Early treatment outcome assessed by the Peer Assessment Rating index

ORIGINAL ARTICLE Early treatment outcome assessed by the Peer Assessment Rating index Valmy Pangrazio-Kulbersh, DDS, MS,a Richard Kaczynski, PhD,b and Michael Shunock, DDSc Detroit, Michigan In this study, the Peer Assessment Rating (PAR) index was used to objectively evaluate early treatment outcomes. Pretreatment and posttreatment casts of 103 consecutively treated patients were analyzed. The mean chronological, skeletal, and dental ages were 9.82, 9.76, and 9.32 years1, respectively. Calibrated examiners2 scored all models using the PAR ruler. PAR scores were weighed by means of a validation exercise. Cronbach α reliability analysis was used to establish the consistency of the subjective rating among 10 orthodontists of the severity of malocclusion. Pearson’s correlation coefficient was used to assess the association among the orthodontists and the total PAR scores. Multiple regression analysis was used to determined the optimum weight of the PAR scores. Pretreatment and posttreatment differences were evaluated with t tests. The association between PAR scores and classification of malocclusions and treatment categories was assessed by means of multivariate analysis of variance (MANOVA). A reduction in the PAR index was observed for the mean raw and weighted scores, from 15.82 to 8.82 and from 5.28 to 3.73, respectively (P < .001). Twenty percent of the sample greatly improved the PAR index, by a 70% reduction.3 Forty-eight percent improved scores by at least a 30% reduction.4 The remaining 32% did not reduce scores by at least 30%. Subjects with both Class I and Class II malocclusions reduced their scores similarly. There was no statistically significant association between reduction of PAR scores and treatment modalities. (Am J Orthod Dentofacial Orthop 1999;115:544-50)

O

rthodontic treatment results are continually being graded subjectively or by some objective method of evaluation, either in clinical settings, study groups, or national or state board examinations. Unfortunately, the variation in criteria used by different orthodontists makes it difficult to compare results of treatment. Several indices have been used to overcome the subjectivity of evaluation for malocclusion5 and treatment outcomes.1-8 The use of precise criteria is essential to uniformly quantify and measure malocclusions and the efficacy of different treatment modalities. The Peer Assessment Rating (PAR) index was developed to fulfill this need and to more objectively evaluate treatment results.9-13 In recent years, it is evident that orthodontists have been recommending early intervention for different types of orthodontic problems; these range from simple space management procedures to more complex ones requiring skeletal and dental changes.14-22 From the Department of Orthodontics, University of Detroit Mercy Dental School. aAssociate Professor. bAssistant Professor. cResearch Assistant. Reprint requests to: Dr. Valmy Pangrazio-Kulbersh, 8200 W. Outer Drive, P.O. Box 19900, Box 189, Detroit, MI 48219 Copyright © 1999 by the American Association of Orthodontists. 0889-5406/99/$8.00 + 0 8/1/93351

544

Some literature supports the need for early treatment in certain types of malocclusion, like early correction of skeletal Class II and Class III problems as well as transverse expansion to “develop” the dental arches and create space for the erupting permanent dentition.23-43 But there is still a lack of consensus among orthodontists regarding the degree of success of different treatment modalities applied during the early to late mixed dentition stages.44-47 The PAR index provides a tool to effectively evaluate treatment outcomes. Its reliability and validity have been established not only in England,10 where it was developed,6 but also in the United States.12 It can record malocclusions at any stage of treatment, and it is flexible because it allows the weighting to change to reflect standards achieved in different treatment centers. Richmond et al10 have found that a 30% reduction in weighted PAR scores is required for a case to be considered as improved, and a change of 22 points is necessary to bring about great improvement with treatment. They have evaluated a PAR score of 10 or less as an acceptable alignment and occlusion, and a PAR score of 5 or less suggests an almost-ideal occlusion. The purpose of this investigation was to evaluate: • The amount of change in PAR scores as a result of early treatment intervention.

