Tooth erosion became a topic of great interest in the last decade of the 20th century, with increasing attention being paid in scientific papers to its prevalence in children, and as part of the aetiological triad in tooth surface loss in adults.

Prevalence studies of tooth erosion in the UK began in 1993 with the National Survey of Children's Dental Health,1 and subsequently they have varied from reports of small convenience samples2 and local population studies,3,4,5,6,7 to national surveys.1,8,9 The age of the children examined has varied from 1.5 to 18 years and some of the studies have examined the difference in prevalence of erosion between different groups within the sample. The 1993 Survey of Children's Dental Health recorded that 52% of 5-year-olds and 27% of 12-year-olds had erosion into enamel.1 The 2000 National Diet and Nutrition Survey (NDNS) of young people aged 4 to 18 years,9 showed that 58% of 4–6-year-olds and 42% of 11–14-year-olds were affected by dental erosion. Two local studies of 12-year-olds recorded erosion present in 44.8%2 and 57%10 of children. The British Association for the Study of Community Dentistry (BASCD) has been unable to agree upon a suitable index for the recording of tooth erosion.

There is some evidence that socio-economic status has an influence on the prevalence of erosion, although this is not conclusive. The majority of studies suggest that males have a higher prevalence of tooth erosion than females, but all have failed to investigate any difference between ethnic groups. Further investigations are required to extend the limited information on the prevalence of tooth erosion in the general population, and to establish differences between groups in an attempt to distinguish those in greatest need, thereby enabling preventive programmes to be properly targeted.

The aim of the present study was to assess the prevalence of tooth erosion in a random sample of 12-year-old children resident in Leicestershire and Rutland, and to consider if there were differences between genders, ethnic groups, levels of material deprivation, and association with caries experience.

Materials And Methods

The community dental service in Leicestershire was due to complete a dental health survey of 12-year-old children in 1997 as part of the BASCD national epidemiology programme. Approximately 2,000 of the 10,500 children attending all 62 secondary schools across the county were randomly selected for inclusion in the survey, with 1,753 attending for examination at their school. This produced a representative sample of the mixed ethnic groups in the population. In addition to recording full BASCD data, each child was examined for tooth erosion. The Leicestershire Research Ethics Committee gave approval for this extension of the BASCD survey, and consent was sought from the parents/guardians of each child.

Every child was examined fully reclined on a seven position garden chair, under standardized illumination from a Daray 4000 series Versatile Dental light. The recorder noted the gender, ethnic group and home postcode, and checked the date of birth of each child before the BASCD data, including caries experience, were recorded. Subsequent to this, tooth erosion was assessed on the labial and palatal surfaces of upper and lower incisors, and on the buccal, occlusal and lingual surfaces of first molars. The index for evaluating tooth erosion was the same as that used in the study of Children's Dental Health in the UK 1993.1 Codes were as follows:

0 – Normal enamel 1 – Loss of surface enamel characteristics 2 – Loss of enamel sufficient to expose dentine 3 – Loss of enamel and dentine resulting in pulp exposure 9 – No assessment possible

All examinations were completed by the same individual (CD), an experienced examiner who had previously used the same tooth erosion criteria when participating in the 1993 study. Training and calibration for the dental health criteria were completed in the regional BASCD (TC1296) exercises, and training in the use of the erosion index was obtained from concurrent participation in the NDNS study of young people aged 4–18 years.9 A calibration exercise with 120 children was used to assess intra-examiner reproducibility, which produced an un-weighted Kappa of 0.80.

The socio-economic background of the children was assessed by allocating a Townsend score based on the postcode of each child.12 Subjects with a Townsend Deprivation Index between −5.2 and −1.1 were graded as advantaged, those between −1 and 1.99 as intermediate and those in the range 2 and above as deprived, indicating material deprivation. The Townsend index performs well in explaining variations in true health measures by measuring material deprivation.11,12 Data were stored, collated and analysed using Survey Plus computer software. Qualitative data were analysed using the Chi-square test, and quantitative data were analysed using Student's t-test. Statistical significance was accepted at the 95% confidence level, P<0.05.


