- 71
Population	Mean (SD)	658 (703)	685 (746)	0.51
Population	Range	11-4,298	4-4,932
Caries (Mean dmft/ NYKO)	Mean (SD)	0.74 (1.31)	0.85 (1.35)	0.15
Caries (Mean dmft/ NYKO)	Range	0.0- 10.0	0.0- 10.5
Proportion of male	Mean (SD)	50.7 (3.4)	51. 1 (3.1)	0.04
Proportion of male	Range	36.0-72.7	37.5-72.7
Proportion born in Sweden	Mean (SD)	91.3 (8.6)	88.3 (11.3)	<0.001
Proportion born in Sweden	Range	32.4-100.0	20.0-100.0
Proportion with foreign migration background	Mean (SD)	11.2 (11.2)	15.3 (14.9)	<0.001
Proportion with foreign migration background	Range	0.0- 79.4	0.0-100.0
Average age of mothers at the birth of their first child	Mean (SD)	27.6 (3.7)	28.4 (3.4)	<0.001
	Range	17-41	19-41
Average age of fathers at the birth of their first child	Mean (SD)	30.4 (3.8)	30.4 (3.9)	0.965
	Range	20-51	20-51
Proportion with employment	Mean (SD)	50.2 (7.1)	51.0 (7.6)	0.045
Proportion with employment	Range	21.3-73.4	0.0-73.5
Proportion with highest education elementary school	Mean (SD)	14.5 (6.6)	10.8 (6.7)	<0.001
Proportion with highest education elementary school	Range	0.0-39.0	0.0-42.1
Proportion with highest education high school	Mean (SD)	50.3 (10.3)	48.4 (11.5)	<0.001
Proportion with highest education high school	Range	8.3-73.3	0.0-81.8
Proportion with higher education	Mean (SD)	33.7 (14.1)	38.9 (14.5)	<0.001
Proportion with higher education	Range	5.4-91.7	0.0-100.0
Proportion of single mothers	Mean (SD)	4.0 (3.6)	3.6 (3.0)	0.045
Proportion of single mothers	Range	0.0-40.0	0.0-25.0
Proportion of single fathers	Mean (SD)	1.5 (1.7)	1.7 (1.6)	0.066
	Range	0.0-20.0	0.0-16.7
Average number of children/ household	Mean (SD)	0.69 (0.28)	0.66 (0.29)	0.03
Average number of children/ household	Range	0.04-1.75	0.04-2.00
Average number of persons/ household	Mean (SD)	2.4 (0.43)	2.3 (0.43)	0.07
Average number of persons/ household	Range	1.3-3.8	1.3-4.0
Median disposable income per consumption unit/ household	Mean (SD)	233,007 (38,829)	450,842 (135,465)	<0.001
	Range	104,424-393,913	134,768-1,123,667
Proportion of high-income households	Mean (SD)	10.8 (4.3)	11.1 (4.4)	0.1
Proportion of high-income households	Range	0.0-30.8	0.0-27.3
Proportion of households with financial assistance	Mean (SD)	3.6 (5.7)	2.4 (4.6)	<0.001
Proportion of households with financial assistance	Range	0.0-43.5	0.0-30.6
Proportions housing renting	Mean (SD)	21.7 (30.1)	21.9 (30.1)	0.899
	Range	0-100	0-100
Proportion owning appartment in appartment building	Mean (SD)	6.5 (17.0)	7.6 (17.9)	0.285
Proportion owning appartment in appartment building	Range	0-100	0-100
Proportion owning house	Mean (SD)	66.0 (36.7)	66.2 (38.)	0.91
	Range	0-100	0-100
	Std.Dev	36.7	38.1
Else	Mean	5.8	4.3
Proportion of households in urban area	Mean	65.7	66.3	0.824
	Range	0-100	0-100
	Std.Dev	46.1	45.8
	Missing	-	1
Residential area CNI	Mean	1.14	1.19	0.148
	Range	0.3-4.0	0.2-6.3
	Std.Dev	0.5	0.7

Main results for descriptive data on caries prevalence and social and socioeconomic variables at residential area level for year 2010 and 2019 and significant differences in proportions between the years.

Association between caries prevalence and socioeconomic variables

Family level

In the univariable regression analysis at the family level, all socioeconomic variables included in the study were associated with the likelihood of caries prevalence in both years (except gender and number of persons/ households in 2019). The association between socioeconomic variables and caries prevalence was consistently stronger for severe caries than for moderate. For the variables, maternal age when the child in the study group was born, parental age when first child was born, maternal educational level, family type, number of persons per household, financial assistance, form of housing OR was significantly lower in 2019 than in 2010. A selection of socioeconomic variables and their association with caries on the family level in 2010 and 2019 are presented in Table 2 (for full table, see appendix 2a).

Table 2. Main results of univariable analysis of the association between caries and socioeconomic variables (family level)