American Journal of Orthodontics and Dentofacial Orthopedics Volume 115, Number 5

• The type of malocclusion that most favorably responded to treatment by decreasing the PAR scores. • What type of orthodontic therapy reduced the PAR scores the most.7

Pangrazio-Kulbersh, Kaczynski, and Shunock 545

Table I. Multiple

regression analysis: weight of the different components of the PAR index.11 Unstandardized. Coefficient

METHODS AND MATERIALS

Two hundred six sets of pretreatment and posttreatment study casts of consecutively treated cases, representing all types of malocclusions, were evaluated from the early treatment clinic files of the University of Detroit Mercy Orthodontic Department. The mean chronological age of the sample was 9.82 years, the mean skeletal age was 9.76 years, and the mean dental age was 9.32 years.8 All casts were assigned numbers and randomly arranged to avoid the bias of comparing pretreatment and posttreatment casts of the same individual at the same time. Pretreatment and posttreatment casts were not identified as such. The sample size for this investigation was selected by using the data from previous studies9-12 that suggested the required number of dental models be 200 with a minimum of 10 raters. Each set of study casts was given a PAR score by 3 raters whose skills were previously calibrated9 in the use of the PAR index. The PAR index ruler was used to score the casts.9-10 The interjudge reliability was 0.831, and the intrajudge reliability was 0.877. These kappa scores demonstrate a good degree of accordance among the PAR raters. A group of 10 orthodontists, members of the parttime faculty at the University of Detroit Mercy and practicing in different geographic areas in the metropolitan Detroit area, were asked to subjectively evaluate the severity of malocclusion of all 206 casts on an 8-point scale. To arrange the casts according to the scale, they were instructed to first choose the casts that fit both ends of the scale according to the severity of malocclusion and from those 2 end points develop the rest of the groups. They were also instructed to avoid evaluating the cases on the difficulty of treatment. Instead, they were given an operational definition of normal occlusion9 and were asked to arrange the dental casts within the scale according to how far they deviated from “normal” occlusion. The cases were classified as skeletal Class I, II, and III by using cephalometric standards. Scale: 12 34 56 78 Slight Moderate Great Very Great DATA ANALYSIS

The data were analyzed in the following steps:

1. Cronbach’s α reliability analysis was used to determine the consistency of opinion among the 10 ortho-

(Constant) Upper RS Upper AS Upper LS Lower RS Lower AS Lower LS RB Occlusion Overjet Overbite CT Line LB Occlusion

β

SE

P

1.858 –.148 –.015 .106 .280 .059 .044 .219 1.012 .174 .134 .270

.182 .129 .027 .092 .093 .025 .112 .069 .068 .085 .137 .075

.000 .250 .589 .250 .003 .017 .694 .002 .000 .041 .329 .000

SE, Standard Error; P, Statistical Significance; RS, Right Segment; AS, Anterior Segment; LS, Left Segment; RB, Right Buccal; LB, Left Buccal.

dontists regarding the severity of malocclusion. 2. Pearson’s correlation coefficient was used to assess the association between the orthodontists’ subjective evaluation of the malocclusions and unweighted total PAR scores. 3. Multiple regression analysis was used to determine the optimum weight for the components of the PAR index. 4. Pretest and posttest differences were evaluated with t tests. 5. Multivariate analysis of variance (MANOVA) was calculated to assess differences in PAR scores in association with types of malocclusions and treatment categories. The treatment categories were defined as follows: • Category 1: Class II treatment • Category 2: Treatment of maxillary constriction • Category 3: Space management. RESULTS The Weighted PAR Index

Reliability analysis indicated a high level of agreement among the subjective rankings of malocclusion made by the 10 orthodontists acting as judges in this study (Cronbach α = 0.955). This confirmed that our judges were evaluating the casts on similar criteria. However, when the average of these rankings was correlated with the unweighted sum of the 11 items making up the PAR index, the level of agreement was much lower, albeit still significant (r = 0.550, P < .001. Correlations with individual rankings were lower, ranging from 0.387 to 0.581). Multiple regression analysis was performed to bring the components of the PAR index into line with the highly consistent criteria of our panel of judges.

