Age estimation of an Indian population by using the Kim's scoring system of occlusal tooth wear.

CONTEXT: Age is one of the prime factors employed to establish the identity of an individual and the use of teeth for this purpose has been considered reliable. Tooth wear is widely accepted as a physiological consequence of aging and evaluation of tooth wear can be a simple and convenient tool to estimate age in adults. AIMS: The present study was conducted to record the degree of tooth wear among Indian adults and to estimate their ages from the degree of tooth wear based on Kim's scoring system.
MATERIALS AND METHODS: Dental stone casts of 120 participants were used to assess the degree of occlusal tooth wear based on the criteria given by Kim et al. STATISTICAL ANALYSIS USED: The age of all subjects was estimated based on these scores using multiple regression analysis function.
RESULTS: The degree of tooth wear showed a significant positive correlation with age in each and every examined tooth of both males and females. The predicted age was within ± 5 years of actual age in 70% of males and 68.3% females, and within ± 3 years of actual age in 50% of males and 50.1% of females.
CONCLUSIONS: Kim's scoring system has proven to be a useful tool in estimation of age using occlusal wear in an Indian population with a high level of accuracy in adults.

Introduction

Age is one of the prime factors employed to establish the identity of an individual and the use of teeth for this purpose has been considered reliable. Tooth wear is widely accepted as a physiological consequence of aging and evaluation of tooth wear has been frequently used as a tool to estimate age of adults.[123] There is ample evidence that the average age at which patients are becoming edentulous is rising sharply, and there is optimism that the problem of extensive caries in young people and periodontal disease may be declining. As patients keep their teeth longer and have them restored less, the problems associated with the dentition wearing out will assume greater importance.[4]

Age estimation of living persons might be required in such cases that chronological age plays a crucial role, e.g., criminal responsibility, school attendance, social benefits, and so on.[5] Tooth wear may thus be used as a reliable marker for estimating the age of such individuals. Tooth wear may be dependent on various factors such as eating and chewing habits, hardness of the teeth, the type of food and method of mastication, existence of artificial teeth, gender, geographic location, environmental conditions and parafunctions.[56789] Despite this, estimation of an individual's age from tooth wear is a very simple and a convenient method. This method does not need any invasive process such as tooth extraction or tissue preparation. However, a low level of accuracy is a factor to limit its usefulness. The purpose of this study was to utilize the scoring system proposed by Kim et al. as a method of recording the degree of tooth wear and determining the individual's age on an Indian population.

Materials and Methods

The study sample constituted 120 participants, 60 males and 60 females in the age group of 13-70 years, reporting to the Department of Oral Medicine and Radiology, J.S.S. Dental College and Hospital, Mysore, India. The participants were subdivided into six age groups in each of the male and female category with 10 individuals in each group [Table 1]. The individuals having non-carious, intact and satisfactorily aligned maxillary and mandibular permanent, posterior teeth were included in the study. Sixteen permanent posterior teeth (first and second premolars and first and second molars namely, 14, 15, 16, 17, 24, 25, 26, 27, 34, 35, 36, 37, 44, 45, 46, and 47 Fédération Dentaire Internationale system) were selected to examine the degree of tooth wear. Tooth wear of anterior teeth (from canine to canine) were disregarded because of very wide individual variations in their angles and rotations. Complete arch maxillary and mandibular impressions were made. Following the guidelines of Kim's scoring system; the degree of tooth wear for each tooth was visually evaluated from the dried casts using a magnifying lens, under good illumination.

Table 1

Age and gender distribution of the study subjects

Tooth wear score was classified into a 0-to-8 point scale based on the pattern and amount of tooth wear on the occlusal surface as proposed by Kim's scoring system [Table 2]. Tooth wear was evaluated by two different criteria. One is the area of tooth wear, which may be termed as the horizontal factor and the other is the degree of dentine exposure, which may be termed as the vertical factor. Combination of both the horizontal and vertical factors was considered to obtain an accurate score. Scoring was performed by two examiners and the intra-examiner variability was tested by blinding the results. The inter-examiner variability was assessed by asking the two examiners to score all the 120 casts individually.