			Comparison dmft 1-3 against 0					Comparison dmft >3 against 0
Variable	Year	Category	P-value	OR	CI Lower	CI Upper	p-trend	p-value	OR	CI Lower	CI Upper	p-trend
(ref category)			P-value	OR	CI Lower	CI Upper	p-trend	p-value	OR	CI Lower	CI Upper	p-trend
Childs´ ethnicity (native)	2010	Europe	0.16	1.76	0.80	3.88		<0.001	6.18	3.36	11.37
		Outside Europe	<0.001	3.87	2.65	5.64		<0.001	10.53	7.52	14.79
	2019	Europe	0.76	1.12	0.54	2.31	0.71	<0.001	6.45	4.04	10.30	0.68
		Outside Europe	<0.001	3.72	2.69	5.14		<0.001	12.77	9.66	16.89
Childs´ migration background (native)	2010	Foreign	<0.001	3.96	3.20	4.92		<0.001	16.01	12.71	20.17
Childs´ migration background (native)	2019	Foreign	<0.001	3.39	2.84	4.03	0.27	<0.001	14.12	11.60	17.18	0.42
Maternal migration background (native)	2010	Foreign	<0.001	3.08	2.53	3.76		<0.001	11.87	9.45	14.92
Maternal migration background (native)	2019	Foreign	<0.001	3.05	2.59	3.59	0.96	<0.001	11.73	9.59	14.36	0.94
Paternal migration background (native)	2010	Foreign	<0.001	3.16	2.60	3.85		<0.001	10.47	8.31	13.19
Paternal migration background (native)	2019	Foreign	<0.001	2.83	2.40	3.33	0.37	<0.001	11.07	9.04	13.56	0.76
Maternal age when child in the study group was born (25-34y)	2010	<20y	<0.01	2.62	1.40	4.92		<0.001	9.17	5.44	15.48
		20-24y	<0.001	1.71	1.32	2.22		<0.001	3.45	2.64	4.50
		>35y	0.21	1.16	0.92	1.46		<0.01	1.48	1.13	1.95
	2019	<20y	<0.01	2.49	1.28	4.87	0.66	<0.001	6.32	3.62	11.03	<0.01
		20-24y	<0.01	1.44	1.15	1.81		<0.001	2.12	1.69	2.66
		>35y	0.96	1.00	0.82	1.21		0.06	0.80	0.63	1.01
Maternal age when first child was born (25-34y)	2010	<20y	<0.001	3.15	2.26	4.39		<0.001	12.53	9.15	17.17
		20-24y	<0.001	2.17	1.78	2.65		<0.001	4.04	3.15	5.18
		>35y	0.50	0.86	0.56	1.33		0.12	1.49	0.90	2.46
	2019	<20y	<0.001	2.96	2.17	4.05	0.83	<0.001	7.74	5.76	10.41	0.03
		20-24y	<0.001	1.92	1.61	2.28		<0.001	3.65	3.00	4.44
		>35y	0.32	0.85	0.61	1.18		0.09	0.65	0.39	1.07
Maternal employment status (Yes)	2010	No	<0.001	2.83	2.32	3.45		<0.001	8.65	6.94	10.77
Maternal employment status (Yes)	2019	No	<0.001	2.80	2.32	3.36	0.94	<0.001	6.82	5.65	8.23	0.11
Paternal employment status (Yes)	2010	No	<0.001	2.46	1.91	3.17		<0.001	8.41	6.61	10.69
Paternal employment status (Yes)	2019	No	<0.001	2.49	1.95	3.18	0.89	<0.001	6.46	5.22	8.00	0.12
Maternal level of education (Higher education)	2010	Elementary school	<0.001	3.94	2.98	5.22		<0.001	14.99	11.15	20.16
		Highschool	<0.001	1.54	1.27	1.87		<0.001	1.89	1.44	2.49
	2019	Elementary school	<0.001	2.94	2.30	3.76	0.11	<0.001	8.01	5.34	12.03	<0.001
		Highschool	<0.001	1.65	1.35	2.02		<0.001	2.08	1.60	2.72
Number of children in the household (2-3)	2010	1	0.12	1.25	0.95	1.66		0.21	1.25	0.88	1.78
		>3	<0.001	2.34	1.79	3.06	0.58	<0.001	5.22	4.039	6.752	0.15
	2019	1	0.07	1.25	0.98	1.58		0.37	0.87	0.64	1.18
		>3	<0.001	2.80	2.261	3.463		<0.001	4.06	3.28	5.04
Number of persons/ household (1-5)	2010	>5	<0.001	2.10	1.58	2.79		<0.001	4.82	3.70	6.27
Number of persons/ household (1-5)	2019	>5	0.26	2.03	0.52	8.04	0.06	0.12	4.08	0.60	28.15	0.01
Family’s´ disposable income (>highest quartile)	2010	<lowest quintile	<0.001	2.52	1.93	3.30		<0.001	11.76	7.85	17.64
		In between	<0.01	1.45	1.15	1.83		<0.001	3.26	2.19	4.86
	2019	<lowest quintile	<0.001	2.76	2.18	3.49	0.85	<0.001	8.74	6.34	12.05	0.15
		In between	<0.001	1.47	1.20	1.81		<0.001	3.22	2.36	4.37
Financial assistance (No)	2010	Yes	<0.001	4.15	3.11	5.54		<0.001	16.39	12.61	21.29
Financial assistance (No)	2019	Yes	<0.001	4.10	2.77	6.06	0.91	<0.001	8.34	5.83	11.91	<0.01
Form of housing (Owning house)	2010	Renting	<0.001	2.66	2.20	3.23		<0.001	9.56	7.54	12.12
		Owning apartment	0.11	1.48	0.92	2.38		<0.001	2.89	1.63	5.13
	2019	Renting	<0.001	2.61	2.11	3.23		<0.001	6.37	4.14	9.80	0.01
		Owning apartment	<0.01	2.13	1.34	3.37	0.22	0.11	1.61	0.89	2.90

Main results of the univariable analysis on the association between caries and a selection of socioeconomic variables on family level year 2010 respectively 2019 and differences in OR between the studied years (p-trend).

The result of the multivariable models of the socioeconomic variables that explained most of the association with caries are presented in Table 3, arranged by the variables contribution of explanation to the 2019s model. The goodness of fit: were McFadden=0.17 (2010), 0.13 (2019); Nagelkerke=0.28 (2010), 0.22 (2019). The variables that explained most of the association both years were maternal age upon having their first child, parental and child migration background, form of housing, paternal employment, maternal education, maternal age when the child in the study group was born, and financial assistance—but with some variation in the proportion of explanation between the years. Financial assistance was included in the 2010 model but not in the 2019. For maternal educational level OR caries experience was significantly lower in 2019 than in 2010 (p ≤0.01). Very young/young maternal age when the first child was born explained most of the association (35% in 2010 and 30% in 2019) both years. Children whose mothers had their first child before the age of 25 were more than twice as likely to suffer severe caries disease than children whose mothers were between 25 and 34 years when they had their first child. Children with multiple risk variables had an even higher likelihood of caries experience. For instance, the OR for severe caries for children with maternal age when first child was born <20 years and with foreign maternal background was 6.0 (2.6*2.3).