546 Pangrazio-Kulbersh, Kaczynski, and Shunock

Table II. Pretreatment

American Journal of Orthodontics and Dentofacial Orthopedics May 1999

and posttreatment change in PAR indices

Dependent Variable

Pretreatment Mean ± SD

Posttreatment Mean ± SD

Mean Change

95% CI Change

t

Raw PAR index Weighted PAR index

15.82 ± 6.60 5.28 ± 1.44

8.82 ± 4.87 3.73 ± 1.15

7.00 1.59

5.79-8.21 1.24-1.86

11.51* 9.81*

*P < .001 CI, Confidence Interval; SD, Standard Deviation.

Table III. Grouping

of cases by PAR score change (%)

Worse/No Difference Pre Tx PAR 1.00 to 10 11 to 20 21 to 30 31 to 40 Total

Improved

Greatly Improved

Total

>30 52.4 47.5 40.0 66.7 47.6

>70 14.3 16.9 35.0 33.3 20.4

20.4 57.3 19.4 2.9 100.0

<30 33.3 35.6 25.0 32.0

Tx, Treatment.

The average ranking was used as the dependent variable. The 11 components of the PAR index were entered simultaneously into the solution as independent variables. Thus, the unstandardized regression coefficients would provide the weights needed to appropriately align the PAR items with the judges’ subjective rankings. The results of this analysis are shown in Table I. Interestingly, only 6 of the 11 measures achieved statistical significance in this analysis: overjet (P < .001), left buccal occlusion (P < .001), right buccal occlusion (P < .002), lower right segment (P < .003), lower anterior segment (P < .02), and overbite (P < .05). Nevertheless, the complete set of measurements and weights was used to maximize comparability with the original PAR index. The weighted PAR index correlated with the average subjective ranking at 0.822 (P < .001), a great improvement over the unweighted correlation of 0.550. This was very much in accordance with the findings reported by Richmond et al.9 Assessment of Change

A dependent groups t test was run to determine the effectiveness of orthodontic treatment in producing a change in PAR scores. This analysis checked for the presence of significant individual change over time. As shown in Table II, a significant change was observed in both the unweighted and weighted PAR scores. The mean percentage changes in the total sample were 38% and 25%, respectively, for the raw and

weighted PAR scores. Forty-eight percent of the total sample had PAR score reductions of at least 30%, moving to the improved category according to Richmond et al10 and Fox.13 Twenty percent had at least 70% reduction and belonged in the greatly improved category. The remaining 32% failed to decrease their scores by 30%; therefore, they remained in the no difference/worse category (Fig 1, Table III). The cases that started with a Class I dental and skeletal malocclusion showed a 20% mean reduction of their weighted PAR scores. The Class II skeletal and dental malocclusions showed 28% and 26% mean reductions, respectively. The Class III cases were not statistically evaluated because of their small numbers in this sample (Table IV and V, Figs 2 and 3). Change According to Treatment

Finally, MANOVA was conducted to determine whether the amount of change experienced was related to either the type of malocclusion or the type of mechanotherapy. Thus, a 3 × 3 repeated-measures design was tested. Results were nonsignificant for both interaction and main effects, suggesting that all patients experienced similar reductions in their PAR scores regardless of appliance or malocclusion type. Table VI10 illustrates the mean change in PAR scores from pretreatment to posttreatment in the 3 treatment categories. DISCUSSION

The results of this study suggest that a significant improvement in malocclusion occurs with early treatment, bringing the majority of the cases to PAR scores of 1-10. According to Richmond and Shaw,9,10 PAR scores of 1-5 indicate normal or almost normal malocclusion, therefore requiring little or no treatment. In his sample at Pittsburgh, Younis48 found that the cutoff point for treatment need was a PAR score of 15, whereas Richmond et al9,10 suggested a PAR score of 10. It could be speculated from the reduction in PAR scores in this study that very little or no treatment may be necessary after early correction of a developing malocclu-

Pangrazio-Kulbersh, Kaczynski, and Shunock 547

American Journal of Orthodontics and Dentofacial Orthopedics Volume 115, Number 5

Fig 1. Percentage of patients grouped by pretreatment PAR scores and outcomes.

