Download AN ANALYSIS OF VARIABILITY IN CLASS I NON-

AN ANALYSIS OF VARIABILITY IN CLASS I NONEXTRACTION TREATMENT OUTCOMES IN A
RESIDENT CLINIC USING THE AMERICAN
BOARD OF ORTHODONTICS OBJECTIVE
GRADING SYSTEM
John Wesley Fleming, D.M.D.
An Abstract Presented to the Faculty of the Graduate School
of Saint Louis University in Partial Fulfillment
of the Requirements for the Degree of
Master of Science in Dentistry
2007
Abstract
Introduction:
To better understand why the Objective
Grading System (OGS) shows considerable ranges of variation
among post-treatment occlusions, this study was designed to
quantify the variability of OGS scores among similarly
treated patients and determine the factors that explain the
variability observed.
Methods:
One hundred and thirty-
eight subjects were randomly selected from the posttreatment archives of the Department of Orthodontics at
Saint Louis University.
Age, sex, mandibular plane angle,
and ANB angle were patient factors recorded from the
patients’ charts; active treatment time and supervising
orthodontist were treatment factors recorded from patient
charts.
Post-treatment OGS scores for six of the criteria
(excluding interproximal contacts and root angulations) and
anterior Bolton ratio were measured on study casts.
Results:
The mean overall OGS score was 24.9 ± 8.0.
Occlusal contacts was the most important component
contributing to the overall score and variation, followed
by alignment.
Variation in total OGS scores was explained
by pre-treatment mandibular plane angle and treatment
duration.
Overall OGS scores increased by one point for
every four degree increase in the mandibular plane angle
1
and nearly one point for every three additional months of
treatment.
Approximately 16% and 15% of the variation in
alignment and buccolingual inclination, respectively, was
due to the treating orthodontist.
Conclusions:
Moderate
amounts of occlusal variation, which are evident
immediately post-treatment in Class I non-extraction
patients, can be explained by both patient and treatment
related factors.
2
AN ANALYSIS OF VARIABILITY IN CLASS I NONEXTRACTION TREATMENT OUTCOMES IN A
RESIDENT CLINIC USING THE AMERICAN
BOARD OF ORTHODONTICS OBJECTIVE
GRADING SYSTEM
John Wesley Fleming, D.M.D.
A Thesis Presented to the Faculty of the Graduate School
of Saint Louis University in Partial Fulfillment
of the Requirements for the Degree of
Master of Science in Dentistry
2007
COMMITTEE IN CHARGE OF CANDIDACY:
Adjunct Professor Peter H. Buschang
Chairperson and Advisor
Assistant Professor Ki Beom Kim
Assistant Professor Donald R. Oliver
i
Acknowledgments
I would like to acknowledge the following individuals:
Dr. Peter Buschang for chairing my thesis committee.
Thank you for your guidance and your time in the
development and writing of this thesis.
Dr. Ki Beom Kim for serving on my thesis committee.
It has been a privilege to work with you and I appreciate
your guidance throughout the research process.
Dr. Donald Oliver for serving on my thesis committee.
You have been a great teacher and mentor during my time at
Saint Louis University.
It has been a pleasure to know you
and to work with you.
ii
TABLE OF CONTENTS
List of Tables...........................................iv
List of Figures...........................................v
CHAPTER 1: INTRODUCTION...................................1
CHAPTER 2: REVIEW OF THE LITERATURE
Measures of Treatment Outcome..................5
ABO Objective Grading System................9
Variability...................................17
Measures of Variation......................17
Components of Variability..................23
Multilevel Modeling........................25
Variation in Orthodontic Treatment Outcomes...26
Factors Contributing to Variability...........31
Patient Factors............................32
Treatment Factors..........................38
References....................................41
CHAPTER 3: JOURNAL ARTICLE
Abstract......................................46
Introduction..................................47
Materials and Methods.........................49
Results.......................................51
Discussion....................................53
Conclusions...................................58
Literature Cited..............................59
Tables........................................61
Figures.......................................66
Vita Auctoris............................................68
iii
LIST OF TABLES
Table 2.1
Previously reported individual measurements
and PAR scores of immediate post-treatment
occlusions: means and standard
deviations................................27
Table 2.2
ABO OGS scores reflecting mean points lost
and standard deviations in previously
reported samples of post-treatment
occlusions................................30
Table 3.1
ABO OGS scores reflecting mean points lost
and standard deviations in previously
reported samples of post-treatment
occlusions compared to current
research results..........................61
Table 3.2
Sample description........................62
Table 3.3
Objective Grading System scores...........62
Table 3.4
Stepwise multiple regression with
standardized coefficients showing the
contributions of objective grading system
scores (independent variables) to total
score (dependent variable)................63
Table 3.5
Multilevel estimates (Est) and standard
errors (SE) for the effects of age at start
of treatment, sex of the patient, treatment
duration, initial mandibular plane angle
(MPA), initial ANB angle, and initial
anterior Bolton ratio on the objects
grading system scores.....................64
Table 3.6
Absolute and relative between doctor (B/D)
and between patient (B/P) variation in
objective grading system scores...........65
iv
LIST OF FIGURES
Figure 2.1
American Board of Orthodontics Measuring
Gauge.....................................10
Figure 2.2
Normal distribution of a sample with a
mean of 10 and standard
deviation of 2............................21
Figure 2.3
Normal distribution for a sample with a
mean of 10 and a standard
deviation of 4............................21
Figure 3.1
Distribution of Total OGS Scores..........66
Figure 3.2
Component Scores As A Percentage
Of Total Deductions.......................67
v
CHAPTER 1:
INTRODUCTION
There exists a certain, and likely significant, amount
of variability between individuals in occlusal
relationships after orthodontic treatment.
Most studies
evaluating post-treatment occlusions focus on mean values
of specific aspects of occlusion or on global indices of
occlusion.
They often quantify the variability that exists
in the form of standard deviations.
Standard deviations
describe the individual differences observed among treated
cases.
Because differences in treatment outcomes among
individual patients are rarely addressed, the casual reader
focusing on average values could be led to assume that
differences do not exist.
It would be both important and helpful to know how
much variability exists among treated cases. Which
components of occlusion are the most variable, and what
factors determine variability?
Such information would
allow orthodontists to more critically evaluate their own
completed patients, focusing on the measures more likely to
vary in light of information regarding the variability of
post-treatment occlusions.
This approach emphasizes the
importance of evaluating the treatment effects on
individual patients, placing less importance on “average”
1
treatment effects that have little or no bearing on
individual patients.
With information about factors that
determine variability, clinicians could make treatment
decisions to reduce the variation in treatment outcomes.
There are two basic types of factors that could
explain post-treatment variability.
First, there are
patient factors such as genetic predisposition, age, sex,
patient compliance, severity of malocclusion, skeletal
characteristics, differences in facial growth, and dental
characteristics.
Most of these can only be controlled by
patient selection.
However, their contribution must be
understood because they likely add to differences in posttreatment occlusions.
In addition to the patient factors,
the diagnostic, technical, and motivational skills of the
doctor are treatment factors that could contribute to the
variability of orthodontic finishing.
Treatment factors
such as these, in addition to treatment duration, could
represent a substantial portion of the variation in posttreatment occlusions.
It is this source of variability
that orthodontists should attempt to minimize in their
efforts to provide the best treatment possible for each
patient.
The purpose of this study is to evaluate patient and
treatment related variability of treatment outcomes among
2
patients with Class I malocclusions that underwent nonextraction orthodontic treatment at Saint Louis University
(SLU).
Although Class I malocclusions might be expected to
show less post-treatment variation due to their reduced
complexity, as compared for example with Class II and Class
III malocclusions, they represent the most prevalent type
of malocclusion and comprise the majority of cases in the
typical orthodontic practice.
This makes them the group
that is most practical to study.
Patient factors and
treatment factors that contribute to outcome variability
will be partitioned and statistically evaluated.
The study
will attempt to identify the diagnostic criteria and
patient differences that account for variability in
orthodontic finishing, with an emphasis on the variation
introduced by the treating doctor.
The following review of the literature will first
examine various methods that have been used for quantifying
treatment outcomes.
This will be necessary in order to
validate the methods chosen to evaluate post-treatment
occlusions in the current study.
Next, variability will be
defined and put into the context of treatment outcomes.
This is important because it will allow the reader to
better understand the terms and methods used in the current
study.
It will then be shown that considerable variability
3
exists among post-treatment occlusions. This section is
important because it will provide the basis for performing
the current study.
Lastly, the various patient and
treatment factors that might contribute to the overall
variability in orthodontic treatment outcomes will be
reviewed.
They will provide orthodontists with insightful
information regarding individual patient care.
4
CHAPTER 2:
REVIEW OF THE LITERATURE
Measures of Treatment Outcome
Measurements taken from dental casts have been used by
orthodontists to describe malocclusions prior to treatment
and to evaluate treatment results.
They have also been
used to evaluate treatment progress and monitor posttreatment stability or the lack thereof.
Measures of
malocclusion are used to estimate severity or to establish
a need for orthodontic treatment.
Measures of treatment
outcome, on the other hand, examine dental casts after
completion of treatment.
Over the years, many methods have been employed to
evaluate malocclusion in either an attempt to determine
treatment need or estimate treatment difficulty and/or
complexity.
An example of a measure of malocclusion is the
Treatment Priority Index (TPI), which was developed by
Grainger in 1967.1
The TPI is a composite measure that
includes overjet, overbite, the transverse dimension of the
posterior occlusion, and tooth displacements or rotations.
The TPI was used as a screening tool for public health
programs to determine orthodontic treatment need among
children.