The proportion of children and parents with foreign migration background had increased between 2010 and 2019 (Table 1a), and migration background remained one of the most important explanatory variables in the models, explaining the association between caries and socioeconomic variables in 2010 and 2019 (Table 3). Figure 1 present a comparison of the caries prevalence between 2010 and 2019 for children with foreign migration background and native migration background. The proportion of caries-free children with foreign migration background was unchanged between 2010 and 2019 (p =0.97), the proportion of caries-free among children with native migration background was 2pp lower in 2010 than in 2010 (p =0.13). The unadjusted OR for caries prevalence (dmft >0) for children with foreign migration backgrounds in relation to children with native migration backgrounds was 7.3 in 2010, and 6.7 in 2019.

In the models (Table 3), OR for caries prevalence for children with low maternal level of education was significantly lower in 2019 than in 2010 in comparison to higher level of education. Figure 2 present the caries prevalence in relation to maternal level of education in 2010 respective to 2019 at the family level. The proportion of caries-free children with low maternal levels of education had increased from 44% in 2010 to 52% in 2019 (p <0.01). The proportion of caries-free children with higher maternal levels of education was 2pp lower in 2019 (83%) compared to 2010 (85%) (p =0.15). Hence, the difference in the proportion of caries-free children between children with the lowest and highest maternal level of education was 10pp lower in 2019 than in 2010, from 41 (85%-44%) to 31 percentage points (83%-52%). The unadjusted OR for children with low maternal education, in relation to children with high maternal education, was 7.38 in 2010 and decreased to 4.62 in 2019 (p <0.001).

Area level

In the univariable analysis at area level, all socioeconomic variables included in the study showed associations with the likelihood of caries prevalence in both 2010 and 2019 and the associations were stronger for severe caries than moderate (Table 4). However, for ethnicity or migration background (foreign), employment (proportion of unemployed), level of education (proportion with elementary or high-school education), family type (proportion of single mothers), (proportions of households with) financial assistance, or housing allowance, the OR for severe caries was lower in 2019 than in 2010. For paternal age when first child was born, (median) disposable income per consumption unit/ household, (proportion) of high-income households, the OR was less than 1 both years and had increased. Children growing up in areas with CNI >1 were significantly more likely (OR 5.98 y2010, OR 4.72 y2019) to suffer severe caries compared to children growing up in areas with CNI <1. Full table can be found in Appendix 4.

Table 4. Main results of univariable analysis of the association between caries and socioeconomic variables (area level)

			Comparison 1-3 against 0					Comparison >3 against 0
Variable (Reference)	Year	Category	p-value	OR	CI Lower	CI Upper	p-interaction	p-value	OR	CI Lower	CI Upper	p-interaction
Ethnicity (Proportion born in Sweden)	2010	Proportion born outside Sweden	<0.001	1.04	1.04	1.05		<0.001	1.09	1.08	1.10
Ethnicity (Proportion born in Sweden)	2019		<0.001	1.03	1.03	1.04	0.02	<0.001	1.06	1.05	1.06	<0.001
Migration background (Proportion with native migration background)	2010	Proportion with foreign background⁴	<0.001	1.03	1.03	1.04		<0.001	1.07	1.06	1.07
	2019		<0.001	1.02	1.02	1.03	0.03	<0.001	1.04	1.04	1.04	<0.001
Maternal age when first child was born	2010	Average age of mothers at the birth of their first child	<0.001	0.90	0.87	0.92		<0.001	0.80	0.77	0.83
Maternal age when first child was born	2019		<0.001	0.94	0.92	0.97	0.02	<0.001	0.82	0.79	0.85	0.350
Paternal age when their first child was born	2010	Average age of fathers at the birth of their first child	<0.01	0.96	0.93	0.984		<0.001	0.86	0.83	0.90
Paternal age when their first child was born	2019		0.76	1.00	0.97	1.02	0.03	<0.01	0.96	0.93	0.98	<0.001
Employment status	2010	Proportion without employment	<0.001	1.06	1.05	1.07		<0.001	1.12	1.10	1.13
Employment status	2019		<0.001	1.04	1.03	1.05	0.06	<0.001	1.08	1.07	1.10	<0.01
Education level (Proportion with higher education)	2010	Proportion with highest education elementary school	<0.001	1.05	1.03	1.07		<0.001	1.10	1.09	1.12
		Proportion with highest education high school	0.19	1.01	1.00	1.02		<0.001	0.97	0.96	0.99
	2019	Proportion with highest education elementary school	<0.001	1.04	1.03	1.05	0.25	<0.001	1.07	1.06	1.08	<0.01
		Proportion with highest education high school	0.80	1.00	0.99	1.01	0.21	0.56	1.00	0.99	1.01	<0.001
		Else	<0.001	1.08	1.05	1.12	0.41	<0.001	1.14	1.10	1.17	0.46
Family type (not single)	2010	Proportion of single mothers	<0.001	1.12	1.09	1.15		<0.001	1.23	1.20	1.27
Family type (not single)	2019		<0.001	1.10	1.07	1.12	0.64	<0.001	1.17	1.15	1.20	<0.01
Family type (not single)	2010	Proportion of single fathers	0.57	0.98	0.90	1.06		0.01	0.86	0.77	0.96
Family type (not single)	2019		0.01	0.91	0.84	0.98	0.22	<0.001	0.77	0.71	0.84	0.14
Number of persons/ household	2010	Average number of persons/ household	<0.001	0.55	0.45	0.68		<0.001	0.41	0.32	0.52
Number of persons/ household	2019		<0.001	0.69	0.58	0.83	0.12	<0.001	0.44	0.36	0.54	0.62
Disposable income per consumption unit/ household	2010	Median disposable income per consumption unit/ household (divided by 10<0.001)	<0.001	0.32	0.25	0.39		<0.001	0.07	0.05	0.09
Disposable income per consumption unit/ household	2019		<0.001	0.41	0.35	0.48	0.04	<0.001	0.16	0.13	0.19	<0.001
Form of housing (Proportion owning house)	2010	Proportions renting	<0.001	1.01	1.01	1.02		<0.001	1.03	1.03	1.04
Form of housing (Proportion owning house)	2019		<0.001	1.01	1.01	1.015	0.65	<0.001	1.03	1.02	1.03	0.65
Residential area CNI (<1)	2010	>1	<0.001	1.98	1.65	2.38		<0.001	5.99	4.55	7.87
Residential area CNI (<1)	2019		<0.001	1.88	1.60	2.21	0.68	<0.001	4.72	3.79	5.86	0.18