Table IV. Percentage Molar Classification Class I

Class II

Total

change in raw and weighted PAR scores according to pretreatment molar classification .

T1 PAR

T2 PAR

% Change Raw PAR

PAR WT1

PAR WT2

% Change Wt. PAR

Mean N SD Mean N SD Mean N SD

15.55 33 6.99 16.26 66 6.44 16.02 99 6.60

8.06 33 5.02 9.35 66 4.75 8.92 99 4.85

−44.6600 33 33.1446 −34.6989 66 59.3797 −38.0193 99 52.1501

4.6468 33 1.1836 5.6770 66 1.4110 5.3336 99 1.4199

3.4402 33 1.0189 3.9266 65 1.1817 3.7628 98 1.1477

−19.4431 33 38.1742 −28.1733 65 27.4111 −25.2335 98 31.5230

SD, Standard Deviation.

sion. Of the remaining 32 subjects in the worse/no difference category, 5 cases remained the same while 3 increased their scores posttreatment. The remaining cases did not reduce their PAR scores by the 30% minimum. When the cases that remained unchanged or became worse were independently examined, lack of cooperation and an increasing number of missed appointments were observed. The pretreatment mean PAR score was 16, which could be interpreted as the sample not having a severe malocclusion. When the mean age of this sample (9.82 years) was compared with that of previous studies,11,13, 55 a large difference in age and pretreatment PAR scores

was present. Their samples’ mean age was approximately 14 years, and the mean pretreatment PAR index was 29. The large difference in pretreatment PAR scores could be due to the young age of our subjects. It is known from the literature and from clinical appraisals that a malocclusion rarely improves with time without orthodontic intervention.49-54 Therefore, older patients could have more severe malocclusion. Early orthodontic intervention could prevent malocclusions from getting progressively worse as the permanent teeth erupt and as growth and development continue. When the success rate of early treatment was compared with that obtained during late treatment,

548 Pangrazio-Kulbersh, Kaczynski, and Shunock

American Journal of Orthodontics and Dentofacial Orthopedics May 1999

Fig 2. Mean change in weighted PAR scores according to pretreatment molar classification.

Fig 3. Mean change in weighted PAR scores according to pretreatment skeletal classification.

higher reduction in the mean percentage change in PAR score was reported. Fox13 reported a 66% reduction in the PAR index in his mixed sample of fixed, removable, and functional appliances. Richmond11 and Buchanan et al55 reported a 74% mean reduction in their samples, including full fixed appliance treatment. In our sample, a 38% mean decrease was observed. Because the per-

centage reduction associated with treatment is related partly to the start PAR scores, this comparison should be interpreted with caution because higher start PAR scores tend to have greater percentage reduction.55 In this study, the patients with the highest pretreatment PAR scores also had the largest posttreatment decreases. Furthermore, the limited treatment goals associated

Pangrazio-Kulbersh, Kaczynski, and Shunock 549

American Journal of Orthodontics and Dentofacial Orthopedics Volume 115, Number 5

Table V. Percentage

change in raw and weighted PAR scores according to pretreatment skeletal classification

