Download - The Angle Orthodontist

Original Article
The relationship between the ABO discrepancy index and treatment duration
in a graduate orthodontic clinic
Laura D. Parrisha; W. Eugene Robertsb; Gerardo Maupomec; Kelton T. Stewartd;
Robert W. Bandye; Katherine S. Kulaf
ABSTRACT
Objective: To test the hypothesis that there is no relationship between the components of the
American Board of Orthodontics (ABO) discrepancy index (DI) and duration of orthodontic
treatment.
Materials and Methods: A retrospective review of 732 patient records with permanent dentition
was performed. Pretreatment radiographs and casts were used to determine the DI score. Other
data collected were total treatment duration, age, sex, ethnicity, and the date fixed appliances were
removed. Reliability tests showed substantial agreement between examiners (Cohen’s kappa
0.68–0.94). Pearson and Spearman correlation coefficients were used to assess the association
between the DI scores and length of treatment. A multiple variable regression analysis was used to
determine which variables predict treatment duration (P , .05 significant).
Results: There was a significant association between the DI and treatment duration. There was a
significant multivariate association for DI components (occlusions, crowding, overjet, cephalometrics, overbite, lateral open bite, and tooth transposition) and treatment duration.
Conclusions: The hypothesis was rejected. This retrospective study of university clinical records
showed that the average increase in treatment duration was about 11 days for each point increase in
total DI score. Treatment duration was differentially increased by various components of the DI:
approximately 6.5 months for tooth transposition; approximately 1 month for crowding, overjet, or
overbite; approximately 3 weeks for occlusion discrepancies; approximately 2 weeks for lateral open
bite; and approximately 5 days for cephalometric discrepancies. (Angle Orthod. 2011;81:192–197.)
KEY WORDS: Discrepancy index; Treatment duration; Malocclusion
INTRODUCTION
Orthodontic Resident, School of Dentistry, Indiana University, Indianapolis, Ind.
b
Professor Emeritus, Department of Orthodontics and
Oral Facial Genetics, School of Dentistry, Indiana University,
Indianapolis, Ind.
c
Professor, Department of Preventive and Community Dentistry, School of Dentistry, Indiana University, and The Regenstrief Institute Inc, Indianapolis, Ind.
d
Assistant Professor, Department of Orthodontic and Oral
Facial Genetics, School of Dentistry, Indiana University, Indianapolis, Ind.
e
Biostatistician II, Division of Biostatistics, School of Medicine,
Indiana University, Indianapolis, Ind.
f
Chair and Program Director, Joseph R. and Ada Jarabak
Endowed Professor, Department of Orthodontics and Oral Facial
Genetics, School of Dentistry, Indiana University, Indianapolis,
Ind.
Corresponding author: Katherine S. Kula, MS, DMD, MS,
1121 W Michigan St, Indianapolis, IN 46202
(e-mail: [email protected])
a
Evaluating the likely duration of a patient’s orthodontic
treatment remains an elusive component of clinical
management. Although a finished orthodontic case
could have an esthetic result and meet American Board
of Orthodontics (ABO) standards, the patient might
consider it unsatisfactory if treatment duration exceeded
the initial prediction. Patients who receive accurate
information about predicted treatment duration are
found to be more satisfied with their treatment and have
more realistic expectations of outcomes.1 Satisfied
patients also are more likely to refer other patients.2
An accurate estimate of treatment length is not only
valuable to the patient to determine time absent from
work or school, but also allows the orthodontist to predict
overhead costs and determine appropriate fees.3–5 It is
for these reasons that it is critical for practicing clinicians
to predict accurate treatment duration.
One possible method of prediction utilizes the
discrepancy index (DI), a pretreatment scoring system
Accepted: August 2010. Submitted: June 2010.
2011 by The EH Angle Education and Research Foundation,
Inc.
G
Angle Orthodontist, Vol 81, No 2, 2011
192
DOI: 10.2319/062210-341.1
193
DISCREPANCY INDEX AND TREATMENT DURATION
developed by the ABO for Phase III of the orthodontic
board certification exam. The DI has become an
accepted and reliable index for quantifying the
complexity of cases based on pretreatment orthodontic
record analysis and measurements from dental casts
and radiographs.6,7 However, the DI scoring method
has been modified during the past few years in the
‘‘other’’ category.
Multiple studies have sought to determine which
factors best predict treatment duration. Poor patient
compliance, missed appointments, broken appliances,
and poor oral hygiene can contribute to longer
duration.1,2,4,5,8–12 Furthermore, excessive duration
may lead to poor oral hygiene, patient compliance,
and case outcomes.13
To date, no studies have quantified the relationship
of DI score to treatment duration. The purpose of this
study was to determine if there was a significant
association between total DI score or DI components
with treatment duration in a graduate orthodontic
clinic.