OR=Odds Ratio, CI=95% Confidence Interval

The results of the multivariable models the socioeconomic variables and the associated categories that explained most of the association with caries at the area level are presented in Table 5, arranged by the d variables contribution of explanation to the 2019 model. The goodness of fit were: McFadden=0.10 (2010), 0.08 (2019); Nagelkerke=0.17 (2010), 0.14 (2019). The variables that explained most of the association between caries and socioeconomic variables were the same in both 2010 and 2019, but with some variation in the proportion of model explanation. These were proportions with foreign background, proportions with low education, and average number of persons per household. Proportions with foreign background explained most of the association in both years (82% in 2010 and 52% in 2019). The OR for proportions with foreign backgrounds was significantly lower in 2019 than in 2010 (p <0.001), and the OR for the proportion with low education were significantly higher in 2019 than in 2010 (p <0.001). There was a positive correlation (R=0.62, p<0.001) between the average number of persons per household in the area and the proportion of residents owning their own house. In 2019, the OR for severe caries was 1.03 for the proportion with a foreign migrant background and the proportion with the highest level of education compulsory school (low education), which means that the odds increase by 1.03 for every pp with an increased proportion of residents with a foreign migration background or low education in the area. For example, for children growing up in an area with an increased proportion of inhabitants of 10pp with low education OR for severe caries was 1.34 (1.03^10), compared to an area with a lower proportion. If the proportion of inhabitants with foreign migration background also was 10pp higher, then OR was 1.81 (1.03^10*1.03^10).

There was a gradient of increased prevalence of caries for areas with an increased proportion of inhabitants with foreign migration background from the areas with the lowest proportion (<10%) to the areas with the highest proportion (≥40%) of inhabitants with foreign backgrounds (Figure 3). Between 2010 and 2019, there was a decrease of caries-free children in the third group (20%–29.9%), and an increase in the fourth group (30%–39.9%). The proportion of caries-free children in the areas with the lowest proportions with foreign migration background was 85% in 2010 and 84% in 2019 (p =0.20), and 41% in 2010 and 46% (p =0.15) in 2019 in the areas with the highest proportion; hence the difference in caries prevalence between the areas was 6pp less in 2019 than in 2010. The unadjusted OR for caries experience in areas with the highest proportion with foreign migration background in comparison to areas with the least was 6.3 in 2019 compared to 8.3 in 2010 (p =0.07).

Figure 3. Caries prevalence by residential area based on the proportion of inhabitants with foreign background

The gradient was also evident when analysing groups in areas with increased proportions of inhabitants with low education (Figure 4). Between 2010 and 2019, the proportion of caries-free children remained high; 87% in 2010 and 85% in 2019, in the areas with least proportion of inhabitants with low education (<5%). In the four other areas with higher proportion of inhabitants with low education, the proportion of caries-free decreased and in the areas with the highest proportion (>20%) the corresponding figures were 55% and 46% (p<0.001). Hence the difference in caries prevalence between areas with the highest proportion of inhabitants with low education, compared to areas with the lowest proportion, was 7pp higher in 2019 compared to 2010. In 2010, the unadjusted OR for caries experience in areas with the highest proportions of inhabitants with low education was 5.5, in comparison to areas with the lowest proportions, and 6.6 in 2019 (p =0.41).

Figure 4. Caries prevalence by residential area based on the proportion of inhabitants with low educational level

The overall aim of this paper was to study the development of 6-year-olds´ caries prevalence between 2010 and 2019 in relation to socioeconomic variables at family and residential area levels. The variables explaining the association between caries and socioeconomic variables were largely the same within each level and partly the same between the levels studied years. Despite changes in associations between studied years, the associations with caries prevalence and socioeconomic variables remained significant both years. The overall results indicate that socioeconomic variables on both levels continued to be associated with caries in 6-year-olds and the associations were consistently stronger for severe caries experience (dmft >3) than moderate (dmft 1-3), gradually increasing with higher levels of socioeconomic vulnerability.

When adjusting for confounders in the multivariable analysis, the models at both levels contained several explanatory variables in the association with caries, and children having multiple risk variables had even higher likelihood of caries experience. At the family level, the explanatory socioeconomic variables were maternal age when the first child was born, maternal age when child in the study group was born, parental employment, migration background of both parents and children, type of housing, and financial assistance (2010 only); and at area level migration background, parental educational level and number of persons per household.
The ORs in the models were reduced compared to the univariable analysis. The results illustrates the complex relationship between the different socioeconomic variables and health.(42)

On the family level, maternal age when the first child was born explained most of the association for both years, and at the area level, mean parental age when having the first child was associated with caries in the univariable analysis. Maternal age at childbirth may not be a socioeconomic variable in and of itself, but is associated with maternal socioeconomic situation and health outcome in offspring as well as being a risk indicator for caries, as both our and previous studies have shown.(39, 43) In our study, maternal age <25 of the first child was associated with a higher likelihood of caries in the child and the associations were even stronger for mothers having their first child <20 years of age. In addition, maternal age when having the child in the study group also showed an increased likelihood of severe caries if the mother's age was <25 years, but also at >34 years at childbirth. This finding is supported by an earlier study from Sweden demonstrating a U-shaped association between caries in children and maternal age, where mothers <25 or >35 years at childbirth had an increased risk of caries in their children.(44) The fact that the U-shape is not observed for the age at the first child may be because the number of children in the family is also a predictor for caries, as caries risk increases with higher birth order among siblings; perhaps because this means that more children need to share the same limited parental resources.(45, 46)

At the area level, migration background explained most of the relationship in both years, followed by the educational variable. At the family level, children's and parents' migration backgrounds were associated with caries, as was maternal level of education. Our result is in line with an analysis performed by the National Board of Health and Welfare in Sweden of the oral health development in children of preschool age and the interaction between children's oral health and their social and demographic background between 2011 and 2019.(47) According to their analysis country of birth, education and income were the three factors most clearly associated with children's oral health. Country of birth had the strongest connection; children whose parents were born abroad had almost three times the risk of suffering from caries compared to children of native-born parents.

