Comparison of HLD CAL-MOD scores obtained from digital versus plaster models

PDF

Download

Open document

Flip pages

Copy asset link

Request this asset

Transcript (if available)

Content Comparison
of
HLD
Cal-‐Mod
Scores
Obtained
From
Digital
Versus
Plaster
Models

By

Hany
Youssef

A
Thesis
Presented
to
the

FACULTY
OF
THE
USC
GRADUATE
SCHOOL

UNIVERSITY
OF
SOUTHERN
CALIFORNIA

In
Partial
Fulfillment
of
the
Requirements
for
the
Degree

MASTER
OF
SCIENCE

(CRANIOFACIAL
BIOLOGY)

May
2014

2

Acknowledgements

I
would
like
to
express
my
greatest
appreciation
and
gratitude
to
Dr.
Stephen
Yen
for
all
his
time
and

valuable
mentorship,
throughout
this
project.
I
would
also
like
to
thank
Dr.
Frederick
Schubert
from
Denti-‐
Cal
and
Dr.
David
Noel
from
the
Department
of
Health
Care
Services
for
sharing
their
insights
and
instruction

on
the
HLD
index.
I
am
grateful
to
Dr.
Hussein
Ebrahim,
who
scored
the
models,
worked
with
Denti-‐Cal
and

helped
me
with
this
project.
I
would
also
like
to
thank
Christiane
Lane,
who
performed
the
statistical

analysis.

I
would
like
to
express
special
gratitude
to
my
director,
Dr.
Glenn
Sameshima,
who
provided
my

colleagues
and
myself
with
the
best
opportunities
to
learn
and
grow
professionally.
Because
of
his

leadership,
my
residency
experience
at
USC
has
been
phenomenal.
My
appreciation
extends
to
the
full-‐time

and
part-‐time
faculty
for
sharing
their
knowledge
and
helping
me
to
become
a
competent
orthodontist.

Special
thanks
to
my
family:
words
cannot
express
how
grateful
I
am
to
my
parents
for
all
of
the

sacrifices
that
they
have
made
and
for
their
unlimited
and
unconditional
love
and
support.
Thank
you
to
my

dear
sister
and
best
friend,
Caroline,
who
has
been
always
there
for
me.

3

Table
of
Contents

Chapter
1:
Background
.............................................................................................................................
5

The
Development
of
Orthodontic
Indices
.............................................................................................
6

Angle’s
Classification
.............................................................................................................................
6

Index
of
Tooth
Position
.........................................................................................................................
7

Malalignment
Index
..............................................................................................................................
9

Summers
Occlusal
Index
.....................................................................................................................
10

Handicapping
Labio-‐lingual
Deviation
................................................................................................
10

Requirements
for
Orthodontic
Indices
...............................................................................................
13

World
Health
Organization
Requirements
..........................................................................................
13

Purpose
of
Orthodontic
Indices
..........................................................................................................
14

HLD
Score
Sheet
..............................................................................................................................
16

HLD
Data
Sheet
d-‐10
.......................................................................................................................
17

HLD
California
Modification
............................................................................................................
18

Development
of
Laser
Scanners
and
Digital
Models
..........................................................................
21

Denti-‐Cal
Authorization
Process
.........................................................................................................
21

Chapter
2:
Objective
...............................................................................................................................
23

Chapter
3:
Materials
and
Methods
.........................................................................................................
24

Sample
................................................................................................................................................
24

Raters
and
Calibration
........................................................................................................................
24

Experimental
Design
and
Data
Collection
...........................................................................................
25

Overjet
and
negative
overjet
measurements
are
made
using
the
back
end
of
the
caliper
which
sticks
of

the
ruler
like
and
extension
(Figure
6
and
7)
......................................................................................
30

Labio-‐lingual
spread
is
measured
from
the
most
labial
to
the
most
lingual
anterior
tooth
(figure
11).30

Openbite
measurement
is
made
from
contact
area
of
upper
incisors
to
the
contact
area
of
the
lower

incisors.
(Figure.9)
...............................................................................................................................
30

Ectopic
eruption
is
only
scored
when
there
are
2
or
more
ectopically
erupted
teeth
otherwise
crowding

is
scored
instead.
................................................................................................................................
30

A
tooth
must
be
50%
blocked
to
be
considered
ectopically
erupted.
................................................
30

Figure
11:
Labio-‐lingual
spread
measurement
...................................................................................
31

Statistical
Methods
.............................................................................................................................
32

Chapter
4:
Results
...................................................................................................................................
33

Descriptive
Statistics:
..........................................................................................................................
33

Inter-‐rater
Reliability
comparing
Raters’
Plaster
Model
Scores
to
Dr.
Schubert’s
HLD
scores
...........
33

Inter-‐rater
Agreement
between
Raters
1
and
2
HLD
scores
...............................................................
33

Intra-‐rater
Reliability
of
Plaster
versus
Digital
HLD
scores
..................................................................
33

Statistical
Analysis
...............................................................................................................................
34

Inter-‐rater
Comparison
of
Rater
1
scores
versus
Rater
2
scores
........................................................
36

Intra-‐rater
Reliability
of
Plaster
versus
Digital
Model
Scores
.............................................................
36

Chapter
5:
Discussion
..............................................................................................................................
37

References
..............................................................................................................................................
48

4

Abstract

Objective:
The
Handicapping
Labio-‐lingual
Deviation
Index
(HLD)
is
an
orthodontic
treatment
need
index,

which
is
used
to
assess
the
severity
of
a
malocclusion.
States
such
as
California
use
the
HLD
as
a
screening
tool

to
determine
whether
patients
qualify
for
subsidized
treatment.
The
objective
of
this
study
was
to
compare

HLD
scores
using
plaster
versus
digital
models.

Methods:
Seventy-‐eight
duplicate
study
models
that
were
sent
from
Children’s
Hospital
Los
Angeles
(CHLA)
to

Denti-‐Cal
for
evaluation
were
scored
by
an
orthodontic
fellow
(rater
1),
a
senior
orthodontic
resident
(rater
2)

and
by
the
Denti-‐Cal
orthodontic
consultant.
The
scores
of
a
Denti-‐Cal
consultant
were
considered
the
gold

standard
in
this
study.
Fifty
out
of
these
models
were
then
scanned
using
the
3D
Ortho
Insight
laser
scanner.

Rater
1
and
rater
2
scored
the
50
digital
models.
Kappa
statistics
were
used
to
compare
inter-‐
and
intra-‐rater

reliability.
Comparisons
included
(a)
inter-‐rater
reliability
between
plaster
model
scores
of
both
raters
versus

the
gold
standard,
(b)
inter-‐rater
reliability
between
rater
1
versus
rater
2
for
both
plaster
and
digital
model

scores
and
(c)
intra-‐rater
reliability
between
digital
and
plaster
model
scores
for
each
rater.

Results:
Inter-‐rater
reliability
of
Raters
1
and
2
(post-‐calibration)
versus
gold
standard
was
almost
perfect

(combined
K=
0.89).
Inter-‐rater
reliability
of
rater
1
versus
rater
2
was
almost
perfect
for
both
plaster
and

digital
model
scores
(combined
K=
0.85).
Intra-‐rater
reliability
was
substantial
for
both
Rater
1
(K=
0.79)
and

Rater
2
(K
=
0.67).

Conclusion:
There
was
substantial
intra-‐rater
agreement
between
HLD
Cal-‐Mod
digital
and
plaster
model

scores.
Differences
in
HLD
Cal-‐Mod
scores
between
digital
and
plaster
models
were
as
high
as
13%,
and
the

combined
kappa
for
intra-‐rater
agreement
was
0.73.
HLD
specific
software
is
needed
to
correct
the
sources

of
difference
in
HLD
score.

Digital
models
present
a
promising
alternative
to
plaster
models
for
use
with
the

HLD
Index.
Training
and
calibrating
raters
on
proper
usage
of
digital
model
software
and
its
tools
are

recommended,
prior
to
HLD
digital
scoring.

5

Chapter
1:
Background

The
Handicapping
Labio-‐lingual
Deviation
Index
(HLD)
is
an
orthodontic
treatment
need
index

used
to
quantify
the
severity
of
a
malocclusion
and
demonstrate
the
presence
or
absence
of
a
handicap

by
using
a
cutoff
point.
Harry
Draker
developed
the
HLD
in
1960,
when
public
health
orthodontists

highlighted
the
need
for
an
objective
simple
index
to
complement—and
possibly
replace—the
clinical

judgment
needed
to
decide
whether
or
not
a
patient
qualifies
for
free
treatment
through
government-‐
funded
health
programs
(Draker,
1960).
Currently,
the
HLD
is
used
by
state-‐supported
entitlement

programs
in
many
states
such
as
California
and
New
York
to
help
determine
eligibility
for
free
orthodontic

treatment.
Due
to
the
financial
constraints
to
these
programs,
not
all
patients
who
apply
would
receive

free
orthodontic
treatment.
The
HLD
index
is
used
as
a
screening
tool
to
quantify
malocclusion,
prioritize

need
for
treatment,
and
identify
only
those
patients
who
demonstrate
a
handicapping
malocclusion
for

free
treatment.

In
the
state
of
California,
Denti-‐Cal
accepts
second
pour
diagnostic
models
as
documentation
for

authorization
in
all
handicapping
malocclusion
and
cleft
palate
cases.
Providers
must
send
diagnostic

models
to
the
Denti-‐Cal
office
in
Sacramento
in
order
to
obtain
authorization
for
subsidized
orthodontic

treatment
(Denti-‐Cal
Provider
Hand
Book
2014).
But
these
models
are
subject
to
loss
and/or
damage

during
the
shipping
process,
which
could
force
the
patient
to
return
to
the
provider
to
make
a
new
set
of

models
to
send
to
Sacramento
for
authorization.
Denti-‐Cal
does
not
return
the
models;
therefore,
once
a

case
is
approved,
the
orthodontist
is
required
to
take
an
additional
set
of
impressions
in
order
to
acquire

models
for
his
own
records.

With
advances
in
digital
technology,
three-‐dimensional
models
are
becoming
more
popular
in

orthodontics.
Our
presupposition
is
that
digital
models
for
HLD
scoring
streamline
the
authorization
and

reduce
costs,
in
comparison
to
the
current
use
of
plaster
models.
However,
it
is
not
yet
known
whether

6

digital
and
plaster
models
will
yield
the
same
HLD
scores.
Therefore,
this
study
was
undertaken
to

compare
HLD
scores
between
digital
and
plaster
models.

The
Development
of
Orthodontic
Indices

Occlusion
has
been
of
major
interest
to
dentists
since
the
late
nineteenth
century.
In
1899,

Edward
H.
Angle
realized
the
need
to
identify
normal
occlusion
and
assess
deviations
from
normal

occlusion,
which
he
classified
into
three
classes
of
malocclusion
(Angle,
1900).
Since
then,
researchers

have
developed
various
indices
to
measure
and
assess
malocclusion.
An
orthodontic
index
can
be

qualitative,
in
that
it
classifies
and
categorizes
malocclusion,
or
quantitative,
when
it
measures
the

amount
of
malocclusion
and
gauges
its
severity.

