Smile characterization by U.S. white, U.S. Asian Indian, and Indian populations

326 Volume 107 Issue 5

The Journal of Prosthetic Dentistry Sharma et alAleisa et al

rated materials were easily removed without additional dentin prepara-tion.15,18 Additionally, to eliminate the factor of the differences in the sizes of the teeth, the post spaces were pre-pared with large diameter and paral-lel ParaPost drills.12,18 In this way, post spaces were uniformly cylindrical throughout their length and were of the same dimensions.12,18

CONCLUSIONS

Within the limitations of this study, it was concluded that:

1. The type of root canal sealer had an effect on the bond strength of ce-mented fiber posts.

2. Eugenol-based sealers (Endofil and Tubli-seal) significantly reduced the bond strength of prefabricated fi-ber posts luted with resin cement.

3. Posts luted with Rely X Unicem res-in cement into root canals obturated with eugenol-based sealers exhibited higher bond strengths than those luted with Paracore and Variolink II resin cements.

REFERENCES

1. Morgano SM. Restoration of pulpless teeth: application of traditional principles in pres-ent and future contexts. J Prosthet Dent 1996;75:375-80.

2. Bateman G, Ricketts DN, Saunders WP. Fiber-based post systems: A review. Br Dent J 2003;195:43-8.

3. Qualtrough AJ, Chandler NP, Purton DG. A comparison of the retention of tooth-colored posts. Quintessence Int 2003;34:199-201.

4. Colley IT, Hampson EL, Lehman ML. Reten-tion of post crowns: An assessment of relative efficiency of posts of different shapes and sizes. Br Dent J 1968;124:63-9.

5. Johnson JK, Sakumura JS. Dowel form and tensile force. J Prosthet Dent 1978;40:645-9.

6. Standlee JP, Caputo AA, Hanson EC. Reten-tion of endodontic dowels: Effects of cement, dowel length, diameter, and design. J Prosthet Dent 1978;39:401-5.

7. Assif D, Ferber A. Retention of dowels using a composite resin as a cementing medium. J Prosthet Dent 1982;48:292-6

8. Chapman KW, Worley JL, von Fraunhofer JA. Retention of prefabricated posts by cements and resins. J Prosthet Dent 1985;54:649-52.

9. Boschian-Pest L, Cavalli G, Bertani P, Gagliani M. Adhesive post-endodontic restorations with fiber posts: Push-out tests and SEM observations. Dent Mater 2002;18:596-602.

10.Kurtz JS, Perdigao J, Geraldli S, Hodges JS, Bowles WR. Bond strengths of tooth-colored posts, effect of sealer, dentin adhesive, and root region. Am J Dent 2003;16:31-6.

11.Cohen BI, Pagnillo MK, Newman I, Musikant BL, Deutsch AS. Retention of three endodon-tic posts cemented with five dental cements. J Prosthet Dent 1998;79:520-5.

12.Al-Ali K, Talic Y, Abduljabbar T, Omar R. Influence of timing of coronal preparation on retention of cemented cast post and cores. Int J Prosthodont 2003;16:290-4.

13.Tjan AH, Nemetz H. Effect of eugenol-containing endodontic sealer on retention of prefabricated posts luted with an adhesive composite resin cement. Quintessence Int 1992;23:839-4.

14.Schwartz RS, Murchison DF, Walker WA. Ef-fects of eugenol and noneugenol endodontic sealer cements on post retention. J Endodon 1998;24:564-7.

15. Boone KJ, Murchison DF, Schindler WG, Walker WA. Post retention: The effect of sequence of post space preparation, cementa-tion time, and different sealers. J Endodon 2001;27:768-71.

16.Hagge MS, Wong RD, Lindemuth JS. Effect of three root canal sealers on the retentive strength of endodontic posts luted with resin cement. Int Endod J 2002;35:372-8.

17.Hagge MS, Wong RD, Lindemuth JS. Reten-tion strengths of five luting cements on prefabricated dowels after root canal obturation with a zinc oxide/eugenol sealer: 1. Dowel space preparation/cementation at one week after obturation. J Prosthodont 2002;11:168-75.

18.Al-Ali K. Effect of eugenol-based root canal sealers on retention of prefabricated metal posts luted with resin cement. Saudi Dent J 2009;21:69-73.

19.Demarchi MG, Sato EF. Leakage of in-terim post and cores used during labora-tory fabrication of custom posts. J Endod 2002;28:328-9.

20.Stewardson DA. Non-metal post systems. Dent Update 2001;28:334-6.

21.Asmussen E, Peutzfeldt A, Heitmann T. Stiff-ness, elastic limit, and strength of newer types of endodontics posts. J Dent 1999;27:275-8.

22.Hedlund SO, Johansson N, Sjogren G. Retention of prefabricated and individu-ally cast root canal posts in vitro. Br Dent J 2003;195:155-8.

23.Al-harbi F, Nathanson D. In vitro assessment of retention of four esthetic dowels to resin core foundation and teeth. J Prosthet Dent 2003;90:547-55.

24.Cohen BI, Musikant BL, Deutsch AS. Com-parison of the retentive properties of four post systems. J Prosthet Dent 1992;68:264-8.

25.Dejak B, Młotkowski A. Finite element analysis of strength and adhesion of cast posts compared to glass fiber-reinforced composite resin posts in anterior teeth. J Prosthet Dent 2011;105:115-26.

26.Chuang SF, Yaman P, Herrero A, Dennison JB, Chang CH. Influence of post material and length on endodontically treated incisors: an in vitro and finite element study. J Prosthet Dent 2010;104:379-88.

27.Albashaireh ZS, Ghazal M, Kern M. Effects of endodontic post surface treatment, dentin conditioning, and artificial aging on the reten-tion of glass fiber-reinforced composite resin posts. J Prosthet Dent 2010;103:31-9.

28.Bouillaguet S, Troesch S, Wataha JC, Krejci I, Meyer JM, Pashley DH. Microtensile bond strength between adhesive cements and root canal dentin. Dent Mater 2003;19:199-205.

29.Ferrari M, Vichi A, Grandini M. Efficacy of dif-ferent adhesive techniques on bonding to root canal walls: an SEM investigation. Dent Mater 2001;17:422-9.

30.al-Wazzan KA, al-Harbi AA, Hammad IA. The effect of eugenol-containing temporary cement on the bond strength of two resin composite core materials to dentin. J Prostho-dont 1997;6:37-42.

31.Schwartz R, Davis R, Hilton TJ. Effect of temporary cements on the bond strength of a resin cement. Am J Dent 1992;5:147-50.

32.Woody TL, Davis RD. The effect of eugenol-containing and eugenol-free temporary cements on microleakage in resin bonded restorations. Oper Dent 1992;17:175-80.

33.Ganss C, Jung M. Effect of eugenol-containing temporary cements on bond strength of com-posite to dentin. Oper Dent 1998;23:55-62.

34.Mayer T, Pioch T, Duscner H, Staehle HJ. Den-tinal adhesion and histomorphology of two dentinal bonding agents under the influence of eugenol. Quintessence Int 1997;28:57-62.

35.Dilts WE, Miller RC, Miranda FJ, Duncanson MG Jr. Effect of zinc oxide-eugenol on shear bond strengths of selected core/cement com-binations. J Prosthet Dent 1986;55:206-8.

36.Carvalho CN, Oliveira B Jr, Loguercio AD, Reis A. Effect of ZOE temporary restoration on resin-dentin bond strength using different adhesive strategies. J Esthet Restor Dent 2007;9:137-41.

37.Hagge MS, Wong RDM, Lindemuth JS. Retention of posts luted with phosphate monomer-based composite cement in canals obturated using a eugenol sealer. Am J Dent 2002; 15:378-82.

38.Abo-Hamar SE, Federlin M, Hiller KA, Friedl KH, Schmalz G. Effect of temporary cements on the bond strength of ceramic luted to dentin. Dent Mater 2005;21:794-803.

39.Davis ST, O’Connell BC. The effect of two root canal sealers on the retentive strength of glass fiber endodontic posts. J Oral Rehabil 2007;34:468-73.

40.Teixeira C, Pasternak-Junior B, Borges AH, Paulino SM, Sousa-Neto MD. Influence of endodontic sealers on the bond strength of carbon fiber posts. J Biomed Mater Res Part B: Appl Biomater 2008;84B:430-5.

41.Gatewood RS. Endodontic materials. Dent Clin North Am 2007;51:695-712.

42.Patierno JM, Rueggeberg FA, Anderson RW, Weller RN, Pashley DH. Push-out strength and SEM evaluation of resin composite bonded to internal cervical dentin. Endod Dent Trauma-tol 1996;12:227-36.

43.Purton DG, Love RM, Chandler NP. Rigidity and retention of ceramic root canal posts. Oper Dent 2000;25:223-7.

Corresponding author:Dr Steven M. MorganoHenry M. Goldman School of Dental MedicineBoston University100 East Newton Street Rm. G-219Boston, MA 02118Fax: 617-638-5434E-mail: [email protected]

Copyright © 2012 by the Editorial Council for The Journal of Prosthetic Dentistry.

Clinical ImplicationsThe study suggests there are ethnic differences for buccal corridor and smile arc.

