INTRODUCTION

In orthodontics, it is important to objectively assess whether a worthwhile improvement has been achieved in terms of overall alignment and occlusion for an individual patient and the greater proportion of a practitioner's caseload.1 Several indices have been developed specifically to assess the success of orthodontic treatment by comparing pre- and posttreatment records to register the outcome of orthodontic care.2–4 Summers occlusal index,5 a treatment index, has also been used to assess the outcome of treatment.67

In 1987, the Peer Assessment Rating (PAR) Index was developed over a series of six meetings with a group of 10 experienced orthodontists (British Orthodontic Standards Working Party).8 The PAR Index provides a single summary score for all the occlusal anomalies which may be found in a malocclusion. The score provides an estimate of how much a case deviates from normal alignment and occlusion. The difference in scores between the pre- and posttreatment patients reflects the degree of improvement and, therefore, the success of the treatment.

The PAR Index has been used to evaluate treatment standard among general practitioners and orthodontic specialists.910 The PAR Index has its limitations, principally because of the high weight assigned to overjet, and difficulties still arise from the application of one weighting system to all malocclusions because occlusal features vary in importance in different classes of malocclusion.11 However, it remains a valid and reliable index in the assessment of orthodontic treatment outcome. Treatment outcome can be assessed using the PAR Index in two ways: using the numerical reduction in the weighted PAR score or using the percentage reduction in the weighted PAR score.

Unlike in Europe, there is paucity of information on the outcome of orthodontic treatment in the United States assessed using the PAR Index. The report of Holman et al12 on extraction vs nonextraction orthodontic treatment using the PAR Index applied a US weighting system which was developed specifically for white United States patients.

Therefore, this article is aimed at assessing the orthodontic treatment outcome in an accredited graduate orthodontic clinic in the United States, using the original PAR Index.

MATERIALS AND METHODS

A sample of 100 patients was randomly selected from the model store of an accredited graduate orthodontic clinic in the United States. No consideration was given to race or sex in the selection of patients for the sample. The duration of treatment was calculated by subtracting the age at the onset of treatment as recorded on the pretreatment models from the age at the end of treatment as indicated on the posttreatment models.

The pretreatment and posttreatment PAR scores were recorded according to Richmond et al1 All the components of the PAR Index were assessed. Treatment outcome can be assessed using the PAR Index in two ways: using the numerical reduction in the weighted PAR score or using the percentage reduction in the weighted PAR score.

Several conventions have been derived from the panel of 74 examiners (members of the British Orthodontic Society) to categorize improvement:

1. “Worse or no different,” patients who show less than a 30% reduction in weighted PAR score.

2. “Improved,” patients showing greater than or equal to 30% reduction in weighted PAR score.

3. “Greatly improved,” generally a reduction of 22 weighted PAR points (greatly improved reflects the severity of the initial malocclusion).

In assessing the standard of orthodontic treatment, a high standard of care is achieved when the proportion of patients treated by an individual practitioner falling into the worse or no different category is negligible and the mean percentage reduction in weighted PAR score is high (eg, greater than 70%). If a high proportion of patients also fall into the category “greatly improved” (eg, greater than 40%), then this suggests that the practitioner is treating a substantial proportion of patients with significant malocclusion to high standard of care.

Intraexaminer reliability

The errors of the method for the recording of the PAR Index were evaluated from double recordings of 20 randomly selected patients from the original sample. The pretreatment and posttreatment study casts were evaluated a second time; 40 double recordings were performed. The random or accidental error for the weighted and unweighted PAR Index scores was evaluated with the formula:

Where di is the difference between the double determinations and n is the number of double determination.13 For further measure of the reliability of the measurements, the intraclass correlation coefficient was calculated using the Spearman rank-order correlation coefficient.14

RESULTS

From the 40 double recordings, the Si value was 0.30 for unweighted total PAR Index and 0.67 for the weighted total PAR Index. There was no systematic bias between the two sets of PAR scores for the double recordings.

Intraexaminer correlation coefficient results revealed that pretreatment (a) vs pretreatment (b) showed excellent correlation (r = .99; P < .0001) and posttreatment (a) vs posttreatment (b) indicated very good correlation (r = .83; P < .0001).

Pretreatment and posttreatment unweighted and weighted PAR scores are shown in Tables 1 and 2. The upper anterior segment had the highest mean for both unweighted and weighted pretreatment PAR scores. In the posttreatment PAR score, the right buccal occlusion had the highest mean for unweighted PAR score, whereas overjet had the highest for weighted.

TABLE 1.  Mean Unweighted and Weighted Pretreatment PAR Scores^a

View Fullscreen

TABLE 2.  Mean Unweighted and Weighted Posttreatment PAR Scores^a

View Fullscreen

Table 3 shows the PAR Index descriptors for the whole sample. The mean start weighted PAR score was 23.83 ± 11.55 (SD) and the finish PAR score was 1.72 ± 3.79 (SD). The weighted PAR point reduction was 22.11 ± 11.72 (SD), whereas the mean percentage reduction was 86.20 ± 27.44 (SD).

TABLE 3.  PAR Index Descriptives for the Whole Sample^a

View Fullscreen

The numbers in the right-hand column in Table 4 indicate the number corrected out of the number of patients in whom improvement was indicated or possible, as shown on the first column on the left-hand side. In parentheses are the corresponding percentages. The upper anterior segment malocclusions had the highest percentage improvement (94.7%) followed by overjet (92.8%), as shown in Table 4.

