Perception of tooth color.

TOP ABSTRACT Tooth color distribution. Perception of tooth color. Statement of the problem. Objectives. SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

A unique study demonstrated that perceived tooth whiteness and attractiveness are influenced by the color of surrounding lips and gingivae.¹⁶ For perceived tooth whiteness, the most popular image chosen had a magenta hue added to the gingivae.

	Statement of the problem.

TOP ABSTRACT Tooth color distribution. Perception of tooth color. Statement of the problem. Objectives. SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Advertisements imply that white, bright teeth are more attractive than yellow teeth, and that exceptionally white teeth are normal.¹⁷ As a result, the demand for esthetic dentistry has increased during the last decade.¹⁸ A number of methods are available to improve the color of teeth, including whitening toothpastes, internal and external bleaching, microabrasion and placement of crowns and veneers.¹⁴

Tooth whitening is the most popular esthetic dental treatment, which is aided by the ease of at-home bleaching and the increased availability of over-the-counter tooth whitening products.¹⁹ However, are all skin colors equally attractive with bright, white teeth? Previous studies investigating the influence of tooth color on dental attractiveness have included only white subjects.^1,4,5,20 To date, no study, to our knowledge, has investigated this effect on a variety of skin colors. Reno and colleagues¹⁶ showed that modifying lip and gum color influenced respondents’ perceptions of tooth whiteness and attractiveness. Would variations in surrounding skin color have the same effect?

	Objectives.

TOP ABSTRACT Tooth color distribution. Perception of tooth color. Statement of the problem. Objectives. SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The principal aim of this study was to determine, for a given tooth shade value, whether variations in skin color influenced perceptions of smile attractiveness.

In addition, we wanted to

– compare the perceptions of laypeople and dentists;

– assess the effects of participant’s age and sex on attractiveness perception ratings;

– assess whether the brightest tooth shade value was considered the most attractive for all skin tones.

	SUBJECTS, MATERIALS AND METHODS

TOP ABSTRACT Tooth color distribution. Perception of tooth color. Statement of the problem. Objectives. SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Subjects. We included two groups of subjects in this study: 70 dentists (36 female) and 70 laypeople (49 female). The dentist group consisted of general dental practitioners and dental hospital residents. We excluded orthodontists and prosthodontists. Laypeople consisted of teachers, business owners and others, none of whom were affiliated with the dental profession. We gave participants as little information as possible about the study, and no one was aware of the purpose of the study. We obtained informed consent from all participants.

Study design. Questionnaire. One of us (R.S.S.) gave each participant a questionnaire in which we asked for his or her age, sex and ethnic group. In addition, we asked participants to rate their dental appearance ("very unattractive," "unattractive," "attractive" or "very attractive") and the importance of an attractive smile ("very unimportant," "unimportant," "important" or "very important"). We asked dentists to state the number of years they had been in practice.

Image attractiveness. We selected from a digital archive a photograph of a woman’s smile that exhibited good dental alignment and tooth-size symmetry. The photograph showed the teeth, lips and surrounding skin, but we cropped out the nose and chin to reduce the number of confounding variables. One of us (F.B.N.) manipulated the photographic image digitally (Adobe Photoshop CS2 software, Adobe Systems, San Jose, Calif.) to create a range of images with varying skin colors and tooth shades.

The investigator (F.B.N.) altered the skin and lip shade to create four tones (fair, fair/medium, medium/dark and dark). He varied the tooth shade value (brightness) to range from 00 (brightest) to 05 (darkest). The result was 24 images. The investigator duplicated one of the images to assess intraexaminer reliability; hence, subjects viewed a total of 25 images (Figure 1).

View larger version (107K): [in this window] [in a new window]   Figure 1. Smile images produced by computer manipulation.

View larger version (58K): [in this window] [in a new window]   Figure 2. Example of an image viewed by study participants during the slide show presentation.

Statistical analysis. One hundred forty participants yielded 80 percent power to detect an effect of size 0.25 (small effect size) at the 5 percent significance level. The questionnaire responses were descriptive and consisted of frequencies, means or medians where appropriate. The main outcome was smile attractiveness ratings of the images. We used mixed linear regression to model the rating of attractiveness for each tooth shade value in terms of skin color and all other explanatory factors (such as age or sex). We tested intraexaminer reliability by using the intraclass correlation coefficient (ICC). We analyzed data by using statistical software (Stata 9, STATA, College Station, Texas).