Other measurements such as upper molar arch
width, mandibular plane angle, lower arch length, and the
5
ANB angle were also deemed to be useful in determining
treatment need.2
The Peer Assessment Rating (PAR) is an index that has
been used before, during, and after treatment.3 It uses 11
component characteristics to classify a malocclusion,
including
upper right segment alignment, upper anterior
segment alignment, upper left segment alignment, lower
right segment alignment, lower anterior segment alignment,
lower left segment alignment, right buccal occlusion,
overjet, overbite, midline discrepancies, and left buccal
occlusion.4
These characteristics are measured and summed
to derive a score representative of the malocclusion.
While considered by most to be a reliable and valid method,
it lacked precision and utilizes a subjectively based
weighted scale to calculate its score.
More recently, the Discrepancy Index (DI) has been
adopted by the American Board of Orthodontics (ABO) for use
in its certification process.5
It comprises measurements of
overjet, overbite, anterior openbite, lateral openbite,
crowding, occlusion, posterior crossbite, and cephalometric
values for ANB, mandibular plane, and lower incisor
angulation.
Similar to the PAR, a composite score is
assigned to each malocclusion, indicating case complexity,
rather than outcome.
6
Measures of treatment outcome are based on an
evaluation of post-treatment status or on a comparison of
pre- and post-treatment status.
The most basic way to
describe post-treatment occlusal relationships is to use
individual measurements.
While simple measures of overjet,
overbite, or incisor irregularity do not provide a
comprehensive description of treatment outcomes, they can
be reliably measured and compared to pre-treatment
measurements to evaluate treatment outcome and its
variation.
Indices have also been used to compare pre-treatment
and post-treatment records.
In 1974, Eismann measured 15
pre-treatment traits, both intra-orally and on dental casts
to derive a composite score that he compared with the same
post-treatment measurements.6
In 1975, Gottlieb evaluated
10 dental relationships before and after treatment and
scored each based on how completely it was corrected.7
Later, Berg and Fredlund8 used a weighted index to quantify
malocclusion.
Each component of malocclusion was evaluated
after treatment and the “degree of improvement” was noted.
The Occlusal Index developed by Summers was a very complex
measure designed specifically to evaluate pre-treatment
occlusions rather than interpret treatment results.9
While
those indices were helpful in examining the effects of
7
orthodontic treatment, they were not very precise and were
never proven to be either valid or reliable.
The Irregularity Index is a measure of outcome in
orthodontics developed to monitor stability.
It is based
on the contact points between the lower anterior teeth from
the mesial surface of the left canine to the mesial surface
of the right canine.10
The sum of the five linear
distances, in millimeters, between anatomic contact points
in perfectly aligned incisors is zero and a finished
orthodontic case should approach this score.
In post-
treatment studies, an increase in this value during the
retention phase of orthodontic treatment or after retention
indicates instability and/or relapse.
During Phase III of the ABO certification process,
Board members evaluate candidates’ clinical competency by
scrutinizing their treated patients’ records.
In the past,
this was done with a degree of subjectivity based on
accepted standards of occlusion, such as Andrews’ six keys
to normal occlusion11 and the ideals of the Board examiners.
Existing indices, while useful for various purposes, were
deemed not to be precise enough to critically and
objectively evaluate the occlusal characteristics.
In this
endeavor, the ABO developed the Objective Grading System.
8
ABO Objective Grading System
From 1995 to 1999, the American Board of Orthodontics
developed and improved upon a system for evaluating posttreatment records called the objective grading system
(OGS).
The original model for this grading system, which
was developed in 1995, evaluated 15 criteria on 100 sets of
dental casts and panoramic radiographs.
The first field
test of the system found that seven of the 15 criteria
accounted for 85% of the occlusal inadequacies.
These
criteria included alignment, marginal ridges, buccolingual
inclination, overjet, occlusal relationship, occlusal
contacts, and root angulation.
In a second field test conducted in 1996, 300 sets of
dental casts and panoramic radiographs were evaluated by a
committee of four Board Directors. The majority of point
deductions were again described by the same seven criteria.
Because this field test showed that inter-examiner
reliability and consistency were inadequate, the committee
recommended the development of a measuring tool to make the
process more objective and reliable.
In 1997, a third
field test led to modifications of the measuring tool and
the addition of interproximal contacts as an eighth
criterion.
In 1998, a fourth and final field test was
performed successfully and the OGS was officially adopted
9
in February of 1999 for Phase III examination of ABO
certification.
The current OGS utilizes a measuring tool (Figure 2.1)
to more accurately and reliably measure seven tooth
relationships on dental casts: alignment, buccolingual
inclination of posterior teeth, occlusal contacts of
posterior teeth, overjet, marginal ridge heights, occlusal
relationships of posterior teeth, and interproximal
contacts.12
Figure 2.1 American Board of Orthodontics Measuring Gauge
A This portion of the gauge is 1 mm in width and is used to measure
discrepancies in alignment, overjet, occlusal contact, interproximal
contact, and occlusal relationships.
B This portion of the gauge has steps measuring 1 mm in height and is
used to determine discrepancies in mandibular posterior buccolingual
inclination.
C This portion of the gauge has steps measuring 1 mm in height and is
used to determine discrepancies in marginal ridges.
D This portion of the gauge has steps measuring 1 mm in height and is
used to determine discrepancies in maxillary posterior buccolingual
inclination.
10
The OGS also uses the panoramic radiograph to evaluate
root parallelism, even though its accuracy for assessment
of root angulation remains questionable.13,14
Root
angulation is considered ideal if adjacent roots are
parallel to each other and perpendicular to the occlusal
plane.12
The rationale for the use of each of eight criteria is
based on the orthodontic goals of esthetics, function,
health, and stability.
Alignment is a primary goal of
orthodontic treatment and has a great impact on anterior
esthetics. To measure alignment according to the OGS, the
functional surfaces of teeth are marked with a graphite
pencil (i.e. maxillary anterior marginal ridges continuing
posteriorly along the central fossa line to form a
continuous arch and mandibular anterior labial incisal
edges continuing posteriorly along the buccal cusps to form
a continuous arch).
No points are deducted if all teeth
are properly aligned or within 0.5 mm of proper alignment.
A single point is deducted for deviations of 0.5 mm to 1.0
mm as measured with the measuring gauge (Figure 2.1, A).
Two points are deducted if the deviation is greater than
1.0 mm.
No more than two points can be deducted per tooth.
A maximum of 56 points can be deducted for this criterion.
This measure excludes the third molars.
11
Marginal ridges indicate proper vertical positioning
of posterior teeth.
All marginal ridges should be at the
same level in patients with no restorations, periodontal
bone loss, and little or no attrition.
This will produce
flat bone levels between teeth, which is the ideal
architecture for sustainable periodontal health.
Marginal
ridges are measured at each interproximal contact point
using the OGS measuring tool (Figure 2.1, C).
No points
are deducted if adjacent marginal ridges are within 0.5 mm.
Discrepancies between 0.5 mm and 1.0 mm warrant a deduction
of one point.
Two points are deducted for marginal ridges
with discrepancies greater than 1.0 mm.
No more than two
points can be subtracted for any one interproximal contact
area.
The contact area between mandibular first and second
premolars is not included.
Twenty points is the maximum
that can be deducted for this criterion.
The correct buccolingual angulation of posterior teeth
allows maximum intercuspation while avoiding balancing
interferences.
There should not be a significant
difference between the buccal and lingual cusps of
posterior teeth.
In the maxillary arch, buccolingual
inclination is measured with the OGS measuring tool (Figure
2.1, D).
The maxillary buccal cusps should be within 1.0
mm vertically of the palatal cusps.
12
In the mandibular
arch, this is also measured with the OGS measuring tool
(Figure 2.1, B).
The mandibular lingual cusps should be
within 1.0 mm vertically of the buccal cusps.
No points
will be deducted if this relationship exists.
Deviations
between 1.0 mm and 2.0 mm will lead to a deduction of one
point.
Two points are subtracted for deviations greater
than 2.0 mm.
tooth.
No more than two points can be deducted per
Mandibular first premolars are not included.
28
points is the maximum that can be deducted for this
criterion.
Overjet deals with the buccolingual relationship of
the maxillary and mandibular arches.
The maxillary arch
should just contain the mandibular arch.
This is manifest
as anteroposterior relationships of the anterior teeth and
as transverse relationships in the posterior segment when.
The horizontal overlap of anterior teeth in proper
occlusion allows for proper anterior guidance upon lateral
and protrusive mandibular movements.
Overjet is also
measured using the OGS measuring tool (Figure 2.1, A).
Articulated models are utilized to examine the labiolingual
relationship of opposing teeth.
Mandibular incisors and
canines should contact the lingual surface of maxillary
incisors and canines while the buccal cusps of mandibular
posterior teeth should occlude in the buccolingual center
13
of maxillary posterior teeth.
If these relationships
exist, no points are deducted.
One point is deducted for
deviations of 1.0 mm or less.
Two points are deducted for
deviations greater than 1.0 mm.
No more than two points
can be deducted per tooth. 28 points is the maximum that
can be deducted for this criterion.
Occlusal contacts provide a measure of intercuspation,
another primary goal in orthodontic treatment.
The buccal
cusps of mandibular posterior teeth and the lingual cusps
of maxillary posterior teeth should be in contact with the
occlusal surfaces of the opposing teeth.
This is again
measured with the OGS measuring tool (Figure 2.1, A).
If
each contact is present, then no points are deducted.
If a
cusp is out of contact by 1 mm or less, one point is
deducted.