Statement of problem. With growing demand for high esthetic standards, dentists must understand patient percep-tion and incorporate their preferences into treatment. However, little is known about how cultural and ethnic differ-ences influence esthetic perception. Purpose. The purpose of this study was to determine whether differences in ethnic background, including the possibil-ity of assimilation, affected a layperson’s perception of esthetic and smile characteristics.

Material and methods. A survey was developed containing images that were digitally manipulated into a series of barely perceptible steps, changing 1 smile parameter to form a strip of images that displayed that parameter over a wide range. Data were collected with a customized program which randomly displayed a single image and allowed the subject to use the mouse to adjust an on-screen slider according to displayed instructions, that is, “Please move the slider to select the image you find to be most ideal”; or “Please move the slider to select the first image that you find unattractive.” A convenience sample (n=288) comprised of U.S. whites, U.S. Asian Indians, and Indians living in India was surveyed. This sample provided a power of .86 to detect a difference of ±1.5 mm. Subjects evaluated images showing the smile arc, buccal corridor, gingival display, vertical overlap, lateral incisal step, maxillary midline to mid-face, and maxillary to mandibular midline. Rater reliability was assessed with the Fleiss-Cohen weighted Kappa (Kw) statistic and corresponding 95% confidence interval after each question was repeated in a random sequence. Choice differences due to ethnicity were assessed with a multiple randomization test and the adjusted P value with the step-down Bonferrroni method of Holm (α=.05).

Results. The Kw for the 17 variables in all 3 groups ranged from 0.11 for ideal vertical overlap to 0.64 for ideal buccal corridor space. Overall reliability was fair to moderate. Differences attributed to ethnicity were demonstrated between the Asian Indians and U.S. whites. Differences attributed to assimilation were demonstrated between U.S. Asian Indi-ans and Asian Indians. Differences between U.S. Asian Indians and U.S. whites can be instructive and demonstrate the relative power of ethnicity and assimilation. A difference between these groups shows the power of ethnicity and no difference between these groups shows the power of assimilation. The ratings of the Asian Indians and the U.S. whites showed a clinically significant difference for Ideal Buccal Corridor and Maximum Smile Arc. There were no significant differences between the U.S. Asian Indians and Asian Indians. There were clinically significant differences between the U.S. Asian Indians and U.S. whites only for Ideal Buccal Corridor.

Conclusions. Ethnicity had a significant effect on the esthetic choices for Buccal Corridor and Smile Arc. There is no conclusive evidence for assimilation. (J Prosthet Dent 2012;107:327-335)

Smile characterization by U.S. white, U.S. Asian Indian, and Indian populations

Neeru Sharma, BDS, DDS, MS,a Stephen F. Rosenstiel, BDS, MSD,b Henry W. Fields, DDS, MS, MSD,c and F. Mike Beck DDS, MAd

The Ohio State University, College of Dentistry, Columbus, Ohio

Supported by a grant by the Dental Master’s Thesis Award Program sponsored by Delta Dental Foundation, a philanthropic affili-ate of Delta Dental of Michigan, Ohio, and Indiana.Presented at the 88th General Session of the International Association for Dental Research, Barcelona, Spain, July 2010.

aPrivate practice, Hilliard Ohio. bProfessor, Restorative and Prosthetic Dentistry.cProfessor, Orthodontics.dAssociate Professor, Emeritus, Oral Biology.


The Journal of Prosthetic Dentistry

329May 2012

Sharma et alSharma et al

The importance of dental esthetics on psychosocial well-being has been suggested1; however, other long-term data do not support that.2 Neverthe-less, patients have come to expect their dentists to address their esthetic con-cerns as well as their oral disease. Es-thetics, more than any area of dentist-ry, is based on individual perception.

Patients may have a different percep-tion of beauty from their dentist or the generally accepted set of ideal norms. Since the ultimate measure of treatment success depends upon the patients’ ac-ceptance of the esthetic outcome, it is interesting to examine their subjective judgment and include their preferences into treatment planning.

A smile can have different meanings in different cultures and social environ-ments.3 Numerous factors, such as de-mographic, racial, ethnic, and cultural differences, may have a significant role in the perception of facial esthetics. Cultural influences are a product of a person’s environment, and cultural esthetic values are defined by factors such as the attitudes of a communi-ty, economics, familiarity with dental treatment, and general esthetic pref-erences.4 Cultural influences on smile design can be subtle or dominating, depending on the individual and their society, and there are accepted norms that guide a dentist when creating an esthetically pleasing smile.5-8 How-ever, despite these norms, the prefer-ences of the patient and the percep-tion of his or her smile must be given careful consideration.4

The authors identified no study that has comprehensively examined whether differences in ethnic back-ground can alter the perception of smile characteristics. More clarifi-cation is needed for ideal and ac-ceptable smile esthetics of different ethnicities from a nondental profes-sional perspective so that there is an awareness of potentially unique es-thetic preferences. The ethnic groups chosen for comparison were whites of European descent and Asians of Indian descent. Within the Asian-In-dian group, a comparison was made

between persons of Indian origin liv-ing in the U.S. for 5 or more years and those living in India. This was to deter-mine whether assimilation into a dif-ferent culture affected the perception of the layperson. Among the various factors that affect assimilation into a culture are socioeconomic status, spa-tial concentration, language attain-ment, and intermarriage.9 Depalo et al10 recognized that the length of stay in the host country represents a key factor when assessing immigrants’ in-tegration. The authors distinguished 4 groups of immigrants as a function of the time they spent in the receiving country: less than 5 years, between 6 and 15 years, between 15 and 25 years, and more than 25 years. They found that migrants’ social relations tend to increase with their length of stay in the host country. The percent-age of “socializers” rises from 62.3% for immigrants who have been in the host country for less than 5 years to 70% for those who have been there for more than 15 years. The correlation flattens out, however, for immigrants who have been in the receiving country for a long time (more than 25 years).10 Subjects with a minimum of 5 years of stay would be an appropriate choice in determining how a different culture af-fects esthetic perceptions.

The first Asian Indians or Indian Americans arrived in America as early as the middle of the nineteenth cen-tury.11 According to the 2000 U.S. census, the number of Asian Indians in the U.S. was approximately 1.7 mil-lion,12 growing to over 2.5 million in 2007.13 It is considered to be the fast-est growing ethnic community in the U.S.13 Cultural assimilation is defined as a process by which members of an ethnic minority group lose cultural characteristics that distinguish them from the dominant cultural group or take on the cultural characteristics of another group.14 The purpose of this study was to assess whether differ-ent ethnic background and potential cultural assimilation made a differ-ence in the lay perception of the ideal and the acceptable range of several

important smile characteristics mea-sured in the context of the lower face. The null hypotheses tested were that there would be no differences in the perception of the ideal and accept-able values of smile characteristics between white and Indian popula-tions and that there would be no dif-ference in the perception of ideal and acceptable values of smile character-istics between Indians living in India and Indians living in the U.S.

MATERIAL AND METHODS Esthetic preferences were measured

with a computer program that pre-sented images that could be manipu-lated with an on-screen slider. Subjects from 3 populations were sampled: whites of European descent, Asian In-dians living in the United States, and Asian Indians living in India. The study was reviewed and approved by the In-stitutional Review Board of The Ohio State University and granted a waiver of written informed consent.

A power analysis was performed to determine study sample size. Of the dependent variables in this study, the vertical overlap was reported by Ker et al15 to have the highest vari-ance, so it was used to determine the sample size. With a nondirectional al-pha risk of .05 and assuming a stan-dard deviation of 3.5 (chosen based on the variable with the largest stan-dard deviation from the raw data of Ker et al15), a sample size of 87 sub-jects was needed to detect a differ-ence of ±1.5 mm with a power of .86. Ten percent was added to this sample size in the event that nonparametric analysis would be needed. As a result, the final sample size per group was 96 subjects. Since there were 3 groups, a total of 288 subjects was required.

The target population of this study included whites of European descent living in the United States, Asian Indi-ans living in the U.S. for more than 5 years, and Indians living in India. There were 288 English speaking raters older than 18 years, familiar with the use of a mouse to control a computer and with

no association with the dental profes-sion (96 subjects per group). This was a convenience sample.

White subjects were recruited at a central university campus sports fa-cility. Asian Indians living in the U.S. for 5 or more years were recruited by displaying posters at religious gather-ing points. Indians living in India were recruited by displaying a poster in the patient reception area of an Indian dental college. The respective institu-tions granted permission for the sur-vey. The subjects were provided first

with a script to briefly explain the study and then presented with questions to answer. Subjects voluntarily consent-ed to participation and to answering demographic questions, including gender, ethnicity, and length of stay in the U.S. (asked only of the Asian In-dians living in the U.S.). Rater reliabil-ity for each variable was assessed by means of the Fleiss-Cohen weighted Kappa (Kw) statistic and correspond-ing 95% confidence interval.