At the area level, living in areas with an increased average of number of persons per household was associated with decreased likelihood of caries. At the family level, however, more than five people per household were associated with an increased likelihood of caries in the univariable analysis, which, as mentioned, previous studies also identified.(44, 45) The relative inconsistency of this, and of other variables, between family and area levels, may be due to the heterogeneity within the areas. For example, in our study, areas with a higher average of number of persons per household also had an increased proportion of inhabitants owning their own house, which was associated with a reduced likelihood of caries in the univariable analysis. In areas with many houses, in addition to families with children, many elderly people also live alone in their houses. In socially disadvantaged areas, there is a large proportion of apartments, where, in addition to families, there are also many students and single households, which lowers the average number per household of the area. This heterogeneity within areas may be the reason why the results at area and individual levels diverges to some extent.

Composite indices describing multiple socioeconomic conditions in an area have been argued to be more stable and robust.(26, 48) In our study, the CNI was used as a composite socioeconomic index at the area level. CNI is a well-used composite index for resource allocation which takes into account the composition of the inhabitants in an area based on several constituent variables (proportions of inhabitants within the area aged +65 living alone, born abroad, unemployed, aged 16–64 years, single parent with children <18, person who moved into the area).(20, 49) In the univariable analysis, CNI stood out in terms of the likelihood of severe caries; however, when adding other variables, which were partly the ones that build up the CNI variable, CNI was not significant in the model.

The contextual social conditions in areas and the impact they have on the families living there—for example on children’s oral health—were highlighted in a recent study from the Netherlands.(50) After adjusting for individual and neighborhood characteristics neighborhood deprivation remained significantly associated with severe caries and low odds of visiting the dentist yearly in 6-years-olds. In the analysis of oral health development in children of preschool age made by the National Board of Health and Welfare in Sweden, however, no neighbourhood effects after adjusting for individual social factors were seen, suggesting family characteristics has greater impact than area on caries in 6-years-olds. (47) In our study, based on goodness of fit, the models on family level were better than the area level models at estimating caries prevalence. However, the National Board of Health and Welfare in Sweden also noted that in general, more children with an increased risk of caries live in vulnerable areas, as individual socioeconomic conditions are linked to an increased risk for caries; therefore, targeting efforts to these areas is a way to reach children at a greater risk of caries. Area level socioeconomic status is useful for measuring geographical inequalities, for distributing and evaluating health preventive efforts and as a proxy for individual SES; for example, where individual data is lacking due to confidentiality.(51) However, when using area level this way it is important to acknowledge the described heterogenicity within the areas and that the association between socioeconomic prerequisites and dental caries is at a population level, which is not the caries situation on the family level.(52)

On the family level, there was an increased proportion of caries-free 6-years-olds with low level of maternal education during the years studied. This improvement was not seen among the group of children with a foreign migration background. It is likely that the group of children with foreign migration background have had fewer opportunities to take part in the preventive measures offered by the region, since they have lived in the region for a shorter time. This improvement in oral health among children at potentially higher risk of tooth decay may indicate that the preventive measures offered by the region have reached children with a potentially higher risk of caries. However, during the same time period, there was an increased proportion of children with caries experience among children whose mothers’ highest attained education level was high school. This finding and the gradient in health found in our study, as well as in previous research, indicate the importance of oral health efforts being universal, to all, but also proportional, gradually increased with increased risk of disease.(4)

A repeated cross-sectional study comparing time trends in caries-prevalence for 15-year-old Danish adolescents concluded persisting associations between socioeconomic position and dental caries from 1995 up until 2013.(53) During that period, there was a steep decline in caries prevalence for all social groups in Denmark and a decrease in absolute inequality as the difference in caries prevalence between the groups decreased in both groups improving in terms of caries prevalence. However, relative inequality between different social groups had increased as the difference in likelihood of caries disease (OR) had increased. In our study, there was no general reduction in the prevalence of caries between years studied—rather the opposite. However, for maternal level of education at the family level, both absolute and relative inequalities had decreased, as the difference in the proportion of caries-free children and the likelihood of developing caries between children with the lowest and highest levels of maternal education had decreased. Despite these positive trends, no change in absolute or relative inequality (or the opposite) was observed for migration background at either level or for the educational variable at the area level. Furthermore, the association between caries prevalence in 6-year-olds and low level of education, migration background, and all the other variables included in the models (except financial assistance) continued to be significantly associated with caries at both levels during the period studied. The result is in accordance with the Lancet series on oral health, declaring persistent oral health inequalities disproportionately affect poorer and marginalised groups in society.(7, 54) Furthermore, as caries in early childhood is a risk factor of poor oral health, continued caries disease and need for dental care as an adult, childhood caries can affect an individual throughout life and established inequalities in health that persist over time.(55, 56) Oral health shares the same risk factors as other non-communicable diseases (NCD) closely linked to the social determinants of health—for example, diabetes and obesity—hence the development in caries prevalence in deciduous teeth may serve as a measurable indicator of these and for the development of health inequalities in society.(8, 12, 57, 58) In the Lancet series, Watt et al. argue that a fundamentally different approach than the current treatment-dominated approach is needed with upstream, midstream, and downstream policies and interventions targeting the underlying social determinants of health in joint action with the prevention of other NCDs.(54)

The increase in caries prevalence among 6-year-olds between 2010 and 2019 observed at the family level in the present study was also observed nationwide.(23) During the same period, the proportion of children and parents with foreign migration backgrounds had increased in our study. The proportion with caries experience in the groups with foreign respective native migrant background remained unchanged, but with a significantly larger proportion of 6-year-olds with caries experience in the group with a foreign migration background. As already described, caries prevalence is heavily skewed worldwide, ranged from 18.7%–53.2% between different countries, and Sweden is among the countries with an average better oral health in the population.(8) The increase in caries prevalence among 6-year-olds during the studied period can therefore partly be explained by the population with foreign migration background being larger in 2019 than in 2010. In the national in-depth analysis of oral health development in children up to 6 years of age, it was also assessed that the increased proportion of children with a migrant background could partly explain the increase in caries prevalence in 6-year-olds between 2013 and 2019. (47)

Since Region Östergötland is the fourth largest region in Sweden, with both rural and urban areas, and since regional epidemiological caries data follow the national development, the results are considered representative of the Swedish population and can also be generalised to other countries with similar conditions as, for example, Scandinavian countries with similar welfare systems.