MATERIALS AND METHODS
Institutional review board approval was obtained for
this retrospective study of patient records from 1720
consecutively completed cases started and finished by
orthodontic residents in a 2-year university clinic from
1998–2004. Cases were excluded if: (1) they were in
the mixed dentition, (2) casts were broken, or (3) other
records were incomplete. A total of 732 cases were
analyzed.
Data collection was completed in two parts. In part
one, DI scores for 716 cases were collected by the
senior author and five undergraduate dental students.
Records were randomized for each year and divided
among the six examiners. Pretreatment cephalometric
tracings (ANB, SN-GO-GN, and IMP angles), panoramic radiographs (axial inclinations) and casts (overjet, overbite, anterior open bite, lateral open bite,
crowding, occlusion, lingual posterior crossbite, buccal
posterior crossbite, and other) were used to collect the
DI score. Treatment duration, age, sex, ethnicity, and
year debonded (removal of fixed appliances) were
also recorded. The DI scores calculated followed the
formula outlined by Cangiolosi et al6 and established in
the ABO November 2006 guidelines.
Calibration of the six examiners was achieved by
scoring 20 cases at 1–2 week intervals until a 65%
margin of error was achieved. Cohen’s kappa coefficients ranged from 0.68–0.94 depending on the
variable scored. Data collection began once the kappa
coefficient was greater than 0.61, indicating substantial
agreement.14 In part two, the senior author reviewed
the first data set and found an additional 16 cases that
were not scored. These cases were scored the same
as part one and increased the sample size to 732. Of
the 732 cases, 189 had a score in the ‘‘other’’ category
of the DI. Since the DI was revised in 2009 to specify
components of the ‘‘other’’ category with definitive
point values after the original scoring, these components and point values were recorded.14,15 The DI
scores were updated to reflect the revised scoring.
Pretreatment radiographs were used to evaluate
supernumerary teeth, impacted teeth, missing teeth,
skeletal asymmetries, and additional treatment complexities (ie, ectopic eruption). Patient photos and
casts were used to evaluate midline discrepancy,
anomalous morphology, spacing, ankylosis of permanent teeth, and tooth transposition.
The second part of data collection was completed by
the senior author. Twenty randomly selected cases
included in the 732 cases were rescored to assess
reliability. During the 2 months of data collection, the
same 20 cases were scored at the beginning, after
2 weeks, and at the end of data collection. Test-retest
reliability was excellent (ICC . 0.90).
Pearson and Spearman correlation coefficients were
calculated along with plots to evaluate the associations
of the DI score and the individual DI components with
treatment duration. A multiple variable regression
analysis was performed for significant DI score (or
individual components) to assess significance of
additional possible predictors of treatment duration
(age, sex, ethnicity, and the year treatment finished)
and significance of individual DI components after
adjusting for other DI components. With a sample size
of 732, the study had .99% power to detect a
difference in the correlation between total DI and
treatment duration if the true correlation was 0.30,
assuming a two-sided 5% significance level for the
test.
RESULTS
The population had 59% women and 41% men;
mean age was 19 6 9.4 years. The majority were
White (84%), followed by African American (8%), other
(4%), and Asian and Hispanic (2% each). Between 80
and 123 cases (11%–17%) were used each year from
1998–2004 (Table 1). None of the demographic
variables were significant in any of the models.
The mean total DI score was 15.7; mean treatment
duration was 2.6 years (Table 2). Since the magnitude
and inference were similar between Pearson and
Spearman coefficients for each DI variable, only
Pearson correlation coefficients were reported in
Table 2. Total DI score explained 9% of the variability
in treatment duration, occlusion 5%, overjet 5%,
crowding 3%, and none of the remaining individual
Angle Orthodontist, Vol 81, No 2, 2011
194
PARRISH, ROBERTS, MAUPOME, STEWART, BANDY, KULA
Table 1. Demographics (N 5 732)a
Variable
Mean
SD
Min
Max
Age, years
Variable
Sex
19
n (%)
9.35
9.7
66.7
Female
Male
435 (59%)
297 (41%)
Race/ethnicity
White
African American
Asian
Hispanic
Other
Year
1998
1999
2000
2001
2002
2003
2004
a
618
59
18
11
26
(84%)
(8%)
(2%)
(2%)
(4%)
123
115
89
80
85
117
123
(17%)
(16%)
(12%)
(11%)
(12%)
(16%)
(17%)
Table 2. Descriptive Statistics and Pearson Correlation Coefficient
vs Treatment Time for Treatment Time and DI Components (N
5 732)a
Variable
Mean
SD
Treatment time, years
DI total
DI components
2.6
15.7
1.07
10.51
Overjet
Overbite
Anterior open bite
Lateral open bite
Crowding
Occlusion
Lingual posterior
crossbite
Buccal posterior
crossbite
Cephalometrics
Other
1.9
1.6
0.8
0.5
2.2
2.5
Min Max Correlation
0.3
1
7.5
78
N/A
0.30****
2.15
1.64
3.38
1.93
1.99
2.99
0
0
0
0
0
0
25
5
44
26
7
16
0.22****
0.12***
0.03
0.07
0.16****
0.23****
0.5
1.18
0
9
0.10**
0.2
4.8
0.8
0.75
6.60
1.49
0
0
0
8
55
9
0.07
0.13***
0.11**
a
SD indicates standard deviation; Min, minimum; Max, maximum;
N/A, not applicable; DI, discrepancy index.