Two points are deducted if the cusp is more than
1 mm from the opposing contact.
can be deducted per tooth.
No more than two points
Maxillary first premolars and
diminutive distolingual cusps on maxillary second molars
are not included.
Twenty-eight points is the maximum
possible deduction for this criterion.
Occlusal relationships are determined by the
anteroposterior positions of maxillary and mandibular
posterior teeth.
For maximum intercuspation and ideal
function, the mesiobuccal cusp of the maxillary first molar
14
should align with the buccal groove of the mandibular first
molar and all other buccal cusps of maxillary posterior
teeth, from canine back, should align with their
corresponding interproximal embrasures.
This is measured
with the OGS measuring tool (Figure 2.1, A).
If each
maxillary buccal cusp is aligned properly within 1 mm, no
points are subtracted.
If there is a deviation between
1 mm and 2 mm, one point is deducted.
Two points are
deducted for deviations greater than 2 mm.
two points can be deducted per tooth.
No more than
20 points is the
maximum that can be deducted for this criterion.
Interproximal contacts are assessed by viewing the
casts from the occlusal perspective.
The mesial and distal
surfaces of teeth should be in contact for ideal function
in the posterior region and for esthetics in the anterior
region.
This is again measured with the OGS measuring tool
(Figure 2.1, A).
If no interproximal space exists, no
points are subtracted.
point is deducted.
If space of up to 1 mm exists, one
Two points are deducted for
interproximal spaces greater than 1 mm.
No more than two
points are subtracted for any contact area.
Fifty-two
points is the maximum that can be deducted for this
criterion. (Note: this criterion was not used in the
current study)
15
Root angulation is the final component of the OGS.
If
roots are parallel to each other and properly angulated,
then sufficient bone will be present between teeth to
maximize periodontal support.
When viewed using a
panoramic radiograph, roots should be both parallel to each
other and perpendicular to the occlusal plane.
are deducted if this relationship exists.
No points
If a root
deviates mesially or distally at the apex more than 1 mm
but less than 2 mm, one point is subtracted.
Two points
are subtracted if the deviation is greater than 2 mm.
Fifty-six points is the maximum that can be deducted for
this criterion.
(Note: this criterion was also not used in
the current study)
The OGS score is calculated by summing all of the
component deductions.
There are 288 possible point
deductions based on all 8 criteria.
The current study
considers 6 of the 8 criteria and has 180 possible point
deductions.
Generally, cases that have more than 30 points
deducted will fail the Phase III examination; cases with
fewer than 20 points deducted will pass.
because of intra-examiner reliability.
This range exists
Other factors such
as appropriateness of treatment plan, quality of records,
facial profile, and proper positioning of the maxilla,
mandible, and their respective dentitions.
16
Variability, Measures of Variation, Components of
Variability, and Multilevel Modeling
While orthodontists and craniofacial biologists
typically focus on measures of central tendency, such as
the mean, it is equally as important to understand
dispersion or variability.
If, for example, different
malocclusion groups are extremely heterogeneous, then small
differences in their mean values would be of little
practical significance.
The following section will first
describe several different measures of variability, how
they are computed, how they are interpreted, and when they
are useful.
variance.
It will then describe the components of
Both explained and unexplained sources of
variation will be defined and put into context.
Lastly,
multilevel modeling, a procedure that partitions variation
at multiple levels will be described because it is used in
the current study.
Measures of Variation
There are several measures used to describe
variability.
The simplest measure of variability is the
range, which is the difference between the highest and
lowest values within a sample. However, it is based on only
the two most extreme values. Because the extremes of any
17
distribution of values are by definition rare or unlikely,
it is simply a matter of chance whether or not any given
range correctly describes the sample’s variability.
Another measure of variability is the quartile
deviation, which is defined as half the difference between
the first and third quartiles.
In other words, it
represents half of the range covered by the middle 50% of
the cases in a sample or the middle 25% of the sample.
Since the first and third quartiles vary less across
samples than the extreme values, the quartile deviation is
a more stable and reliable measure of variability than the
range.
A more precise way of describing variability relates
the individuals’ values to the mean.
This is accomplished
by the mean deviation, which is the defined as the
arithmetic mean of the absolute differences of the values
from the mean. It provides a good descriptive measure of
how large of a deviation is expected, on average.
Variance is another measure of variability closely
related to the mean deviation.
Variance is calculated as:
Variance = Σdeviations2/n
18
Deviations refer to the differences between the individual
values and the overall mean and n refers to the sample
size. Because the deviations are squared, individuals with
extreme values have a much greater influence on estimates
of variation than individuals with small deviations from
the mean. The fact that the deviations are squared also
makes the variance statistic hard to interpret
meaningfully. To change the variance statistic back to
meaningful units, standard deviations are typically used to
describe variability.
The most useful and frequently utilized measure of
variability is the standard deviation (SD), which is
defined as the square root of the arithmetic mean of the
squared deviations from the mean. It is also the square
root of the variance.
Standard deviations are calculated
as:
S.D. = √ deviations2/n
First, the deviation of each subject’s value from the mean
is calculated, each of the deviations is squares, a sum of
the squared deviations is computed, the sum is divided by
the number of cases, and then the square root of that
average is derived.
19
Standard deviations have several important properties
that should be emphasized.
First, they describe the
dispersion of values around the mean; the larger the
standard deviation, the larger the spread of values. More
importantly, the standard deviation makes it possible to
apply probabilities associated with the normal curve, a
special type of symmetrical curve (a bell shape) that makes
it possible to estimate the likelihood of any value within
a normally distributed sample.
Regardless of the mean and
standard deviation, the normal curve will always have the
same proportion of values that fall between the mean and
any given number of standard deviation units. Knowing the
standard deviation makes it possible to predict the
probability of any measure relative to the mean. Normal
tables, which can be found at the back of any statistics
book, provide the exact probability associated with any
standard deviation score. For example, approximately 68.3%
of the individuals in a sample are distributed within
of the mean and 95.5% are distributed within
± 1SD
± 2SD of the
mean. As such, standard deviations provide probabilities
for all possible values around the mean.
On that basis,
the distribution of values for a sample with a mean of 10
with a standard deviation of 2 (Figure 2.2) will look
20
entirely different than the distribution for a sample with
a mean of 10 with a standard deviation of 4 (Figure 2.3).
Mean 10 SD 2
-10
-5
0
5
10
15
20
25
30
Figure 2.2 Normal distribution of a sample with a mean of
10 and standard deviation of 2
Mean 10 SD 4
-10
0
10
20
30
Figure 2.3 Normal distribution for a sample with a mean of
10 and a standard deviation of 4
21
Statistics provide parameter estimates (e.g. means) of
a population’s true value. A standard error, another
measure closely related to the standard deviation, provides
the probabilities of the distribution of values associated
with parameter estimates.
It is based on a theorem that if
repeated random samples of a given size are drawn from a
normal population, with a given mean and variance, the
sampling distribution of sample means will be normal and
their variance will be described by the standard error.
In
other words, assuming an unlimited number of samples, each
with a sample size of n, drawn from one population, the
standard error of the mean of any one of those samples
provides the theoretical distribution of all of the other
mean values about that mean. For example, given a mean of 5
with a standard error of 2, the mean might be expected to
actually range between 3 and 7 approximately 68% of the
time and between 1 and 9 approximately 95% of the time.
Standard errors can therefore be used to determine when an
estimate derived for a sample differs significantly from
zero.
The standard error is closely related to the
standard deviation because it provides the standard
deviation of the normal distribution of sample means.
is calculated as:
22
It
Standard Error = SD/√ n
Coefficients of variation (CV) are used to compare the
variability of measures that have different means. Since
measures with larger means typically have larger standard
deviations, it might be misleading to directly compare the
absolute values of the standard deviations.
The CV
measures the size of the standard deviation relative to the
mean. CVs are especially useful for measures that differ in
the types of units recorded (e.g. mm, kg, deg, etc.). A CV
of variation is calculated as:
CV = SD/Mean
Components of Variability
It is important to understand that there are
components of variance that combine to make up total
variance.
At the simplest level, variance can be divided
into explained and unexplained variation.
The explained
component pertains to factors that have been shown to
account for variance.
Homogeneous groups that differ
considerable from each other might be expected to explain a
high proportion of the total variance. For example, a
23
sample composed of males and females might be expected to
show substantially higher variance than separate samples of
males and females.
Unexplained sources of variation
pertain to variation that is left unaccounted for by the
explained sources. Unexplained variation is sometimes
referred to a residual or error variance.
Variation can be explained by all types of measures,
including variables measured on dichotomous, nominal,
ordinal and interval scales.
Sex and malocclusion are
examples of variables previously shown to explain variance
measured on dichotomous and nominal scales, respectively.
T-test and analyses of variance are typically used to
determine the significance of dichotomous and nominal
sources of variation.
Variables measured on interval
scales can also be used to explain variation, using
correlation, regression or analyses of covariance.
Treatment duration, for example, might be expected to
explain variance, or individual differences, in treatment
outcomes.
Unexplained or residual variation is what remains
after all of the explained sources of variability have been
accounted for. It pertains to random variation without any
pattern (i.e. it has not or cannot be related to other
variables). Importantly, unexplained random variation can
24
also be partitioned or organized.
can occur at different levels.
Unexplained variation
There are many different
possible levels of random variation.
For example, every
sample has random variability between subjects. Betweensubject variability describes differences among individuals
in a sample.
The more individuals differ genetically,
epigenetically, or in their environmental circumstances,
the greater the between-subject variability might be
expected. It is also possible to have within-subject
variability. Within-subject variability refers to changes
or differences that occur in the same subjects. For
example, as individuals grow and mature, they become larger
and change shape. Longitudinal treatment data might
therefore be expected to show both within- and betweensubject variability. Within- and between-subject variations
combine to produce the overall variability typically seen
among subjects during growth.