Initial image selection and ma-nipulation followed the protocol of

Ker et al.15 An extraoral photograph, cropped to show only the lips, nasal tip, and labiomental fold of a male and female orthodontically treated patient, was combined with a bi-sected and mirrored frontal intraoral image with the features of photo edit-ing software (Photoshop CS2; Adobe Systems, San Jose, Calif ) to create a gender-ambiguous composite image. Once a series of modification values that appeared to represent the range of visually realistic smiles was estab-lished, the tooth images were stored

1 Buccal corridor (median values in mm).

2 Smile arc (mm).

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable

Same lowercase superscript letters in rows denote ethnic groups that do not differ significantly (P<.05)

Average Depiction

7.92ab

21.78%

4.50cd

12.3%

2.34f

6.4%

U.S. Indians

8.28b

22.77%

4.86d

13.3%

1.98f

5.4%

Asian Indians

7.38a

20%

3.06c

8.4%

1.35e

3.7%

U.S. Whites

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable

Same lowercase superscript letters in rows denote ethnic groups that do not differ significantly (P>.05)

Average Depiction

2.0 mm at

second molarsab

1.97 mm at canines

5.50 mm at

second molarsd

3.35 mm at canines

8.00 mm at

second molarsf

4.50 mm at canines

U.S. Indians

1.5 mm at

second molarsb

1.5 mm at canines

5.50 mm at

second molarsd

3.35 mm at canines

8.00 mm at

second molarsf

4.50 mm at canines

Asian Indians

2.5 mm at

second molarsa

2.0 mm at canines

7.00 mm at

second molarsc

4.04 mm at canines

8.50 mm at

second molarse

4.75 mm at canines

U.S. Whites



329May 2012


The importance of dental esthetics on psychosocial well-being has been suggested1; however, other long-term data do not support that.2 Neverthe-less, patients have come to expect their dentists to address their esthetic con-cerns as well as their oral disease. Es-thetics, more than any area of dentist-ry, is based on individual perception.

Patients may have a different percep-tion of beauty from their dentist or the generally accepted set of ideal norms. Since the ultimate measure of treatment success depends upon the patients’ ac-ceptance of the esthetic outcome, it is interesting to examine their subjective judgment and include their preferences into treatment planning.

A smile can have different meanings in different cultures and social environ-ments.3 Numerous factors, such as de-mographic, racial, ethnic, and cultural differences, may have a significant role in the perception of facial esthetics. Cultural influences are a product of a person’s environment, and cultural esthetic values are defined by factors such as the attitudes of a communi-ty, economics, familiarity with dental treatment, and general esthetic pref-erences.4 Cultural influences on smile design can be subtle or dominating, depending on the individual and their society, and there are accepted norms that guide a dentist when creating an esthetically pleasing smile.5-8 How-ever, despite these norms, the prefer-ences of the patient and the percep-tion of his or her smile must be given careful consideration.4

The authors identified no study that has comprehensively examined whether differences in ethnic back-ground can alter the perception of smile characteristics. More clarifi-cation is needed for ideal and ac-ceptable smile esthetics of different ethnicities from a nondental profes-sional perspective so that there is an awareness of potentially unique es-thetic preferences. The ethnic groups chosen for comparison were whites of European descent and Asians of Indian descent. Within the Asian-In-dian group, a comparison was made

between persons of Indian origin liv-ing in the U.S. for 5 or more years and those living in India. This was to deter-mine whether assimilation into a dif-ferent culture affected the perception of the layperson. Among the various factors that affect assimilation into a culture are socioeconomic status, spa-tial concentration, language attain-ment, and intermarriage.9 Depalo et al10 recognized that the length of stay in the host country represents a key factor when assessing immigrants’ in-tegration. The authors distinguished 4 groups of immigrants as a function of the time they spent in the receiving country: less than 5 years, between 6 and 15 years, between 15 and 25 years, and more than 25 years. They found that migrants’ social relations tend to increase with their length of stay in the host country. The percent-age of “socializers” rises from 62.3% for immigrants who have been in the host country for less than 5 years to 70% for those who have been there for more than 15 years. The correlation flattens out, however, for immigrants who have been in the receiving country for a long time (more than 25 years).10 Subjects with a minimum of 5 years of stay would be an appropriate choice in determining how a different culture af-fects esthetic perceptions.

The first Asian Indians or Indian Americans arrived in America as early as the middle of the nineteenth cen-tury.11 According to the 2000 U.S. census, the number of Asian Indians in the U.S. was approximately 1.7 mil-lion,12 growing to over 2.5 million in 2007.13 It is considered to be the fast-est growing ethnic community in the U.S.13 Cultural assimilation is defined as a process by which members of an ethnic minority group lose cultural characteristics that distinguish them from the dominant cultural group or take on the cultural characteristics of another group.14 The purpose of this study was to assess whether differ-ent ethnic background and potential cultural assimilation made a differ-ence in the lay perception of the ideal and the acceptable range of several

important smile characteristics mea-sured in the context of the lower face. The null hypotheses tested were that there would be no differences in the perception of the ideal and accept-able values of smile characteristics between white and Indian popula-tions and that there would be no dif-ference in the perception of ideal and acceptable values of smile character-istics between Indians living in India and Indians living in the U.S.

MATERIAL AND METHODS Esthetic preferences were measured

with a computer program that pre-sented images that could be manipu-lated with an on-screen slider. Subjects from 3 populations were sampled: whites of European descent, Asian In-dians living in the United States, and Asian Indians living in India. The study was reviewed and approved by the In-stitutional Review Board of The Ohio State University and granted a waiver of written informed consent.

A power analysis was performed to determine study sample size. Of the dependent variables in this study, the vertical overlap was reported by Ker et al15 to have the highest vari-ance, so it was used to determine the sample size. With a nondirectional al-pha risk of .05 and assuming a stan-dard deviation of 3.5 (chosen based on the variable with the largest stan-dard deviation from the raw data of Ker et al15), a sample size of 87 sub-jects was needed to detect a differ-ence of ±1.5 mm with a power of .86. Ten percent was added to this sample size in the event that nonparametric analysis would be needed. As a result, the final sample size per group was 96 subjects. Since there were 3 groups, a total of 288 subjects was required.

The target population of this study included whites of European descent living in the United States, Asian Indi-ans living in the U.S. for more than 5 years, and Indians living in India. There were 288 English speaking raters older than 18 years, familiar with the use of a mouse to control a computer and with

no association with the dental profes-sion (96 subjects per group). This was a convenience sample.

White subjects were recruited at a central university campus sports fa-cility. Asian Indians living in the U.S. for 5 or more years were recruited by displaying posters at religious gather-ing points. Indians living in India were recruited by displaying a poster in the patient reception area of an Indian dental college. The respective institu-tions granted permission for the sur-vey. The subjects were provided first

with a script to briefly explain the study and then presented with questions to answer. Subjects voluntarily consent-ed to participation and to answering demographic questions, including gender, ethnicity, and length of stay in the U.S. (asked only of the Asian In-dians living in the U.S.). Rater reliabil-ity for each variable was assessed by means of the Fleiss-Cohen weighted Kappa (Kw) statistic and correspond-ing 95% confidence interval.

Initial image selection and ma-nipulation followed the protocol of

Ker et al.15 An extraoral photograph, cropped to show only the lips, nasal tip, and labiomental fold of a male and female orthodontically treated patient, was combined with a bi-sected and mirrored frontal intraoral image with the features of photo edit-ing software (Photoshop CS2; Adobe Systems, San Jose, Calif ) to create a gender-ambiguous composite image. Once a series of modification values that appeared to represent the range of visually realistic smiles was estab-lished, the tooth images were stored

1 Buccal corridor (median values in mm).

2 Smile arc (mm).

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable

Same lowercase superscript letters in rows denote ethnic groups that do not differ significantly (P<.05)

Average Depiction

7.92ab

21.78%

4.50cd

12.3%

2.34f

6.4%

U.S. Indians

8.28b

22.77%

4.86d

13.3%

1.98f

5.4%

Asian Indians

7.38a

20%

3.06c

8.4%

1.35e

3.7%

U.S. Whites

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable


Average Depiction

2.0 mm at

second molarsab

1.97 mm at canines

5.50 mm at

second molarsd

3.35 mm at canines

8.00 mm at

second molarsf

4.50 mm at canines

U.S. Indians

1.5 mm at

second molarsb

1.5 mm at canines

5.50 mm at

second molarsd

3.35 mm at canines

8.00 mm at

second molarsf

4.50 mm at canines

Asian Indians

2.5 mm at

second molarsa

2.0 mm at canines

7.00 mm at

second molarsc

4.04 mm at canines

8.50 mm at

second molarse

4.75 mm at canines

U.S. Whites



331May 2012


as sequences that showed small and gradual incremental changes in 1 vari-able. The survey was presented on a laptop computer with a numerical computing environment (MATLAB R2008a; The Mathworks, Inc, Natick, Mass). A customized mouse-operat-ed graphical interface was developed that combined demographic data collection and survey administration. The slider on the screen made the im-age on the screen change in small in-crements to alter the variable of inter-est. When the subject made a choice,

this was translated, according to the image number, into values that were summarized so that an ideal value ex-pressed as a median for each variable could be established.