** P , .01; *** P , .001; **** P , .0001.
SD indicates standard deviation; Min, minimum; Max, maximum.
components explained more than 2% of the variability
in treatment duration. Spearman correlation coefficients are reported in Table 3 for the DI ‘‘other’’
subcategories because of the large proportion of zero
values (ie, at least 94% for each subcategory). None of
the individual ‘‘other’’ subcategories explained more
than 1% of the variability in treatment duration.
A multiple variable regression analysis including DI
score failed to identify demographic variables as
significant, so they were excluded from the model.
There was a significant, but weak, correlation between
the DI and treatment duration (P , .0001; r 2 5 0.09).
The average increase in treatment duration in years
was 0.031 (approximately 11 days) for each point
increase in total DI score. A scatter plot of treatment
duration vs total DI showed the regression equation:
treatment duration 5 0.031 3 DI + 2.098 (Figure 1).
A multiple variable regression analysis including DI
components, DI ‘‘other’’ components, and demographic variables identified the components of occlusion,
crowding, overjet, cephalometrics, overbite, lateral
open bite, and tooth transposition as significant. The
results of the multiple variable regression analyses and
the individual regressions with each significant DI
component are reported in Table 4. There was a
significant association between the components of the
DI and treatment duration (P , .0001; r 2 5 0.14). After
adjusting for the other significant DI components, the
average increase in treatment duration per point
increase was as follows: tooth transposition approximately 199 days, crowding 30 days, overbite 27 days,
overjet 25 days, occlusion 21 days, lateral open bite
14 days, and cephalometrics 5 days.
Angle Orthodontist, Vol 81, No 2, 2011
DISCUSSION
The large population used in this study allowed for a
precise estimation of the correlation between the DI
score and treatment duration (95% CI: 0.23, 0.36).
Although various factors contribute to the length of
orthodontic treatment, the DI is one method that can be
used for prediction. The DI is a reliable index for
measuring complexity of malocclusion.16 In agreement
with Vu et al.,17 the DI score is correlated with increased
treatment duration. The current study showed that the
average increase in treatment duration is about 11 days
for each point increase in total DI score, so a 10-point
increase in DI score will increase treatment duration by
110 days on average.