Angle’s
Classification

In
1899
Angle
defined
and
classified
malocclusion
based
on
the
relationship
of
the
upper
and

lower
first
molars.
As
an
orthodontist,
Angle
focused
on
the
diagnosis
and
treatment
of
malocclusion.
His

classification
is
considered
to
be
the
first
method
for
describing
and
defining
malocclusion.
However,

Angle’s
classification
was
criticized
because
it
only
considered
the
antero-‐posterior
deviation
(Tang,

1993).
Case
pointed
out
the
fact
that
Angle
ignored
the
relationship
between
teeth
and
face
(Case,
1963).

Gravely
and
Johnston
(1974)
further
found
Angle’s
method
to
be
unreliable,
due
to
high
inter-‐
and
intra-‐
examiner
error,
when
assessing
Class
II
Division
2
cases.
As
a
clinician,
Angle
was
more
concerned
with

diagnosis
and
treatment
planning
and
therefore
developed
his
classification
to
serve
these
purposes.

7

Index
of
Tooth
Position

Until
1950,
the
majority
of
occlusal
indices
were
qualitative,
practitioner-‐focused,
and
primarily

concerned
with
diagnosis
and
treatment
methods.
Massler
and
Frankel
(1951)
recognized
those

deficiencies
and
proposed
a
method
of
assessing
malocclusion
that
could
produce
statistical
data.
They

proposed
the
Index
of
Tooth
Position,
which
was
meant
to
serve
the
epidemiologic
needs
of
quantifying

the
prevalence
rate
and
incidence
of
malocclusion.
The
index
was
based
on
quantifying
the
number
of

teeth
displaced
or
rotated
as
all
or
none
(Massler
and
Frankel
1951).
Since
1950,
several
innovative

orthodontic
indices
have
been
developed.
The
most
popular
and
widely
acceptable
indices
are

summarized
in
Table
1.

Table
1:
Summary
of
Orthodontic
Indices

Year
Index
Summary

1951
Index
of
Tooth
Position

(Massler
and
Frankel)

Counts
the
number
of
displaced
and/or
rotated
teeth

qualitatively
(all
or
none).

1959
Malalignment
Index

(Vankirk
and
Pennell)

Quantitatively
measures
rotated
and
displaced
teeth.

1961
Occlusal
Feature
Index

(Poulton
and
Aaronson)

Measures
lower
anterior
crowding,
cuspal
interdigitation,

vertical
overbite,
and
horizontal
overjet.

1961

Malocclusion
Severity

Estimate
(Grainger)

Seven
weighted
and
defined
measurements:

1. Overjet

2. Overbite

3. Anterior
open
bite

4. Congenitally
missing
maxillary
incisors

5. First
permanent
molar
relationship

6. Posterior
crossbite

7. Tooth
displacement

1964
Bjork
Method
Records
malocclusion
based
on
symptoms
and
detailed

definitions.

1. Anomalies
of
the
dentition:
tooth
anomalies,
abnormal

eruption
and
malalignment
of
individual
teeth.

2. Occlusal
anomalies:
deviations
in
the
positional

relationship
between
the
upper
and
lower
dental

arches
in
the
sagittal,
vertical,
and
transverse
planes.

3. Deviation
in
space
conditions:
spacing
or
crowding.

8

1966
Occlusal
Index

(Summers)

Nine
weighted
and
defined
measurements

1. Molar
relation

2. Overbite

3. Overjet

4. Posterior
crossbite

5. Posterior
open
bite

6. Tooth
displacement

7. Midline
relation

8. Maxillary
median
diastema

9. Congenitally
missing
maxillary
incisors

1973

Five-‐Point
System

(Ackerman
and
Proffit)

1. Alignment:
ideal,
crowding,
spacing,
mutilated

2. Profile:
mandibular
prominence,
mandibular
recession,

lip
profile
relative
to
nose
and
chin

3. Crossbite:
Relationship
of
the
dental
arches
in
the

transverse
plane

4. Angle
classification:
antero-‐posterior
relationship
of
the

dental
arches

5. Bite
depth:
Relationship
of
the
dental
arches
in
the

vertical
plane

1966-‐
1974

Swedish
Medical
Board

Index
(SMBI)

Relies
on
the
subjective
judgment
and
taking
patient’s
wish
in

consideration.
Originally
it
had
4
categories
of
treatment
need

(1966)
then
in
1974
Linder-‐Aronson
and
co-‐workers
added
the

‘zero’
or
no
need
for
treatment
category.

0. No
need:
Normal
occlusion

1. Little
need:
Mild
deviations
from
normal
occlusion,

such
as
class
III
with
little
crossbite,
class
II
without

other
anomalies,
mild
deep
or
open,
mild
crowding,

spacing
or
rotations.

2. Moderate:
Esthetically/functionally
disturbing

proclined
or
retroclined
incisors,
deep
bite
with
gingival

contact
but
without
gingival
irritation,
severe
crowding

or
spacing
and
moderate
frontal
rotations.

3. Urgent:
Class
III
with
forced
bite,
deep
bite
with

gingival
irritation
not
only
on
incisive
papilla,
large

overjet
with
lower
lip
behind
upper
centrals,
extremely

open
bite,
crossbite
causing
transverse
forced
bite,

scissors
bite
interfering
with
articulation,
severe
frontal

crowding
or
spacing,
retained
canines,
aesthetically

and/or
functionally
disturbing
rotations.

4. Very
urgent:
Esthetically
and/or
functionally

handicapping
anomalies
such
as
cleft
lip
and
palate,

extreme
class
II
or
III
malcocclusion.

9

1992
Peer
Assessment
Rating

Index
(PAR)

(Richmond)

Evaluates
occlusal
traits
in
11
segments
of
the
dentition
and

the
scores
are
added
to
make
the
final
PAR
score.
A
score
of

zero
means
perfect
occlusion
and
the
higher
the
score
indicate

more
deviation
from
ideal.

1. Upper
right
segment

2. Upper
anterior
segment

3. Upper
left
segment

4. Lower
right
segment

5. Lower
anterior
segment

6. Lower
left
segment

7. Right
buccal
occlusion

8. Overjet

9. Overbite

10. Centerline

11. Left
buccal
segment

2000
Index
of
Complexity,

Outcome,
and
Need

(ICON)

(Daniels
and
Richmond)

Five
components
are
measured
on
the
cast
and
their
scores

are
weighted
as
follow:

1. Dental
esthetics
(multiplied
by
7)

2. Crowding
or
spacing
(multiplied
by
5)

3. Crossbite
(multiplied
by
5)

4. Incisor
open
bite/overbite
(multiplied
by
5)

5. Buccal
segments
anterior-‐posterior
relationship

(multiplied
by
3)

Complexity
is
interpreted
from
scores
as
follow:

Less
than
29
are
classified
as
easy

29
to
50
is
mild

51
to
63
is
moderate

64
to
77
is
difficult

Greater
than
77
is
very
difficult.

The
treatment
outcome
is
evaluated
by
this
formula:

Pre-‐Treatment–
4
x
Post-‐Treatment
=
Improvement

Improvement
grades
are
interpreted
as
follow:

Greater
than
-‐1
indicates
great
improvement

-‐25
to
-‐1
indicates
substantial
improvement

53
to
-‐26
indicates
moderate
improvement

-‐85
to
-‐54
indicates
minimal
improvement

Less
than
-‐85
is
no
improvement
or
worse

Malalignment
Index

In
1959
Van
Kirk
and
Pennell
developed
the
Malalignment
Index,
which
measured
the
degree
of

tooth
displacement
and
rotation
in
a
quantitative
fashion.

10

Summers
Occlusal
Index

In
1966,
Summers
described
a
scoring
system
that
measured
nine
characteristics
of
malocclusion:

(a)
dental
age,
(b)
molar
relation,
(c)
overbite,
(d)
overjet,
(e)
posterior
crossbite,
(f)
posterior
open
bite,

(g)
tooth
displacement,
(h)
midline
relations,
and
(i)
missing
permanent
teeth.
Summers
aimed
to
make

the
Occlusal
Index
score
valid
over
time
by
taking
into
consideration
dental
age
and
molar
relationship.

According
to
the
Occlusal
Index,
molar
relationship
yields
different
values,
depending
on
the
particular

dental
stage
of
the
patient
(Summers
1971).

Handicapping
Labio-‐lingual
Deviation

In
1960,
H.
L.
Draker
proposed
the
HLD
Index
as
an
approach
to
scoring
and
weighting
selected

deviations
from
ideal
occlusion.
The
HLD
is
a
quantitative
treatment
need
index.
Essentially,
Draker’s
goal

was
to
develop
a
simple
and
objective
administrative
tool
to
aid
public
health
workers
in
identifying

malocclusions
that
were
severe
enough
to
be
considered
as
handicapping
(Draker
1960).
The
HLD
and
its

evolution
are
further
discussed
within
this
chapter.

Dental
Aesthetic
Index

In
1980,
Jenny
et
al.
developed
the
Dental
Aesthetic
Index
(DAI),
to
overcome
a
deficiency

recognized
in
many
orthodontic
treatment
need
indices.
Because
other
indices
ignored
the
aesthetic

handicap
and
its
possible
psychosocial
impact
on
patients,
the
DAI
provided
an
objective
quantitative

method
to
assess
the
aesthetic
handicap
and
the
resulting
psychosocial
handicap.
The
DAI
uses
initial

diagnostic
models
to
evaluate
occlusal
traits
thought
to
contribute
to
dental
aesthetics:
missing
teeth,

incisal
crowding,
antero-‐posterior
molar
relationship,
anterior
tooth
diastema,
anterior
tooth

irregularities,
overjet
and
open
bite.
The
traits
are
differentially
weighted
relative
to
their
contributions
to

the
aesthetic
handicap.
The
DAI
uses
a
threshold
score,
or
cutoff
point,
to
determine
subsequent
need
for

11

treatment.
This
threshold
score
can
be
adjusted
according
to
the
resources
available
in
publicly
funded

health
programs.
The
DAI
traits
and
weights
are
summarized
in
Table
2
(Cons
et
al.
1986).

Table
2:
DAI
Traits
and
their
Weights

DAI
component
Weight

Number
of
missing
visible
teeth
(incisors,
canines,
and
premolars)
5.76

Assessment
of
incisal
segment
crowded:
0,
no
segment
crowded,
1:
1
segment

crowded,
2:
2
segments
crowded

1.15

Assessment
of
incisal
segment
spacing:
0,
no
segment
spaced,
1:
1
segment

spaced,
2:
2
segments
spaced

1.31

Measurement
of
midline
diastema
in
mm
3.13

Measurement
of
the
largest
anterior
irregularity
in
mm
on
the
maxilla
1.34

Measurement
of
the
largest
anterior
irregularity
in
mm
on
the
mandible
0.75

Overjet
in
mm
3.68

Open
bite
in
mm
3.69

Antero-‐posterior
molar
relationship
on
the
most
deviated
side,
0
for
normal,
1

for
half
cusp
and
2
for
full
cusp,
either
mesial
or
distal

2.69

Index
of
Orthodontic
Treatment
Need

In
1989,
Brook
and
Shaw
developed
the
Index
of
Orthodontic
Treatment
Need
(IOTN),
which

incorporated
two
components.
The
first
component
consisted
of
the
dental
health
and
functional

indications
for
treatment.
The
second
included
the
aesthetic
impairment
caused
by
the
malocclusion.