Table 2

Kim's new scoring system

These scores were then subjected to statistical analysis in four steps as follows:

Means and standard deviation of tooth wear scores for each tooth were calculated for males and females separately and the t-test was used to examine gender difference.

Correlation analysis was done to show the correlation between the degrees of tooth wear of all the teeth.

Univariate regression analysis of age against tooth wear score for each tooth was done to examine the relationship between the ages and tooth wear score. The linear equation was derived, which was Y = aX + b, where Y = estimated age in years, X = tooth wear score, a = intercept, b = regression coefficient, r2 = coefficient of correlation.

Multiple regression analysis was adopted to predict the age using the tooth wear scores of 16 teeth in each and every individual. The SPSS package was used to derive the β-coefficient. The multiple linear regression function provided a linear equation which would predict the dependent variable with the number of independent variables. In the present study the age was the dependent variable and all 16 variables (tooth wear scores) were taken as independent variables. The linear equation for estimating the age was:

where Y = Estimated age, a = Intercept, β = Coefficient, and x = Tooth wear score.

Results

On comparing the values, it was observed that the wear scores of molars were consistently more than the scores of the premolars. The mean wear scores of the first premolars were found to be more than the scores of the second premolars. Similarly the mean wear scores of the first molars were found to be higher than the scores of the second molars. Also the mean wear scores of every individual tooth of the males were found to be higher than those of the females of the group, except for those in group VI. Statistically significant difference between the tooth wear scores of males and females was observed to vary from group to group [Tables 3–8].

Table 3

Mean values and Standard deviation of tooth wear score for each tooth and levels of statistical significance between tooth wear scores of Males and Females of Group I (13-20 years)

Table 8

Mean values and Standard deviation of tooth wear score for each tooth and levels of statistical significance between tooth wear scores of Males and Females of Group VI (61-70 years)

The correlation analysis showed that the degrees of correlation were relatively strong between the tooth wear scores of all examined teeth. The coefficient of correlation (r2) values were between 0.807 and 0.920 [Table 9].

Table 9

Correlation coefficients (r2) observed in multiple regression function between age and tooth wear

The univariate regression analysis of age against tooth wear score were done to examine the relationship between the age and tooth wear score [Table 10]. The ranges of coefficient of correlation (r2) were between 0.65 and 0.81 in males and 0.75 and 0.85 in females.

Table 10

Univariate regression analysis of age with tooth wear score for each tooth

The multi-variant analysis was carried out to predict the age by using the 16 variables (tooth wear scores of 14, 15, 16, 17, 24, 25, 26, 27, 34, 35, 36, 37, 44, 45, 46, and 47). The intercept and β-coefficients are stated in Table 11. The values were calculated based on the tooth wear scores for each individual, separately for males and females. The multiple linear regression function provided a linear equation which would predict the dependent variable with the number of independent variables. In the present study age was the dependent variable and all 16 variables (tooth wear scores) were taken as independent variables. The coefficient of correlation (r2) was very high for the males (r2 = 0.959) and females (r2 = 0.971) [Table 12].

Table 11

The intercept and ß coefficients observed for the multiple regression analysis

Table 12

Correlation coefficients (r2) observed in multiple regression function between age and tooth wear

The predicted age values for each individual were estimated using multiple linear regression function. It was observed that the predicted age was within ±5 years of the actual age in 70% of males and 68.3% of females. It was also observed that the predicted age was within ±3 years of the actual age in 50% of the males [Table 13] and 50.1% of the females and within ±2 years of actual age in 36.6% of males and 33.3% of females [Table 14]. For example, the actual age of case 1 of group I of males was 15 years and the estimated age (Y) was calculated using the linear regression formula as follows:

Table 13

Actual age and predicted ages of males derived through multiple regression analysis

Table 14

Actual age and predicted ages of females derived through multiple regression analysis

14.8637 = 3.489 (+) −0.546x (2)2 + 0.391x (1)2 + 0.914x (2)2 (+) −3.652x (0)2 (+) −0.245x (2)2 (+) −0.565x (2)2 (+) 2.448x (3)2 (+) −2.194x (1)2 (+) 2.584x (1)2 (+) 1.741x (1)2 (+) 0.762x (5)2 (+) 2.976x (1)2 (+) 2.434x (1)2 (+) 0.919x (1)2 (+) −0.715x (5)2 (+) 2.977x (1)2.

When the total of 120 subjects was divided into two groups, the accuracy of estimation of age increased. In the group less than or equal to the age of 40 years, the age could be estimated within ±5 years in 23 male subjects (76.6%) and 22 females (73.3%). In the group above 40 years, the estimated age was within ±5 years of actual age in 19 males (66.3%) and 23 females (76.6%). It was further noted that in the group ≤40 years, the age could be estimated within ±3 years in 17 male (566%) and 17 female subjects (56.6%) and within ±2 years in 11 males (36.6%) and 9 females (30%). Whereas, in the group above 40 years, the age could be estimated within ±3 years in 13 males (43.3%) and 14 females (46.6%) and within ±2 years in 11 males (36.6%) and 11 females (36.6%) [Table 15 and Graph 1].

Table 15

Accuracy of age estimation

Graph 1

Comparision of actual age and predicted age of all subjects

Discussion

Occlusal tooth wear is one of the degenerative changes related to age, and increased levels of tooth wear with age have been consistently reported in the literature.[3561011121314151617] The process of occlusal tooth wear may be dependent on various factors such as eating and chewing habits, hardness of the teeth, the type of food and the degree of pressure transmitted by the jaws to the teeth during mastication, presence or absence of teeth in the opposing arch, existence of artificial teeth, gender, geographic location, environmental conditions, parafunctions like bruxism and malocclusion. Hence, it has been suggested by various studies that age estimation based on occlusal tooth wear alone may not be accurate.[5789] However, in cases where age estimation has to be carried out inevitably with limited information, especially in living persons over the age of 20-30 years, occlusal tooth wear scoring may be used as an effected tool. This has been demonstrated to be accurate and reliable by the use of Kim's scoring system.[35] Our study also focuses on the scoring system itself in the manner of selecting limited number of sound teeth to prove its validity in an Indian population.

It was observed that the degree of tooth wear score increased with the age in each and every tooth in both males and females. This fact agrees with the results of Kim et al.[3] and Ekfeldt et al.[13] In group I of the present study the first premolars showed greater wear compared to the second premolars, which agrees with the findings of Kim et al. In their study the first premolars of all four quadrants consistently showed increased wear compared with that of the second premolars.[3] However, in the rest of the groups of the present study (groups II, III, VI, V, and VI), the second premolars showed more wear than the first premolars, contradictory to the findings of Kim et al. There are a number of factors which could contribute to this finding. Firstly, it is well understood that the occlusal forces increase from the anterior to posterior teeth as we move distally.[1819] The second premolar is placed distal to the first premolar in the arch and may thus show more amount of wear. Secondly, the second premolar is positioned adjacent to the first molar, on which the maximum occlusal forces are exerted during grinding of food. This position of the second premolar may cause it to take more active part in the grinding of food than does the first premolar. In addition, the maxillary second premolars occlude partly with the mandibular first molars which exert more forces during the grinding of food.[18] This could be another reason for amount of occlusal force being exerted on this tooth, attributing to more wear. However, further studies are warranted to observe this pattern of wear and confirm this finding.