Skeletal Class Class I

Mean N SD Mean N SD Mean N SD

Class II

Total

T1 PAR

T2 PAR

% Change Raw PAR

PAR WT1

PAR WT2

% Change Wt. PAR

17.47 17 7.47 15.35 77 5.86 15.73 94 6.19

9.18 17 5.83 8.68 77 4.51 8.77 94 4.75

–49.1594 17 22.8817 −35.6600 77 57.7442 38.1014 94 53.3127

4.9936 17 1.2479 5.4566 77 1.4394 5.3729 94 1.4118

3.7456 17 1.0455 3.8125 76 1.1923 3.8003 93 1.167

–19.9598 17 35.2987 −26.4999 76 29.9441 −25.3044 93 30.8888

SD, Standard Deviation. Table VI. Mean

raw and weighted PAR scores by treatment category ± SD

Treatment Type

N

Pretreatment PAR

Posttreatment PAR

Pretreatment Weighted PAR

Posttreatment Weighted PAR

Class II Treatment Maxillary Expansion Space Management

47 18 5

15.64 ± 7.40 17.89 ± 6.28 15.59 ± 5.98

9.40 ± 5.24 9.83 ± 5.40 7.59 ± 4.09

5.64 ± 1.47 5.32 ± 1.42 4.54 ± 0.99

3.92 ± 1.25 3.72 ± 1.18 3.35 ± 0.85

SD, Standard Deviation.

with early treatment could have influenced the posttreatment mean percentage scores. If we also evaluate the cephalometric changes of the sample, further justification could be made toward the benefits of early treatment. Unfortunately, the PAR index does not measure skeletal changes or the psychological implications associated with the correction of dentoskeletal problems during this stage of development. Further studies relating cephalometric changes to occlusal scores will be beneficial to further validate the advantage of early intervention. CONCLUSIONS 1. A significant reduction of the mean PAR index was obtained from pretreatment to posttreatment (P = .000). 2. Forty-eight percent of the total sample improved PAR scores by a reduction of at least 30%. 3. Twenty percent of the sample greatly improved the PAR index by a reduction of 70%. 4. Thirty-two percent of the sample did not decrease the PAR index by 30%. They remained in the no difference/worse category. 5. The Class I cases had a 20% mean reduction in the PAR index, and the Class II dental and skeletal cases had 28% and 26% mean reductions, respectively, in PAR scores from pretreatment to posttreatment. 6. The 3 different treatment categories studied showed a significant reduction in the PAR index, but when com-

pared with each other there was no statistically significant difference in the reduction of the PAR scores.

We thank Drs. Hamed Haerian, Cameron George, Shelley Kavanaugh, and Geoffrey Stefko for their assistance with the statistical analysis and data collection. REFERENCES 1. Eismann D. A method of evaluating efficiency of orthodontic treatment. Trans Eur Orthod Soc 1974;223-32. 2. Eisman D. Reliable assessment of morphological change results from orthodontic treatment. Eur J Orthod 1980;2:19-25. 3. Elderton RJ, Clark JD. Orthodontic treatment in General Dental Service assessed by the Occlusal Index. Br J Orthod 1983; 10:178-86. 4. Elderton RJ, Clark JD. An investigation of treatment in the General Dental Service for patients with Class II div 1 malocclusions. Br J Orthod 1984;11:2-8. 5. Gottlieb EL. Grading your orthodontic treatment results. J Clin Orthod 1975;9:156-68. 6. Summers CJ. The occlusal index: a system for identifying and scoring occlusal disorders. Am J Orthod 1971;59:552-66. 7. Berg R. Post retention analysis of treatment problems and failures in 264 consecutively treated cases. Eur J Orthod 1979;1:55-68. 8. Pickering EA, Vig P. The occlusal index used to assess orthodontic treatment. Br J Orthod 1975;2:47-51. 9. Richmond S, Shaw WC, O’Brien KD, Buchanan IB, Jones R, Stephens CD, et al. The development of the P.A.R. index: reliability and validity. Eur J Orthod 1992;14:125-39. 10. Richmond S, Shaw WC, Roberts CT, Andrews M. The P.A.R. Index (Peer Assessment Rating: methods to determine outcome of orthodontic treatment in terms of improvements and standards. Eur J Orthod 1992;14:180-7. 11. Richmond S. Personal audit in orthodontics. Br J Orthod 1993;20:135-45. 12. De Guzman L, Bahiraei D, Vig KWL, Vig P, Weyant RJ, O’Brien K. The validation of the Peer Assessment Rating index for malocclusion severity and treatment difficulty. Am J Orthod Dentofac Orthop 1995;107:172-6. 13. Fox NA. The first 100 cases: a personal audit of orthodontic treatment assessed by the P.A.R. (Peer Assessment Rating) index. Br Dent J 1993;174:290-7. 14. Bishara SE, Fustus R, Graber TM. Proceedings of The Workshop Discussion on Early Treatment. Am J Orthod Dentofac Orthop 1978;113:5-6. 15. Grosfeld O, Migdalska Chojcka M. Interception of malocclusions in the deciduous dentition: long term results. Am J Orthod 1978;73:73-8. 16. Freeman JD. Preventive and interceptive orthodontics: a clinical review and the results of a clinical study. J Prev Dent 1977;4:7-23.