The smile characteristics mea-sured were the smile arc, buccal cor-ridor, gingival display, vertical over-lap, maxillary central to lateral incisal edge discrepancy, maxillary midline to midface, and maxillary to mandibular midline. Of the 7 variables, 5 (buccal corridor fill, smile arc, maxillary gin-gival display, incisal edge discrepancy,

and vertical overlap) had 3 requests associated with them: choose the ideal image and the upper and lower level of acceptability for the individu-al variable being manipulated by the slider. The other 2 had only 1 request: the greatest acceptable deviation from 0, because the ideal was defined as no deviation. These were maxillary midline to face and maxillary to man-dibular midline discrepancies. Each variable was repeated to assess rater reliability. The subject evaluated a to-tal of 34 images. One image and the

3 Gingival display (mm).

4 Lateral step (Central to lateral incisal edge discrepancy) (mm).

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable


Average Depiction

-3.06a

-0.72b

2.70d

U.S. Indians

-3.42a

-0.54b

2.88d

Asian Indians

-3.42a

-0.72b

2.34c

U.S. Whites

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable


Average Depiction

2.34a

1.80b

0.72d

U.S. Indians

2.52a

1.80b

0.45c

Asian Indians

2.52a

1.80b

0.54cd

U.S. Whites

slider were presented on the screen at a time. Each image was presented with 1 of 2 statements: “Please move the slider to select the image you find to be most ideal”; or “Please move the slider to select the first image that you find unattractive.” A new random trial order was generated for each rater as follows: every time the program was run, it randomly shuffled the order, using system time at startup to seed a Mersenne twister random number generator.17 It took most subjects 10 to 15 minutes to complete the survey. Each subject in the U.S. was compen-sated with a $10 gift card and those in India with 100 rupees in cash.

The following were the 7 depen-dent variables that were included as defined by Ker et al15: buccal corridor space, smile arc, gingival display, lat-eral incisor step, vertical overlap, max-illary midline to face, and maxillary to mandibular midline discrepancy.

The buccal corridor space (Fig. 1) was defined by the amount of dark space displayed between the facial surfaces of the posterior teeth and the corner of the mouth. Buccal corridor spaces were manipulated by altering the amount of black space between the lip commissure and the most buc-cal tooth in the smile by moving the posterior teeth medially or laterally.

Images showed variations of approxi-mately 0.18 mm for this study.

The smile arc (Fig. 2), defined as the curvature formed by an imaginary line tangential to the incisal edges of the teeth, was modified with varying degrees of curvature in relationship to the lower lip. The range of modi-fication was from no curvature to an accentuated curvature and was ex-pressed in millimeters at the maxillary canines and maxillary second molars.

The gingival display (Fig. 3) was defined as the amount of gingival tis-sue visible above the maxillary central incisor crowns and below the central portion and border of the upper lip. The gingival display on smiling was approached by modifying the skeletal position of the dental arches in 0.18 mm increments.

The lateral incisor step (Fig. 4), the difference between the incisal edges of the maxillary central and the lateral incisors, was assessed by moving the entire lateral incisor tooth up or down in 0.18 mm increments.

The vertical overlap (Fig. 5) was defined as the projection of the maxil-lary anterior teeth over the mandibu-lar anterior teeth in the maximum intercuspation position of the jaws. It was modified by incrementally al-tering the mandibular layer of the

photographic image in the vertical dimension. The vertical movement of the mandibular layer produced an increased or decreased vertical over-lap. The layer was moved in 0.18 mm increments.

The maxillary midline to face (Fig. 6), in this study, was considered to be ideal when the maxillary midline was coincident with the philtrum. The maxillary midline was moved to the left of the face in 0.18 mm incre-ments. The right and left buccal cor-ridor was maintained throughout the movement of the dentition.

The maxillary to mandibular mid-line discrepancy (Fig. 7) was consid-ered to be ideal when this discrepancy was zero. While maintaining the maxil-lary midline, the mandibular dentition was moved to the left in 0.18 mm in-crements. The right and left buccal cor-ridors were maintained throughout the movement of the mandibular dentition.

The independent variable was the ethnic background of the subjects. Differences attributed to ethnic-ity were demonstrated between the Asian Indians and U.S. whites. Differ-ences attributed to assimilation were demonstrated between U.S. Asian In-dians and Asian Indians. Differences between U.S. Asian Indians and U.S. whites can be instructive and dem-

5 Vertical overlap (mm).

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable


Average Depiction

7.56a

5.58b

3.60d

U.S. Indians

8.10a

6.48c

3.60d

Asian Indians

7.92a

5.94bc

3.96d

U.S. Whites



331May 2012


as sequences that showed small and gradual incremental changes in 1 vari-able. The survey was presented on a laptop computer with a numerical computing environment (MATLAB R2008a; The Mathworks, Inc, Natick, Mass). A customized mouse-operat-ed graphical interface was developed that combined demographic data collection and survey administration. The slider on the screen made the im-age on the screen change in small in-crements to alter the variable of inter-est. When the subject made a choice,

this was translated, according to the image number, into values that were summarized so that an ideal value ex-pressed as a median for each variable could be established.

The smile characteristics mea-sured were the smile arc, buccal cor-ridor, gingival display, vertical over-lap, maxillary central to lateral incisal edge discrepancy, maxillary midline to midface, and maxillary to mandibular midline. Of the 7 variables, 5 (buccal corridor fill, smile arc, maxillary gin-gival display, incisal edge discrepancy,

and vertical overlap) had 3 requests associated with them: choose the ideal image and the upper and lower level of acceptability for the individu-al variable being manipulated by the slider. The other 2 had only 1 request: the greatest acceptable deviation from 0, because the ideal was defined as no deviation. These were maxillary midline to face and maxillary to man-dibular midline discrepancies. Each variable was repeated to assess rater reliability. The subject evaluated a to-tal of 34 images. One image and the

3 Gingival display (mm).

4 Lateral step (Central to lateral incisal edge discrepancy) (mm).

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable


Average Depiction

-3.06a

-0.72b

2.70d

U.S. Indians

-3.42a

-0.54b

2.88d

Asian Indians

-3.42a

-0.72b

2.34c

U.S. Whites

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable


Average Depiction

2.34a

1.80b

0.72d

U.S. Indians

2.52a

1.80b

0.45c

Asian Indians

2.52a

1.80b

0.54cd

U.S. Whites

slider were presented on the screen at a time. Each image was presented with 1 of 2 statements: “Please move the slider to select the image you find to be most ideal”; or “Please move the slider to select the first image that you find unattractive.” A new random trial order was generated for each rater as follows: every time the program was run, it randomly shuffled the order, using system time at startup to seed a Mersenne twister random number generator.17 It took most subjects 10 to 15 minutes to complete the survey. Each subject in the U.S. was compen-sated with a $10 gift card and those in India with 100 rupees in cash.

The following were the 7 depen-dent variables that were included as defined by Ker et al15: buccal corridor space, smile arc, gingival display, lat-eral incisor step, vertical overlap, max-illary midline to face, and maxillary to mandibular midline discrepancy.

The buccal corridor space (Fig. 1) was defined by the amount of dark space displayed between the facial surfaces of the posterior teeth and the corner of the mouth. Buccal corridor spaces were manipulated by altering the amount of black space between the lip commissure and the most buc-cal tooth in the smile by moving the posterior teeth medially or laterally.

Images showed variations of approxi-mately 0.18 mm for this study.

The smile arc (Fig. 2), defined as the curvature formed by an imaginary line tangential to the incisal edges of the teeth, was modified with varying degrees of curvature in relationship to the lower lip. The range of modi-fication was from no curvature to an accentuated curvature and was ex-pressed in millimeters at the maxillary canines and maxillary second molars.

The gingival display (Fig. 3) was defined as the amount of gingival tis-sue visible above the maxillary central incisor crowns and below the central portion and border of the upper lip. The gingival display on smiling was approached by modifying the skeletal position of the dental arches in 0.18 mm increments.

The lateral incisor step (Fig. 4), the difference between the incisal edges of the maxillary central and the lateral incisors, was assessed by moving the entire lateral incisor tooth up or down in 0.18 mm increments.

The vertical overlap (Fig. 5) was defined as the projection of the maxil-lary anterior teeth over the mandibu-lar anterior teeth in the maximum intercuspation position of the jaws. It was modified by incrementally al-tering the mandibular layer of the

photographic image in the vertical dimension. The vertical movement of the mandibular layer produced an increased or decreased vertical over-lap. The layer was moved in 0.18 mm increments.

The maxillary midline to face (Fig. 6), in this study, was considered to be ideal when the maxillary midline was coincident with the philtrum. The maxillary midline was moved to the left of the face in 0.18 mm incre-ments. The right and left buccal cor-ridor was maintained throughout the movement of the dentition.

The maxillary to mandibular mid-line discrepancy (Fig. 7) was consid-ered to be ideal when this discrepancy was zero. While maintaining the maxil-lary midline, the mandibular dentition was moved to the left in 0.18 mm in-crements. The right and left buccal cor-ridors were maintained throughout the movement of the mandibular dentition.