	RESULTS

TOP ABSTRACT Tooth color distribution. Perception of tooth color. Statement of the problem. Objectives. SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Descriptive statistics. Demographic data for the subjects, including sex, age, ethnic group and dentists’ experience are shown in Table 2. The proportion of women in the group of laypeople was significantly higher than the proportion in the dentists’ group (P = .02). The results show no significant difference in mean age between the two groups (P = .09) and no significant difference in the proportion of white subjects.

View larger version (49K): [in this window] [in a new window]   Figure 3. Participants’ responses to the question "How would you rate the attractiveness of your dental appearance?"

View larger version (47K): [in this window] [in a new window]   Figure 4. Participants’ responses to the question "How important do you think it is to have an attractive smile?"

Attractiveness ratings. The attractiveness ratings for all images are shown in Figure 5. Dentists and laypeople gave the lowest rating for image BM (that is, fair skin, darkest teeth [05 tooth shade value]). Dentists gave the highest rating to the duplicated image FM and OV (fair/medium skin, 02 tooth shade value), while laypeople rated image GR (fair/medium skin, 01 tooth shade value) highest.

View larger version (82K): [in this window] [in a new window]   Figure 5. Attractiveness ratings for all images. Bars indicate mean (± standard deviation).

Tooth shade 01. For tooth shade 01, respondents rated the image of fair/medium skin higher in comparison with the images of fair skin (P < .001), medium/dark skin (P = .01) and dark skin (P < .001). White subjects rated the images higher in comparison with other ethnic groups (P = .04).

Tooth shade 02. For tooth shade 02, the results show no significant association between skin color and respondents’ ratings (P = .16) in the multiple mixed linear regression model.

Tooth shade 03. For tooth shade 03, subjects rated the image of fair skin lower in comparison with the images of fair/medium skin (P < .001), medium/dark skin (P < .001) and dark skin (P < .001). They rated the image of dark skin lower than that of medium/dark skin (P = .01).

Tooth shade 04. For tooth shade 04, subjects rated the image of fair skin lower in comparison with the images of fair/medium skin (P < .001), medium/ dark skin (P < .001) and dark skin (P < .001). They rated the image of fair/medium skin higher in comparison with the images of medium/dark skin (P < .001) and dark skin (P < .001). White participants rated the images higher in comparison with other ethnic groups (P = .01).

Tooth shade 05 (darkest). For tooth shade 05, respondents rated the image of fair skin lower in comparison with the images of fair/medium skin (P < .001), medium/dark skin (P < .001) and dark skin (P < .001). They rated the image of fair/medium skin lower in comparison with the images of medium/dark skin (P < .001) and dark skin (P < .001).

In general, attractiveness ratings increased with increasing participant age. Moreover, dentists rated the images higher in comparison with laypeople, and women gave higher ratings than did men. The images that were rated "natural" received higher ratings than those rated "too dark" or "too bright."

	DISCUSSION

TOP ABSTRACT Tooth color distribution. Perception of tooth color. Statement of the problem. Objectives. SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Judgments regarding a person’s characteristics are influenced by his or her dental appearance.²¹ With the increasing interest in esthetic dentistry, greater knowledge is required regarding perceptions of dental appearance. Scientific data regarding the relationship between variations in skin color and tooth shade values in terms of perceived attractiveness are limited. Such information would be valuable for the dentist, both in the selection of dentures and anterior restorations and in deciding the degree of tooth whitening required to complement the patient’s facial skin tone. Previous studies¹³ examining the effect of tooth shade on perceived attractiveness have included only white subjects. No study, to our knowledge, has investigated the influence of variations in skin color and tooth shades on perceived attractiveness.

Study design. Thomas and colleagues²² highlighted the problems associated with using full-face photographs for assessing attractiveness, as background stimuli may influence attractiveness ratings. In our study, we eliminated most of the face, including the nose and chin, from the image to reduce the number of confounding variables. Computer-aided image modification allowed us to maintain the exact same image, apart from changes in skin color and tooth shade value.