Multilevel Modeling
Multilevel modeling procedures were developed in order
to estimate both explained and unexplained sources of
variation, with an emphasis on the unexplained sources of
variation.
The explained sources of variation represent
the fixed portion of variability.
25
This is represented in
the current study by ANB, MPA, treatment duration and
anterior Bolton.
The unexplained, or random, variation is
represented by patient and doctor differences.
The term
multilevel refers to multiple levels of variation, with one
level nested in another, which is in turn nested in
another, and so on.
In the current study, this model
allows the residual variation, that which has not been
accounted for by the fixed portion, to be partitioned at
the level of the doctor and the patient.
Variation in Orthodontic Treatment Outcomes
Substantial variation exists in orthodontic treatment
outcomes.
This indicates that not all cases are completed
to the same specifications.
Post-treatment occlusal
characteristics, when evaluated using individual
measurements or global indices, show considerable and
clinically significant variation (Table 2.1).15-20
26
27
Vaden20 showed that maxillary incisor irregularity
immediately post treatment was 1.6 mm with a standard
deviation of 1.4 mm.
This indicates that 16 percent of
patients, or about six of his thirty-six subjects, finished
with 3 mm or more of maxillary incisor crowding.
Furthermore, one patient, or 2.5 percent of his sample,
finished with 4.4 mm or more of maxillary incisor crowding.
This demonstrates the value of examining standard
deviations and placing less emphasis on mean outcomes.
Similar observations can be made based on Rossouw et
al.’s19 description of post-treatment overbite and overjet.
According to his standard deviations, 14 patients (16%)
displayed overbites of more than 3.8 mm and at least two
patients (2.5%) had overbites of nearly 5 mm immediately
following orthodontic treatment.
By the same token, 14
patients had post-treatment overjets of more than 3.8 mm;
two patients had overjets that were more than 4.7 mm
immediately following orthodontic treatment.
Using the PAR index as a global measure of occlusion
following orthodontic treatment, Ormiston et al.18 showed
considerable variation in their sample.
They observed a
mean PAR score of 5.5 (0 is ideal) with a standard
deviation of 4.1.
This indicates that about seven of the
subjects had post-treatment PAR scores above 9.6 and one
28
individual finished treatment with a PAR score approaching
14, a score indicative of malocclusion.
In a comprehensive review of previously reported ABO
OGS scores, significant variation can also be readily
observed (Table 2.2).18,21-28
29
30
The standard deviations for total OGS scores range
from 8.8 points to 18.3 points, representing considerable
variation.
The Yang-Powers et al.28 university sample
comprised of 92 patients showed the largest overall mean
OGS score of 45.5 and the largest standard deviation at
18.3.
This indicates that 29 patients (16%) had an OGS
score of 64 or greater and 29 patients had an OGS score of
27 or lower, a score potentially in the passing range
according to the ABO.
This broad dispersion represents
clinically significant variation.
Correspondingly, the
Nett and Huang27 sample (n=100) showed the smallest overall
mean OGS score (21.5) and the smallest standard deviation
at 8.8.
Even in this less variable sample, 16 subjects had
OGS scores over 30, a failing score when evaluated by the
ABO.
Factors Contributing to Variability
The goal of orthodontic diagnosis and treatment
planning should be to consistently produce “ideal” (i.e.
not perfect) results for each patient within a reasonable
time period.
If “ideal” occlusion is not achieved in every
patient, it would be helpful to know what factors are
31
responsible for the variability of treatment outcomes.
This would make it possible to reasonably anticipate and
address these issues in order to effectively minimize any
requisite deviation from “ideal”.
Sources of variability can be categorized as patient
factors and treatment factors.
There are no studies that
have directly associated patient or treatment factors with
post-treatment variability.
However, there is indirect
evidence that various factors contribute to post-treatment
variation.
For example, several patient factors might be
expected to increase variability including anteroposterior
skeletal relationships, vertical skeletal relationships,
the anterior Bolton ratio.29
The treatment factors that
might be expected to contribute to variability are duration
and doctor.
Patient Factors
Anteroposterior Skeletal Relationship
Orthodontic treatment of anteroposterior skeletal
discrepancies requires dental compensations to achieve
correction at the level of the occlusion.
The
compensations change the geometry of the anterior tooth
angulations.
The differences in the buccolingual
inclination of anterior teeth consequently affect the
32
entire posterior occlusion.
For instance, dental
correction of a Class II skeletal discrepancy commonly
results in proclined lower incisors and upright upper
incisors.
While excessive overjets are resolved and the
skeletal malocclusion camouflaged, posterior tooth
angulations are consequently affected and the occlusal
scheme can be expected to be more variable.
Specifically,
changes in posterior tooth angulations might be expected to
affect marginal ridge heights, occlusal contacts and cuspembrasure relationships which are all accounted for in the
OGS.
Conversely, treatment of Class III skeletal pattern
produces upright lower incisors and proclined upper
incisors.
While anteroposterior skeletal relationships have not
been directly related to differences in post-treatment
occlusion, Polk and Buchanon30 related pretreatment
anteroposterior skeletal discrepancies to treatment
duration and outcome.
They showed that the pre-treatment
ANB and Wits measurements combined provided better
predictions of treatment time than either one alone.
Treatment time generally increased with increases in
anteroposterior discrepancies.
Fink and Smith31 also showed
that pretreatment measures of Wits and ANB were related to
treatment success or failure.
The criterion for success,
33
however, was the correction of the skeletal relationship.
Outcome was not evaluated at the occlusal level.
Vertical Skeletal Relationship
Clearly, dental compensations occur as a result of a
skeletal imbalance in the vertical dimension just as they
do with discrepancies in the sagittal plane.
As the
mandibular plane angle increases, mandibular incisors must
be positioned more upright to achieve proper anterior
occlusal relationships (i.e. overbite); the opposite is
true for low mandibular plane angles.
Andrews11 described
the effects that differences in the buccolingual
inclination of anterior teeth would have on posterior
occlusion.
It was noted that posterior teeth should be
mesially inclined to achieve proper occlusion.
This is not
geometrically possible if incisors are too upright.
Marginal ridge heights, occlusal contacts, and occlusal
relationships, which are all measured in the OGS, would be
affected.
In addition, orthodontic management of patients with
increased or decreased vertical dimension often requires
biomechanics that are different from those used to treat
vertically “normal” patients.
As such, it is reasonable to
34
expect vertical relationships will have an effect on posttreatment occlusions.
Similarly, there is no literature that directly
relates vertical skeletal relationships to differences in
post-treatment occlusions.
However, vertical relationships
might be expected to increase treatment difficulty or
duration and indirectly cause differences in post-treatment
occlusions.
In their extensive study on treatment
duration, Fink and Smith31 showed that subjects with higher
mandibular plane angles (MPA) had shorter average treatment
times.
However, the correlation, while statistically
significant, was small.
Furthermore, the relationship
between MPA and treatment duration was not shown to be
linear.
Anterior tooth-size ratio
Differences in the ratio of the size of anterior
mandibular teeth to the size of the anterior maxillary
teeth might also be expected to produce variation in posttreatment occlusions.
For example, if maxillary lateral
incisors are too small mesiodistally and no spacing is
present, the maxillary canine will necessarily be forward
of its ideal Class I position creating a mild Class II
posterior occlusion.
By the same token, if the mandibular
35
incisors are relatively wide and no interproximal reduction
is done, the posterior occlusion will be affected in the
same way.
Conversely, if maxillary central incisors are
relatively too wide, each properly aligned posterior
maxillary tooth would theoretically be in a mild Class III
relationship relative to a normally sized and properly
aligned mandibular arch.
These dental compensations can be addressed during
treatment by altering tooth size, overjet, overbite, and
anterior buccolingual and mesiodistal tooth angulations.
Each of these would be expected to produce variation in
post-treatment occlusions.
The effect of anterior tooth-size ratios on occlusion
has been studied by several investigators.
Neff32 computed
the “anterior coefficient” by dividing the sum of the six
maxillary anterior teeth by the sum of the six mandibular
anterior teeth and used this to mathematically determine
each normal occlusion’s individual anterior overbite.
Bolton29 later used the reciprocal of this ratio in his
analysis of malocclusion and popularized its use.
The mean
ratio of mandibular anterior teeth to maxillary anterior
teeth, according to Bolton, is 77.2 percent with a standard
deviation of 1.65 percent.
He suggested the use of his
ratio as an adjunct to or replacement of a diagnostic
36
setup. For example, the ratio would be applicable when the
removal of a lower incisor, interproximal reduction, or the
build-up of a maxillary lateral incisor is considered to
achieve normal anterior and consequently posterior
occlusion.
Redahan and Lagerstrom33 examined post-treatment
anterior dental relationships and their correlation to pretreatment anterior inter-arch tooth size discrepancies.
While it cannot be determined if the Bolton discrepancies
were addressed during treatment in their study, they found
no significant association between the two.
If an anterior
“Bolton discrepancy” is addressed during treatment by the
addition or removal of tooth structure, it seems reasonable
that differences in post-treatment occlusions might not be
evident.
However, if anterior tooth-size discrepancies are
not addressed during treatment, it seems also reasonable
that they could lead to variation in post-treatment
occlusions.
This is the reason that the anterior tooth-
size ratio was calculated post-treatment in the current
study.
