Malocclusion is any incorrect position of the teeth that results in imperfect contact between mandibular and maxillary teeth (1). According to a US study (2), only 40% of Americans have ideal overjet, 50% ideal incisor overlap and 40% ideal incisor irregularity. The prevalence of malocclusion in the US between 1930 and 1935 was variously estimated as 35% to 95%, depending on investigators’ differing criteria for “how much the deviation from the ideal” is acceptable. Orthodontic treatment globally consists in moving and straightening teeth, to align and level them and to correct the bite (3), and thus achieve a long-lasting balanced occlusion. Nonetheless all malocclusions do not need to be corrected (4). There is no evidence of an association between malocclusions and dental caries (5, 6). Whether malocclusion has a negative effect on periodontal health remains both unclear (7, 8) and controversial (9, 10). Several reviews have concluded that there is no evidence of an improvement in periodontal health associated with orthodontic treatment (10–12). A recent 20-year cohort study in Wales (13), intended to assess the effects of orthodontic treatment on oral health, provided evidence that lack of orthodontic treatment is not associated with psychological difficulties in later life (13, 14) and that orthodontic treatment neither causes nor prevents temporomandibular disorders (15). But it has been suggested that some severe malocclusions should be reduced to enhance the longevity of the dentition (7, 16–18). French guidelines recommend orthodontic treatment when oral and nasal functions are impaired, anterior teeth are exposed to trauma, incorrect tooth alignment increases plaque accumulation risk, or poor aesthetics is an issue for the patient (4). Most orthodontic treatment takes place around puberty (approximately from the age of 12 to 15 years) (19), in young adult dentition.

In France, the cost of orthodontic treatment is not fixed. National health insurance covers only a part of the cost, 20% on average (20, 21), and then only if treatment starts before the patient’s 16^th birthday and for a maximum of 3 years. In addition to national health insurance coverage, most families have supplementary private health insurance policy that can reimburse all of the balance but most often covers only some of it. Families with extremely low income have supplementary government insurance, which is supposed to ensure total coverage of the cost of orthodontic treatment, which is fixed for these families at 928 euros per year. Otherwise, costs range approximately from 1000 to 3000 euros a year.

In France as elsewhere, there are social inequalities in oral health care (22, 23). Expenditures are closely related to socio-economic status: managers spend much more on oral health care than manual workers do (22). The particularity of the French system for oral health lies in the distinction between conservative care on the one hand and prosthetic and orthodontic care on the other. The national health insurance offers relatively good coverage for conservative care, which is low-priced whereas its coverage for the much more costly prosthetic and orthodontic care is poor (24). There remain social disparities in conservative care, despite the coverage by the national health insurance of 70% of its cost. Because orthodontic treatment is covered much less completely, we suppose that the social and economic inequalities are greater in orthodontic treatment than in conservative care.

In European and American studies, the prevalence of orthodontic treatment ranges from 10 to 35% and is lower in less privileged groups (2, 25–27). A study describing inequalities in orthodontic treatment in Northern Ireland where orthodontic treatment is free for teenagers showed that several factors were associated with orthodontic treatment including regular dentist visits and good oral hygiene, but did not find any relation between socio-economic factors and orthodontic treatment (28). An English study showed that discontinuation of orthodontic treatment was more frequent among children in less privileged social classes (29). In France, two studies estimated that 11% of schoolchildren aged in average 10–11 years and 27% of schoolchildren aged in average 14–15 years had orthodontic treatment (26, 27). In both studies, the treatment rate was twice as high among managers’ as among manual workers’ children. Economic issues do not appear to explain fully this disparity (22, 28). Education and culture may play a specific role (30, 31), as may accessibility of care. To our knowledge, the specific roles of all these dimensions on orthodontic treatment have not been explored.

The aim of this study was to estimate the prevalence of orthodontic treatment in France among children and teenagers aged 8 to 18 years, by sex and by age, and to investigate the specific role of social and economic characteristics on use of orthodontic treatment.