While the current study was the first to quantify the
connection between DI and treatment duration, other
studies found similar results. Buccal occlusion,9 specifically Class II division 1, increased treatment duration
by approximately 150 days.18 As documented by the
current data, a patient who is a full step Class II would
score eight points for occlusion, and the treatment
would increase by approximately 180 days, which
agrees with other studies that found crowding, 2
increased overbite, and excessive overjet extend
treatment.19 The cephalometric variables (ANB, SNMP, and IMPA) prolonged treatment by 5 days for each
point in this category. Previous studies reported
increases in the ANB angle lengthens treatment
duration.8,20
While numerous studies reported that cases requiring extractions, particularly premolars, have a significant increase on treatment duration,2,8,10,11,18,19,21,22 the
current study could not evaluate this influence because
195
DISCREPANCY INDEX AND TREATMENT DURATION
Table 3. Frequency and Percentages for DI Other Subcategories and Spearman Correlation vs Treatment Time (N 5 732)a
DI
0
2
4
6
DI Other Subcategory
n (%)
n (%)
n (%)
n (%)
Correlation
Additional treatment complexities
Ankylosis of permanent teeth
Anomalous morphology
Impaction
699 (95%)
732 (100%)
703 (96%)
27 (4%)
0 (0%)
24 (3%)
6 (1%)
0 (0%)
5 (1%)
0 (0%)
0 (0%)
0 (0%)
0.07
N/A
0.01
689
704
723
726
(94%)
(96%)
(99%)
(99%)
35
27
3
5
(5%)
(4%)
(,1%)
(1%)
7
1
5
1
(1%)
(,1%)
(1%)
(,1%)
1
0
1
0
(,1%)
(0%)
(,1%)
(0%)
0.09*
0.09*
0.01
0.04
690
693
704
732
727
721
727
726
(94%)
(95%)
(96%)
(100%)
(99%)
(98%)
(99%)
(99%)
42
25
20
0
5
11
4
5
(6%)
(3%)
(3%)
(0%)
(1%)
(2%)
(1%)
(1%)
0
11
6
0
0
0
1
1
(0%)
(2%)
(1%)
(0%)
(0%)
(0%)
(,1%)
(,1%)
0
3
2
0
0
0
0
0
(0%)
(,1%)
(,1%)
(0%)
(0%)
(0%)
(0%)
(0%)
0.09*
0.05
0.02
N/A
20.07
0.01
0.04
0.11**
All
Unilateral canine
Bilateral canine
Not canine
Midline discrepancy ($3 mm)
Missing teeth (except 3rds)b
Missing teeth congenital
Skeletal asymmetry
Spacing (4 or more mm per arch)
Spacing (diastema $ 2 mm)
Supernumeraryb
Tooth transposition
a
DI indicates discrepancy index; N/A, not applicable.
DI 2 includes 1 and 2; DI 4 includes 3 and 4; DI 6 includes 5 and 6.
* P , .05; ** P , .01.
b
Figure 1. Scatter plot of treatment duration (years) and total DI (correlation 5 0.30) regression equation: y 5 0.031x + 2.098.
Angle Orthodontist, Vol 81, No 2, 2011
196
PARRISH, ROBERTS, MAUPOME, STEWART, BANDY, KULA
Table 4. Regression Results for Analysis of Treatment Time with
Significant DI Components Combined and DI Total Individuallya
Variable
Parameter Estimatea
Model r 2
DI total
DI component model
.031****
0.09
0.14
Occlusion
Crowding
Overjet
Cephalometrics
Overbite
Lateral open bite
Tooth transposition
.058****
.082****
.069***
.015**
.075**
.039*
.545**
a
DI indicates discrepancy index. DI component model excludes DI
total.
b
Parameter estimates are based on multiple regression for the DI
component model estimates and an individual model for the DI total
estimates.
* P , .05; ** P , .01; *** P , .001; **** P , .0001.
potential extractions are not specifically scored with
the DI. Transposed teeth, although composing only a
small portion of the population (,1%), posed the
greatest increase in treatment duration (approximately
199 days). This study also found a 14-day increase for
each point increase in lateral open bite, which had not
been previously reported. This small amount of
increased treatment duration may not adequately
represent the difficulty clinically because lateral open
bite is heavily weighted in the DI scoring.
Contrary to the findings of Stewart et al.23 that
impacted maxillary canines lengthened treatment by
approximately 72 days for the unilateral-impacted
canine group, and approximately 297 days for the
bilateral-impacted canine group, the current study was
unable to identify a significant influence on treatment
duration after adjusting for the other DI components.
There was a significant bivariate correlation between
unilateral canine impaction and treatment time (Table 3), but it was not significant in the multiple variable
model (Table 4).
Even though the high variability in treatment duration
is often related to compliance, practitioners can use
the results of this retrospective study as an aid to
predict treatment duration. Though the DI was constructed by the ABO as a measure of ‘‘case complexity,’’ the index also has value beyond its original intent
by predicting time in treatment.
CONCLUSIONS
N This retrospective study of university clinical records
comprising diverse clinicians showed that treatment
duration increased 11 days for each point in the DI
score. The components with the strongest influence
were about:
N 119 days for each point increase in tooth transposition;
Angle Orthodontist, Vol 81, No 2, 2011
N 28–30 days for each point increase in crowding,
overjet, and overbite;
N 21 days for each point increase in occlusion;
N 14 days for each point increase in lateral open bite;
and
N 5 days for each point increase in cephalometrics.
ACKNOWLEDGMENTS
The authors would like to thank George Eckert, James
Shanks, James Baldwin, and Gayle Massa for their support on
the project and Brandon Parrish, Ryan Pulfer, Carl Drake, Sean
Schafer, and Kristi Donnelly for helping with part one of data
collection. Partial funding was provided by the Indiana University
School of Dentistry Graduate Education Fund.