The sample consisted of 906 (51.7%) boys and 847 (48.3%) girls with a mean age of 12.51 (SD 0.28) years. Of these, 1,379 (78.7%) were Caucasian and 316 (18.0%) were Asian, with both ethnic groups having similar proportions of boys and girls. The mean DMFT was 1.03, with 31% of children having active decay, and 68% lifetime decay experience.

Tooth erosion was recorded in 1,046 (59.7%) children, with 47 (2.7%) exhibiting exposed dentine and no child having erosion exposing the pulp. Significantly more males (63.9%) than females (55.3%) (P<0.01) had tooth erosion, as did Caucasian (62.7%) compared with Asian (48.1%) children, (P<0.01) (Table 1). Significantly more children with caries experience also had erosion present (66%) compared with those who had no caries experience (54.9%) in their permanent dentition. Conversely children with erosion had a higher mean DMFT score of 1.17 compared with the 0.83 of those with no erosion (P<0.01).

Table 1 The prevalence of tooth erosion in the sample

Surface data revealed that tooth erosion was symmetrical about the midline and uncommon on lower incisors (13.5%). Erosion occurred most frequently on the palatal aspects of upper incisors (49%) and upper molars (53%), and the buccal surface of lower molars (50%). Dentine was exposed to the greatest extent (2.2%) on the occlusal surfaces of lower molars with less than 1% of the palatal aspects of central incisors exhibiting exposed dentine.

Analysis of Townsend groupings shows that the results indicate that the majority of children did not live in conditions of deprivation (Table 2). The proportions of Caucasian to Asian children differed greatly in each deprivation category; the relationship between ethnicity, deprivation and erosion experience is shown in more detail in Table 3. Different levels of deprivation had no significant effect on erosion experience for the total sample or for Asian children. However, a lower proportion of Caucasian children in the low deprivation group had tooth erosion than did those in the two less deprived groups, P<0.05.

Table 2 Numbers and proportions of children by Townsend deprivation score
Table 3 Tooth erosion prevalence by Townsend deprivation score


Twelve-year-old children were selected for the present study for several reasons. At this age the index teeth should have been present in the mouth for approximately six years, primarily under aetiological influences experienced in the home environment. Twelve-year-olds are usually more co-operative during epidemiological studies than older children and there are fewer refusals. Also this age group could be identified and revisited at a later date whilst still at school.

Prevalence of tooth erosion has been reported in ten studies in the UK (Table 4). Different indices have been employed, different teeth and surfaces have been examined, and results have been variously presented. Comparability between studies is thus made difficult. Several studies4,6,10,13 used modifications of the Tooth Wear Index of Smith and Knight,14 whilst others2,3,5,7 used assessments based on that used in the national children's study of 1993.1 Sample sizes have varied from 1252 in a single school, to 17,061.1 The prevalence of erosion in 3 year olds was found to be 29% by both Jones and Nunn3 and Hinds and Gregory8 although the former study identified a slightly higher proportion of subjects with erosion into dentine; 17% compared with 14%. Several studies have assessed erosion in children around the age of 5 years. In the 1993 national study1 52% of subjects had erosion present, with 24% having exposed dentine. Taylor5 found that 98% of 5-year-olds in North Warwickshire had erosion but in the latter study both deciduous molars were included in the assessment. Walker et al.9 scored only primary molars and found 58% of 4–6-year-olds to have erosion, with dentinal involvement present in 19%. Millward et al.4 found that 48% of 4–5-years-olds had erosion into dentine but in this study all deciduous teeth were included. Prevalence of erosion in 12-year-olds has been reported as 2.7%, with 1% of subjects having exposed dentine,1 whilst Deery et al.2 found 45% with loss of enamel, but none with dentine exposed. This compares with the 60% and 27% respectively in the present study, which included assessment of first molars. Other studies of erosion on permanent teeth in children have found 52% and 57% respectively to have erosion of enamel surfaces.9,10 In the NDNS study of 11–14-year-olds, 3% had dentine exposure.9