In addition, we standardized the conditions for rating the images. All participants viewed the images in the same sequence and were not allowed to confer with each other regarding their ratings or revisit previously viewed images.

The 10-point attractiveness scale (created by F.B.N.) allowed participants to express their perception of relative attractiveness by assigning a numerical value to each image. We deemed this to be more appropriate than using a visual analog scale, which is a continuous scale in which raters might not commit to a decision and all of their ratings could be the same. Raters also may use certain portions of the scale and ignore others, particularly the extremes.²³

The results of this study indicate that we used a reliable method, with an overall ICC of 0.66. Agreement was better among the group of dentists (0.81) than it was among the laypeople (0.55).

Questionnaire responses. Comparison by participant group. Sixty-five dentists (93 percent) rated their own dental appearance as either attractive or very attractive, as did 42 laypeople (60 percent) (Figure 3); 28 laypeople (40 percent) rated themselves as unattractive or very unattractive. These differences were statistically significant (P < .001). These findings validate those of a previous population survey, which revealed that 37 percent of laypeople were dissatisfied with their dental appearance.²

One hundred thirty-two participants (94 percent) considered an attractive smile to be important or very important (Figure 4). However, eight laypeople (11 percent) considered it unimportant or very unimportant compared with only one dentist (1 percent) who considered it so. These differences are statistically significant (P = .04). Our results are inconsistent with those of Carlsson and colleagues,⁵ who found that dentists and non-dentists placed a similar importance on dental appearance.

Comparison by sex. We found no differences between the responses of men and women to the two questions (P = .26 for the first question, P = .27 for the second question). These results agree with those of Carlsson and colleagues,⁵ who reported no significant differences between men and women with regard to the importance of dental appearance. In contrast, a survey by Vallittu and colleagues³ revealed that dental appearance was more important to women than to men. Samorodnitzky-Naveh and colleagues² reported that more women than men were satisfied with their dental appearance, but more women than men had undergone tooth-whitening procedures.

Role of skin color. Table 3 reveals the principal finding of our study: variation in skin color for a given tooth shade value influenced perceived smile attractiveness. This was highly significant for tooth shade values 00, 01, 03, 04 and 05 (P < .001), but not for tooth shade value 02 (P = .16). However, we found tooth shade value 02 to be significant in the simple mixed linear regression model.

For the brightest tooth shade (00), fair, fair/medium and medium/dark skin images were rated 0.51, 0.49 and 0.32 higher (that is, the higher the coefficient, the higher the attractiveness rating), respectively, than the dark skin image (P < .01) (Table 3). For the dark skin image (LJ), 92 percent of participants stated that the teeth were "too bright." This perception may be explained by the optical illusion resulting from the greater contrast between skin color and tooth shade.²⁴

For images in the 01 group, subjects rated the fair/medium skin higher (more attractive) than the other skin colors (P < .01) (Figure 5). One hundred ten (79 percent) of the 140 participants responded that the appearance of the teeth was "natural."

As the brightness of teeth decreased (tooth shade values 03, 04 and 05), subjects rated fair skin images lowest among the four skin tones (P < .001) (Figure 5). For fair skin images (MK, OU and BM), 89 percent, 93 percent and 97 percent of participants, respectively, reported that the teeth appeared "too dark." For the darkest tooth shade value (05), participants preferred the medium/dark and dark skin images to fair and fair/medium skin images (P < .001) (Figure 5). Of all 24 images, image BM (fair skin, tooth shade value 05) received the lowest overall score on the 10-point scale (mean score, 3.37; range, 3.03 to 3.70). Again, the difference in perceived attractiveness may be explained by the optical illusion resulting from the greater contrast between skin color and tooth shade.²⁴

Because this study involves original research, it is not possible for us to compare our results directly with those of other studies. However, Reno and colleagues¹⁶ reported that perceived attractiveness and perceived tooth whiteness were influenced by the color of the surrounding lips and gums. Our study findings show that for most tooth shades, surrounding skin color influenced perceptions of tooth brightness and attractiveness. Reno and colleagues¹⁶ also found that images with the whitest teeth were not always judged to be the most attractive. Our study results validate this finding, as the most attractive images selected by dentists and laypeople were not in the group with the brightest tooth shade value. Dentists gave the highest ratings to the duplicate images FM and OV, which belonged to the 02 group, and laypeople chose image GR, which belonged to the 01 group (Figure 1).