Age and Sex
The influence of age and sex on orthodontic treatment
is not clear and much of the literature devoted to these
37
relationships is contradictory.34
Biologically, neither the
amount or rate of tooth movement is dependent on age.35
Furthermore, Robb et al.36 showed that treatment duration
and effectiveness, as assessed by the PAR index using mean
values, were no different for adolescents and adults. This
has been corroborated by other investigators.31,34,37
There
is no reason to expect that sex would increase posttreatment variability.
Treatment Factors
Treatment Duration
Treatment outcome as a function of treatment duration
has not been directly evaluated.
Treatment duration is
often studied as a function of treatment severity or
difficulty.
Pae et al.38 defined the severity of a
malocclusion as the degree of its deviation from ideal and
difficulty as the probability of attaining an ideal
occlusion.
DeGuzman et al.39 positively correlated severity
of malocclusion (rated by the PAR index) and difficulty as
perceived by a panel of 11 orthodontists.
Malocclusions determined to have greater severity and
treatment need have been shown to require more appointments
and longer treatment times.4,40,41
And while there is no
accepted objective measure of difficulty, Casinelli et al.42
38
compared post-treatment cases deemed by 10 orthodontists to
be difficult to those determined to be easy.
The
“difficult” cases had PAR scores indicating more residual
malocclusion and additional need for treatment than the
“easy” cases. They also required more appointments and had
longer treatment times.
It is then reasonable to expect
that severe malocclusions and occlusions deemed difficult
to treat may take longer.
Therefore, differences in
treatment time, as a reflection of severity or difficulty,
could then contribute to the variability of post-treatment
occlusions.
Doctor Variation
Finally, the treating orthodontist might be expected
to influence the variability of post-treatment occlusions
in several ways.
The first step of orthodontic treatment
involves diagnosis and treatment planning.
Orthodontists
will vary in this regard based on their training, clinical
experience, and specific treatment goals.
For instance,
one orthodontist may treatment plan a crowded case for
extraction treatment whereas another may plan for
expansion, another for distalization, and still another for
interproximal reduction.
It is clear that the occlusal
schemes following each of these treatments would not be the
39
same.
As a related example, Ross43 showed long term
differences in facial growth of cleft patients was based
more on the surgery (and the surgeon) than the nature or
extent of the facial deformity.
Treatment techniques, including the choice of
appliance prescription, the positioning of brackets and
bands, bracket slot size, and archwire materials and sizes,
will vary among practitioners.
While similar results are
theoretically possible with differences in any of the above
aspects of treatment, variation in the technical skills
among doctors is to be expected.
Furthermore, the
treatment mechanics chosen by the orthodontist,
specifically those that demand patient cooperation such as
headgear or intermaxillary elastics, can affect both
treatment outcome and treatment duration.34,37,44
40
References
1. Grainger RM. Orthodontic treatment priority index. Vital
Health Stat 2 1967:1-49.
2. Kowalski CJ, Prahl-Andersen B. Selection of dentofacial
measurements for an orthodontic treatment priority index.
Angle Orthod 1976;46:94-97.
3. Richmond S, Shaw WC, O'Brien KD, Buchanan IB, Jones R,
Stephens CD et al. The development of the PAR Index (Peer
Assessment Rating): reliability and validity. Eur J Orthod
1992;14:125-139.
4. Richmond S, Shaw WC, Roberts CT, Andrews M. The PAR
Index (Peer Assessment Rating): methods to determine
outcome of orthodontic treatment in terms of improvement
and standards. Eur J Orthod 1992;14:180-187.
5. Cangialosi TJ, Riolo ML, Owens SE, Jr., Dykhouse VJ,
Moffitt AH, Grubb JE et al. The ABO discrepancy index: a
measure of case complexity. Am J Orthod Dentofacial Orthop
2004;125:270-278.
6. Eismann D. A method of evaluating the efficiency of
orthodontic treatment. Trans Eur Orthod Soc 1974:223-232.
7. Gottlieb EL. Grading your orthodontic treatment results.
J Clin Orthod 1975;9:155-161.
8. Berg R, Fredlund A. Evaluation of orthodontic treatment
results. Eur J Orthod 1981;3:181-185.
9. Summers CJ. The occlusal index: a system for identifying
and scoring occlusal disorders. Am J Orthod 1971;59:552567.
10. Little RM. The irregularity index: a quantitative score
of mandibular anterior alignment. Am J Orthod 1975;68:554563.
11. Andrews LF. The six keys to normal occlusion. Am J
Orthod 1972;62:296-309.
12. Casko JS, Vaden JL, Kokich VG, Damone J, James RD,
Cangialosi TJ et al. Objective grading system for dental
41
casts and panoramic radiographs. American Board of
Orthodontics. Am J Orthod Dentofacial Orthop 1998;114:589599.
13. Lucchesi MV, Wood RE, Nortje CJ. Suitability of the
panoramic radiograph for assessment of mesiodistal
angulation of teeth in the buccal segments of the mandible.
Am J Orthod Dentofacial Orthop 1988;94:303-310.
14. McKee IW, Williamson PC, Lam EW, Heo G, Glover KE,
Major PW. The accuracy of 4 panoramic units in the
projection of mesiodistal tooth angulations. Am J Orthod
Dentofacial Orthop 2002;121:166-175.
15. Dyken RA, Sadowsky PL, Hurst D. Orthodontic outcomes
assessment using the peer assessment rating index. Angle
Orthod 2001;71:164-169.
16. Glenn G, Sinclair PM, Alexander RG. Nonextraction
orthodontic therapy: posttreatment dental and skeletal
stability. Am J Orthod Dentofacial Orthop 1987;92:321-328.
17. Onyeaso CO, Begole EA. Orthodontic treatment-improvement and standards using the peer assessment rating
index. Angle Orthod 2006;76:260-264.
18. Ormiston JP, Huang GJ, Little RM, Decker JD, Seuk GD.
Retrospective analysis of long-term stable and unstable
orthodontic treatment outcomes. Am J Orthod Dentofacial
Orthop 2005;128:568-574.
19. Rossouw PE, Preston CB, Lombard CJ, Truter JW. A
longitudinal evaluation of the anterior border of the
dentition. Am J Orthod Dentofacial Orthop 1993;104:146-152.
20. Vaden JL, Harris EF, Gardner RL. Relapse revisited. Am
J Orthod Dentofacial Orthop 1997;111:543-553.
21. Abei Y, Nelson S, Amberman BD, Hans MG. Comparing
orthodontic treatment outcome between orthodontists and
general dentists with the ABO index. Am J Orthod
Dentofacial Orthop 2004;126:544-548.
22. Cook DR, Harris EF, Vaden JL. Comparison of university
and private-practice orthodontic treatment outcomes with
the American Board of Orthodontics objective grading
system. Am J Orthod Dentofacial Orthop 2005;127:707-712.
42
23. Cook MK. Evaluation of Board-certified orthodonist's
sequential finished cases with the ABO objective grading
system Orthodontics. Saint Louis: Saint Louis University;
2003: p. 56.
24. Costalos PA, Sarraf K, Cangialosi TJ, Efstratiadis S.
Evaluation of the accuracy of digital model analysis for
the American Board of Orthodontics objective grading system
for dental casts. Am J Orthod Dentofacial Orthop
2005;128:624-629.
25. Deguchi T, Honjo T, Fukunaga T, Miyawaki S, Roberts WE,
Takano-Yamamoto T. Clinical assessment of orthodontic
outcomes with the peer assessment rating, discrepancy
index, objective grading system, and comprehensive clinical
assessment. Am J Orthod Dentofacial Orthop 2005;127:434443.
26. Djeu G, Shelton C, Maganzini A. Outcome assessment of
Invisalign and traditional orthodontic treatment compared
with the American Board of Orthodontics objective grading
system. Am J Orthod Dentofacial Orthop 2005;128:292-298.
27. Nett BC, Huang GJ. Long-term posttreatment changes
measured by the American Board of Orthodontics objective
grading system. Am J Orthod Dentofacial Orthop
2005;127:444-450.
28. Yang-Powers LC, Sadowsky C, Rosenstein S, BeGole EA.
Treatment outcome in a graduate orthodontic clinic using
the American Board of Orthodontics grading system. Am J
Orthod Dentofacial Orthop 2002;122:451-455.
29. Bolton W. Disharmonies in tooth size and its relation
to the analysis and treatment of malocclusions. Angle
Orthod 1958:113-120.
30. Polk CE, Buchanan D. A new index for evaluating
horizontal skeletal discrepancies and predicting treatment
outcomes. Am J Orthod Dentofacial Orthop 2003;124:663-669.
31. Fink DF, Smith RJ. The duration of orthodontic
treatment. Am J Orthod Dentofacial Orthop 1992;102:45-51.
32. Neff CW. Tailored occlusion with anterior coefficient.
Am J Orthod 1949;35:309-313.
43
33. Redahan S, Lagerstrom L. Orthodontic treatment outcome:
the relationship between anterior dental relations and
anterior inter-arch tooth size discrepancy. J Orthod
2003;30:237-244.
34. Skidmore KJ, Brook KJ, Thomson WM, Harding WJ. Factors
influencing treatment time in orthodontic patients. Am J
Orthod Dentofacial Orthop 2006;129:230-238.
35. Harris EF. Effects of patient age and sex on treatment:
correction of Class II malocclusion with the Begg
technique. Angle Orthod 2001;71:433-441.
36. 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.
37. 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.
38. Pae EK, McKenna GA, Sheehan TJ, Garcia R, Kuhlberg A,
Nanda R. Role of lateral cephalograms in assessing severity
and difficulty of orthodontic cases. Am J Orthod
Dentofacial Orthop 2001;120:254-262.