Table 4 Prevalence studies of tooth erosion / wear in children resident in the United Kingdom

In the present study boys had a significantly higher prevalence of erosion than girls, which confirms the findings of several authors,5,6,13,15 although others9,10 found no difference between the genders. Adult studies of tooth wear,16,17,18,19,20 of which erosion is an integral part in combination with attrition and abrasion, have found that wear is present in significantly more men than women, indicating that trends established in youth continue into adulthood.

The link between ethnic group and erosion prevalence has not previously been recorded. The present study revealed that a significantly higher proportion of Caucasian (62.7%) than Asian (48.1%) children had erosion present (P<0.01), with this reflected in the respective proportions of subjects with dentine exposed. Exposure to potential aetiological factors and different modifying factors may explain this.

In eight studies which investigated the association between socio-economic status and tooth erosion or tooth wear in children (Table 5), five different indices have been employed to determine socio-economic status/deprivation. This reflects the indices available to different authors at the time of their research. Overall this present study found no difference between socio-economic groups, which supports other findings.5 Several studies have found a lower level of tooth erosion in the high socio-economic group, as was the case for Caucasian children in the present study.3,6,13,21 Conversely Millward et al.4 recorded a higher prevalence in higher socio-economic groups. Walker et al.9 reported that although there was no difference between the socio-economic groups in 4–6 and 7–10-year-olds, erosion prevalence was higher in lower socio-economic groups of 11–14 and 15–18-year-olds. As there is no clear association established to date, this should be further investigated, preferably using agreed socio-economic and erosion indices, with consideration being given to the ethnic make up of the population investigated.

Table 5 Socio-economic status and prevalence of tooth erosion

A significantly higher proportion of 12-year-old children with caries experience (66.0%) had erosion present than those with no caries experience (54.9%, p<0.01). Only one study has previously considered this association, when it was found that a slightly higher proportion of 2–5-year-olds in Saudi Arabia with caries also had erosion present (33%), compared with caries free individuals (27%), the difference was not significant.22 In 12-year-olds, it may be that those individuals who fail to care for their teeth by failing to maintain a potentially non cariogenic diet, also fail to maintain a potentially non erosive diet.

The present study demonstrates a high prevalence of erosion in 12-year-olds, although the proportion of children with exposed dentine was less than 3%. Boys had significantly more erosion than girls, and a significant difference in the prevalence of tooth erosion was observed between Caucasian and Asian children. No difference in the prevalence of erosion between socio-economic groups was apparent in the overall sample. This was however influenced by the ethnic proportions of each group, as a significantly lower proportion of the least deprived Caucasian children had tooth erosion. There was a positive relationship between caries experience and erosion. However, since 68% of children had caries involving dentine whilst only 2.7% had erosion into dentine, it may be concluded that tooth erosion in this age group is not a serious public health problem at the present time. Prevalence studies completed to date in the UK reveal a lack in unity of approach and reporting. It is desirable to rectify this in order to achieve improved clarity and comparability of results in future. Further research is required to monitor prevalence and establish clearer associations between erosion and socio-economic status, ethnic groups and other variables, to enable resources to be targeted towards preventive programmes. Commencement of data collection in BASCD co-ordinated studies would be of immense benefit, as a unified and consistent approach would be established, allowing valid comparability of nationwide results.


Of a random sample of 12-year-old Leicestershire children, 59.7% had suffered dental erosion, with 2.7% having exposed dentine. Erosion was significantly more prevalent in boys than girls, in Caucasian than Asian children and in subjects with caries experience. Social deprivation had no overall effect on erosion experience, although socio-economically advantaged Caucasian children had significantly less erosion than other groups.