Clinical implications. Our study findings are relevant to today’s popular tooth whitening market, as well as to the prosthodontic rehabilitation of the anterior dentition. Subjects did not perceive the brightest tooth shade to be the most attractive, and they did not perceive all skin colors to be equally attractive with bright white teeth. They perceived dark skin with bright white teeth and fair skin with dark teeth to be relatively unattractive (Figure 5).

The findings of this study may be particularly significant to patients requesting tooth whitening procedures. For example, clinicians may advise dark-skinned patients requesting very bright white teeth of the perceived unattractive and perhaps unnatural appearance.

Explanatory factors. Sex. Men rated images with tooth shade values 01, 02, 03, 04 and 05 lower than did women (P < .05) (Table 3). The results show no statistically significant difference between men and women with regard to images in the 00 group. We can deduce that as tooth brightness decreases, men become more critical in their ratings. This finding corroborates those of Grosofsky and colleagues.²⁰ Other studies have shown that women tend to give higher ratings of attractiveness than do men.^25,26

Bar-Tal and Saxe²⁷ suggested that physical attractiveness is a more important factor—among both men and women—in evaluating women than men. Men regard physical attractiveness as an important determinant in mate selection; consequently, they are more critical of women’s physical attractiveness than women are of men’s.²⁷ It would be interesting to compare female and male images in a future study.

Age. Attractiveness ratings for images with tooth shade values 01, 02, 03, 04 and 05 increased by 0.02, 0.02, 0.03, 0.03 and 0.03, respectively, with an increase in the participant’s age (P < .05) (Table 3). This suggests that as subjects grow older, they become more accepting of darker teeth. Studies have shown that the lightness of teeth decreases with age and teeth become yellower at cervical sites.^6–8,11 The participants in our study may have been less critical in their ratings as they see changes in their own tooth color with age. Our results are similar to those of Vallittu and colleagues,³ who found that as the age of patients increased, the perception that "very white teeth are the most beautiful teeth" decreased. In addition, their survey results revealed that younger patients expressed a greater preference for white teeth than did older patients.

Dentists versus laypeople. Images with tooth shade values 01, 02, 03, 04 and 05 were rated 0.50, 0.94, 0.96, 0.92 and 1.10 higher, respectively, by dentists than by laypeople (P < .05) (Table 3). We found no statistically significant difference between the groups for tooth shade value 00 (P = .29). This suggests that laypeople are less tolerant of darker tooth shades than are dentists.

Differences in esthetic perceptions among dentists and laypeople have been examined in the literature. In general, studies have found dentists to be more sensitive in identifying deviations from an ideal appearance than are laypeople.^22,28–31

Our findings are consistent with those of Wagner and colleagues⁴ and Carlsson and colleagues,⁵ who found that nondentists had a stronger preference for white teeth than did dentists. Dunn and colleagues¹ also reported that tooth shade was one of the most important variables in assessing smile attractiveness among the general population.

Future studies might include variations in lip morphology to assess the influence of morphological ethnic differences on perceptions of the relationship between tooth shade value and skin tone.

	CONCLUSIONS

TOP ABSTRACT Tooth color distribution. Perception of tooth color. Statement of the problem. Objectives. SUBJECTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The results of this study show that variations in skin color for most tooth shade values influenced perceived smile attractiveness. For the brightest tooth shade (00), subjects rated the dark skin image lowest in comparison with the other three skin colors. As the brightness of teeth decreased, subjects rated fair skin images most poorly. One possible explanation for these findings is the optical illusion resulting from the greater contrast between the skin color and tooth shade value.

Dentists and laypeople did not perceive the brightest tooth shade as the most attractive, and they did not perceive all skin colors as equally attractive with bright white teeth. Men gave lower attractiveness ratings than did women, and attractiveness ratings increased with an increase in participants’ ages. Laypeople were more critical of tooth shade than were dentists, as they perceived the darker tooth shades to be less attractive.

Dentists need to educate patients requesting tooth whitening procedures or alternative esthetic enhancement of the anterior teeth about the possibility that excessively bright white teeth may be perceived as too bright depending on their skin tone.