Brook
and
Shaw
(1989)
recognized
the
importance
of
aesthetic
assessment
in
any
treatment
need
index

and
stated
that
the
primary
benefit
to
the
patient
from
orthodontic
treatment
is
aesthetic
improvement,

and
results
in
the
consequent
improvement
of
a
patient’s
social
well-‐being.

1. The
aesthetic
component
utilizes
the
standardized
continuum
of
aesthetic
need
(SCAN),
which

was
developed
from
pictures
of
1,000
12-‐year-‐olds.
Six
non-‐dental
judges
rated
these
photos
on
a

visual
analogue
scale
and
on
a
10-‐point
scale
from
0.5
to
5,
with
0.5
being
most
attractive
and
5

being
least
attractive.

2. The
dental
health
and
functional
component
categorizes
occlusion
using
five
grades:
Grade
1

represents
no
or
little
need
for
treatment
and
Grade
5
means
that
great
treatment
is
needed

(Table
3).

12

Table
3:
Traits
Measured
in
the
Dental
Health
and
Functional
Component
of
the
IOTN

Grade
5—very
great
need

Defects
of
cleft
lip
and/or
palate.

Increased
overjet
greater
than
9
mm.

Reverse
overjet
greater
than
3.5
mm
with
reported
masticatory
or
speech
difficulties.

Impeded
eruption
of
teeth
due
to
crowding,
displacement,
presence
of
supernumerary
teeth,

retained
deciduous
teeth,
and/or
any
other
pathological
cause.

Extensive
hypodontia
with
restorative
implications
(more
than
one
tooth
missing
in
any
quadrant),

requiring
pre-‐restorative
orthodontics.

Grade
4—great
need

Increased
overjet
greater
than
6
mm
but
less
than
or
equal
to
9
mm.

Reverse
overjet
greater
than
3.5
mm
with
no
reported
masticatory
or
speech
difficulties.

Reverse
overjet
greater
than
1
mm
but
less
than
or
equal
to
3.5
mm
with
reported
masticatory
or

speech
difficulties.

Anterior
or
posterior
crossbites
with
greater
than
2
mm
displacement
between
retruded
contact

position
and
intercuspal
position.

Posterior
lingual
crossbites
with
no
occlusal
contact
in
one
or
both
buccal
segments.

Severe
displacement
of
teeth
greater
than
4
mm.

Extreme
lateral
or
anterior
open
bite
greater
than
4
mm.

Increased
overbite
causing
notable
indentations
on
the
palate
or
labial
gingiva.

Patient
referred
by
colleague
for
collaborative
care
(e.g.,
periodontal,
restorative,
or
TMJ).

Less
extensive
hypodontia
requiring
pre-‐restorative
orthodontics
or
orthodontic
space
closure
to

obviate
the
need
for
a
prosthesis
(not
more
than
one
tooth
missing
in
any
quadrant).

Grade
3—moderate
need

Increased
overjet
more
than
3.5
mm
but
less
than/equal
to
6
mm
with
incompetent
lips.

Reverse
overjet
greater
than
1
mm
but
less
than
or
equal
to
3.5
mm.

Increased
overbite
with
gingival
contact
but
without
indentations
or
trauma.

Anterior
or
posterior
crossbite
with
less
than
or
equal
to
2
mm
but
greater
than
1
mm.
Displacement

between
retruded
contact
position
and
intercuspal
position.

Moderate
lateral
or
anterior
open
bite
greater
than
2
mm
but
less
than
or
equal
to
4
mm.
Moderate

displacement
of
teeth
greater
than
2
mm
but
less
than
or
equal
to
4
mm.

Grade
2—little
need

Increased
overjet
greater
than
3.5
mm
but
less
than/equal
to
6
mm
with
lips
competent.

Reverse
overjet
greater
than
0
mm
but
less
than
or
equal
to
1
mm.

Increased
overbite
greater
than
3.5
mm
with
no
gingival
contact.

Anterior
or
posterior
crossbite
with
less
than
or
equal
to
1
mm
displacement
between
retruded

contact
position
and
intercuspal
position.

Small
lateral
or
anterior
open
bites
greater
than
1
mm
but
less
than
or
equal
to
2
mm.

Pre-‐normal
or
post-‐normal
occlusions
with
no
other
anomalies.

Mild
displacement
of
teeth
greater
than
1
mm
but
less
than
or
equal
to
2
mm.

Grade
1—no/little
need

Other
variations
in
occlusion
including
displacement
less
than
or
equal
to
1
mm.

13

Requirements
for
Orthodontic
Indices

In
late
1950s,
Jamison
and
McMillan
(1950)
suggested
a
list
of
requirements
for
any
index
used
by

a
public
orthodontic
program
as
a
screening
tool.
The
list
included
the
following
criteria
for
any
such

index:

1. It
should
be
simple,
accurate,
reliable,
and
reproducible.

2. It
should
be
objective
in
nature
and
yield
quantitative
data
that
may
be
analyzed
using
current

statistical
methods.

3. It
must
be
designed
so
as
to
differentiate
between
handicapping
and
non-‐handicapping

malocclusions.

4. The
examination
required
must
be
one
that
can
be
performed
quickly,
even
by
examiners
without

special
instruction
in
orthodontic
diagnosis.

5. It
should
lend
itself
to
modification
for
the
collection
of
epidemiological
data
regarding

malocclusion
other
than
prevalence,
incidence,
and
severity
(e.g.,
frequency
of
malpositioning
of

individual
teeth).

Draker
added
two
more
criteria
to
the
list,
which
are
as
follows:

6. It
should
be
usable
on
either
patients
or
study
models.

7. It
should
measure
the
degree
of
handicap
and
avoid
classifying
the
malocclusion.

World
Health
Organization
Requirements

In
1966,
the
World
Health
Organization
(WHO)
reported
on
international
epidemiological
studies

of
oral
diseases,
defining
nine
requirements
for
indices.
In
1979,
in
response
to
the
increased
public
and

institutional
funding
of
orthodontic
services,
the
WHO
developed
a
simplified
method
for
recording

14

malocclusion.
The
WHO
index
was
made
to
assess
the
prevalence
of
malocclusion
and
estimate
the

treatment
need
of
a
population.

The
WHO
requirements
for
an
ideal
index
are
as
follows:

1. The
index
should
be
expressed
as
a
single
number
that
corresponds
to
a
relative
position
on
a

finite
scale,
with
definite
upper
and
lower
limits,
running
by
progressive
gradation
from
zero

(absence
of
disease)
to
the
ultimate
point
(disease
in
its
terminal
stage).

2. The
index
should
be
sensitive
throughout
the
scale.

3.

The
score
should
correspond
to
the
clinical
importance
of
the
disease
stage
it
represents.

4.

The
index
value
should
be
amendable
to
statistical
analysis.

5.

The
classification
must
be
reproducible.

6. The
index
should
be
simple
and
accurate
and
yield
itself
to
modification
for
data
collection.

7. The
examination
procedure
should
require
minimal
judgment.

8. The
index
should
be
simple
enough
to
permit
the
study
of
a
large
population
without
undue
cost

in
time
or
energy.

9. The
examination
required
should
be
performed
quickly
to
show
a
group
variation.

Summers
added
a
tenth
criterion:

10. The
index
should
be
valid
over
time,
meaning
that
it
focuses
on
dental
abnormalities
that
are

constant
and
do
not
improve
with
age.

Purpose
of
Orthodontic
Indices

Shaw

et
al.
(1995)
classified
orthodontic
indices
based
on
their
purpose,
into
the
following

categories:

1.

Diagnostic
index—a
qualitative
tool
used
to
describe
and
classify
malocclusion,
in
order
to

develop
a
treatment
prescription.

15

2.

Epidemiologic
index—used
to
quantify
the
prevalence
rate
and
incidence
of
malocclusion
in
a

population.

3.

Treatment
need
index—used
to
quantify
the
malocclusion
and
consequently,
the
need
for

treatment.

4.

Orthodontic
treatment
outcome
index—measures
and
evaluates
post-‐orthodontic
treatment

results.

5.

Orthodontic
treatment
complexity
index—used
to
evaluate
the
level
of
complexity
of
a
case,

prior
to
treatment.

A
summary
of
the
indices
and
their
purpose
is
given
in
Table
4
(Shaw
et
al.
1995).

Table
4:
Classification
of
Indices
Based
on
Their
Purpose
of
Use

Diagnostic
indices
Angle
Classification
System

Incisal
Categories
of
Ballard
and
Wayman
(1964)

Five-‐Point
System
of
Ackerman
and
Proffit
(1969)

Epidemiologic
indices

Index
of
Tooth
Position
(Massler
and
Frankel
1951)

Malalignment
Index
(Van
Kirk
and
Pennel
1959)

Occlusal
Feature
Index
(Poulton
and
Aaronson
1961)

Bjork
Method
(Bjork
et
al.
1964)

Summer’s
Occlusal
index
(Summers
1971)

FDI
Method
(Baume
et
al.
1973)

Little’s
Irregularity
Index
(Little
1975)

Orthodontic

treatment
need

indices

Handicapping
Labio-‐lingual
Deviation
Index
(Draker
1960)

Swedish
Medical
Board
Index
(Swedish
Medical
Health
Board
1966;

Linder-‐Aronson
1974,
1976)

Dental
Aesthetic
Index
(Cons
et
al.
1986)

Index
of
Orthodontic
Treatment
Need
(Brook
and
Shaw
1989)

Index
of
Complexity,
Outcome,
and
Need
(Daniels
and
Richmond

2000)

Orthodontic

treatment
outcome

indices

Peer
Assessment
Rating
Index
(Richmond
et
al.
1992)

16

Orthodontic

treatment
complexity

indices

Index
of
Orthodontic
Treatment
Complexity
(Llewellyn
et
al.
2007)

Evolution
of
the
HLD
and
the
California
Modification

In
1959
Draker
developed
an
index
that
would
measure
particular
deviations
from
the
ideal

occlusion,
which
were
scored
and
weighted.
His
goal
was
to
develop
a
simple,
objective
administrative

tool
to
help
public
health
workers
identify
malocclusions
that
were
severe
enough
to
qualify
as

handicapping.
The
result
was
the
HLD
Index
(Draker
1960).

HLD
Score
Sheet

Draker
initially
proposed
a
score
sheet
for
the
HLD
Index
(Figure
1),
which
evaluated
nine

components
of
malocclusion:
(a)
cleft
plate,
(b)
severe
traumatic
occlusion,
(c)
overjet
in
mm,
(d)
overbite

in
mm,
(e)
mandibular
protrusion,
(f)
open
bite,
(g)
ectopic
eruption
of
anterior
teeth,
(h)
anterior

crowding
in
maxilla
and
mandible,
and
(i)
labio-‐lingual
spread.

17

Figure
1:
Draker’s
score
sheet.

According
to
the
score
sheet,
a
score
of
13
or
higher
is
considered
handicapping.
A
cleft
palate
and

severe
traumatic
occlusion
automatically
received
a
score
of
15.
Measurements
are
multiplied
for
the

following
conditions
as
follows;
mandibular
protrusion
by
five,
open
bite
by
four,
and
ectopic
eruption
by

three.