Since this study was conducted, recent national data on caries prevalence has indicated that the negative trend in decrease of caries-free children seems to have broken after 2019. In 2021, the proportion of caries-free 6-year-olds was 75%.(23, 59) However, national data and research have shown that health inequalities are at risk of worsening after the COVID-19 pandemic, as the pandemic has affected individuals in vulnerable situations more than individuals in more favourable social conditions. The effects of the pandemic were both direct (disease and death), but also indirect through changes in living conditions, health habits, and through reduced range of support and care opportunities in health care.(60, 61) The pandemic’s impact on health over time remains to be evaluated, but overall, this indicates that the risk of continued inequality remains.(62)

Strengths and weaknesses

A strength of this study was the coverage of the study population, where more than 94% of relevant individuals were included in the study. Reporting epidemiological caries data is mandatory in the remuneration system, contributing to the comprehensive coverage. However, there is a risk of bias, as 6-year-olds excluded from this study due to missing data may to a greater extent be children living in social vulnerability, as socially disadvantaged people are less likely to visit the dentist and are at greater risk of poorer oral health. This dropout may have affected the outcome of our study and underestimated the relationship between social and socioeconomic variables and caries.(14, 50)

Despite extensive national registrys, at SCB there were missing values for several socioeconomic variables on the family level. Missing data could be unfortunate, since there is an increased risk children living in social vulnerability could be overrepresented in this group. This was avoided by using multiple imputation to compensate for missing values at the family level. However, no multiple imputation was performed on area level as the only variables with missing values were paternal (n=463) and maternal (n=498) age when having the first child, and this was due to no first-born children were born in the areas studied years.

Socioeconomic variables can be strongly correlated with each other, which can cause multicollinearity that affects the results and make it difficult to distinguish the individual effects of the included variables in the analysis. To avoid multicollinearity in the present study, VIF analyses were performed, and if two variables were highly correlated the variable with the least association with caries was removed.

The composition and limitations of the area may affect the results and a low number of inhabitants may have caused uncertainty in the results. Neither health nor socioeconomic conditions follow administrative boundaries, and as described, this heterogeneity within areas can hide unequal conditions between, for example, socioeconomic groups within cities. NYKO4 was chosen as the definition of smaller areas based on the assumption that smaller areas expose more of the socioeconomic differences in society compared to larger areas. NYKO4 has the advantage of areas being defined and managed by the municipalities, which are well familiar with the social conditions and are rather small to expose more of the socioeconomic disparities.

Identifying inequalities is important to make them visible and thereby provide the conditions to even them out. This study adds a longitudinal perspective on children’s oral health and the relationship between caries in children and socioeconomic conditions within families and areas.However, despite efforts directed to families and areas at greater risk of caries, and minor signs of improvement noted for children with low maternal level of education, the overall results indicate that socioeconomic variables on both family and area levels have continued to be associated with caries in 6-year-olds in Sweden. Associations were consistently stronger for severe caries experience than for moderate, and were gradually increasing, with higher levels of social or socioeconomic vulnerability. Multiple socioeconomic risk variables meant even greater likelihood of caries. For instance, the OR for severe caries for children with maternal age when first child was born <20 years, and with foreign maternal background was 6.0 compared to children with maternal age of 25-34 years and native maternal background. The results can be used to identify families with a higher caries risk in the individual risk assessment or to identify areas useful for targeted interventions. Future oral health strategy must be supportive of families in socially vulnerable contexts to even out health disparities and the persisting oral health inequalities indicate a different approach, targeting the social determinants of health, is needed. In addition, it is important to continue to monitor oral health based on socioeconomic conditions to ensure good and equal oral health. These persistent oral health disparities in caries risk follows children as they grow and affects both their oral and general health, cementing health inequalities.

Ethics approval:

The study received approval from the Ethical Review Board 201101 (No. 2020-05054). Before SCB retrieved and linked the socioeconomic information to the caries prevalence, they conducted an internal confidentiality and review process. At the area level, values for fewer than four inhabitants were randomised according to Statistics Sweden’s routine.

Consent for publication:

Not applicable

Availability of data and materials:

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests:

The authors declare that they have no competing interests.

Funding:

This study was supported by Public Dental Service Östergötland and Region Östergötland.

Authors' contributions:

All authors contributed substantially to the design of the work, CB and JL performed the analysis, and all authors contributed in the interpretation of data, CB drafted the work and all authors substantively revised it and approved the submitted version.

Acknowledgements:

We would like to thank Kerstin Aronsson for her diligent work in mapping children's oral health in relation to socioeconomic conditions in the region, which laid the foundation for the present study. Thanks also to Magnus Husberg, who contributed extensive knowledge and work in structuring the data files from Statistics Sweden. We would also like to thank Region Östergötland and Public Dental Service Östergötland for the opportunity to conduct this study.