It was observed in the present study that the molars in general showed more wear when compared with the premolars in all the groups. The first molars of all the groups showed more degrees of wear when compared with the second molars which can be attributed to the eruption pattern of permanent teeth. Teeth that erupt early are exposed to more physiological wear with age. This finding was consistent with the study findings of Kim et al.[3] The molars are more bulky, have larger occlusal tables and greater anchorage and receive greater occlusal forces when compared with the premolars.[1819] Thus irrespective of the pattern of eruption the molars tend to show more wear than the premolars.

In the present study when all the males and females irrespective of their age groups were considered together, it was found that the males had greater degree of tooth wear than the females. This is in accordance with Kim et al.,[3] Seligman et al.,[11] and Donachie and Walls[12] who observed that males show better development of masticatory muscles, muscle fiber mass and stronger ligaments than females and thus males could exert a stronger biting force than females. A general observation (in five of the six groups) was that, males had higher tooth wear scores than the females of the same group in the present study, but this difference was statistically significant with respect to only few teeth. In case of group VI (61-70 years) although the females of showed more wear compared with the males of the group, only a few teeth (6 out of 16) showed statistically significant difference.

In groups IV, V, and VI, it was seen that the tooth wear scores of females were very close to the scores of the males. Keeping the above observations in mind, it could possibly be inferred that as the age advances the wear scores of males and females parallel each other.

The tooth wear scoring system implemented in the present study was similar to the method used by Kim et al. The regression analysis for the whole data was carried out and multiple regression function provided a linear equation which was used to predict the ages of all the subjects. The age could be calculated by adding the intercept to the sum of the β-coefficients multiplied by the individual tooth wear scores. In the present study, in 70% of males and 68.3% of females, the age could be estimated within ±5 years of the actual age and in 50% of the males and 50.1% of the females the predicted age was within ±3 years of the actual age. On further observation in 36.6% of males and 33.3% of females, the age could be estimated within ±2 years of actual age. It was further observed that in 31.3% males and 41.6% females, the age could be estimated within ±2 years of actual age.

It was also noted that the accuracy of estimation of age increased when the total of 120 subjects was divided into two groups. In the group less than or equal to 40 years of age, in 23 male subjects (76.6%) and 22 females (73.3%), the age could be estimated within ±5 years of actual age. In the group above 40 years, in 19 males (66.3%) and 23 females (76.6%) the estimated age was within ±5 years of actual age. It was further noted that in the group less than or equal to 40 years, in 17 male (56.6%) and 17 female subjects (56.6%) the age could be estimated within ±3 years and in 11 males (36.6%) and 9 females (30%) it could be estimated within ±2 years. Whereas, in the group above 40 years, the age could be estimated within ±3 years in 13 males (43.3%) and 14 females (46.6%) it was within ±2 years in 11 males (36.6%) and 11 females (36.6%).

Conclusion

Several scoring systems for the evaluation of degree of tooth wear have been presented in literature. However, there is no universally applicable tooth wear scoring system and moreover a low level of accuracy limits their usefulness. Kim's scoring system has proven to be a useful tool in estimation of age using occlusal wear in an Indian population with a high level of accuracy adults. It has been observed that once the individual is classified into the young or old age group, age can be estimated with a higher level of accuracy. We recommend that further studies using larger samples may be needed to observe the trend in various populations.

Table 4

Mean values and Standard deviation of tooth wear score for each tooth and levels of statistical significance between tooth wear scores of Males and Females of Group II (21-30 years)

Table 5

Mean values and Standard deviation of tooth wear score for each tooth and levels of statistical significance between tooth wear scores of Males and Females of Group III (31-40 years)

Table 6

Mean values and Standard deviation of tooth wear score for each tooth and levels of statistical significance between tooth wear scores of Males and Females of Group IV (41-50 years)

Table 7

Mean values and Standard deviation of tooth wear score for each tooth and levels of statistical significance between tooth wear scores of Males and Females of Group V (51-60 years)