HLD
Data
Sheet
d-‐10

Through
his
research,
Draker
found
that
measurements
of
ectopic
eruptions
and
crowding
are
not

statistically
reliable.
He
also
found
that
labio-‐lingual
deviation
is
the
most
accurate
measure
of
a

handicap.
In
May
1959,
he
developed
and
published
HLD
data
sheet
d-‐10,
to
replace
the
original
score

sheet.
The
HLD
data
sheet
d-‐10
(Figure
2)
eliminates
the
conditions
of
ectopic
eruption
and
crowding,
and

has
only
seven
components
of
malocclusion.

Moreover,
the
new
sheet
does
not
use
weighted
factors.

18

Figure
2:
HLD
data
sheet
d-‐10

HLD
California
Modification

In
1989
the
State
of
California
was
sued
for
not
complying
with
the
orthodontic
provisions
of

Medicaid
statutes.
William
Parker,
an
orthodontist
in
Sacramento,
was
recruited
to
help
design
an
index

to
select
patients
who
qualified
for
treatment.
Parker
wrote
a
paper
identifying
the
factors
that
should
be

weighed
when
determining
the
medical
necessity
of
orthodontic
treatment.
He
suggested
the
use
of
the

HLD
Index
with
some
additions.
He
added
two
automatically
qualifying
conditions:
(a)
destruction
of

tissue
due
to
deep
impinging
bites
and
(b)
destruction
of
tissue
due
to
crossbites
of
individual
anterior

teeth.
As
a
result
of
the
lawsuit,
another
category
was
added
to
the
scoring
portion
of
the
index;
the

unilateral
posterior
crossbite.

The
original
HLD
(Cal
Mod)
was
created
and
put
into
use
in
1991.
In
1994
California
was
sued
again

by
Duran
Belche,
which
opened
the
door
for
more
modifications.
An
overjet
greater
than
9
mm
was

added
as
an
automatic
qualifying
condition,
as
previously
suggested
by
Parker’s
research
results.
To

resolve
the
lawsuit,
a
reverse
overjet
greater
than
3.5
mm
was
added
as
a
qualifying
condition
(Parker

1998).

19

The
current
form
of
the
HLD
index
(Figure
3)
has
13
parameters
or
conditions
that
are
taken
into

consideration
in
the
scoring
process.
The
first
six
(1-‐6A)
are
automatically
qualifying
conditions,
meaning

that
if
the
model
shows
that
the
patient
has
one
of
these
conditions,
then
no
more
scoring
is
necessary

and
the
patient
automatically
qualifies
for
subsidized
treatment.
These
automatic
qualifying
conditions

are
as
follows:
(a)
cleft
palate,
(b)
craniofacial
anomaly,
(c)
deep
impinging
overbite,
(d)
anterior
crossbite

causing
attachment
loss,
(e)
severe
traumatic
deviation,
and
(f)
overjet
greater
than
9
mm.
The
conditions

from
6B
to
13
are
used
to
score
the
malocclusion
and
include
the
following:
(a)
overjet
equal
to
or
less

than
9
mm,
(b)
overbite,
(c)
mandibular
protrusion,
(d)
open
bite,
(e)
ectopic
eruption,
(f)
anterior

crowding,
(g)
labio-‐lingual
spread,
and
(h)
posterior
unilateral
crossbite.

20

Figure
3.
The
current
HLD
form.

21

Development
of
Laser
Scanners
and
Digital
Models

Digital
models
were
introduced
in
the
1990s
as
a
result
of
improvements
in
laser
scanning

technology.
Advances
in
technology
decreased
the
price
of
both
hardware
and
software,
enabling
more

companies
to
produce
three-‐dimensional
laser
scanners
which
in
turn,
eventually
made
this
technology

affordable
to
orthodontist
consumers.

A
digital
study
models
is
produced
by
a
method
of
laser
scanning
or
a
combination
of
laser

scanning
and
stereophotogrammetry,
in
order
to
produce
a
three-‐dimensional
representation
of
a
study

model.
The
Motion
View
system
uses
three
lasers
to
scan
a
dental
model
placed
on
a
rotating
platform.

The
proprietary
software
allows
the
practitioner
to
perform
linear
measurements,
Bolton
analysis,
and

space
analysis
on
the
digital
model.
The
validity
and
reproducibility
of
digital
models
have
been

demonstrated
in
clinical
studies.
Mayers
and
co-‐workers
(2005)
found
the
peer
assessment
rating
(PAR)

scores
of
plaster
versus
digital
models
to
be
highly
correlated.
Jooseong
Kim
and
co-‐workers
(2013)

compared
the
plaster
models
to
the
3-‐D
models
scanned
by
the
Ortho
Insight
Motion
View
Scanner
and

found
the
digital
models
had
high
accuracy.
However,
it
was
unclear
whether
three-‐dimensional
models

would
produce
the
same
HLD
results,
which
combines
measurements
with
weighted
scoring.

Denti-‐Cal
Authorization
Process

Denti-‐Cal
allows
an
orthodontic
provider
to
send
initial
second
pour
diagnostic
models
to
its
main

office
in
Sacramento,
for
scoring
by
one
of
Denti-‐Cal
three
calibrated
orthodontic
raters.
These
models

may
be
subjected
to
loss
and/or
physical
damage
during
shipping.
If
the
models
are
lost
or
damaged,
the

orthodontist
can
then
copy
the
retained
original
cast
set
to
be
resent
to
Sacramento.
Denti-‐Cal
does
not

return
the
sent
models.

22

Denti-‐Cal
Raters
Calibration
and
Quality
Management

Denti-‐Cal
quality
management
randomly
draws
a
statistical
amount
of
scored
casts
every
month

from
each
of
the
three
orthodontic
consultants.
The
consultants
are
expected
to
score
98%
accurate

grades
each
month.

If
there
are
any
differences
in
scoring
the
consultant
reviews
the
scoring
with
quality

management
and
can
defend
that
score
of
if
not
accurate,
be
calibrated
on
that
case.
Every
3
months
all

three
orthodontic
consultants
grade
the
same
3
cast
set
and
later
meet
with
quality
management
to

review
and
discuss
any
differences.
This
calibration
insures
all
three
consultants
approach
the
scoring
in

an
accurate
and
uniform
way.

23

Chapter
2:
Objective

In
this
study,
the
use
of
laser-‐scanned
digital
models
for
HLD
scoring
was
compared
to
plaster
models.

This
substitution
could
save
time
and
decrease
the
cost
of
the
authorization
process.
However,
it
is

unknown
whether
digital
and
plaster
models
would
yield
the
same
HLD
scores.

Therefore,
this
study
was

undertaken
to
compare
the
HLD
scores
and
to
analyze
why
digital
and
plaster
models
might
differ
in
HLD

scores.

24

Chapter
3:
Materials
and
Methods

Sample

This
study
compared
two
types
of
study
models:
digital
vs.
plaster
models.
The
plaster
model

group
consisted
of
78
sets
of
plaster
models
obtained
from
the
Children’s
Hospital
Los
Angeles
(CHLA)

dental
department.
Each
set
of
models
consisted
of
upper
and
lower
casts
that
were
trimmed
according

to
a
wax
bite
taken
in
centric
occlusion.
The
digital
model
group
consisted
of
fifty
sets
of
models
taken

from
the
plaster
group
which
were
scanned
using
the
Ortho
Insight
3-‐D
laser
scanner
(Motion
View

Software
LLC,
Hixson,
Tenn.).

The
HLD
Index–California
Modification
(HLD
Cal-‐Mod)
was
used
for
score

the
plaster
and
digital
models.
All
measurements
on
study
casts
were
made
with
a
6
inch
Mitotoyo
digital

caliper
that
is
calibrated
to
an
accuracy
of
0.01mm.

Overjet
measurements
were
made
using
the
back

end
of
the
caliper
which
sticks
of
the
ruler
like
and
extension.

Raters
and
Calibration

Using
the
HLD
Cal-‐Mod,
two
raters
scored
the
two
groups
of
models.
In
addition,
an
official
rater

from
Denti-‐Cal
traveled
from
Sacramento
to
Los
Angeles
in
order
to
score
the
Group
1
plaster
models.
His

scores
were
used
as
the
gold
standard
for
comparison.
Rater
1
was
an
orthodontic
fellow,
and
Rater
2

was
a
third-‐year
postgraduate
orthodontic
resident.
The
Denti-‐Cal
rater
was
Dr.
Frederick
Schubert,
one

of
three
orthodontists
who
score
models
submitted
to
Denti-‐Cal
for
authorization.
He
has
scored
over

70,000
casts
for
the
State
of
California
using
the
HLD
Cal-‐Mod.

Prior
to
calibration,
Rater
1
scored
Group
1
using
the
HLD
Cal-‐Mod.
Dr.
Schubert
trained
and

calibrated
both
Raters
1
and
2
for
one
day.
Denti-‐Cal
provided
a
calibration
DVD
made
by
Dr.
Donald

Poulton.
For
8
hours,
Raters
1
and
2
observed
Dr.
Schubert
scoring
78
sets
of
plaster
models.

There
were

some
unwritten
rules
for
measuring
models
with
digital
calipers
that
were
taught
to
Rater
1
and
2
by
Dr.

25

Schubert
(Table
5).
These
unwritten
instructions
used
by
Denti-‐Cal
were
recorded
with
a
Kodak
digital

camcorder.
After
calibration,
Raters
1
and
2
scored
Group
1
(78
plaster
models)
and
Group
2
(50
digital

models)
using
the
HLD
Cal-‐Mod.
Dr.
Schubert’s
scores
for
Group
1
were
used
for
comparison
in
this
study.

Experimental
Design
and
Data
Collection

The
current
form
of
the
HLD
Cal-‐Mod
includes
13
conditions
that
are
taken
into
consideration
in

the
scoring
process.
The
first
six
(1-‐6A)
(Figure
4)
are
automatically
qualifying
conditions.
If
the
model

displays
that
a
patient
has
one
of
these
conditions,
then
no
further
scoring
would
be
required:
the
patient

would
automatically
qualify
for
receiving
subsidized
treatment.

Figure
4:
HLD
Cal-‐Mod
automatic
qualifying
conditions.

These
automatic
qualifying
conditions
are
as
follows:

1.

Cleft
palate

2.

Craniofacial
anomaly

3.

Deep
impinging
overbite
(when
lower
incisors
are
destroying
the
soft
tissue
of
the
palate)

4.

Anterior
crossbite
when
clinical
attachment
loss
and
recession
of
the
gingival
margin
are
present.

(Figure
5)

5.

Severe
traumatic
deviation
(e.g.,
loss
of
a
premaxilla
segment
by
burns
or
accident
or
as
the
result

of
osteomyelitis
or
other
gross
pathology)

26

6A.
Overjet
greater
than
9mm
or
mandibular
protrusion
(reverse
overjet)
greater
than
3.5mm.
Overjet

is
recorded
with
the
patient’s
teeth
in
centric
occlusion
and
is
measured
from
the
labial
of
the

lower
incisors
to
the
labial
of
the
corresponding
upper
central
incisors.
This
measurement
should

record
the
greatest
distance
between
any
one
upper
central
incisor
and
its
corresponding
lower

central
or
lateral
incisor.
An
automatic
qualifying
condition
exists
when
the
overjet
is
greater
than
9

mm
or
mandibular
protrusion
(reverse
overjet)
is
greater
than
3.5
mm.
(Figures
6
and
7)

Figure
5.
Anterior
cross-‐bite
with
gingival
recession

Figure
6:
Overjet
measurement

27

Figure
7:
Reverse
Overjet
measurement

When
a
cast
has
one
of
the
aforementioned
automatic
qualifying
conditions,
the
rater
indicates
the

condition
on
the
form,
and
no
further
scoring
is
required.
If
none
of
the
automatic
qualifying
conditions

exist
then
conditions
6B
to
13
are
assessed
(Figure
8).