The independent variable was the ethnic background of the subjects. Differences attributed to ethnic-ity were demonstrated between the Asian Indians and U.S. whites. Differ-ences attributed to assimilation were demonstrated between U.S. Asian In-dians and Asian Indians. Differences between U.S. Asian Indians and U.S. whites can be instructive and dem-

5 Vertical overlap (mm).

Maximum

Tolerable

Value

Ideal

Minimum

Tolerable

Value

Variable


Average Depiction

7.56a

5.58b

3.60d

U.S. Indians

8.10a

6.48c

3.60d

Asian Indians

7.92a

5.94bc

3.96d

U.S. Whites



333May 2012

Sharma et al Sharma et al

7 Maxillary to mandibular midline (mm).

Maximum

Tolerable

Value

Ideal

Variable


Average Depiction

1.80b

0

U.S. Indians

2.16c

0

Asian Indians

1.26a

0

U.S. Whites

6 Maxillary midline to midface (mm).

Maximum

Tolerable

Value

Ideal

Variable


Average Depiction

1.44b

0

U.S. Indians

1.62c

0

Asian Indians

0.90a

0

U.S. Whites

onstrate the relative power of ethnic-ity and assimilation. A difference be-tween these groups shows the power of ethnicity, and no difference be-tween these groups shows the power of assimilation.

Differences were considered clini-cally significant if they were < .05, the confidence interval of the Kw did not include 0, and the difference was > 1 mm (arbitrarily set as a perceptible difference at a conversational dis-tance of approximately 60 cm).

Gender differences were assessed nonparametrically with the random-ization test because of the skewing of the data. As no differences were found between male and female sub-jects, their data were combined for further statistical analysis. Because of correlations between rater age and many of the dependent variables, a rank analysis of covariance,18 with age

as the covariate, was used to compare the different ethnic groups. Nonpara-metric statistics were used because all 17 variables were not normally distributed as indicated by the Sha-piro-Wilk statistic. Since this analysis involved multiple comparisons, the step-down Bonferrroni method of Holm was used to adjust the P values. This adjustment was done to ensure that any significant differences found were not due to random occurrence from testing multiple variables. The level of significance was set at an al-pha value of .05.

RESULTS The total sample surveyed was

288 with 96 raters in each group: U.S. whites, U.S. Asian Indians, and Indians living in India. All the re-spondents completed the survey.

Rater demographics

There were 138 female (48%) and 150 male raters (52%). U.S. whites had a mean age of 32 years with a range of 20 to 56 years. U.S. Asian Indians had a mean age of 39 years with a range of 24 to 72 years. For Indians living in In-dia, the mean age was 44 years with a range of 19 to 68 years. Median ages among the 3 groups were not equiva-lent (Kruskal-Wallis Chi-square = 193, df=2, P<.001). For U.S. Asian Indians, the mean for the years of residency in the U.S. was 10 years with a range of 5 to 26 years. There were no statisti-cally significant differences between the ratings for male and female raters for any of the variables examined. P values ranged from a low of .14 for lat-eral step ideal to a high of .84 for buc-cal corridor maximum. Therefore, gen-der data were combined for analysis.

Table I. Reliability measured by Fleiss-Cohen weighted kappa statistics

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

Variable

0.70

0.67

0.48

0.39

0.30

0.38

0.41

0.39

0.46

0.31

0.48

0.44

0.22

0.40

0.33

0.66

0.72

UCB.95

All Groups

0.57

0.46

0.20

0.15

0.05

0.11

0.16

0.13

0.25

0.12

0.23

0.22

-0.01

0.07

0.07

0.36

0.56

LCB.95

0.63

0.65

0.42

0.33

0.25

0.25

0.24

0.37

0.45

0.32

0.37

0.51

0.35

0.32

0.29

0.61

0.56

UCB.95

U.S. Indians

0.36

0.36

0.05

-0.04

-0.07

-0.02

-0.13

-0.07

0.16

-0.03

-0.01

0.13

-0.07

-0.14

-0.05

0.29

0.21

LCB.95

0.66

0.76

0.50

0.34

0.32

0.51

0.67

0.42

0.48

0.46

0.56

0.55

0.29

0.65

0.53

0.65

0.75

UCB.95

Asian Indians

0.36

0.34

0.01

-0.17

-0.16

0.01

0.32

-0.05

0.11

0.20

0.10

0.16

-0.04

0.33

0.06

0.17

0.43

LCB.95

0.76

0.63

0.50

0.60

0.47

0.37

0.37

0.58

0.60

0.26

0.63

0.41

0.19

0.35

0.35

0.86

0.83

UCB.95

U.S. Whites

0.51

0.32

0.07

0.25

-0.02

0.14

-0.03

0.19

0.26

-0.09

0.21

0.06

-0.23

-0.22

-0.17

0.05

0.59

LCB.95

0.49

0.51

0.24

0.14

0.09

0.14

0.06

0.15

0.31

0.15

0.19

0.32

0.14

0.09

0.12

0.45

0.38

ê1

0.51

0.55

0.25

0.08

0.08

0.26

0.49

0.18

0.29

0.33

0.33

0.36

0.13

0.49

0.29

0.41

0.59

ê1

0.64

0.47

0.29

0.42

0.23

0.25

0.17

0.38

0.43

0.08

0.42

0.24

-0.02

0.07

0.09

0.45

0.71

ê1

0.64

0.57

0.34

0.27

0.17

0.25

0.28

0.26

0.36

0.21

0.35

0.33

0.11

0.24

0.20

0.51

0.64

κ1

1Fleiss-Cohen weighted kappa coefficient

No agreement<0

0.0-0.20

0.21-0.40

0.41-0.60

0.61-0.80

0.81-1.00Slight agreement

Fair agreement

Moderate agreement

Substantial agreement

Almost perfect agreement

Table II. Multiple rank analyses of covariance with age as covariate

BC I

BC Min

BC MX

SA I

SA Min

SA Max

GD I

GD Mx

GD Mn

Lat Step I

Lat St Mini

Lat St Mx

OB I

OB Min

OB Mx

Mid Face

MxMidl

Variable

1.53

1.17

1.08

3.00

3.00

1.00

1.44

1.26

0.72

0.54

0.36

0.36

1.17

0.90

0.99

0.45

1.17

RangeQuartile

U.S. Indian Asian IndianGroup OmnibU.S. Null Hypothesis Post Hoc Testing

U.S. White

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

N

4.50

2.34

7.92

8.00

3.00

16.00

-0.72

-3.06

2.70

1.80

0.72

2.34

5.58

3.60

7.56

1.44

1.80

Median

1.80

1.53

1.26

3.50

3.00

2.50

1.44

1.44

1.08

0.36

0.36

0.36

1.62

0.72

0.81

0.99

1.08

RangeQuartile

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

N

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

DF

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

.0022

.0681

.0006

1.0000

.0279

.0192

.0280

P1

-0.36

0.36

-0.36

0.00

0.00

-1.00

-0.18

0.36

-0.18

0.00

0.27

-0.18

-0.90

0.00

-0.54

-0.18

-0.36

Delta

<.0001

<.0001

<.0001

<.0001

.0016

<.0001

.6121

.1518

.0032

.0002

.0003

.0222

<.0001

.0190

.0026

<.0001

<.0001

P

4.86

1.98

8.28

8.00

3.00

17.00

-0.54

-3.42

2.88

1.80

0.45

2.52

6.48

3.60

8.10

1.62

2.16

Median

1.80

1.80

0.72

2.00

1.00

2.00

1.08

0.90

0.81

0.36

0.36

0.54

1.26

0.90

0.90

0.54

0.54

RangeQuartile

HypothesisTest

U.S. Indian vs Asian Indian

<.001

.0003

<.0001

<.0001

.0251

.0002

1.0000

.7203

.0132

.0032

1.0000

1.0000

.4120

.3333

.4877

.0017

.0010

P1

1.44

0.99

0.54

3.00

1.00

0.50

0.00

0.36

0.36

0.00

0.18

-0.18

-0.36

-0.36

-0.36

0.54

0.54

Delta

U.S. Indian vs U.S. White

<.0001

.0073

.0075

.0015

.0600

.0064

.9587

.9587

.0802

.3147

.4910

.7474

.0102

.0600

.6477

<.0001

<.0001

P1

1.80

0.63

0.90

3.00

1.00

1.50

0.18

0.00

0.54

0.00

-0.09

0.00

0.54

-0.36

0.18

0.72

0.90

Delta

Asian Indian vsU.S. White

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

N

44.6

28.5

24.1

26.1

12.9

21.4

1.0

3.8

11.5

16.8

16.0

7.6

20.0

7.9

11.9

34.1

39.7

x2

3.06

1.35

7.38

5.00

2.00

15.50

-0.72

-3.42

2.34

1.80

0.54

2.52

5.94

3.96

7.92

0.90

1.26

Median

1Step-down Bonferroni method of Holm adjusted P-valuesBold Delta scores indicate that a 95% confidence interval included zero for one or both weighted kappa coefficients



333May 2012

Sharma et al Sharma et al

7 Maxillary to mandibular midline (mm).

Maximum

Tolerable

Value

Ideal

Variable


Average Depiction

1.80b

0

U.S. Indians

2.16c

0

Asian Indians

1.26a

0

U.S. Whites

6 Maxillary midline to midface (mm).

Maximum

Tolerable

Value

Ideal

Variable


Average Depiction

1.44b

0

U.S. Indians

1.62c

0

Asian Indians

0.90a

0

U.S. Whites

onstrate the relative power of ethnic-ity and assimilation. A difference be-tween these groups shows the power of ethnicity, and no difference be-tween these groups shows the power of assimilation.