39. DeGuzman L, Bahiraei D, Vig KW, Vig PS, Weyant RJ,
O'Brien K. The validation of the Peer Assessment Rating
index for malocclusion severity and treatment difficulty.
Am J Orthod Dentofacial Orthop 1995;107:172-176.
40. Brook PH, Shaw WC. The development of an index of
orthodontic treatment priority. Eur J Orthod 1989;11:309320.
41. Shaw WC, Richmond S, O'Brien KD, Brook P, Stephens CD.
Quality control in orthodontics: indices of treatment need
and treatment standards. Br Dent J 1991;170:107-112.
42. Cassinelli AG, Firestone AR, Beck FM, Vig KW. Factors
associated with orthodontists' assessment of difficulty. Am
J Orthod Dentofacial Orthop 2003;123:497-502.
43. Ross RB. Testing surgical success. Plast Reconstr Surg
1988;82:921-922.
44
44. Chew MT, Sandham A. Effectiveness and duration of twoarch fixed appliance treatment. Aust Orthod J 2000;16:98103.
45
27
---
Onyeaso
100
and Begole17
Ormiston et al.18
Group 1
41
--1.6
Rossouw et al.19 88
Vaden et al.20
36
---
45
Group 2
---
14
---
---
51
Grad. sample
Glenn et al.16
Cl. I sample
1.4
---
---
---
---
---
---
1.1
0.4
---
---
---
1.1
---
1.0
0.7
---
---
---
0.6
---
2.5
2.8
2.3
2.4
---
2.4
---
0.9
1.0
1.1
1.1
---
0.6
---
2.7
2.9
2.1
2.5
---
2.2
---
0.6
0.9
0.7
0.9
---
0.7
---
---
---
2.9
5.5
1.0
---
4.0
---
---
2.3
4.1
1.8
---
2.8
Mx. Incisor
Md. Incisor
Overbite
Overjet
Irregularity
(mm)
(mm)
PAR score
Irregularity
Study
n
mean
SD
mean
SD
mean
SD
mean
SD
mean
SD
________________________________________________________________________________________________
Dyken et al.15
Board sample
54
----------------3.1
2.0
________________________________________________________________________________________________
Table 2.1 Previously reported individual measurements and PAR scores of immediate posttreatment occlusions: means and standard deviations15-20
Table 2.2 ABO OGS scores reflecting mean points lost and standard deviations in
previously reported samples of post-treatment occlusions18,21-28
__________________________________________________________________________________________________________
Marginal
Buccolingual
Occlusal
Occlusal
Alignment
Ridges
Inclination
Overjet
Contacts
Relationship
Total
Study
n
mean
SD
mean SD
mean SD
mean SD
mean SD
mean SD
mean
SD
__________________________________________________________________________________________________________
Abei et al.21
Ortho
126
5.4
4.4
3.9 2.9
4.5 4.0
4.2 4.0
4.1 4.3
3.1 3.4
26.0 11.4
GP
70
7.8
5.2
4.4
2.9
4.2
3.0
3.8
3.1
4.9
5.2
3.3
3.5
29.6
12.8
115
6.0
2.8
3.7
2.5
2.4
2.4
3.7
3.1
3.5
3.2
3.6
3.0
28.5
10.0
77
6.1
3.1
2.9
2.2
1.5
1.4
5.7
4.4
2.5
3.4
5.0
4.3
25.1
11.9
62
5.4
2.8
2.2
1.9
1.8
2.0
5.8
4.2
4.7
3.8
3.7
3.5
26.0
9.7
Costalos et al.24 24
Deguchi et al.25
Japan
72
7.8
3.9
4.0
2.6
6.7
3.1
4.7
2.8
5.3
5.3
2.2
2.6
31.2
10.5
5.5
2.3
3.2
2.2
6.9
3.7
6.6
2.5
4.7
3.4
3.1
3.0
33.6
13.6
54
6.1
3.8
2.9
3.5
5.6
3.3
4.5
3.2
5.8
3.4
3.7
3.8
32.8
10.3
48
6.8
3.3
4.4
2.6
2.8
2.6
3.6
2.5
5.7
4.7
5.5
4.7
32.2
11.7
Nett and Huang27 100
Yang-Powers et al.28
Univ
92
5.0
2.8
3.6
2.5
3.0
2.0
4.1
2.8
3.9
3.5
1.8
3.2
21.5
8.8
8.8
5.1
5.4
3.4
9.4
5.0
6.5
5.0
8.2
7.0
4.6
4.1
45.5
18.3
7.3
4.3
5.1
3.4
7.9
4.8
2.6
3.0
2.5
3.3
3.2
3.3
33.9
9.7
31.6
12.3
25.2
9.0
Cook et al.22
Cook23
Univ.
30
Priv.
Ameri
Djeu et al.26
ABO
32
Ormiston et al.18
Group 1
41
Group 2
45
CHAPTER 3:
JOURNAL ARTICLE
Abstract
Introduction:
To better understand why the Objective
Grading System (OGS) shows considerable variation among
post-treatment occlusions, this study was designed to
quantify the variability of OGS scores among similarly
treated patients and determine the factors that explain the
variability observed.
Methods:
One hundred and thirty-
eight subjects were randomly selected from the posttreatment archives of the Department of Orthodontics at
Saint Louis University.
Age, sex, mandibular plane angle,
and ANB angle were patient factors recorded from the
patients’ charts; active treatment time and supervising
orthodontist were treatment factors recorded from the
patients’ charts.
Post-treatment OGS scores for six of the
criteria (excluding interproximal contacts and root
angulations) and anterior Bolton ratio were measured on
study casts.
± 8.0.
Results:
The mean overall OGS score was 24.9
Occlusal contacts was the most important component
contributing to the overall score and variation, followed
by alignment.
Variation in total OGS scores was explained
by pre-treatment mandibular plane angle and treatment
duration.
Overall OGS scores increased by one point for
46
every four degree increase in the mandibular plane angle
and nearly one point for every three additional months of
treatment.
Approximately 16% and 15% of the variation in
alignment and buccolingual inclination, respectively, was
due to the treating orthodontist.
Conclusions:
Moderate
amounts of occlusal variation, which are evident
immediately post-treatment in Class I non-extraction
patients, can be explained by both patient and treatment
related factors.
Introduction
The American Board of Orthodontics (ABO) developed the
Objective Grading System (OGS) as an occlusal index to
evaluate post-treatment dental casts.1
Using this grading
system, recent studies have reported considerable ranges of
variation among post-treatment occlusions (Table 3.1).2-10
Cook et al.3 reported an average total OGS score of 25.1,
with individual variation ranging up to almost 45 for a
university sample; Yang-Powers et al.10 showed that the
post-treatment total OGS scores could average as high as
45.5 and range up to over 80.
Because orthodontists strive
to achieve the best occlusal results possible, it would be
both helpful and important to understand how much
variability actually exists among treated patients, which
47
components of occlusion are most variable, and what factors
determine variability.
There are various patient related factors, such as
skeletal discrepancies and anterior Bolton discrepancies,
that might be expected to explain post-treatment
variability.
Orthodontic treatment of anteroposterior or
vertical skeletal discrepancies often requires
dentoalveolar compensations to achieve correction at the
level of the occlusion.
Anterior teeth are frequently
positioned at different angles within the alveolus to
compensate for the skeletal disharmonies.
Andrews,11 for
example, demonstrated how third order angulation of
anterior teeth affect posterior occlusion.
Marginal
ridges, overjet, occlusal relationships, and occlusal
contacts, all components of the ABO OGS, might be expected
be affected by such compensations.
Similarly, differences
in the sizes of the anterior maxillary and mandibular teeth
might also be expected to produce variation in posttreatment occlusions.
In addition to patient factors, there are treatment
factors such as the diagnostic, technical, or motivational
skills of the doctor, and treatment duration that could
contribute to the variation in post-treatment occlusions.
Treatment factors are controllable to an extent and their
48
contribution to variability must be understood so that
orthodontists can attempt to minimize it in their efforts
to provide the best treatment possible for each patient.
The purpose of the current study was to evaluate
patient and treatment related variability among patients
with Angle Class I malocclusions that underwent nonextraction treatment at the postgraduate orthodontic clinic
at Saint Louis University.
These relationships have not
been previously explored.
Class I malocclusions comprise
the majority of cases in the typical orthodontic practice,
making them the most practical to study.
Materials and Methods
The sample includes 138 (81 females, 57 males)
randomly selected Class I cases who were treated without
extractions in the Department of Orthodontics at Saint
Louis University.
To be included in the study, patients
had to have Class I molar relationships at the beginning of
treatment (T1), non-extraction treatment as determined
immediately following treatment (T2), second molars in
occlusion, only one supervising instructor, and no more
than two treating residents.
Exclusion criteria included
missing teeth, craniofacial anomalies or syndromes,
patients treated in two phases or with surgery, and
49
retreated subjects.
The patients were chosen without
regard to age, race, or sex.
The data collected from the patients’ charts included
age, sex, time in fixed appliances, the ANB angle at T1,
the T1 mandibular plane angle as defined by SN-GoGn (FMA
plus 7 degrees was used in the 7% of cases for whom this
value was unavailable), and the supervising instructor.
The anterior Bolton ratio was measured at T2 by the primary
investigator using a Boley gauge. The ABO Objective Grading
System, as defined by Casko et al.,1
the casts of all subjects at T2.
was used to evaluate
The primary investigator
measured all of the casts after being calibrated for the
ABO OGS.
To control for bias, the examiner was blinded to
subject and instructor identity when scoring the casts.