6B.
Overjet
equal
to
or
less
than
9
mm
is
recorded
as
in
condition
6A.
The
measurement
is
rounded

off
to
the
nearest
millimeter
and
entered
on
the
score
sheet
(Figure
6).

7.

Overbite:
A
pencil
mark
on
the
tooth
indicating
the
extent
of
overlap
facilitates
this
measurement.

It
is
measured
to
the
nearest
millimeter
and
entered
on
score
sheet.

8.

Mandibular
protrusion
(reverse
overjet)
less
than
or
equal
to
3.5mm
is
recorded,
as
in
condition

6A.
The
measurement
is
rounded
off
to
the
nearest
millimeter,
recorded
on
score
sheet,
and
then

multiplied
by
five
(Figure
7).

9.

Open
bite:
This
condition
is
defined
as
the
absence
of
occlusal
contact
in
the
anterior
region.
It
is

measured
in
millimeters
from
the
incisal
edge
of
a
maxillary
central
incisor
to
incisal
edge
of
a

corresponding
mandibular
incisor.
The
measurement
is
entered
on
the
score
sheet
and
multiplied

by
four.
(Figure
9)

10.
Ectopic
eruption:
Each
tooth
is
counted,
except
third
molars.
Each
qualifying
tooth
must
be
more

than
50%
blocked
out
of
the
arch.
Only
one
tooth
is
counted
when
two
teeth
are
mutually
blocked

28

out.
The
number
of
qualifying
teeth
is
entered
on
the
score
sheet
and
multiplied
by
three.
If

anterior
crowding
(condition
11)
also
exists
in
the
same
arch,
then
only
the
condition
that
scores

the
most
points
is
scored.
Only
one
of
those
two
conditions
(anterior
crowding
and
ectopic

eruption)
is
scored;
however
posterior
ectopic
teeth
can
still
be
counted
separately
from
anterior

crowding
when
they
occur
in
the
same
arch.

11.
Anterior
crowding
presents
as
arch
length
insufficiency,
which
must
exceed
3.5mm
to
be
scored.

Mild
rotations
that
may
react
favorably
to
stripping
or
mild
expansion
procedures
are
not
to
be

scored
as
crowded.
A
score
of
1
is
given
for
a
crowded
maxillary
arch
and/or
1
for
a
crowded

mandibular
arch.
The
total
is
entered
on
the
score
sheet
and
multiplied
by
five.
If
ectopic
eruption

(condition
10)
exists
in
the
anterior
region
of
the
same
arch,
then
only
the
condition
that
scores

the
most
points
is
scored.
However,
posterior
ectopic
teeth
can
still
be
counted
separately
from

anterior
crowding
when
they
occur
in
the
same
arch.

12.
Labio-‐lingual
spread:
A
Boley
gauge
is
used
to
determine
the
extent
of
deviation
from
a
normal

arch.
When
there
is
only
a
protruded
or
lingually
displaced
anterior
tooth,
the
measurement
is

made
from
the
incisal
edge
of
that
tooth
to
the
normal
arch
line.
Otherwise,
the
total
distance

between
the
most
protruded
anterior
tooth
and
the
most
lingually
displaced
adjacent
anterior

tooth
is
measured.
In
the
event
that
multiple
anterior
crowding
of
teeth
is
observed,
only
the

most
severe
individual
measurement
should
be
entered
on
the
score
sheet.

13.
Posterior
unilateral
crossbite:
This
condition
involves
two
or
more
adjacent
teeth,
one
of
which

must
be
a
molar.
The
crossbite
must
involve
two
or
more
maxillary
posterior
teeth
that
are
palatal

or
buccal
in
relation
to
the
mandibular
posterior
teeth.
The
presence
of
posterior
unilateral

crossbite
is
indicated
by
a
score
of
4
on
the
score
sheet.

29

Figure
8:
HLD
Cal-‐Mod
conditions
6B
through
13.

Figure
9:
Open
bite
measurement

The
cutoff
score
for
the
HLD
index
to
qualify
a
patient
for
treatment
is
26.
When
the
scores
from

conditions
6A
through
13
are
added
and
yield
a
final
score
of
26
or
greater,
the
patient
will
qualify
for

subsidized
treatment.
Conversely,
if
the
total
score
totals
25
or
less,
then
the
patient
will
not
qualify.

Raters
1
and
2
recorded
the
final
scores
on
HLD
form
sheets
and
transferred
the
scores
to
an
Excel

spreadsheet,
where
the
“X”
represents
the
qualifying
models
and
the
“O”
represents
the
non-‐qualifying

models
(Figure
10).

30

Figure
10:
Excel
sheet
used
to
record
the
scores.

Table
5:
Dr.Schubert
Unwritten
Rules
for
using
calipers
for
scoring

Overjet
and
negative
overjet
measurements
are
made
using
the
back
end
of
the

caliper,
which
sticks
out
of
the
ruler
like,
and
extension.
This
extension
should
be

parallel
to
the
occlusal
plane
(Figure
6
and
7).

When
measuring
overbite,
a
mechanical
pencil
is
used
parallel
to
the
occlusal
plane
to

make
a
mark
on
lower
incisors,
then
the
overbite
is
measured
from
the
pencil
mark
to

the
lower
incisor
incisal
edge
(Figures
10,11
and
12).

Labio-‐lingual
spread
is
measured
from
the
labial
surface
of
the
most
labial
to
the
lingual

surface
of
the
most
lingual
anterior
tooth,
with
the
caliper
being
parallel
to
the
occlusal

surface
of
the
arch.
This
measurement
is
made
only
on
one
arch
with
the
highest

measurement
(Figure
13).

Openbite
measurement
is
made
from
contact
area
of
upper
incisors
to
the
contact
area

of
the
lower
incisors,
not
from
incisal
edge
to
incisal
edge
(Figure
9).

Ectopic
eruption
is
only
scored
when
there
are
2
or
more
ectopically
erupted
teeth
per

arch
otherwise
crowding
is
scored
instead
because
it
gives
the
patient
more
points.

A
tooth
must
be
50%
blocked
to
be
considered
ectopically
erupted.

Figure
10:
Mechanical
pencil
used
to
mark
overbite

31

Figure
11:
Overbite
mark

Figure
12:
Overbite
measurement

Figure
13:
Labio-‐lingual
spread
measurement

32

Statistical
Methods

Kappa
statistics
were
used
to
compare
inter-‐
and
intra-‐rater
reliability
of
the
dichotomous

outcome
of
qualifying
versus
non-‐qualifying
for
subsidized
treatment.
Comparisons
included
the

following:

1. Inter-‐rater
reliability
between
gold
standard
versus
Rater
1
plaster
scores
pre-‐calibration.

2. Inter-‐rater
reliability
between
gold
standard
versus
Rater
1
plaster
scores
post-‐calibration.

3. Inter-‐rater
reliability
between
gold
standard
versus
Rater
2
plaster
scores.

4. Inter-‐rater
reliability
between
Raters
1
versus
2
plaster
scores.

5. Inter-‐rater
reliability
between
Raters
1
versus
2
digital
scores.

6. Intra-‐rater
reliability
between
Rater
1
plaster
versus
digital
scores.

7. Intra-‐rater
reliability
between
Rater
2
plaster
versus
digital
model
scores.

Kappa
statistics
were
evaluated
using
the
suggestions
in
Landis
and
Koch
(1977):
<
0
=
poor,
0–0.2

=
slight
agreement,
0.21–0.4
=
fair,
0.41–0.6
=
moderate,
0.61–0.8
=
substantial,
and
0.81–1
=
almost

perfect.

33

Chapter
4:
Results

Descriptive
Statistics:

Dr.
Schubert
scored
78
plaster
models.
Before
calibration,
Rater
1
scored
67
plaster
models.
After

calibration,
Raters
1
and
2
scored
78
plaster
models
and
50
digital
models.

Inter-‐rater
Reliability
comparing
Raters’
Plaster
Model
Scores
to
Dr.
Schubert’s
scores

1. Before
calibration:
Rater
1
plaster
model
scores,
when
compared
to
Dr.
Schubert’s
scores,
showed

agreement
in
47
cases
(70%)
and
disagreement
in
20
cases
(30%)
(Table
6).

2. After
calibration:
Rater
1
plaster
models
scores
(Group
A),
when
compared
to
Dr.
Schubert’s

scores,
showed
agreement
in
74
cases
(95%)
and
disagreement
in
4
cases
(5%)
(Table
7).

3. After
calibration:
Rater
2
plaster
models
scores,
when
compared
to
Dr.
Schubert’s
scores,
showed

agreement
in
73
cases
(93.5%)
and
disagreement
in
5
cases
(6.5%)
(Table
8).

Inter-‐rater
Agreement
between
Raters
1
and
2
HLD
scores

1. Comparison
of
Rater
1
to
Rater
2
plaster
model
scores
for
the
same
group
showed
agreement
in

72
cases
(92%)
and
disagreement
in
6
cases
(8%)
(Table
9).

2. Comparison
of
Rater
1
to
Rater
2
digital
model
scores
for
the
same
group
showed
agreement
in
46

cases
(94%)
and
disagreement
in
3
cases
(6%)
(Table
10).

Intra-‐rater
Reliability
of
Plaster
versus
Digital
HLD
scores

1. Rater
1
plaster
versus
digital
model
scores
showed
agreement
in
45
cases
(90%)
and
disagreement

in
5
cases
(10%)
(Table
11).

2. Rater
2
plaster
versus
digital
model
scores
showed
agreement
in
42
cases
(84%)
and
disagreement

in
8
cases
(16%)
(Table
12).

34

Statistical
Analysis

1. Inter-‐rater
reliability
when
comparing
raters
1
and
2
plaster
models
scores
separately
to
Dr.

Schubert’s
scores.

Before
calibration,
Rater
1
plaster
model
scores
showed
moderate
agreement,

with
a
kappa
value
of
0.41
(Table
6).
After
calibration,
this
value
increased
to
0.92,
showing

perfect
agreement
between
Rater
1
and
Dr.
Schubert
(Table
6).
Rater
2
plaster
model
scores
had

almost
perfect
agreement
compared
to
Dr.
Schubert,
with
a
kappa
value
of
0.86
(Table
8).

2. Inter-‐rater
reliability
(between
Raters
1
and
2)
was
almost
perfect
for
both
plaster
(k
=
0.83)
and

digital
(k
=
0.88)
models
(Tables
9
and
10).

3. Intra-‐rater
reliability
(a
comparison
of
digital
to
plaster
model
scores
of
the
same
rater)
showed

substantial
agreement
for
both
Rater
1
(k
=
0.79)
and
Rater
2
(k
=
0.67)
(Tables
11
and
12).