Black D, Townsend P. N D. Inequalities in health: the Black report. Harmondsworth; 1982.
Mackenbach JP, Stirbu I, Roskam AJ, Schaap MM, Menvielle G, Leinsalu M, et al. Socioeconomic inequalities in health in 22 European countries. N Engl J Med. 2008;358(23):2468–81.
WHO. Global Status Report on noncommunicable diseases 2014. https://www.who.int/publications/i/item/9789241564854 Access Date: 2024-04-19; 2014.
Marmot M, Hlt CSD. Achieving health equity: from root causes to fair outcomes. Lancet. 2007;370(9593):1153–63.
WHO. A Conceptual Framework for Action on the Social Determinants of Health. https://www.who.int/publications/i/item/9789241500852 Access Date: 2024-04-19; 2010.
Grad FP. The Preamble of the Constitution of the World Health Organization. Bull World Health Organ. 2002;80(12):981–4.
Peres MA, Macpherson LMD, Weyant RJ, Daly B, Venturelli R, Mathur MR, et al. Oral diseases: a global public health challenge. Lancet. 2019;394(10194):249–60.
WHO. Global oral health status report: towards universal health coverage for oral health by 2030. https://www.who.int/publications/i/item/9789240061484 Access Date: 2024-04-19; 2022.
Kassebaum NJ, Smith AGC, Bernabe E, Fleming TD, Reynolds AE, Vos T, et al. Global, Regional, and National Prevalence, Incidence, and Disability-Adjusted Life Years for Oral Conditions for 195 Countries, 1990–2015: A Systematic Analysis for the Global Burden of Diseases, Injuries, and Risk Factors. J Dent Res. 2017;96(4):380–7.
Anil S, Anand PS. Early childhood caries: prevalence, risk factors, and prevention. Front Pediatr. 2017;5:157.
Andre Kramer AC, Skeie MS, Skaare AB, Espelid I, Ostberg AL. Caries increment in primary teeth from 3 to 6 years of age: a longitudinal study in Swedish children. Eur Arch Paediatr Dent. 2014;15(3):167–73.
Ruiz B, Broadbent JM, Murray Thomson W, Ramrakha S, Boden J, Horwood J et al. Is childhood oral health the 'canary in the coal mine' for poor adult general health? Findings from two New Zealand birth cohort studies. Community Dent Oral Epidemiol. 2022.
Fejerskov O, Nyvad B, Kidd E. Dental caries. The disease and its clinical management. 3rd ed. Oxford: Wiley-Blackwell; 2015.
Splieth CH, Steffen H, Welk A, Schwahn C. Responder and nonresponder analysis for a caries prevention program. Caries Res. 2005;39(4):269–72.
Schou L, Wight C. Does dental health education affect inequalities in dental health? Community Dent Health. 1994;11(2):97–100.
Davidson T, Blomma C, Bagesund M, Krevers B, Vall M, Warnberg Gerdin E et al. Cost-effectiveness of caries preventive interventions - a systematic review. Acta Odontol Scand. 2020:1–12.
Featherstone JDB, Crystal YO, Alston P, Chaffee BW, Domejean S, Rechmann P, et al. A Comparison of Four Caries Risk Assessment Methods. Front Oral Health. 2021;2:656558.
Meijer M, Engholm G, Grittner U, Bloomfield K. A socioeconomic deprivation index for small areas in Denmark. Scand J Public Health. 2013;41(6):560–9.
Hosseinpoor AR, Bergen N. Area-based units of analysis for strengthening health inequality monitoring. Bull World Health Organ. 2016;94(11):856–8.
Ostberg AL, Kjellstrom AN, Petzold M. The influence of social deprivation on dental caries in Swedish children and adolescents, as measured by an index for primary health care: The Care Need Index. Community Dent Oral Epidemiol. 2017;45(3):233–41.
Koch G, Helkimo AN, Ullbro C. Caries prevalence and distribution in individuals aged 3–20 years in Jonkoping, Sweden: trends over 40 years. Eur Arch Paediatr Dent. 2017;18(5):363–70.
Swedish National Board of Health. and Welfare in Sweden. [Caries in children and adolscents. Epidemiological data for 2010]. www.socialstyrelsen.se Access Date: 2024-04-09; 2011 2011-12-08.
Swedish National Board of Health. and Welfare in Sweden. [Caries in children and adolscents. Epidemiological data for 2019]. www.socialstyrelsen.se; 2021 2021-03-18.
[The Dental Care Law (1985:125)]. (1985).
Guarnizo-Herreno CC, Watt RG, Pikhart H, Sheiham A, Tsakos G. Socioeconomic inequalities in oral health in different European welfare state regimes. J Epidemiol Community Health. 2013;67(9):728–35.
Andre Kramer AC, Petzold M, Hakeberg M, Ostberg AL. Multiple socioeconomic factors and dental caries in swedish children and adolescents. Caries Res. 2018;52(1–2):42–50.
Schwendicke F, Dorfer CE, Schlattmann P, Foster Page L, Thomson WM, Paris S. Socioeconomic inequality and caries: a systematic review and meta-analysis. J Dent Res. 2015;94(1):10–8.
Hou F, Myles J. Neighbourhood inequality, neighbourhood affluence and population health. Soc Sci Med. 2005;60(7):1557–69.
Swedish National Board of Health and Welfare in Sweden. [Social differences in dental health in children and young people. Background report to Children´s and young people´s health and care 2013]. https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/artikelkatalog/ovrigt/2013-5-34.pdf Access Date: 24-04-09; 2013.
Andre Kramer AC, Pivodic A, Hakeberg M, Ostberg AL. Multilevel Analysis of Dental Caries in Swedish Children and Adolescents in Relation to Socioeconomic Status. Caries Res. 2019;53(1):96–106.
Lundberg O. The next step towards more equity in health in Sweden: how can we close the gap in a generation? Scand J Public Health. 2018;46(22suppl):19–27.
WHO. Draft Global Oral Health Action Plan. (2023–2030). https://www.who.int/publications/m/item/draft-global-oral-health-action-plan-%282023-2030%29 Access Date: 2024-04-19; 2022.
Statistic Sweden (SCB). Statistical database 2023. https://www.statistikdatabasen.scb.se/pxweb/en/ssd/START__BE__BE0101/. Access Date 2023-01-24 [.
Mastrovito B, Blomma C, Johansson A, Borgstedt-Risberg M. [Follow-up on oral helath in children and young people in the region of Östergötland 1994–2019]. Linköping: Region Östergötland, Centrum för verksamhetsstöd och utveckling; 2020.
Aronsson K, Mako E. [Dental health in children and adolescents in Östergötland municipalities 2010] Dental health report. Linköping: Folkhälsocentrum. Region Östergötland; 2011.
Klein H, Palmer C, Knutson J. Studies on Dental Caries:1. Dental Status and Dental Needs of Elementary School Children. Public Health Rep. 1938;53(1):751–65.
WHO. Oral health surveys: basic methods – 5th edition. https://www.who.int/publications/i/item/9789241548649 Access Date: 2024-04-09; 2013 Access date: October 18, 2021.
Statistic Sweden (SCB). [Statistics for small areas with keycode system (NYKO)] https://www.scb.se/contentassets/4d5516f7c4504a669fdf242136eacfee/meromnyko-.pdf Access Date: 2024-04-19: Statistics Sweden; https://www.scb.se/en/ [.
Julihn A, Soares FC, Hjern A, Dahllof G. Socioeconomic determinants, maternal health, and caries in young children. JDR Clin Trans Res. 2018;3(4):395–404.
Donders AR, van der Heijden GJ, Stijnen T, Moons KG. Review: a gentle introduction to imputation of missing values. J Clin Epidemiol. 2006;59(10):1087–91.
Eekhout I, van de Wiel MA, Heymans MW. Methods for significance testing of categorical covariates in logistic regression models after multiple imputation: power and applicability analysis. BMC Med Res Methodol. 2017;17(1):129.
Eckersley R. Beyond inequality: Acknowledging the complexity of social determinants of health. Soc Sci Med. 2015;147:121–5.
Saloojee H, Coovadia H. Maternal age matters: for a lifetime, or longer. Lancet Glob Health. 2015;3(7):e342–3.
Soares FC, Dahllof G, Hjern A, Julihn A. U-shaped association between maternal age at delivery and dental caries in offspring. Acta Odontol Scand. 2020;78(8):565–71.
Julihn A, Soares FC, Hammarfjord U, Hjern A, Dahllof G. Birth order is associated with caries development in young children: a register-based cohort study. BMC Public Health. 2020;20(1):218.
Downey DB. Number of siblings and intellectual development. The resource dilution explanation. Am Psychol. 2001;56(6–7):497–504.
Swedish National Board of Health and Welfare in Sweden. [Oral health development among children in preschool age; The interaction between children's oral health and their social and demographic background]. https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/artikelkatalog/ovrigt/2022-6-7991.pdf Access Date: 2024-04-19; 2022.
Bajekal M, Jan S, Jarman B. The Swedish UPA score: an administrative tool for identification of underprivileged areas. Scand J Soc Med. 1996;24(3):177–84.
Sundquist K, Malmstrom M, Johansson SE, Sundquist J. Care Need Index, a useful tool for the distribution of primary health care resources. J Epidemiol Community Health. 2003;57(5):347–52.
van Meijeren-van Lunteren AW, Oude Groeniger J, Wolvius EB, Kragt L. Neighbourhood characteristics and children's oral health: a multilevel population-based cohort study. Eur J Public Health. 2021;31(4):742–8.
Galobardes B, Lynch J, Smith GD. Measuring socioeconomic position in health research. Br Med Bull. 2007;81–82:21–37.
Piantadosi S, Byar DP, Green SB. The ecological fallacy. Am J Epidemiol. 1988;127(5):893–904.
Sengupta K, Christensen LB, Mortensen LH, Skovgaard LT, Andersen I. Trends in socioeconomic inequalities in oral health among 15-year-old Danish adolescents during 1995–2013: A nationwide, register-based, repeated cross-sectional study. Community Dent Oral Epidemiol. 2017;45(5):458–68.
Watt RG, Daly B, Allison P, Macpherson LMD, Venturelli R, Listl S, et al. Ending the neglect of global oral health: time for radical action. Lancet. 2019;394(10194):261–72.
Isaksson H, Alm A, Koch G, Birkhed D, Wendt LK. Caries prevalence in Swedish 20-year-olds in relation to their previous caries experience. Caries Res. 2013;47(3):234–42.
Peres MA, Peres KG, de Barros AJ, Victora CG. The relation between family socioeconomic trajectories from childhood to adolescence and dental caries and associated oral behaviours. J Epidemiol Community Health. 2007;61(2):141–5.
Watt RG, Mathur MR, Aida J, Bonecker M, Venturelli R, Gansky SA. Oral Health Disparities in Children: A Canary in the Coalmine? Pediatr Clin North Am. 2018;65(5):965–79.
G B D Risk Factors Collaborators. Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016;388(10053):1659–724.
Swedish National Board of Health and Welfare. [Caries in children and adolscents. Epidemiological data for 2021] 2022-5-7906. https://www.socialstyrelsen.se/en/; 2022.
The Public Health Agency of Sweden. [Public Health in Sweden. Annual Report 2023]. https://www.folkhalsomyndigheten.se/contentassets/a448b27d603c44f590fc1aff741b0d5d/folkhalsan-sverige-arsrapport-2023.pdf Access Date: 2024-04-19; 2023.
Stennett M, Tsakos G. The impact of the COVID-19 pandemic on oral health inequalities and access to oral healthcare in England. Br Dent J. 2022;232(2):109–14.
Swedish National Board of Health and Welfare in Sweden. [Effects of COVID-19 on oral health and dental visits among children and Adults – Part 4 ]. https://www.socialstyrelsen.se/globalassets/sharepoint-dokument/artikelkatalog/ovrigt/2022-5-7887.pdf Access Date: 2024-04-19; 2022. Contract No.: 2022-5-7887.