The ABO OGS uses eight components:
alignment,
marginal ridges, buccolingual inclination, overjet,
occlusal contacts, occlusal relationships, interproximal
contacts, and root angulation.
Six of the eight criteria
were measured in the current study as described by Nett and
Huang.8
Interproximal contacts and root angulation, the
other two criteria, were not included in this nonextraction study because they pertain primarily when
extraction spaces must be closed.
50
Replicate analyses of 20 randomly chosen models showed
no significant systematic errors.
Method errors for the
six components ranged from 0.45 to 0.77, with buccolingual
and occlusal contacts showing the greatest technical
errors.
Method error for the total score was 1.20.
Stepwise multiple regression was used to evaluate each
component’s contribution to the variation of the overall
OGS score.
Multilevel modelling12 was used to determine the
effects of patient and treatment factors.
Results
Females and males comprised 58.7% and 41.3% of the
sample, respectively.
The median pretreatment age was 13
years, with a range from 10 to 48 years.
The average
treatment duration was 20.6 months and ranged from 8 to 44
months.
The pretreatment ANB and mandibular plane angles
were 2.6 ± 2.4 and 31.6 ± 5.5 degrees, respectively.
The
mean anterior Bolton ratio was 77.3% with a range from 70.7
to 83.3% (Table 3.2).
The total score of the six graded OGS components was
24.94 ± 7.99.
The coefficient of variation for the total
score was 0.321.
The distribution of total scores ranged
from 5 to 48 and was approximately normal (Figure 3.1).
The highest average component score (most points lost) from
51
the model analysis graded using the ABO OGS was occlusal
contacts at 6.25 ± 3.75 (Table 3.3).
The lowest average
component score (fewest points lost) was occlusal
relationships at 1.74 ± 1.83.
Occlusal contacts accounted
for approximately 26% of the total deductions, followed by
alignment (21%), buccolingual inclinations (18%), marginal
ridges (18%), overjet (10%), and occlusal relationships
(7%) (Figure 3.2).
A stepwise multiple regression showed
that the majority (56.2%) of the variation in the total
score was explained by the occlusal contacts component,
followed by alignment (17.7%), marginal ridges (10.3%),
overjet (6.1%), buccolingual inclination (5.4%) and
occlusal relationships (4.2%)(Table 3.4).
Multilevel estimates showed that treatment duration
(.301) and the mandibular plane angle (.275) had
significant effects on total OGS scores (Table 3.5).
Variation in marginal ridges and occlusal relationships was
explained by sex, variation in alignment and buccolingual
inclinations was explained by treatment duration, the MPA
explained variation in buccolingual inclinations and
occlusal contacts, the ANB angle explained variation in
buccolingual inclinations, and anterior Bolton explained
variation in overjet.
Sixteen percent of the variation in
alignment and 15% of the variation in buccolingual
52
inclinations was accounted for by the doctor (Table 3.6).
Only a small portion of the variation (2.3%) in the overall
OGS score was explained by doctor variation.
There was no
significant variation among doctors for marginal ridges,
overjet, occlusal contacts, or occlusal relationships
Discussion
The OGS scores indicated less post-treatment occlusal
variability and lower average total scores than previously
reported.
The coefficient of variation (CV) of this study
was 0.321, which was 2-15% less variable than most previous
studies (Table 3.1).
Reduced variability could be
attributed to sample selection.
Whereas this study focused
on Class I malocclusions that were treated without
extractions, all previous studies included all malocclusion
types.
Furthermore, several of the other studies included
all eight of the OGS components, while this study used only
six of the eight.
Nett and Huang8, who used the same six
criteria and a comparably sized university sample (n=100),
found slightly lower total OGS scores (mean 21.5), but more
variation (CV .409).
Occlusal contacts was the most important component
contributing to and explaining variability in the overall
OGS scores of Class I non-extraction patients.
53
Occlusal
contacts have been previously shown to be important
determinants of the overall OGS scores.2,6,7,10
This suggests
that less attention was given to this detail of finishing.
However, occlusal contacts, particularly those of the
maxillary palatal cusps, are the most difficult of the
occlusal components to clinically inspect prior to
appliance removal.
Furthermore, occlusal contacts
typically improve as occlusion settles after appliance
removal8 indicating that this component is likely to be less
problematic over time.
Alignment was the next most important component of
occlusion contributing to the variation in overall OGS
scores.
Alignment has been previously shown to be the most
important component.2-8
Because alignment is a primary
objective of orthodontic treatment, it is weighted heavily
in the OGS in two ways.
There are 56 possible point
deductions for this component (excluding third molars) and
deviations as small as .5 mm warrant a
deduction.1
Most
other components have half or fewer possible point
deductions and a larger threshold of deviation before
points are deducted.
This could explain its large
contribution to the overall OGS score in this and other
studies.
It suggests that the components of the OGS should
be weighted to ensure a more equal contribution.
54
Patients with higher pretreatment mandibular plane
angles had significantly higher total OGS scores (more
total deductions) than patients with lower mandibular plane
angles.
The multilevel estimates indicate that the overall
OGS score increases by one point for every four degree
increase in the mandibular plane angle.
This suggests that
the pre-treatment mandibular plane angle, as a diagnostic
patient factor, can have an effect on the final occlusion.
This validates the use of SN-GoGn in the ABO Discrepancy
Index13 to estimate pre-treatment case complexity.
The
mandibular plane also accounted for variation in occlusal
contacts and buccolingual inclinations.
Transverse
skeletal discrepancies that often accompany high mandibular
plane often require dentoalveolar compensations that could
explain its effect on these two occlusal components and the
overall score.
Patients with longer treatment times also had
significantly higher total OGS scores (more total
deductions).
All else being equal, the overall OGS score
increases nearly one point for every three additional
months of treatment.
This suggests that prolonged
treatment does not routinely result in a better posttreatment occlusion.
Specifically, alignment and
buccolingual inclinations had more deductions with
55
increased treatment time.
This is likely due to appliance
breakage or other compliance issues.
Understanding this
allows an orthodontist to critically evaluate cases that
have exceeded estimated treatment times and conscientiously
decide if termination of treatment may yield an acceptable
result in light of poor compliance or other factors.
Variation between orthodontists accounted for a
significant portion of the variation in alignment and
buccolingual inclinations.
The differences observed might
be associated with second molars.
Orthodontists have
differing preferences or philosophies regarding second
molars during treatment.
Some routinely band or bond
second molars initially during treatment while others fail
to control these teeth at any point throughout treatment.
While second molars were not examined separately in this
study, they represent a common deficiency in alignment and
buccolingual inclinations1 and differences in orthodontic
techniques regarding second molars might be expected to
explain some of the variation observed between doctors.
Only a small amount of variation in the overall OGS
score (2.3%) was explained by doctor variation.
be due to sample selection.
This could
For example, non-extraction
treatment might be expected to exhibit less variation than
extraction treatment.
By studying only Class I non-
56
extraction treatment, much of the variation introduced by
the orthodontist was therefore removed.
Also, the
variation between supervising orthodontists might have been
mitigated by the variation among the residents responsible
for patient care.
This study was unique in its approach to studying
variation in the OGS scores and is the first of its kind.
To study variation, the sample theoretically having the
least variability, Class I malocclusions treated without
extractions, was chosen.
A better design to evaluate
between doctor variation would utilize randomly or
sequentially selected patients from several private
practices having a specific treatment (i.e. Class II four
premolar extraction, Class II two premolar extraction,
etc.).
This would remove the potentially confounding
effects introduced in this study by the treating residents
and likely demonstrate that the overall OGS scores varies
according to the orthodontist responsible for treatment.
More studies of this kind could lead to a better
understanding of the Objective Grading System and hopefully
to more consistent excellent orthodontic results.
57
Conclusions
•
Class I non-extraction patients show moderate and
variable post-treatment occlusal discrepancies.
Overall discrepancies were correlated primarily to
mandibular plane angle and treatment duration.
•
Patients with higher mandibular plane angles had
significantly higher total OGS scores (more total
deductions); the overall OGS score increased by
approximately one point for every four degree increase
in the mandibular plane angle.
•
Patients with longer treatment times had significantly
higher total OGS scores (more total deductions); the
overall OGS score increased nearly one point for every
three additional months of treatment.
•
Differences among orthodontists accounted for
approximately 16% and 15% of the variation in
alignment and buccolingual inclinations, respectively.
58
Literature Cited
1. Casko JS, Vaden JL, Kokich VG, Damone J, James RD,
Cangialosi TJ et al. Objective grading system for dental
casts and panoramic radiographs. American Board of
Orthodontics. Am J Orthod Dentofacial Orthop 1998;114:589599.
2. Abei Y, Nelson S, Amberman BD, Hans MG. Comparing
orthodontic treatment outcome between orthodontists and
general dentists with the ABO index. Am J Orthod
Dentofacial Orthop 2004;126:544-548.
3. Cook DR, Harris EF, Vaden JL. Comparison of university
and private-practice orthodontic treatment outcomes with
the American Board of Orthodontics objective grading
system. Am J Orthod Dentofacial Orthop 2005;127:707-712.
4. Cook MK. Evaluation of Board-certified orthodonist's
sequential finished cases with the ABO objective grading
system Orthodontics. Saint Louis: Saint Louis University;
2003: p. 56.
5. Costalos PA, Sarraf K, Cangialosi TJ, Efstratiadis S.
Evaluation of the accuracy of digital model analysis for
the American Board of Orthodontics objective grading system
for dental casts. Am J Orthod Dentofacial Orthop
2005;128:624-629.