35

Inter-‐rater
Comparison
of
Raters
1
and
2
Plaster
HLD
scores
versus
Dr.
Schubert’s

Table
6:
Rater
1
pre-‐calibration
plaster
versus
gold
standard
(k
=
.406)

Plaster

Gold
standard
Total

Qualify
No
Yes

Rater
1

before

calibration
No
19
15
34

Yes
5
28
33

Total

24
43
67

Table
7:
Rater
1
post-‐calibration
plaster
versus
gold
standard
(k
=
.917)

Plaster

Gold
standard
Total

Qualify
No
Yes

Rater
1
after

calibration
No
27
3
30

Yes
0
47
47

Total

27
50
77

Table
8:
Rater
2
post-‐calibration
plaster
versus
gold
standard
(k
=
.860)

Plaster

Gold
standard
Total

Qualify
No
Yes

Rater
2,
after

calibration
No
25
2
27

Yes
3
48
51

Total

28
50
78

36

Inter-‐rater
Comparison
of
Rater
1
scores
versus
Rater
2
scores

Table
9:
Plaster
(k
=
.832)

After

calibration

Rater
2
Total

Qualify
No
Yes

Rater
1
No
25
5
30

Yes
1
46
47

Total

26
51
77

Table
10:
Digital
(k
=
.880)

After

calibration

Rater
2
Total

Qualify
No
Yes

Rater
1
No
23
1
24

Yes
2
24
26

Total

25
25
50

Intra-‐rater
Reliability
of
Plaster
versus
Digital
Model
Scores

Table
11:
Rater
1
after
calibration
(k
=
.794)

After

calibration

Digital
Total

Qualify
No
Yes

Plaster
No
19
1
20

Yes
4
25
27

Total

23
26
49

Table
12:

Rater
2
after
calibration
(k
=
.671)

After

calibration

Digital
Total

Qualify
No
Yes

Plaster
No
16
0
16

Yes
8
26
34

Total

24
26
50

37

Chapter
5:
Discussion

The
objective
of
this
study
was
to
compare
the
HLD
scores
obtained
through
the
use
of
digital

versus
plaster
models
in
order
to
determine
whether
digital
models
could
substitute
for
plaster
models
by

Denti-‐Cal
or
by
any
agency
using
the
HLD
index
as
a
treatment
need
index.
This
study
compared
HLD

scores
of
digital
versus
plaster
models
for
each
rater
and
for
each
of
the
scores
against
the
Denti-‐Cal

scores
presented
by
Dr.
Schubert,
the
Denti-‐Cal
consultant.
Kappa
statistics
were
used
to
compare
inter-‐

and
intra-‐rater
agreement
of
the
dichotomous
outcome
of
qualifying
for
subsidized
treatment.

Before
calibration,
the
comparison
of
Rater
1
to
the
Denti-‐Cal
rater
scores
showed
a
kappa
value

of
0.41.
This
value
increased
to
0.92
after
calibration.
Also,
after
calibration,
the
plaster
model
scores
for

Rater
2
when
compared
to
Denti-‐Cal
scores
showed
a
kappa
value
of
0.86.
Kappa
statistics
for
intra-‐rater

comparisons
(plaster
versus
digital
model
scores
of
the
same
rater)
were
0.79
for
Rater
1
and
0.67
for

Rater
2.
Kappa
statistics
for
inter-‐rater
comparisons
(Rater
1
versus
Rater
2
scores)
was
0.83
for
plaster

models
and
0.88
for
digital
models.

According
to
Landis
and
Koch
(1977),
the
following
guidelines
are
used
to
interpret
the
kappa

scores:

k
<
0
=
poor
agreement,
0–0.2
=
slight
agreement,
0.21–0.4
=
fair,
0.41–0.6
=
moderate,
0.61–0.8

=
substantial,
0.81–1
=
almost
perfect.
Using
these
guidelines,
one
could
interpret
that
post-‐calibration,

there
is
almost
perfect
agreement
between
the
plaster
scores
of
both
raters
and
the
Denti-‐Cal
scores.

During
the
calibration
session,
many
unwritten
methods
for
measuring
the
models
with
digital
calipers

were
learned
by
Raters
1
and
2,
which
led
to
the
increase
in
agreement
after
calibration.
There
was

substantial
agreement
between
the
digital
scores
of
both
raters
and
the
Denti-‐Cal
scores.
Similarly,
there

was
substantial
intra-‐rater
agreement
between
the
digital
and
plaster
scores
of
the
same
rater.
There
was

almost
perfect
inter-‐rater
agreement
between
Raters
1
and
2.

38

In
the
literature,
plaster
models
and
laser-‐scanned
models
have
been
compared
for
other
occlusal

indices.

Paul
Major
(2006)
compared
plaster
versus
digital
model
for
PAR
scores
and
concluded
that

digital
models
are
clinically
acceptable
replacement
to
plaster
models.

Efstratiadis
et
al.
(2005)
concluded

that
1)
digital
models
were
a
viable
alternative
to
plaster
models,
2)
digital
models
had
slightly
higher

inter-‐rater
reliability
than
the
digital
models,
3)
software
improvements
were
needed
for
more
accurate

measurements
on
digital
models
and
4)
calibration
of
raters
was
critical.
Mayers
et
al.
research
(2005),

which
concluded
that
there
were
no
significant
differences
between
PAR
scores
of
digital
versus
plaster

models.
Our
findings
with
the
HLD
index
generally
agrees
with
these
studies
in
terms
of
accuracy
of
model

representation
but
may
differ
in
the
final
outcomes
on
who
qualifies
for
treatment
or
not.

In
terms
of
assessing
the
need
for
prior
calibration,
the
kappa
analysis
showed
a
significant

difference
when
comparing
the
agreement
between
Rater
1
and
the
Dr.
Schubert’s
scores,
both
before

and
after
calibration.
The
increase
of
kappa
value
from
0.4
pre-‐calibration
to
0.9
post-‐calibration

confirmed
the
importance
of
calibration,
which
was
reported
in
previous
studies
(Richmond
1993,

Efstratiadis
2005).
The
protocol
for
calibration
in
this
study,
which
used
methods
learned
from
Denti-‐Cal,

could
be
the
basis
for
a
course
to
teach
orthodontists
how
to
score
the
HLD
Index.

An
important
part
of
the
study
was
to
determine
where
differences
might,
occur
when
scoring

digital
versus
plaster
models.

For
Rater
1,
five
cases
out
of
50
yielded
different
digital
and
plaster
scores,

such
that
the
cases
qualified
for
treatment
as
plaster
models
but
not
as
digital
models.
For
Rater
2,
six

cases
out
of
50
yielded
different
digital
and
plaster
scores.

It
should
be
noted
that
five
of
these
six
cases

for
Rater
2
were
the
same
cases
as
Rater
1.

When
these
cases
were
reviewed,
four
sources
of
error
were
identified
that
led
to
differences

between
digital
and
plaster
models
scores.
The
first
source
of
error
was
dental
crowding,
when
crowding

was
not
recorded
in
one
or
both
arches
in
the
digital
version.
The
HLD
Cal-‐Mod
scores
crowding
as
“all-‐or-‐
none”.

Crowding
greater
than
3.5
mm
yields
in
a
score
of
5
points
per
arch.
Depending
on
whether

39

dental
crowding
was
scored
as
present
or
not,
in
borderline-‐crowding
cases,
there
could
be
a
5-‐
to
10-‐
point
difference
in
HLD
scores,
which
would
alter
the
final
decision.
The
second
source
of
error
was
from

the
weighted
categories
for
open
bite
and
anterior
crossbite.
For
an
open
bite,
a
difference
of
1mm

would
be
multiplied
by
four.
For
an
anterior
crossbite,
a
difference
of
1mm
would
be
multiplied
by
five.

Therefore,
seemingly
small
differences
in
measurement
may
lead
to
a
4-‐
to
5-‐fold
magnification
of
error.

The
third
source
of
error
was
due
to
combinations
of
measurements
for
overjet,
overbite,
and
labio-‐
lingual
spread.
For
example,
a
1
mm
difference
in
overjet,
overbite,
and
labio-‐lingual
spread
would
add
up

to
3
points.
According
to
HLD
Cal-‐Mod
protocols,
the
measurements
are
approximated
to
the
nearest

millimeter.
If
an
overjet
or
overbite
was
measured
as
1.4
instead
of
1.5,
then
this
0.10
mm
difference

could
alter
the
score
by
1
point,
depending
on
whether
the
measurement
was
rounded
up
or
down.
The

fourth
source
of
error
was
failure
to
identify
automatically
qualifying
conditions,
such
as
the
anterior

crossbite
with
clinical
attachment
loss
and
gingival
recession.
Failure
to
identify
an
automatically

qualifying
condition
might
be
due
to
inadequate
models,
inadequate
scanning
of
the
gingiva
and
soft

tissue
or
to
failure
of
the
examiner
to
identify
the
condition
on
the
scanned
model.

Our
study
found
that

borderline
cases
that
scored
close
to
the
cutoff
point
of
26
could
be
troublesome
and
show
differences

between
digital
versus
plaster
model
qualification
outcomes
as
marginally-‐small
errors
could
be
amplified

and
lead
to
changes
in
overall
point
total,
altering
the
final
decision.

During
this
study,
we
recognized
that
software
limitations
and
human
error
also
represented
two

potential
problems
that
could
lead
to
differences
between
plaster
and
digital
scores.
Keating
et
al.
(2008)

reported
a
range
of
differences
between
the
plaster
and
digital
model
measurements,
ranging
from
0.10

to
0.19
mm.
Lagrave`re
et
al.
(2008)
reported
that
the
mean
difference
between
digital
and
plaster

models
was
≤
0.23
±
0.169
mm.
In
this
study,
there
were
limitations
to
how
the
Motion
View
software

recorded
overjet.
Motion
View
software
used
a
bisecting
tool
that
split
the
model
in
half
along
the
sagittal

plane,
clearing
the
right
half
of
the
model
while
leaving
the
left
half
(Figures
14
and
15).
If
maximal

40

overjet
was
on
the
right
side,
then
this
measurement
would
be
missed,
as
the
right
side
would
be
cleared

from
the
screen.
To
correct
for
this
limitation,
in
cases
where
the
maximum
overjet
was
on
the
right,
the

bisecting
tool
was
not
used,
and
the
virtual
caliper
was
used
instead
(Figure
16).
However,
the

introduction
of
two
dissimilar
tools
for
measuring
overjet
can
result
in
scoring
differences.
This
problem

can
be
avoided
by
either
improving
the
software
to
provide
options
for
excluding
the
right
or
left
side
of

the
models,
or
by
using
other
types
of
software,
do
not
have
this
problem.

Figure
14:
Ortho
Insight
bisecting
tool
and
caliper
to
measure
overjet.

41

Figure
15:
Ortho
Insight
bisecting
tool
with
a
grid
ruler
to
measure
overjet.

Figure
16:
Caliper
tool
used
to
measure
overjet
when
larger
overjet
is
on
right
side.

Another
potential
problem
is
the
loss
of
detail
in
digital
images,
which
could
make
identifying

landmarks
difficult
(Houston
1983).

For
example,
gingival
attachment
loss
might
be
missed
because
parts

of
the
gingival
margins
were
lost
in
the
scan,
or
the
image
resolution
was
insufficient
for
showing
the

differences
in
gingival
contour.

In
these
cases,
additional
intraoral
color
photographs
would
be
helpful,

42

especially
as
these
conditions
may
lead
to
automatic
qualification
based
on
the
presence
of
soft
tissue

defects.