Table 1 describes the characteristics of the study population; boys accounted for 51% of the study sample. Most children (74%) lived in standard two-parent families and were covered (86%) by private supplementary insurance; 17% lived in a rural area, 15% in a district with a low density of orthodontists and 25% in a district with a high density of orthodontists.

The prevalence of orthodontic treatment was 14% for all children aged 8 to 18 years, 23% for those 12 to 15, and 9% for the youngest and oldest age groups. Orthodontic treatment was slightly more frequent for girls (15%) than for boys (13%) (table 1).

The rate of orthodontic treatment was significantly lower when there were 4 children or more in the household, when the mother or the stepmother was born in Africa, when the parents were service or sales workers or manual workers, when the child had no supplementary insurance, or when the family lived in a rural area (table 1). The orthodontic treatment rate was higher when parents were professionals or managers, had a high income, or when the child lived in an area with a high density of orthodontists (table 1). The frequency of orthodontic treatment did not differ according to the family structure.

The first multivariate model included family social status but not maternal educational attainment (table 2). Children with a mother or a stepmother born in Africa were less likely to have orthodontic treatment (aOR=0.61; 95%CI: [0.44;0.85]) compared with children with mothers or stepmothers born in France. Family social status was associated with orthodontic treatment: compared with children whose parents were managers or professionals, children were less likely to have orthodontic treatment if their parents were service or sales workers (aOR=0.50; 95%CI: [0.34;0.76]), manual workers (aOR=0.56; 95%CI: [0.39;0.81]), or office workers, self-employed, farmers, or shopkeepers, (aOR=0.73; 95%CI: [0.56;0.96]). There was a gradient in the association between the annual income and orthodontic treatment. When the annual income of the family was low, children were less likely to have orthodontic treatment (aOR=0.62; 95%CI: [0.45;0.85]) than children of families with high income. Children without supplementary insurance were less likely to have orthodontic treatment (aOR=0.53; 95%CI: [0.34;0.81]) than children covered by private supplementary insurance. Children who lived in rural areas were less likely to have orthodontic treatment (aOR=0.70; 95%CI: [0.54;0.91]) than those who lived in urban areas. No significant association was found between orthodontic treatment and density of orthodontists (table 2).

The second multivariate model included maternal educational attainment but not family social status (table 2). The aORs showed a gradient in the relation between educational attainment (from no or primary education to university) and orthodontic treatment. Children whose mother had the least education were less likely to have orthodontic treatment (aOR=0.59; 95%CI: [0.43;0.81]) compared with children whose mother or stepmother obtained a university diploma. The other aORs were very similar to those described in table 2.

We conducted a sensitivity analysis to verify that our results for the entire sample were consistent with those for the children aged 12 to 15 years, which is the common age for orthodontic treatment. It showed pretty much the same associations as for the sample as a whole, but the aORs were closer to 1, and the confidence intervals were larger (results not shown). The results of the GEE taking into account the intra-family correlation are very close to those of the multivariate regression analysis and thus indicate that the non-independence of the data of children from the same family does not affect the results (not shown).

The multivariate analysis excluded 283 children because of missing information on the place of birth of mother or stepmother (138 lived with their father and no mother or stepmother), family social status, or supplementary insurance. Twelve percent of these excluded children had orthodontic treatment.

This study of a large national sample of the population living in France explored the relations between social and economic characteristics and orthodontic treatment. The prevalence of orthodontic treatment was 14% for children aged 8 to 18 years and 23% for those aged 12–15. Orthodontic treatment was mainly associated with family social status, maternal educational level, family income and supplementary health insurance. Orthodontic treatment was also associated with sex, number of children in the household, the place of birth of mother or stepmother, and the residence area.

As it was a cross-sectional survey, we have a snapshot of orthodontic treatment at the time of the survey. So we may assume that the prevalence of orthodontic treatment was correctly estimated.

The 47 children with incomplete questionnaires on health were excluded. Only 5% (283) of the children were excluded from the multivariate analysis for missing data on covariates. These exclusions are thus unlikely to have modified the estimated associations.