550 Pangrazio-Kulbersh, Kaczynski, and Shunock

17. Gottleib EL. Preserving the referral source practice of orthodontists. J Clin Orthod 1984;18:870-6. 18. Gianelly AA. Crowding: timing of treatment. Angle Orthod 1994;64:415-8. 19. Gianelly AA. Leeway space and the resolution of crowding in the mixed dentition. Semin Orthod 1995;1:188-94. 20. Ghafari J. Early treatment of dental arch problems: space maintanance, space regaining. Quintessence Int 1986;17:423-32. 21. Hotz R. Orthodontics in Daily Practice: Possibilities and Limitations in the Area of Children’s Dentistry. Bern, Switzerland: Hans Heuber; 1974. 22. Ringenberg QM. Serial extraction: stop, look, and be certain. Am J Orthod 1964;50:327-36. 23. Graber TM, Rakosi T, Petrovic AG. Dentofacial Orthopedics with Functional Appliances. St Louis: CV Mosby; 1985. 24. McNamara JA Jr. Functional determinants of craniofacial size and shape. Eur J Orthod 1980;2:131-59. 25. Harvold EP. The Activator in Interceptive Orthodontics. St Louis: CV Mosby; 1974. 26. Frankel R, Frankel C. Orofacial Orthopedics with the Function Regulator. Basel, Switzerland: Karger; 1989. 27. Lundstrom A. The aetiology of crowding of the teeth (based on studies of twins and on morphological investigations) and orthodontic treatment (expansion or extraction). Trans Eur Orthod Soc 1951;176-89. 28. Frankel R. Guidance of eruption without extraction. Trans Europ Orthod Soc 1971;47:303-15. 29. Frankel R. Decrowding during eruption under the screening influence of vestibular shields. Am J Orthod 1974;65:372-406. 30. Frankel R, Frankel C. Orofacial Orthopedics with the Function Regulator. Munich: S. Karger; 1989. 31. Haas AJ. Palatal expansion: just the beginning of dento-facial orthopedics. Am J Orthod 1970;57:219-55. 32. Kloehn SJ. A new approach to the analysis and treatment of mixed dentition. Am J Orthod 1953;31:161-86. 33. Wieslander L. Early or late cervical traction therapy in the mixed dentition. Am J Qrthod 1975;67:432-9. 34. Kerr WJS, TenHave TR, McNamara JA Jr. A comparison of skeletal and dental change produced by function regulators (FR-2 and FR-3). Eur J Orthod 1989;11:235-42. 35. Mitani H, Fukazawa H. Effects of chincap force on the timing and amount of mandibular growth associated with anterior reverse occlusion (Class III malocclusion) during puberty. Am J Orthod Dentofac Orthop 1986;90:454-63. 36. Sakamoto T. Effective timing for the application of orthopedic force in the skeletal Class III malocclusion. Am J Orthod 1981;80:411-6. 37. Sugawara J, Asano T, Endo N, Mitani H. Long-term effects of chincup therapy on skeletal profile in mandibular prognathism. Am J Orthod Dentofac Orthop 1990;98:127-33. 38. Frankel R. Maxillary retrusion in Class III and treatment with the functional corrector III. Trans Eur Orthod Soc 1970;249-59. 39. Howe RP, McNamara JA Jr, O’Connor KA. An examination of dental crowding

American Journal of Orthodontics and Dentofacial Orthopedics May 1999

40. 41.