Differences were considered clini-cally significant if they were < .05, the confidence interval of the Kw did not include 0, and the difference was > 1 mm (arbitrarily set as a perceptible difference at a conversational dis-tance of approximately 60 cm).

Gender differences were assessed nonparametrically with the random-ization test because of the skewing of the data. As no differences were found between male and female sub-jects, their data were combined for further statistical analysis. Because of correlations between rater age and many of the dependent variables, a rank analysis of covariance,18 with age

as the covariate, was used to compare the different ethnic groups. Nonpara-metric statistics were used because all 17 variables were not normally distributed as indicated by the Sha-piro-Wilk statistic. Since this analysis involved multiple comparisons, the step-down Bonferrroni method of Holm was used to adjust the P values. This adjustment was done to ensure that any significant differences found were not due to random occurrence from testing multiple variables. The level of significance was set at an al-pha value of .05.

RESULTS The total sample surveyed was

288 with 96 raters in each group: U.S. whites, U.S. Asian Indians, and Indians living in India. All the re-spondents completed the survey.

Rater demographics

There were 138 female (48%) and 150 male raters (52%). U.S. whites had a mean age of 32 years with a range of 20 to 56 years. U.S. Asian Indians had a mean age of 39 years with a range of 24 to 72 years. For Indians living in In-dia, the mean age was 44 years with a range of 19 to 68 years. Median ages among the 3 groups were not equiva-lent (Kruskal-Wallis Chi-square = 193, df=2, P<.001). For U.S. Asian Indians, the mean for the years of residency in the U.S. was 10 years with a range of 5 to 26 years. There were no statisti-cally significant differences between the ratings for male and female raters for any of the variables examined. P values ranged from a low of .14 for lat-eral step ideal to a high of .84 for buc-cal corridor maximum. Therefore, gen-der data were combined for analysis.

Table I. Reliability measured by Fleiss-Cohen weighted kappa statistics

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

Variable

0.70

0.67

0.48

0.39

0.30

0.38

0.41

0.39

0.46

0.31

0.48

0.44

0.22

0.40

0.33

0.66

0.72

UCB.95

All Groups

0.57

0.46

0.20

0.15

0.05

0.11

0.16

0.13

0.25

0.12

0.23

0.22

-0.01

0.07

0.07

0.36

0.56

LCB.95

0.63

0.65

0.42

0.33

0.25

0.25

0.24

0.37

0.45

0.32

0.37

0.51

0.35

0.32

0.29

0.61

0.56

UCB.95

U.S. Indians

0.36

0.36

0.05

-0.04

-0.07

-0.02

-0.13

-0.07

0.16

-0.03

-0.01

0.13

-0.07

-0.14

-0.05

0.29

0.21

LCB.95

0.66

0.76

0.50

0.34

0.32

0.51

0.67

0.42

0.48

0.46

0.56

0.55

0.29

0.65

0.53

0.65

0.75

UCB.95

Asian Indians

0.36

0.34

0.01

-0.17

-0.16

0.01

0.32

-0.05

0.11

0.20

0.10

0.16

-0.04

0.33

0.06

0.17

0.43

LCB.95

0.76

0.63

0.50

0.60

0.47

0.37

0.37

0.58

0.60

0.26

0.63

0.41

0.19

0.35

0.35

0.86

0.83

UCB.95

U.S. Whites

0.51

0.32

0.07

0.25

-0.02

0.14

-0.03

0.19

0.26

-0.09

0.21

0.06

-0.23

-0.22

-0.17

0.05

0.59

LCB.95

0.49

0.51

0.24

0.14

0.09

0.14

0.06

0.15

0.31

0.15

0.19

0.32

0.14

0.09

0.12

0.45

0.38

ê1

0.51

0.55

0.25

0.08

0.08

0.26

0.49

0.18

0.29

0.33

0.33

0.36

0.13

0.49

0.29

0.41

0.59

ê1

0.64

0.47

0.29

0.42

0.23

0.25

0.17

0.38

0.43

0.08

0.42

0.24

-0.02

0.07

0.09

0.45

0.71

ê1

0.64

0.57

0.34

0.27

0.17

0.25

0.28

0.26

0.36

0.21

0.35

0.33

0.11

0.24

0.20

0.51

0.64

κ1

1Fleiss-Cohen weighted kappa coefficient

No agreement<0

0.0-0.20

0.21-0.40

0.41-0.60

0.61-0.80

0.81-1.00Slight agreement

Fair agreement

Moderate agreement

Substantial agreement

Almost perfect agreement

Table II. Multiple rank analyses of covariance with age as covariate

BC I

BC Min

BC MX

SA I

SA Min

SA Max

GD I

GD Mx

GD Mn

Lat Step I

Lat St Mini

Lat St Mx

OB I

OB Min

OB Mx

Mid Face

MxMidl

Variable

1.53

1.17

1.08

3.00

3.00

1.00

1.44

1.26

0.72

0.54

0.36

0.36

1.17

0.90

0.99

0.45

1.17

RangeQuartile

U.S. Indian Asian IndianGroup OmnibU.S. Null Hypothesis Post Hoc Testing

U.S. White

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

N

4.50

2.34

7.92

8.00

3.00

16.00

-0.72

-3.06

2.70

1.80

0.72

2.34

5.58

3.60

7.56

1.44

1.80

Median

1.80

1.53

1.26

3.50

3.00

2.50

1.44

1.44

1.08

0.36

0.36

0.36

1.62

0.72

0.81

0.99

1.08

RangeQuartile

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

N

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

DF

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

.0022

.0681

.0006

1.0000

.0279

.0192

.0280

P1

-0.36

0.36

-0.36

0.00

0.00

-1.00

-0.18

0.36

-0.18

0.00

0.27

-0.18

-0.90

0.00

-0.54

-0.18

-0.36

Delta

<.0001

<.0001

<.0001

<.0001

.0016

<.0001

.6121

.1518

.0032

.0002

.0003

.0222

<.0001

.0190

.0026

<.0001

<.0001

P

4.86

1.98

8.28

8.00

3.00

17.00

-0.54

-3.42

2.88

1.80

0.45

2.52

6.48

3.60

8.10

1.62

2.16

Median

1.80

1.80

0.72

2.00

1.00

2.00

1.08

0.90

0.81

0.36

0.36

0.54

1.26

0.90

0.90

0.54

0.54

RangeQuartile

HypothesisTest

U.S. Indian vs Asian Indian

<.001

.0003

<.0001

<.0001

.0251

.0002

1.0000

.7203

.0132

.0032

1.0000

1.0000

.4120

.3333

.4877

.0017

.0010

P1

1.44

0.99

0.54

3.00

1.00

0.50

0.00

0.36

0.36

0.00

0.18

-0.18

-0.36

-0.36

-0.36

0.54

0.54

Delta

U.S. Indian vs U.S. White

<.0001

.0073

.0075

.0015

.0600

.0064

.9587

.9587

.0802

.3147

.4910

.7474

.0102

.0600

.6477

<.0001

<.0001

P1

1.80

0.63

0.90

3.00

1.00

1.50

0.18

0.00

0.54

0.00

-0.09

0.00

0.54

-0.36

0.18

0.72

0.90

Delta

Asian Indian vsU.S. White

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

96

N

44.6

28.5

24.1

26.1

12.9

21.4

1.0

3.8

11.5

16.8

16.0

7.6

20.0

7.9

11.9

34.1

39.7

x2

3.06

1.35

7.38

5.00

2.00

15.50

-0.72

-3.42

2.34

1.80

0.54

2.52

5.94

3.96

7.92

0.90

1.26

Median

1Step-down Bonferroni method of Holm adjusted P-valuesBold Delta scores indicate that a 95% confidence interval included zero for one or both weighted kappa coefficients



335May 2012


Reliability

The Kw for the 17 variables in all 3 groups ranged from .11 for ideal verti-cal overlap to .64 for ideal buccal cor-ridor space (Table I).

Effects of ethnicity on rater preferences

The ideal and the threshold of ac-

ceptability are reported with medians for each smile characteristic (Figs. 1-7). Clinically significant differences for smile esthetics were evident for 2 of the 17 variables examined. These were the ratings of the Asian Indians and the U.S. whites for Ideal Buccal Corridor and Maximum Smile Arc (Table II).

Effects of cultural assimilation on Indian rater preferences

There were no significant differ-

ences between the U.S. Asian Indians and Asian Indians. Clinically signifi-cant differences were evident for 1 of the 17 variables examined. These were the ratings of the U.S. Asian Indians and the U.S. whites for Ideal Buccal Corridor (Table II).