The
caliper
tool
in
the
Motion
View
software
is
problematic.
It
allows
the
rater
to
obtain
linear

measurements
by
clicking
on
any
two
points
in
space.
When
this
tool
is
used
to
measure
dental
crowding

and
labio-‐lingual
deviation
(Figures
17
and
18),
it
is
important
to
take
the
missing
third
dimension
into

consideration.
The
rater
can
only
view
two
dimensions
on
the
screen,
when
clicking
on
the
two
points.

Thus,
when
making
these
clicks,
the
points
should
be
on
the
same
third
plane,
otherwise
the
points
will

be
farther
apart
than
in
the
same
plane,
leading
to
a
larger
measurement
value.
This
difference
in

perspective,
when
measuring
differences
between
two
points
in
space,
can
lead
to
different
linear

measurements.
Zilberman
et
al.
(2003)
showed
that
landmark
identification
was
more
difficult
when

viewing
a
three-‐dimensional
object
on
a
two-‐dimensional
screen.
Another
potential
source
of
error
is
the

scanning
of
the
bite
registration,
which
could
lead
to
changes
in
bite
opening
and
overjet
when
the
digital

maxillary
and
mandibular
dentitions
are
placed
into
occlusion.
A
possible
solution
would
be
to
first
trim

the
plaster
models
with
the
wax
bite
registration
so
that
the
backs
of
the
models
could
define
the
bite

and
then
scan
the
bite.
The
digital
bite
registration
could
be
checked
against
the
backs
of
the
model.

Additional
bite
registrations
could
also
confirm
the
accuracy
of
the
bite
registration.
When
the
casts
are

trimmed
in
the
correct
bite
before
scanning,
the
Motion
View
software
has
the
capability
to
align
the

models
in
the
correct
bite
automatically,
without
the
need
for
scanning
the
bite
registration.

43

Figure
17:
Measuring
crowding
using
the
digital
caliper.

Figure
18:
Measuring
labio-‐lingual
spread
using
the
digital
caliper.

This
study
examined
50
digital
models.
It
is
possible
that
a
larger
study
might
uncover
more
errors

that
could
lead
to
differences
between
digital
and
plaster
HLD
scores.
With
larger
numbers,
certain

patterns
in
errors
might
be
detected
and
help
with
future
corrections
of
the
software
design.

The

statistical
data
for
this
study
was
dichotomous
as
the
final
outcome
measure
for
this
study
was
whether

or
not
a
particular
set
of
models
scored
high
enough
to
qualify
for
treatment,
a
dichotomous
outcome.

44

Yet,
some
of
the
measurements
were
linear
which
could
be
separately
compared
as
continuous
variables.

Further
studies
could
compare
individual
measurements
used
in
the
HLD
index
and
examine
how
the

linear
measurements
differed
between
digital
and
plaster
models.

One
of
the
reasons
for
conducting
this
study
is
based
on
recognizing
potential
benefits
of
using

digital
technology
for
processing
Denti-‐Cal
authorization
claims.

For
California
and
the
states
that
use
the

HLD
Index,
digital
models
can
potentially
save
costs
to
the
Denti-‐Cal
program
and
to
orthodontists.

For

example,
when
cases
are
rejected
by
Denti-‐Cal,
the
models
are
stored
for
twelve
months
in
case
there
are

provider
requests
for
a
rescore.
Rejected
cases
represent
40%
of
all
the
submitted
cases
which
means

tens
of
thousands
of
models
are
placed
into
storage
facilities
by
the
State
of
California.
The
cost
of
storing

rejected
models
could
be
replaced
by
digitizing
the
models
for
long-‐term
recall.
Since
the
models
can
be

stored
as
an
STL
document
(Standard
Tessellation
Language)
on
computers,
these
digital
models
can
be

stored
on
DVDs,
memory
cards,
external
hard
drives,
large
servers
and
cloud-‐based
storage
systems.
The

savings
to
the
state
would
be
the
cost
for
storing
the
models.
The
savings
to
the
orthodontist
would

include
the
cost
of
the
impressions,
plaster,
boxing
and
shipping
of
the
models,
as
well
as
costs
associated

with
retaking
impression
due
to
loss
or
breakage.
Digital
models
may
also
improve
the
efficiency
of
the

authorization
process,
resulting
in
quicker
turnaround
time
for
authorization
decisions
since
digital

models
can
be
moved,
saved
and
accessed
via
the
Internet
by
both
orthodontists
and
Denti-‐Cal

consultants.
Denti-‐Cal
consultants
can
view
the
models
from
any
location
at
any
time
making
the
work

conditions
more
flexible
and
the
communication
of
information
instantaneous.

The
chances
for
model

breakage
or
loss
are
eliminated.

Dr.
Noel
from
the
Department
of
Health
Care
Services
expressed
that
fraud
is
a
major
concern
for

Denti-‐Cal.

At
present,
it
is
difficult
to
assess
whether
fraud
could
be
prevented
by
using
digital

technology;
however,
additional
photographic
data
can
be
useful
for
corroborating
the
identity
and

integrity
of
the
model
data,
whether
the
models
are
plaster
or
digital.

Digital
models
would
have
to
be

45

tamper-‐proof
like
electronic
medical
records.

If
the
digital
platform
is
used,
then
data
recognition

technology
could
be
incorporated
into
the
software
to
prevent
double
submission
of
records.
Commercial

companies
like
e-‐models
(Falcon
Heights,
MN)
and
OrthoCad/itero
(San
Jose,
CA)
might
be
able
to
provide

both
the
digitizing
service
and
security
for
Denti-‐Cal
submissions.

Although
our
data
suggests
that
laser
scanning
can
produce
an
accurate
replica
of
a
plaster
model,

this
study
does
not
support
use
of
digital
models
for
HLD
Cal-‐Mod
scoring
by
Denti-‐Cal
at
this
time.

On

the
one
hand,
the
intra-‐rater
reliability
between
plaster
and
digital
models
is
substantially
high
(k
=
0.73)

and
the
inter-‐rater
reliability
is
higher
than
the
plaster
model
scores,
which
would
favor
the
use
of
digital

models.
But
on
the
other
hand,
the
disagreement
in
HLD
scores
between
digital
and
plaster
scores
was
as

high
as
13%,
which
may
be
too
high
for
Denti-‐Cal
to
accept
as
these
differences
could
potentially
lead
to

additional
lawsuits.
This
study
recommends
that
software
changes
be
implemented
to
correct
the

problems
identified
in
this
thesis
before
it
can
be
used
by
Denti-‐Cal.

An
HLD-‐specific
software
rather
than

proprietary
scanner
software
might
be
co-‐developed
between
the
software
companies
and
Denti-‐Cal

since
both
parties
would
have
a
vested
interest
in
developing
this
technology.

Finally,
it
is
important
to
consider
who
would
pay
for
the
cost
of
using
this
new
laser
scanning

technology,
which
is
still
expensive,
and
time
consuming.
In
today’s
market,
a
laser
scanner
costs

US
$18,000-‐23,000.
It
is
unlikely
that
all
orthodontists
in
California
will
own
a
laser
scanner
when
plaster

models
are
the
current
standard
of
care.

However,
the
cost
of
making
a
single
digital
model
when
the

service
is
provided
by
a
commercial
company,
like
e-‐model
or
OrthoCad,
is
approximately
the
same
as
the

laboratory
fees
for
producing
a
plaster
orthodontic
model
from
dental
impressions.

The
cost
of
the

technology,
if
adopted
by
Denti-‐Cal,
could
be
affordable
if
the
digital
models
are
made
through

commercial
vendors.
These
vendors
could
also
provide
the
quality
control
and
anti-‐fraud
protection

needed
for
Denti-‐Cal
submissions.
Software
changes
for
HLD
applications
could
be
made
by
these
vendors

in
consultation
with
Denti-‐Cal
so
that
these
companies
would
become
designated
centers
for
rendering

46

digital
models
from
alginate
impressions
in
sealed
bags
or
plaster
models.
Any
mailed
plaster
models

would
be
returned
to
the
orthodontist
by
the
digitizing
vendors;
thus,
avoiding
the
need
to
duplicate

models,
one
for
Denti-‐Cal
and
one
for
the
orthodontist.
In
other
words,
there
would
be
no
need
for
Denti-‐
Cal
to
hold
on
to
the
models
needed
for
authorization
and
therefore,
only
one
set
of
diagnostic
casts

would
be
made
for
the
orthodontist’s
own
records.

In
conclusion,
the
digital
model
technology
represents
a
promising
alternative
to
plaster
models

for
use
with
the
HLD
index.
It
is
anticipated
that
software
specific
to
the
HLD
index
will
be
developed
in

order
to
take
advantage
of
the
conveniences
brought
by
digital
technology.

47

Chapter
6:
Conclusion

1-‐ There
was
substantial
intra-‐rater
agreement
between
HLD
Cal-‐Mod
digital
and
plaster
model

scores.

2-‐ Differences
in
HLD
Cal-‐Mod
scores
between
digital
and
plaster
models
were
as
high
as
13%,
and

the
combined
kappa
for
intra-‐rater
agreement
was
0.73.

3-‐ HLD
specific
software
is
needed
to
correct
the
sources
of
difference
in
HLD
score
and
include

features
to
prevent
fraud
like
data
recognition.

4-‐ Digital
models
present
a
promising
alternative
to
plaster
models
for
use
with
the
HLD
Index,
in
the

near
future

5-‐ Effectively
training
and
calibrating
raters
on
proper
usage
of
Motion
View
software
and
its
tools
is

recommended,
prior
to
HLD
digital
scoring.

48

References

1. Ackerman
JL,
Proffit
WR.
(1969).
The
characteristics
of
malocclusion:
a
modern
approach
to

classification
and
diagnosis.
Am
J
Orthod
56,
443-‐454.

2. Angle
EH.
(1900).
Treatment
of
malocclusion
of
the
teeth
and
fractures
of
the
maxillae.
Angle's

System
6.
Philadelphia:
SS
White
Dental
Mfg
Co

3. Ballard
CF;
Wayman
JB.
(1964).
A
report
on
a
survey
of
the
orthodontic
requirements
of
310
army

apprentices.
Transactions
of
the
British
Society
for
the
Study
of
Orthodontics
86,
81-‐86.

4. Baume
LJ,
Horowitz
HS,
Summers
CJ,
Dirks
BO,
Brown
WAB;
Carlos
JP.
(1973).
A
method
for

measuring
occlusal
traits.
Int
Dent
J
23,
530-‐537

5. Bjork
A,
Krebs
A,
Solow
B.
(1964)
A
method
for
the
epidemiological
registration
of
malocclusion.

Acta
Odontol
Scand
22,
27-‐41.

6. Brook
PH,
Shaw
WC.
(1989).
The
development
of
an
index
of
orthodontic
treatment
priority.
Eur
J

Orthod
20,
309-‐320.

7. Case
CS.
(1963).
Techniques
and
principles
of
dental
orthopedia.
New
York:
Leo
Bruder
,16-‐18

8. Cons
NC,
Jenny
J,
Kohout
FJ.
(1986).
DAI:
The
dental
aesthetic
index.
Iowa
City:
University
of
Iowa.

9. Daniels
C,
Richmond
S.
(2000).
The
development
of
the
index
of
complexity,
outcome
and
need

(ICON).
J
Orthod
27,
149-‐162.