REFERENCES
1. Mavreas D, Athanasiou AE. Factors affecting the duration of
orthodontic treatment: a systematic review. Eur J Orthod.
2008;30:386–395.
2. Skidmore KJ, Brook KJ, Thomson WM, Harding WJ. Factors
influencing treatment time in orthodontic patients.
Am J Orthod Dentofacial Orthop. 2006;129:230–238.
3. Bellavia DC. An effective fee schedule. J Clin Orthod. 1991;
25:546–554.
4. Beckwith FR, Ackerman RJ Jr, Cobb CM, Tira DE. An evaluation of factors affecting duration of orthodontic treatment.
Am J Orthod Dentofacial Orthop. 1999;115:439–447.
5. Shia GJ. Treatment overruns. J Clin Orthod. 1986;20:
602–604.
6. Cangialosi TJ, Riolo ML, Owens SE Jr, et al. The ABO
discrepancy index: a measure of case complexity.
Am J Orthod Dentofacial Orthop. 2004;125:270–278.
7. Pulfer RMDC, Maupome G, Eckert GJ, Roberts WE.
Influence of malocclusion complexity on orthodontics treatment outcome. Angle Orthod. 2009;79:468–472.
8. Fink DF, Smith RJ. The duration of orthodontic treatment.
Am J Orthod Dentofacial Orthop. 1992;102:45–51.
9. Robb SI, Sadowsky C, Schneider BJ, BeGole EA. Effectiveness and duration of orthodontic treatment in adults and
adolescents. Am J Orthod Dentofacial Orthop. 1998;114:
383–386.
10. Taylor PJ, Kerr WJ, McColl JH. Factors associated with the
standard and duration of orthodontic treatment. Br J Orthod.
1996;23:335–341.
11. O’Brien KD, Robbins R, Vig KW, Vig PS, Shnorhokian H,
Weyant R. The effectiveness of Class II, division 1 treatment.
Am J Orthod Dentofacial Orthop. 1995;107:329–334.
12. Jarvinen S, Widstrom E, Raitio M. Factors affecting the
duration of orthodontic treatment in children. A retrospective
study. Swed Dent J. 2004;28:93–100.
13. Pinskaya YB, Hsieh TJ, Roberts WE, Hartsfield JK.
Comprehensive clinical evaluation as an outcome assessment for a graduate orthodontics program. Am J Orthod
Dentofacial Orthop. 2004;126:533–543.
14. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1997;33:159–174.
15. ABO Discrepancy Index Instructions Version 2009–
2010. Available at: http://www.americanboardortho.com/
professionals/downloads/Discrepancy%20Index%20Scoring%
20Instructions.pdf. Accessed October 17, 2009.
16. Schafer SM, Maupome G, Eckert GJ, Roberts WE. The
discrepancy index relative to age, sex and the probability of
DISCREPANCY INDEX AND TREATMENT DURATION
completing treatment by a single resident in a two-year
graduate orthodontics program. Am J Orthod Dentofacial
Orthop. In press.
17. Vu CQ, Roberts WE, Hartsfield JK Jr, Ofner S. Treatment
complexity index for assessing the relationship of treatment
duration and outcomes in a graduate orthodontics clinic.
Am J Orthod Dentofacial Orthop. 2008;133:9.e1–13.
18. Vig KW, Weyant R, Vayda D, O’Brien K, Bennett E.
Orthodontic process and outcome: efficacy studies—strategies for developing process and outcome measures:
a new era in orthodontics. Clin Orthod Res. 1998;1:
147–155.
19. Fisher MA, Wenger RM, Hans MG. Pretreatment characteristics associated with orthodontic treatment duration.
Am J Orthod Dentofacial Orthop. 2010;137:178–186.
197
20. Popowich K, Nebbe B, Heo G, Glover KE, Major PW.
Predictors for Class II treatment duration. Am J Orthod
Dentofacial Orthop. 2005;127:293–300.
21. Vig PS, Weintraub JA, Brown C, Kowalski CJ. The duration
of orthodontic treatment with and without extractions: a pilot
study of five selected practices. Am J Orthod Dentofacial
Orthop. 1990;97:45–51.
22. Turbill EA, Richmond S, Wright JL. The time-factor in
orthodontics: what influences the duration of treatments in
National Health Service practices? Community Dent Oral
Epidemiol. 2001;29:62–72.
23. Stewart JA, Heo G, Glover KE, Williamson PC, Lam EW,
Major PW. Factors that relate to treatment duration for
patients with palatally impacted maxillary canines.
Am J Orthod Dentofacial Orthop. 2001;119:216–225.
Angle Orthodontist, Vol 81, No 2, 2011