DISCUSSION

While the data support rejection of the stated null hypotheses that eth-nic background does not affect a lay-person’s perception of smile charac-teristics, the same data demonstrated only minimal differences as the result of ethnicity, with only 2 of the 17 vari-ables tested (12%) being significantly different between Asian Indians and U.S. whites. They also show that, be-cause there were no differences be-tween the U.S. Asians and the Indian Asians or between the Indian Asians and the U.S. Whites, assimilation did not need to occur because the differ-ence in perception of esthetics due to ethnicity was minimal. This is a con-sistent finding among all the compari-sons made in the study. Ethnic differ-ences were few; therefore assimilation did not need to occur. In the few in-

stances where differences did occur between ethnic groups (U.S. whites and Asian Indians), whites preferred a broader smile and smaller smile arc than Indians living in India.

Because there were no differences between the U.S. Asian Indians and Asian Indians, it could be said that no assimilation had occurred. However, between the U.S. Asian Indians and the U.S. whites, 16 of the 17 variables test-ed did not differ significantly. This sug-gests that there was assimilation into the U.S. culture as well. Clearly, these results conflict.

The reliability of the assessment was determined by asking the same question twice in a random order. The Fleiss-Co-hen weighted Kappa (Kw) statistic for the 17 variables in all 3 groups ranged from .11 for ideal vertical overlap to .64 for ideal buccal corridor space. This would be considered fair to mod-est reliability.18,19 Table I indicates that many of the variables, especially in the U.S. Indian group, do not demonstrate significant reliability. As a result, conclu-sions regarding these variables need to be viewed with caution.

Reliability was less than that found in the study by Ker et al,15 where coefficients ranged from .34 to .88. There is no obvi-ous explanation for this difference.

Because of the power of this study, it is possible to show small statisti-cally significant differences that may not be clinically significant. One mil-limeter was chosen as an appropri-ate threshold for clinically significant differences. The following discussion focuses on those differences. Those clinically significant differences are all variables related to the surrounding lower face and are not solely dental characteristics.

When comparing the results for whites in the current study to the re-sults of Ker et al15 for buccal corridor space, the ideal values indicated that the whites in the current study pre-ferred a broader smile. This difference was unexpected. Both Indian groups were much more similar to Ker et al’s15 values, a fact which further mitigates ethnic differences. Whites preferred a

significantly smaller Maximum Smile Arc than the Indian groups, but all groups preferred a much greater Arc than other reported white groups.15

There are limitations with the study design. Since a convenience sample was used instead of a true random sample, care must be taken when making generalizations about the referent populations. Recruit-ment of subjects was different for the 3 groups. The campus sports center was chosen for the white group to allow comparisons with a previous study with similar methodology.15 The religious gathering point and dental school were chosen as convenient lo-cations to recruit the other groups; however, these different locations may have resulted in some differ-ences. For example, patients seeking dental care at a hospital may be more likely to focus on dental details than people at religious gathering points or sport facilities. This may be a study limitation. This group was also pri-marily made up of upper middle class people who were fluent in speaking and understanding English and famil-iar with the computer. This may not be a good representation of the en-tire cross-section of the population. The Asian Indian subjects living in the United States were primarily born and raised in India. A study of native-born Asian Indian Americans may not pro-duce the same findings.

There were differences in the age ranges of the 3 groups. U.S. whites had a mean age of 32 years and a range of 20 to 56 years; the U.S. Asian Indians had a mean age of 39 years and a range of 24 to 72 years; and the Indian residents had a mean age of 44 years and a range of 19 to 68 years. To account for this difference, a rank analysis of covariance17 was used to compare the different ethnic groups. The skin tone of the lower face image that the subject surveyed was kept the same for all 3 groups to limit this as a confounding variable. Future studies are needed to compare the smile and esthetic perception of other ethnic groups such as African-Americans,

Hispanics, East Asians, Native Ameri-cans, and Arab-Americans.

CONCLUSIONS

Within the limitations of this study, the following conclusions were drawn:

1. Rater gender did not affect the perception of smile.

2. There appeared to be little eth-nic difference in the perception of smile esthetics.

3. Because there were few ethnic differences, assimilation was not re-quired, and the limited differences were apparent among all groups re-gardless of ethnicity or location.

REFERENCES

1. Kiyak HA. Does orthodontic treatment affect patients’ quality of life? J Dent Educ 2008;72:886-94.

2. Shaw WC, Richmond S, Kenealy PM, King-don A, Worthinton H. A 20-year cohort study of health gain from orthodontic treatment: psychological outcome. Am J Orthod Dentofacial Orthop. 132:146-157.

3. Goldstein RE. Study of need for esthetics in dentistry. J Prosthet Dent 1969;21:589-98.

4. Chalifoux PR. Practice made perfect. J Esth Restor Dent 1996;8:189-92.

5. Kokich VO Jr, Kiyak HA, Shapiro PA. Com-paring the perception of dentists and lay people to altered dental esthetics. J Esthet Dent 1999;11:311-24.

6. Kokich V. Esthetics and anterior tooth posi-tion: an orthodontic perspective, Part I: crown length. J Esthet Dent 1993;5:19-23.

7. Kokich VG. Esthetics and vertical tooth po-sition: orthodontic possibilities. Compend Contin Educ Dent 1997;18:1225-31; quiz 1232.

8. Kokich VO, Kokich VG, Kiyak HA. Percep-tions of dental professionals and laypersons to altered dental esthetics: asymmetric and symmetric situations. Am J Orthod Dento-facial Orthop 2006;130:141-51.

9. Crispino JA. The assimilation of ethnic groups: The Italian case. Staten Island: Cen-ter for Migration Studies; 1980. p.150-1.

10.Depalo DR, Faini R, Venturini A. The social assimilation of immigrants. Washington, DC: The World Bank Social Protection Discussion Paper No 0701; 2007. p. 5.

11.Jensen J. Passage from India: Asian Indian immigrants in North America. Princeton: Yale University Press; 1988. p. 1-23.

12.U.S. Census Bureau. The Asian population: 2000. Available from: http://www.census.gov/prod/2002pubs/c2kbr01-16.pdf [2010, August 11].

13.U.S. Census Bureau. ACS Demographic and Housing Estimates: 2006-2008. Available from: http://factfinder.census.gov/servlet/ADPTable?_bm=y&-geo_id=01000U.S.&-qr_name=ACS_2008_3YR_G00_DP3YR5&-ds_name=ACS_2008_3YR_G00_&-_lang=en&-_sse=on (Accessed August 11, 2010).

14.Mosby’s Medical Dictionary. 8th ed. St. Louis: Elsevier; 2009. p. 486.

15.Ker AJ, Chan R, Fields HW, Beck M, Rosenstiel S. Esthetics and smile charac-teristics from the layperson’s perspective: A computer-based survey study. J Am Dent Assoc 2008;139:1318-27.

16.Matsumoto M. Nishimura T. Mersenne twister: a 623-dimensionally equidistrib-uted uniform pseudo-random number generator. ACM Transactions on Modeling and Computer Simulation 1998;8:3-30.

17.Stokes ME, Davis CS, Koch GG. Categorical data analysis using the SAS system. 2nd ed. Cary: SAS Institute Inc. 2000. p. 174-9.

18.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159-74.

19.Sim J, Wright CC. The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther 2005;85:257-68.

Corresponding author: Dr Stephen F. Rosenstiel The Ohio State UniversityCollege of Dentistry 305 W. 12th Avenue Columbus, OH 43210 Fax: 614-292-9422E-mail: [email protected]

AcknowledgmentsThe authors thank the management of the Bhartiya Hindu Temple and the Sikh Temple in the U.S. and the Principal of DAV Dental College in Yamunanagar, India for granting permission to conduct the survey on their premises; in addition Dr. Rajiv Gupta, Reader in the section of Orthodontics at DAV Dental College in Yamunanagar, India for his support in conducting research among the Indian population.


Access to The Journal of Prosthetic Dentistry Online is reserved for print subscribers!

Full-text access to The Journal of Prosthetic Dentistry Online is available for all print subscribers. To activate your individu-al online subscription, please visit The Journal of Prosthetic Dentistry Online. Point your browser to http://www.journals.elsevierhealth.com/periodicals/ympr/home, follow the prompts to activate online access here, and follow the instruc-tions. To activate your account, you will need your subscriber account number, which you can find on your mailing label (note: the number of digits in your subscriber account number varies from 6 to 10). See the example below in which the subscriber account number has been circled.

Personal subscriptions to The Journal of Prosthetic Dentistry Online are for individual use only and may not be trans-ferred. Use of The Journal of Prosthetic Dentistry Online is subject to agreement to the terms and conditions as indicated online.

This is your subscription account number

Sample mailing label

1 V97-3 J010 12345678-9J. H. DOE531 MAIN STCENTER CITY, NY 10001-001

*********AUTO**SCH 3-DIGIT 001



335May 2012


Reliability

The Kw for the 17 variables in all 3 groups ranged from .11 for ideal verti-cal overlap to .64 for ideal buccal cor-ridor space (Table I).

Effects of ethnicity on rater preferences

The ideal and the threshold of ac-

ceptability are reported with medians for each smile characteristic (Figs. 1-7). Clinically significant differences for smile esthetics were evident for 2 of the 17 variables examined. These were the ratings of the Asian Indians and the U.S. whites for Ideal Buccal Corridor and Maximum Smile Arc (Table II).