6. Deguchi T, Honjo T, Fukunaga T, Miyawaki S, Roberts WE,
Takano-Yamamoto T. Clinical assessment of orthodontic
outcomes with the peer assessment rating, discrepancy
index, objective grading system, and comprehensive clinical
assessment. Am J Orthod Dentofacial Orthop 2005;127:434443.
7. Djeu G, Shelton C, Maganzini A. Outcome assessment of
Invisalign and traditional orthodontic treatment compared
with the American Board of Orthodontics objective grading
system. Am J Orthod Dentofacial Orthop 2005;128:292-298;
discussion 298.
8. Nett BC, Huang GJ. Long-term posttreatment changes
measured by the American Board of Orthodontics objective
grading system. Am J Orthod Dentofacial Orthop
2005;127:444-450; quiz 516.
59
9. Ormiston JP, Huang GJ, Little RM, Decker JD, Seuk GD.
Retrospective analysis of long-term stable and unstable
orthodontic treatment outcomes. Am J Orthod Dentofacial
Orthop 2005;128:568-574; quiz 669.
10. Yang-Powers LC, Sadowsky C, Rosenstein S, BeGole EA.
Treatment outcome in a graduate orthodontic clinic using
the American Board of Orthodontics grading system. Am J
Orthod Dentofacial Orthop 2002;122:451-455.
11. Andrews LF. The six keys to normal occlusion. Am J
Orthod 1972;62:296-309.
12. Gilthorpe MS, Cunningham SJ. The application of
multilevel, multivariate modelling to orthodontic research
data. Community Dent Health 2000;17:236-242.
13. Cangialosi TJ, Riolo ML, Owens SE, Jr., Dykhouse VJ,
Moffitt AH, Grubb JE et al. The ABO discrepancy index: a
measure of case complexity. Am J Orthod Dentofacial Orthop
2004;125:270-278.
60
61
* total includes 6 of 8 component scores
† total includes 7 of 8 component scores
Current
138
5.2
2.8
4.5 2.3
4.6 2.3
2.6 1.8
6.3 3.7
1.7 1.8
24.9
24.9
8.0
0.321
Study
______________________________________________________________________________________________________________________________________
_________________________________________________________________________________________
Marginal
Buccolingual
Occlusal
Occlusal
Total
Coefficient
Ridges
Inclination
Overjet
Contacts
Relationship
6 comp.
8 comp.
of Variation
Alignment
Study
n
mean
SD
mean
SD
mean
SD
mean SD
mean SD
mean
SD
mean
mean
SD
(CV)
______________________________________________________________________________________________________________________________________
2
Abei et al.
Ortho 126
5.4
4.4
3.9 2.9
4.5 4.0
4.2 4.0
4.1 4.3
3.1 3.4
25.2
26.0† 11.4
0.438
GP
70
7.8
5.2
4.4 2.9
4.2 3.0
3.8 3.1
4.9 5.2
3.3 3.5
28.4
29.6† 12.8
0.432
Cook et al.3
Univ.
77
6.1
3.1
2.9 2.2
1.5 1.4
5.7 4.4
2.5 3.4
5.0 4.3
23.7
25.1 11.9
0.474
Priv.
62
5.4
2.8
2.2 1.9
1.8 2.0
5.8 4.2
4.7 3.8
3.7 3.5
23.6
26.0
9.7
0.373
4
Cook
115
6.0
2.8
3.7 2.5
2.4 2.4
3.7 3.1
3.5 3.2
3.6 3.0
22.9
28.5 10.0
0.351
Costalos et al.5
24
7.8
3.9
4.0 2.6
6.7 3.1
4.7 2.8
5.3 5.3
2.2 2.6
30.7
31.2 10.5
0.337
Deguchi et al.6
Japan
72
5.5
2.3
3.2 2.2
6.9 3.7
6.6 2.5
4.7 3.4
3.1 3.0
30.0
33.6 13.6
0.405
Ameri
54
6.1
3.8
2.9 3.5
5.6 3.3
4.5 3.2
5.8 3.4
3.7 3.8
28.6
32.8 10.3
0.314
7
Djeu et al.
48
6.8
3.3
4.4 2.6
2.8 2.6
3.6 2.5
5.7 4.7
5.5 4.7
28.8
32.2 11.7
0.363
Nett and Huang8
100
5.0
2.8
3.6 2.5
3.0 2.0
4.1 2.8
3.9 3.5
1.8 3.2
21.5
21.5* 8.8
0.409
Ormiston et al.9
Group 1 41
31.6 12.3
0.389
Group 2 45
25.2
9.0
0.357
Yang-Powers et al.10
Univ
92
8.8
5.1
5.4 3.4
9.4 5.0
6.5 5.0
8.2 7.0
4.6 4.1
42.9
45.5 18.3
0.402
ABO
32
7.3
4.3
5.1 3.4
7.9 4.8
2.6 3.0
2.5 3.3
3.2 3.3
28.6
33.9
9.7
0.286
______________________________________________________________________________________________________________________________________
Table 3.1 ABO OGS scores reflecting mean points lost and standard deviations in
previously reported samples of post-treatment occlusions2-10 compared to current research
results.
Tables
Table 3.2
n=138
Mean
SD
Range
Median
Mode
Table 3.3
Sample description
Age at
T1
(years)
15.4
6.8
10 - 48
13
13
Tx Time
(months)
ANB
(degrees)
Sn-GoGn
(degrees)
Ant. Bolton
(%)
20.6
6.0
8 - 44
20
17
2.6
2.4
-3 - 11
3
2
31.6
5.5
17 - 48
31.5
35
77.3
2.3
70.7 – 83.3
77.15
76
Objective Grading System scores
n=138
Align
Mean
SD
Min
Max
5.20
2.76
0
14
Marg
Ridges
4.50
2.32
0
11
Buclin
Inclin
4.64
2.31
0
13
OJ
2.62
1.85
0
8
62
Occlus
Contact
6.25
3.75
0
18
Occlus
Relat
1.74
1.83
0
11
Total
24.94
7.99
5
48
Table 3.4 Stepwise multiple regression with standardized
coefficients showing the contributions of objective grading
system scores (independent variables) to total score
(dependent variable).
Standardized
Coefficient
Step
Variable
R
R2
1
Occlusal
Contacts
Alignment
Marginal Ridges
Overjet
Buccolingual
Inclination
Occlusal
Relationships
.750
.562
R2
Change
.562
.860
.918
.951
.979
.739
.842
.904
.958
.177
.103
.061
.054
.345
.291
.232
.289
1.000 1.000
.042
.228
2
3
4
5
6
63
.469
64
Table 3.6 Absolute and relative between doctor (B/D) and
between patient (B/P) variation in objective grading system
scores.
n=138
Alignment
Marginal Ridges
Buccolingual
Inclination
Overjet
Occlusal Contacts
Occlusal Relation
Total Scores
B/D
B/P
%B/D
%B/P
1.17
0.001
0.78
6.101
5.284
4.373
16.1%
0.0%
15.1%
83.9%
100.0%
84.9%
0.002
0.08
0.001
1.399
3.372
13.787
3.303
59.835
0.1%
0.6%
0.0%
2.3%
99.9%
99.4%
100.0%
97.7%
65
64
0.197
4.26
1.362
10.76
0.157
2.623
1.836
1.187
0.986
2.366
0.255
4.838
Sex
-
0.896
-
-
-
-.803
-
-
.304
-
-
-
.394
-
.297
-
-
-
.076
-
.109
-
-
-
.030
-
Duration
.084 .038
- = Not statistically significant
Examples: Alignment= 3.469 + (.084 * Duration).
Occlusal
Contacts
Occlusal
Relations
Total
Scores
Marginal
Ridges
Buccolingual
Inclination
Overjet
Alignment
Constant
Est
SE
3.469
0.847
.256
-
.123
.087
-
-
-
.119
-
.057
.032
MPA
ANB
-
-
-.252
-
-
-
-
.073
-
-
-
-
-
-
-
-
.067
Bolton
-.131
-
-
Table 3.5 Multilevel estimates (Est) and standard errors (SE) for the effects of age at
start of treatment, sex of the patient, treatment duration, initial mandibular plane
angle (MPA), initial ANB angle, and initial anterior Bolton ratio on the objects grading
system scores.
Frequency
66
0
2
4
6
8
10
12
5
6
7
Figure 3.1
Distribution of Total OGS Scores
9 10 11 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 41 46
Total OGS Deductions
Figures
Occlusal Relationships
7%
Alignment
21%
Occlusal Contacts
26%
Marginal Ridges
18%
Overjet
10%
Buccolingual Inclination
18%
Figure 3.2
Component Scores As A Percentage Of
Total Deductions
67
VITA AUCTORIS
John Wesley Fleming was born on September 9, 1977 in
Bowling Green, KY to Shirley and John Theron Fleming II.
In 1995, he received his high school diploma from Allen
County-Scottsville High School in Scottsville, KY.
From
there he went to Western Kentucky University where he
graduated Summa Cum Laude with a Bachelor of Science degree
in chemistry, and minors in biology and finance, in May,
1999.
From 1999 to 2003, he attended dental school at the
University of Kentucky in Lexington.
He was awarded a
Doctor of Dental Medicine degree with honors in 2003.
The
following year, Dr. Fleming completed a residency in
general practice at the University of Kentucky and
Lexington VA hospitals.
In 2004, he began his graduate
studies in orthodontics at the Center for Advanced Dental
Education at Saint Louis University in Saint Louis,
Missouri where he is currently a candidate for the degree
of Master of Science in Dentistry.
graduate in January of 2007.
He is expected to
Dr. Fleming is moving to
Atlanta, Georgia to begin his career.
68