Tables 3 and 5 are available in the Supplementary Files section.

No competing interests reported.

Table3.docx
Table 3. The multivariable models for 2010 and 2019´s association between caries and socioeconomic variables (family level)
Appendix1.docx
Appendix2a.docx
Appendix2b.docx
Appendix3.docx
Appendix4.docx
Table5.docx
Table 5. The multivariable models for 2010 and 2019 association between caries and socioeconomic variables at area level

Download PDF

Editorial decision: Revision requested
08 Oct, 2024
Reviews received at journal
07 Oct, 2024
Reviews received at journal
02 Oct, 2024
Reviews received at journal
27 Sep, 2024
Reviewers agreed at journal
16 Sep, 2024
Reviewers agreed at journal
11 Sep, 2024
Reviewers agreed at journal
09 Sep, 2024
Reviewers invited by journal
09 Sep, 2024
Editor invited by journal
22 Aug, 2024
Editor assigned by journal
21 Aug, 2024
Submission checks completed at journal
21 Aug, 2024
First submitted to journal
21 Aug, 2024

You are reading this latest preprint version

Persistent oral health inequality in children—repeated cross-sectional studies in 2010 and 2019

Status:

Version 1

Abstract

Figures

Background

Methods

Study group

Data collection

Analyses and statistics

Results

Discussion

Conclusions

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1