Effects of cultural assimilation on Indian rater preferences

There were no significant differ-

ences between the U.S. Asian Indians and Asian Indians. Clinically signifi-cant differences were evident for 1 of the 17 variables examined. These were the ratings of the U.S. Asian Indians and the U.S. whites for Ideal Buccal Corridor (Table II).

DISCUSSION

While the data support rejection of the stated null hypotheses that eth-nic background does not affect a lay-person’s perception of smile charac-teristics, the same data demonstrated only minimal differences as the result of ethnicity, with only 2 of the 17 vari-ables tested (12%) being significantly different between Asian Indians and U.S. whites. They also show that, be-cause there were no differences be-tween the U.S. Asians and the Indian Asians or between the Indian Asians and the U.S. Whites, assimilation did not need to occur because the differ-ence in perception of esthetics due to ethnicity was minimal. This is a con-sistent finding among all the compari-sons made in the study. Ethnic differ-ences were few; therefore assimilation did not need to occur. In the few in-

stances where differences did occur between ethnic groups (U.S. whites and Asian Indians), whites preferred a broader smile and smaller smile arc than Indians living in India.

Because there were no differences between the U.S. Asian Indians and Asian Indians, it could be said that no assimilation had occurred. However, between the U.S. Asian Indians and the U.S. whites, 16 of the 17 variables test-ed did not differ significantly. This sug-gests that there was assimilation into the U.S. culture as well. Clearly, these results conflict.

The reliability of the assessment was determined by asking the same question twice in a random order. The Fleiss-Co-hen weighted Kappa (Kw) statistic for the 17 variables in all 3 groups ranged from .11 for ideal vertical overlap to .64 for ideal buccal corridor space. This would be considered fair to mod-est reliability.18,19 Table I indicates that many of the variables, especially in the U.S. Indian group, do not demonstrate significant reliability. As a result, conclu-sions regarding these variables need to be viewed with caution.

Reliability was less than that found in the study by Ker et al,15 where coefficients ranged from .34 to .88. There is no obvi-ous explanation for this difference.

Because of the power of this study, it is possible to show small statisti-cally significant differences that may not be clinically significant. One mil-limeter was chosen as an appropri-ate threshold for clinically significant differences. The following discussion focuses on those differences. Those clinically significant differences are all variables related to the surrounding lower face and are not solely dental characteristics.

When comparing the results for whites in the current study to the re-sults of Ker et al15 for buccal corridor space, the ideal values indicated that the whites in the current study pre-ferred a broader smile. This difference was unexpected. Both Indian groups were much more similar to Ker et al’s15 values, a fact which further mitigates ethnic differences. Whites preferred a

significantly smaller Maximum Smile Arc than the Indian groups, but all groups preferred a much greater Arc than other reported white groups.15

There are limitations with the study design. Since a convenience sample was used instead of a true random sample, care must be taken when making generalizations about the referent populations. Recruit-ment of subjects was different for the 3 groups. The campus sports center was chosen for the white group to allow comparisons with a previous study with similar methodology.15 The religious gathering point and dental school were chosen as convenient lo-cations to recruit the other groups; however, these different locations may have resulted in some differ-ences. For example, patients seeking dental care at a hospital may be more likely to focus on dental details than people at religious gathering points or sport facilities. This may be a study limitation. This group was also pri-marily made up of upper middle class people who were fluent in speaking and understanding English and famil-iar with the computer. This may not be a good representation of the en-tire cross-section of the population. The Asian Indian subjects living in the United States were primarily born and raised in India. A study of native-born Asian Indian Americans may not pro-duce the same findings.

There were differences in the age ranges of the 3 groups. U.S. whites had a mean age of 32 years and a range of 20 to 56 years; the U.S. Asian Indians had a mean age of 39 years and a range of 24 to 72 years; and the Indian residents had a mean age of 44 years and a range of 19 to 68 years. To account for this difference, a rank analysis of covariance17 was used to compare the different ethnic groups. The skin tone of the lower face image that the subject surveyed was kept the same for all 3 groups to limit this as a confounding variable. Future studies are needed to compare the smile and esthetic perception of other ethnic groups such as African-Americans,

Hispanics, East Asians, Native Ameri-cans, and Arab-Americans.

CONCLUSIONS

Within the limitations of this study, the following conclusions were drawn:

1. Rater gender did not affect the perception of smile.

2. There appeared to be little eth-nic difference in the perception of smile esthetics.

3. Because there were few ethnic differences, assimilation was not re-quired, and the limited differences were apparent among all groups re-gardless of ethnicity or location.

REFERENCES

1. Kiyak HA. Does orthodontic treatment affect patients’ quality of life? J Dent Educ 2008;72:886-94.

2. Shaw WC, Richmond S, Kenealy PM, King-don A, Worthinton H. A 20-year cohort study of health gain from orthodontic treatment: psychological outcome. Am J Orthod Dentofacial Orthop. 132:146-157.

3. Goldstein RE. Study of need for esthetics in dentistry. J Prosthet Dent 1969;21:589-98.

4. Chalifoux PR. Practice made perfect. J Esth Restor Dent 1996;8:189-92.

5. Kokich VO Jr, Kiyak HA, Shapiro PA. Com-paring the perception of dentists and lay people to altered dental esthetics. J Esthet Dent 1999;11:311-24.

6. Kokich V. Esthetics and anterior tooth posi-tion: an orthodontic perspective, Part I: crown length. J Esthet Dent 1993;5:19-23.

7. Kokich VG. Esthetics and vertical tooth po-sition: orthodontic possibilities. Compend Contin Educ Dent 1997;18:1225-31; quiz 1232.

8. Kokich VO, Kokich VG, Kiyak HA. Percep-tions of dental professionals and laypersons to altered dental esthetics: asymmetric and symmetric situations. Am J Orthod Dento-facial Orthop 2006;130:141-51.

9. Crispino JA. The assimilation of ethnic groups: The Italian case. Staten Island: Cen-ter for Migration Studies; 1980. p.150-1.

10.Depalo DR, Faini R, Venturini A. The social assimilation of immigrants. Washington, DC: The World Bank Social Protection Discussion Paper No 0701; 2007. p. 5.

11.Jensen J. Passage from India: Asian Indian immigrants in North America. Princeton: Yale University Press; 1988. p. 1-23.

12.U.S. Census Bureau. The Asian population: 2000. Available from: http://www.census.gov/prod/2002pubs/c2kbr01-16.pdf [2010, August 11].

13.U.S. Census Bureau. ACS Demographic and Housing Estimates: 2006-2008. Available from: http://factfinder.census.gov/servlet/ADPTable?_bm=y&-geo_id=01000U.S.&-qr_name=ACS_2008_3YR_G00_DP3YR5&-ds_name=ACS_2008_3YR_G00_&-_lang=en&-_sse=on (Accessed August 11, 2010).

14.Mosby’s Medical Dictionary. 8th ed. St. Louis: Elsevier; 2009. p. 486.

15.Ker AJ, Chan R, Fields HW, Beck M, Rosenstiel S. Esthetics and smile charac-teristics from the layperson’s perspective: A computer-based survey study. J Am Dent Assoc 2008;139:1318-27.

16.Matsumoto M. Nishimura T. Mersenne twister: a 623-dimensionally equidistrib-uted uniform pseudo-random number generator. ACM Transactions on Modeling and Computer Simulation 1998;8:3-30.

17.Stokes ME, Davis CS, Koch GG. Categorical data analysis using the SAS system. 2nd ed. Cary: SAS Institute Inc. 2000. p. 174-9.

18.Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159-74.

19.Sim J, Wright CC. The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther 2005;85:257-68.

Corresponding author: Dr Stephen F. Rosenstiel The Ohio State UniversityCollege of Dentistry 305 W. 12th Avenue Columbus, OH 43210 Fax: 614-292-9422E-mail: [email protected]

AcknowledgmentsThe authors thank the management of the Bhartiya Hindu Temple and the Sikh Temple in the U.S. and the Principal of DAV Dental College in Yamunanagar, India for granting permission to conduct the survey on their premises; in addition Dr. Rajiv Gupta, Reader in the section of Orthodontics at DAV Dental College in Yamunanagar, India for his support in conducting research among the Indian population.


Access to The Journal of Prosthetic Dentistry Online is reserved for print subscribers!

Full-text access to The Journal of Prosthetic Dentistry Online is available for all print subscribers. To activate your individu-al online subscription, please visit The Journal of Prosthetic Dentistry Online. Point your browser to http://www.journals.elsevierhealth.com/periodicals/ympr/home, follow the prompts to activate online access here, and follow the instruc-tions. To activate your account, you will need your subscriber account number, which you can find on your mailing label (note: the number of digits in your subscriber account number varies from 6 to 10). See the example below in which the subscriber account number has been circled.

Personal subscriptions to The Journal of Prosthetic Dentistry Online are for individual use only and may not be trans-ferred. Use of The Journal of Prosthetic Dentistry Online is subject to agreement to the terms and conditions as indicated online.

This is your subscription account number

Sample mailing label

1 V97-3 J010 12345678-9J. H. DOE531 MAIN STCENTER CITY, NY 10001-001

*********AUTO**SCH 3-DIGIT 001

Documents

Smile characterization by U.S. white, U.S. Asian Indian, and Indian populations