TABLE 4.  Improvement of Individual PAR Components^a

View Fullscreen

Table 5 shows the relationship of pre- and posttreatment PAR scores, reduction, and percentage reduction in weighted PAR scores for the three categories of improvement. The greatly improved category had 50 patients (50%), and 47 (47%) were in the improved category, while only 3 (3%) were found to belong to the worse or no different category (Figure 1).

TABLE 5.  The Relationship of Pre- and Posttreatment, Reduction, and Percentage Reduction in Weighted PAR Scores for the Three Classifications of Improvement^a

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

A multiple comparison test was used, which indicated statistically significant differences (Scheffe < .05) between the three groups for all the variables recorded (pretreatment, posttreatment, reduction, and percentage reduction in weighted PAR scores).

Patients having less than 22 pretreatment weighted PAR scores were 46 in number (46%). The average duration of treatment was 26 months. Finish weighted PAR scores between 0 and 5 points accounted for 88%.

DISCUSSION

For a practitioner to demonstrate consistently high standards, the proportion of the caseload in the worse or no different category should be negligible, and the mean percentage reduction should be as high as possible.

A high standard of treatment may be judged according to the mean percentage reduction in weighted PAR score for an individual practitioner's caseload, eg, greater than 70%.1 The present US study has revealed a mean percentage reduction in weighted PAR score of over 86%. This parameter was reported to be about 47% in the General Dental Service Orthodontics in England and Wales.15 In a pilot study to assess the standard of orthodontic treatment in Norway, the mean percentage reduction in weighted PAR score for all the patients was 78%, with four of the patients in the worse or no different category. In a related previous study in the United States,12 the mean percentage reduction in weighted PAR score was 79.41% for the extraction group and 77.63% for the nonextraction group.

For a practitioner to produce high standards and treat those patients who have perhaps a greater need for treatment, not only must the mean percentage reduction for the caseload be high (eg, greater than 70%), but the percentage of patients having been greatly improved should also be high (eg, greater than 40%).1 In the present study, the greatly improved category was 50%. Firestone et al16 reported, for the greatly improved category, an average of 27% for patients treated by postgraduate and undergraduate students in Bern, Switzerland. In the same Swiss report,16 it was 28% in 1983 and 40% in 1993 for the postgraduate students. The same Swiss report indicated 12% of patients in worse or no different group, as against 3% in the present US study. Kerr et al17 reported 89.3% of patients as either improved or greatly improved with 16 (10.7%) classified as worse or no different, in assessing the effectiveness of removable orthodontic appliances.

When considering the change in PAR score for a group of subjects, it must be borne in mind that the amount of improvement possible under the nomogram system is influenced by the initial PAR score.17 If this is less than 22, it is impossible for the patients to become greatly improved. In the UK (Glasgow) study by Kerr et al,17 this was found to apply to one third (49) of the patients in their study, as against 46% in the present US study. This high percentage of patients with initial PAR score of less than 22 in the present study contributed in lowering the percentage of patients belonging to the greatly improved category.

Although this study revealed many patients with posttreatment PAR score indicating ideal occlusions, most of them could not be classified as greatly improved. According to Kerr et al,17 if all the patients in a given study had initial PAR scores of less than 22, none could be greatly improved, and one might, therefore, infer that the treatment regimen used was in some way defective. Consequently, it would seem reasonable to identify the proportion of patients in this category to allow a more objective assessment of the outcome.

Longer dual arch fixed appliances were reported as having the tendency to reduce PAR scores to lower levels.15 The present study assessed patients treated with dual arch fixed appliances (preadjusted edgewise technique) because it is the practice at the center. The average duration of treatment for the present study sample was 26 months. The random sample of 1010 patients treated with removable or fixed appliances (or both) in the General Dental Services of England and Wales showed that one out of every five patients would be categorized as worse or no different, with an overall mean percentage reduction in weighted PAR score of 55%.1 When the standard of treatment was assessed in relation to the appliance used, the use of upper and lower fixed appliances produced the best standard of treatment (71.4% reduction in weighted PAR score). In addition, the patient was less likely to be worse after treatment than other patients treated with removable or functional appliances. Single arch removable and fixed appliances showed a mean percentage reduction of around 50%, with a substantial proportion of patients not benefiting from orthodontic treatment.

The present study showed that the mean pretreatment weighted PAR score was 23.83 and the posttreatment weighted PAR score was 1.72. Firestone et al16 reported 25 in 1983 and 24.8 in 1993 for pretreatment PAR scores, and 8.3 in 1983 and 5.3 in 1993, for posttreatment PAR scores in Switzerland. Turbill et al15 reported pretreatment weighted PAR scores of 26.94, 26.74, and 24.74 for patients treated with removable appliances, respectively. Their corresponding posttreatment weighted PAR scores were 12.79, 15.19, and 11.40, respectively. The study of Holman et al12 showed pretreatment PAR score of 30.01 for extraction group as against 25.21 for the nonextraction group in a related study that used the American weighting system.

CONCLUSIONS

• Improved and greatly improved categories of improvement constituted about 97% of the patients, with only 3% in the worse or no different group.

• The mean weighted PAR reduction was 22.11, whereas the mean weighted percentage reduction was 86.20, indicating excellent improvement of the patients according to the PAR Index.

• This accredited graduate orthodontic clinic in the United States is treating a great proportion of patients with a clear need for treatment to a high standard. This could be a reflection of the standard of treatment in the United States.