The observed prevalence of orthodontic treatment, 23% among children aged 12 to 15 years, and 9% among those aged 8 to 11, is not very different to the prevalence observed in other French studies in 2004 (26, 27), considering that the age groups are not similar. These findings are also consistent with those of the international literature. In Europe, 28% of teenagers in Northern Ireland (15–16 years old) had or were about to have orthodontic treatment (28). In the NHANES III study of a representative sample of the US population conducted between 1988 and 1991, 30% of white Americans aged 12 to 17 years had orthodontic treatment.

As in other international and French studies, we found that the prevalence of orthodontic treatment was slightly higher among girls than boys (25, 26, 28, 35). Nonetheless, the literature does not show any sex difference in the need for orthodontic treatment (25, 36–39). Requests for aesthetic reasons are probably higher among girls, and parents might be more attentive to the appearance of their daughters than to their sons. Girls may also be more concerned about their oral health than boys, and may go more easily and regularly to the dentist. They would be thus more likely to accept orthodontic treatment. That was suggested in one study of American adults (40).

Orthodontic treatment was associated with maternal level of education, family social status, income and supplementary insurance. In other words, in equal income groups, and with equal supplementary coverage, family social status or maternal education still influenced orthodontic treatment. This treatment costs approximately € 1000 to € 3000 a year. Our data provide no information about the amount covered by private insurance, but it can vary from € 100 to € 1000 a year, according to the policy. The group of children with private insurance is therefore probably heterogeneous, mixing children with excellent coverage and those with basic insurance. The association between orthodontic treatment and income is therefore not surprising but it is interesting to estimate it, especially when taking into account supplementary insurance and social status. Social status is clearly linked to income (as well as maternal education) but the persistence of the association after adjustment for income probably reflects the influence of the social environment and maternal education on orthodontic treatment.

As every country has its own health care system (41), comparisons between countries are difficult. Nevertheless, social differences in orthodontic treatment have been observed in the United States, where both treatment and insurance are expensive (2, 42), and in France (26, 27), but also in England, where treatment is free for teenagers (25, 29, 43, 44).

Children whose mother or stepmother was born in Africa had orthodontic treatment at a lower rate than those whose mother or stepmother was born in France. This is consistent with a German study that reported that immigrant status of either parent or child was associated with a lower prevalence of orthodontic treatment in a population with low socio-economic status (30). Women born in Africa may have been less exposed to orthodontic treatment than their French counterparts and thus may be less aware of its potential benefits.

Orthodontic treatment was less frequent in rural areas, possibly because longer distances to reach an orthodontist may be an obstacle to orthodontic treatment, which requires regular visits. However, there was no significant association between orthodontic treatment and density of orthodontists, although such a link has been observed in England (31). In our study, density of orthodontists was calculated as a mean density in the district of residence, but districts are large administrative divisions and density may therefore vary substantially within the district.

This study dealt with orthodontic treatment but we had no information about either the type of malocclusion or the need for treatment. Orthodontic treatment needs seem to vary between populations, for instance they have been estimated at around 21% in a population aged 9 to 12 in France (37), and 35% among 12 year-olds in England (25). Both European and American studies found that immigrant background (30, 45), lower income (2), and lower socio-economic group (31, 46) are associated with higher orthodontic treatment need. The same factors thus appear to be associated with both a higher rate of orthodontic treatment need and a lower rate of orthodontic treatment, and the actual social inequalities are probably even greater than that we observed.

Several mechanisms may explain these social differences: fewer spontaneous requests for orthodontic treatment from less privileged families, fewer dental visits, thus fewer occasions for need assessment, and when need is identified, either a more restrictive attitude by dentists to recommending treatment for patients from low income families because of its cost or more refusals from these families for the same reason. Exploring the respective role of each of these mechanisms could help trying to bring needs and treatment closer together.

This study revealed social inequalities in orthodontic treatment in France. Our findings point to the importance of assessing orthodontic treatment needs and of improving access to orthodontic treatment; to reach the goal of equal opportunity for oral health.