42. 43. 44.

45.

46.

47.

48.

49. 50.

51. 52.

53. 54. 55.

and its relationship to tooth size and arch dimension. Am J Orthod 1983;83: 363-73. McNamara JA Jr. An orthopedic approach to the treatment ot Class III malocclusion in young patients. J Clin Orthod 1987;21:598-608. Keeling S, Wheeler T, King G, Garvan C, Cohen DA, Cabassa S, et al. Anteroposterior skeletal and dental changes after early Class II treatment with bionators and headgear. Am J Orthod Dentofac Orthop 1998;113:40-50. Roberts CA, Subtelny JD. Use of the face mask in the treatment of maxillary skeletal retrusion. Am J Orthod Dentofac Orthoped 1988;93:388-94. Turley PK. Orthopedic correction of Class III malocclusion with palatal expansion and custom protraction headgear. J Clin Orthod 1988;22:314-25. Livieratos FA. Class II treatment: a comparison of one and two stage non extraction alternatives. In: McNamara JA, editor. Orthodontic treatment outcome and effectiveness. Craniofacial Growth Series. Ann Arbor Center for Growth and Development, University of Michigan, 1995:163-93. Keeling SD, King GJ, Wheeler TT, McGovray S. Timing of class II treatment: rationale, methods and early results of an ongoing randomized clinical trial. In: McNamara JA, editor. Orthodontic treatment: outcome and effectiveness. Craniofacial Growth Series. Ann Arbor Center for Growth and Development, University of Michigan, 1995:81-112. Tullock JFC, Phillips C, Proffit WR. Early versus late treatment of Class II malocclusion: Preliminary results from the University of North Carolina clinical trial. In: McNamara JA, editor. Orthodontic treatment: outcome and effectiveness. Craniofacial Growth Series. Ann Arbor Center for Growth and Development, University of Michigan 1995:113-38. Tullock JFC, Phillips C, Proffit WR. Benefits of early Class II treatment: Progress report of a phase randomized clinical trial. Am J Orthod Dentofac Orthoped 1998;113:62-72. Younis JW. Validation of the Index of Orthodontic Treatment Need in the United States, submitted to the graduate faculty of the University of Pittsburgh in partial fulfillment of the requirements for the degree of Master of Dental Sciences, 1995. Brodie AG. Behavior of normal and abnormal facial growth patterns. Am J Orthod Oral Surg 1941;27:633-47. McNamara JA Jr. Neuromuscular and Skeletal Adaptations to Altered Orofacial Function. Monograph 1. Craniofacial Growth Series, Center for Human Growth and Development, The University of Michigan, Ann Arbor, 1972. Moorrees CFA. The Dentition of the Growing Child. Cambridge, Mass: Harvard University Press; 1959. Moyers RE, van der Linden FPGM, Riolo ML, McNamara JA Jr. Standards of Human Occlusal Development. Monograph 5, Craniofacial Growth Series, Center for Human Growth and Development, The University of Michigan, Ann Arbor, 1976. Summers CJ. Some effects of developmental changes on the indices of malocclusion, J Public Health Dent 1966;26:212-20. Salzmann JA. Orthodontics in practice and perspective, Am J Orthod 1969;55:556-65. Buchanan IB, Russel JI, Clark JD. Practical Application of the P.A.R. Index: An Illustrative Comparison of the Outcome of Treatment Using Two Fixed Appliance Techniques, Br J Orthod 1996;23:231-357.