10. Denti-‐Cal
provider
hand
book
[Internet].
California;
Denti-‐Cal;
February
2014.
Available
from:

http://www.denti-‐cal.ca.gov/provsrvcs/manuals/handbook2/handbook.pdf

11. Draker
HL.
(1960).
Handicapping
labio-‐lingual
deviations:
a
proposed
index
for
public
health

purposes.
Am
J
Orthod
46,
295-‐305.

12. Efstratiadis
S,
Costalos
PA,
Sarraf
K,
Cangialosi
TJ.
(2005).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
128(5),
624-‐29.

13. Grainger
RM.
(1967).
Orthodontic
treatment
priority
index
(Public
Health
Service
Publication
No.

1000).
Washington
DC:
US
Government
Printing
Office.

14. Gravely
JF
and
Johnson
DB.
(April
1974).
Angle's
classification
of
malocclusion:
an
assessment
of

reliability.
J
Orthod
1,
79-‐86

15. Houston
WJ.
(1983).
The
analysis
of
errors
in
orthodontic
measurements.
Am
J
Orthod
83,
382-‐90.

16. Jamison
HC,
McMillan
RS.
(1950).
An
index
of
malocclusion
for
use
in
multiphasic
screening
and

epidemiological
investigations.
University
of
Michigan
School
Public
Health.

49

17. Jenny
J,
Cons
NC,
Kohout
FJ,
Frazier
PJ.
(1980).
Test
of
a
method
to
determine
socially
acceptable

occlusal
conditions.
Community
Dent
Oral
Epidemiol
8,
424-‐433.

18. Jenny
J,
Cons
NC.
(1996).
Comparing
and
contrasting
two
orthodontic
indices:
the
index
of

orthodontic
treatment
need
and
the
dental
aesthetic
index.
Am
J
Orthod
Dentofacial
Orthop
110,

410-‐416.

19. Jooseong
Kim,
Giseon
Heo,
Manuel
O
Lagravère.
(2013).
Accuracy
of
laser-‐scanned
models

compared
to
plaster
models
and
cone-‐beam
computed
tomography.
The
Angle
Orthodontist;
In-‐
Press.
doi:
http://dx.doi.org/10.2319/051213-‐365.1

20. Keating
AP,
Knox
J,
Bibb
R,
Zhurov
AI.
(2008).
A
comparison
of
plaster,
digital
and
reconstructed

study
model
accuracy.
J
Orthod
35,
191-‐201.

21. Major
PW,
Heo
G,
Raboud
DW,
Nebbe
B,
Flores-‐Mir
C,
Stevens
DR,
Landis
JR,
Koch
GG.
(June
2006)

Validity,
reliability,
and
reproducibility
of
plaster
vs.
digital
study
models:
comparison
of
peer

assessment
rating
and
Bolton
analysis
and
their
constituent
measurements.
Am
J
Orthod

Dentofacial
Orthop
129(6),
794-‐803.

22. Landis
JR.,
Koch
GG.
(1977).
The
measurement
of
observer
agreement
for
categorical
data.

Biometrics
33,
159-‐174.

23. Leifert
MF,
Leifert
MM,
Efstratiadis
SS,
Cangialosi
TJ.
(2009).
Comparison
of
space
analysis

evaluations
with
digital
models
and
plaster
dental
casts.
Am
J
Orthod
Dentofacial
Orthop
136(16),

1-‐4.

24. Linder-‐Aronson
S.(
1974).
Orthodontics
in
the
Swedish
public
dental
health
service.
Transactions
of

the
European
Orthodontic
Society,
233-‐240.

25. Linder-‐Aronson
S,
Fridh
G,
Jensen
R.
(1976).
Need
of
orthodontic
treatment
and
orthodontic

specialists
in
Sweden.
Swed
Dent
J
68,
383-‐402.

26. Little
RM.
(1975).
The
irregularity
index.
Am
J
Orthod
68,
554-‐
563.

27. Llewellyn
SK,
Hamdan
AM,
Rock
WP.
(2007).
An
index
of
orthodontic
treatment
complexity.
Eur
J

Orthod
29,
186-‐92.

28. Massler
M,
Frankel
J.
(1951).
Prevalence
of
malocclusion
in
children
aged
14
to
18
years.
Am
J

Orthod
Dentofacial
Orthop
37,
751-‐768.

29. Mayers
M,
Firestone
AR,
Rashid
R,
Katherine
WL
Vig.
(2005).
Comparison
of
peer
assessment

rating
(PAR)
index
scores
of
plaster
and
computer-‐based
digital
models.
Am
J
Orthod
Dentofacial

Orthop
128(4),
431-‐34.

30. Parker
WS.
(1998).
The
HLD
(Cal-‐Mod)
Index
and
the
index
question.
Am
J
Orthod
Dentofacial

Orthop
114,
134-‐141.

31. Parker
WS.
(1999).
A
study
of
1000
malocclusions
selected
by
the
HLD
(CalMod)
Index.
Am
J

Orthod
Dentofacial
Orthop
115,
343-‐351.

50

32. Poulton
DR;
Aaronson
SA.
(1961).
The
relationship
between
occlusion
and
periodontal
Status.
Am

J
Orthod
47,
690-‐699

33. Richmond
S,
Shaw
WC,
O’Brien
KD,
Buchanan
IB,
Jones
R,
Stephens
CD,
Roberts
CT,
Andrews
M.

(1992).
The
development
of
the
PAR
Index
(Peer
Assessment
Rating):
reliability
and
validity.
Eur
J

Orthod
14,
125-‐139.

34. Richmond
S,
Turbill
E.
(1993).
Calibration
o
f
Dental
and
non-‐dental
personnel
in
the
use
of
the

PAR
index.
Br
J
Orthod
20,
231-‐234.

35. Salzmann
JA.
(1968)
Handicapping
malocclusion
assessment
to
establish
treatment
priority.
Am
J

Orthod
Dentofacial
Orthop
54,
749-‐765.

36. Santoro
M,
Galkin
S,
Teredesai
M,
Nicolay
O,
Cangialosi
TJ.
(2003).
Comparison
of
measurements

made
on
digital
and
plaster
models.
Am
J
Orthod
Dentofacial
Orthop
124,
101-‐105.

37. Shaw
WC,
Richmond
S,
O’Brien
KD.
(1995).
The
use
of
occlusal
indices:
a
European
perspective.

Am
J
Orthod
Dentofacial
Orthop
107,
1-‐10.

38. Shaw,
WC,
Richmond,
S,
O’Brien,
K,
Brook,
P,
Stephens,
C.
(1991).
Quality
control
in
orthodontics:

indices
of
treatment
need
and
treatment
standards.
Br
Dent
J
170,
107-‐112.

39. Summers
CJ.
(1971).
The
occlusal
index:
a
system
for
identifying
and
scoring
occlusal
disorders.

Am
J
Orthod
Dentofacial
Orthop
59,
552-‐567.

40. Swedish
Medical
Health
Board
(1966);
Socialstyrelsen
Kungl
Medicinalstyrelsens
circular
den
21

februari
angående
anvisningarna
för
journalföring
inom
folktandvårdens
tandregleringsvård.

Stockholm.

41. Tang
E,
Lisa
LY.
(1993).
A
comparative
study
using
the
occlusal
index
and
the
index
of
orthodontic

treatment
need.
Angle
Orthod
62,
57-‐64.

42. Van
Kirk
LK
and
Pennell
EH.
(1959).
Assessment
of
malocclusion
in
population
groups.
Am
J
Orthod

45,
752-‐758.

43. World
Health
Organization.
An
international
methodology
for
epidemiological
studies
of
oral

disease.
(1966).
Manual
No.
5:
Epidemiological
studies
of
periodontal
disease.
First
draft.
Geneva:

World
Health
Organization.

44. Zilberman
O,
Huggare
J,
Parikakis
K.
(2003).
Evaluation
of
the
validity
of
tooth
size
and
arch
width

measurements
using
conventional
and
three-‐dimensional
virtual
orthodontic
models.
Angle

Orthod
73,
301-‐306.

Bibliography:

Dr.
Barry
Dugger,
D.M.D,
Chief
Dental
Consultant,
State
Government
Programs,
Denti-‐Cal,
Delta
Dental

Plan
of
California.
(2002).
HLD
Calibration
Instruction
DVD.

Asset Metadata

Creator Youssef, Hany (author)

Core Title Comparison of HLD CAL-MOD scores obtained from digital versus plaster models

Contributor Electronically uploaded by the author (provenance)

School School of Dentistry

Degree Master of Science

Degree Program Craniofacial Biology

Publication Date 04/11/2014

Defense Date 03/17/2014

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag digital models,HLD,laser scanner,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Advisor Yen, Stephen (committee chair), Grauer, Dan (committee member), Paine, Michael L. (committee member), Sameshima, Glenn T. (committee member)

Creator Email hsyousse@gmail.com,hsyousse@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-375879

Unique identifier UC11296087

Identifier etd-YoussefHan-2345.pdf (filename),usctheses-c3-375879 (legacy record id)

Legacy Identifier etd-YoussefHan-2345.pdf

Dmrecord 375879

Document Type Thesis

Format application/pdf (imt)

Rights Youssef, Hany

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Abstract (if available)

Abstract Objective: The Handicapping Labio‐lingual Deviation Index (HLD) is an orthodontic treatment need index, which is used to assess the severity of a malocclusion. States such as California use the HLD as a screening tool to determine whether patients qualify for subsidized treatment. The objective of this study was to compare HLD scores using plaster versus digital models. ❧ Methods: Seventy‐eight duplicate study models that were sent from Children’s Hospital Los Angeles (CHLA) to Denti‐Cal for evaluation were scored by an orthodontic fellow (rater 1), a senior orthodontic resident (rater 2) and by the Denti‐Cal orthodontic consultant. The scores of a Denti‐Cal consultant were considered the gold standard in this study. Fifty out of these models were then scanned using the 3D Ortho Insight laser scanner. Rater 1 and rater 2 scored the 50 digital models. Kappa statistics were used to compare inter‐ and intra‐rater reliability. Comparisons included (a) inter‐rater reliability between plaster model scores of both raters versus the gold standard, (b) inter‐rater reliability between rater 1 versus rater 2 for both plaster and digital model scores and (c) intra‐rater reliability between digital and plaster model scores for each rater. ❧ Results: Inter‐rater reliability of Raters 1 and 2 (post‐calibration) versus gold standard was almost perfect (combined K= 0.89). Inter‐rater reliability of rater 1 versus rater 2 was almost perfect for both plaster and digital model scores (combined K= 0.85). Intra‐rater reliability was substantial for both Rater 1 (K= 0.79) and Rater 2 (K = 0.67). ❧ Conclusion: There was substantial intra‐rater agreement between HLD Cal‐Mod digital and plaster model scores. Differences in HLD Cal‐Mod scores between digital and plaster models were as high as 13%, and the combined kappa for intra‐rater agreement was 0.73. HLD specific software is needed to correct the sources of difference in HLD score. Digital models present a promising alternative to plaster models for use with the HLD Index. Training and calibrating raters on proper usage of digital model software and its tools are recommended, prior to HLD digital scoring.