C O V E R S T O R Y
Background. Dentists play a critical role
Knowledge of oral
cancer risk factors and
diagnostic concepts
among North Carolina
dentists
Implications for diagnosis and
referral
LAUREN L. PATTON, D.D.S.; JOHN R. ELTER,
D.M.D., Ph.D.; JANET H. SOUTHERLAND, D.D.S.,
M.P.H., Ph.D.; RONALD P. STRAUSS, D.M.D., Ph.D.
early 30,000 new cases of oral and pharyn-
N
geal cancer (OPC) are diagnosed each year,
and five-year survival rates are dramatically
improved when cases are diagnosed in local-
ized rather than in distant stages (81 percent
versus 30 percent).1
Dentists play a critical role in the early diagnosis of
OPC and generally recognize that this is part of their
professional responsibilities.2 Consistently performing
thorough oral cancer screening exami-
nations for all patients (including high-
More education risk tongue and floor-of-mouth oral
is needed to sites), paying careful attention to suspi-
enhance dental cious red, white and ulcerative mucosal
professionals’ lesions, and being aware of patients’
high-risk tobacco- and alcohol-use
knowledge
behaviors offer the best potential for
of oral and dentists to detect oral cancer at an
pharyngeal early stage.3
cancer risk
BACKGROUND
factors and
4
diagnostic In July 1995, Yellowitz and colleagues
conducted a random mail survey of gen-
concepts.
eral dentists in the United States to
determine their knowledge of, and opin-
ions about, oral cancer. They found deficits in knowledge
of location and appearance of cancerous and precan-
cerous lesions, with recent dental school graduates
having the highest odds of scoring high on knowledge
602 JADA, Vol. 136 www.ada.org/goto/jada May 2005
Copyright ©2005 American Dental Association. All rights reserved.
ABSTRACT
in the early detection of oral
and pharyngeal cancer
(OPC). The authors admin- ✷
J
A D A

✷
istered a survey that
N
CON
IO
assessed the level of knowl-
T
T
A
edge among dentists N
I
U C
U
regarding risk factors and A ING ED
RT 2
diagnostic concepts. I C LE
Methods. In 2002, the authors mailed a
38-item, pretested survey to a random sample
of 1,115 licensed dentists practicing in North
Carolina. Three-level (low, medium, high)
composite index scores for knowledge of risk
factors and diagnostic concepts were created
using previously developed scales. The
authors formulated multivariable models for
risk factor and diagnostic knowledge indexes.
Results. Of the 584 respondents, only 181
(31 percent) had consistent medium-to-high
levels of knowledge on both highly correlated
indexes. Dentists who had higher risk factor
and diagnostic knowledge scores were signifi-
cantly (P < .05) more likely to have heard of
one or more diagnostic aids (odds ratio [OR],
2.7), to have graduated from dental school
within the previous 20 years (OR, 1.8) and to
have performed biopsies or referred five or
more patients with suspicious lesions per
year (OR, 1.7 and 1.5, respectively) than were
less-knowledgeable respondents.
Conclusions. More education is needed in
dental schools, postgraduate programs and
continuing education programs to enhance
dental professionals’ knowledge of OPC risk
factors and diagnostic concepts. Such pro-
grams should include information about
adjunctive diagnostic aids.
Practice Implications. Greater knowl-
edge of risk factors and diagnostic concepts
may result in more frequent patient referrals,
biopsy procedures or both, thus aiding in the
early diagnosis and treatment of patients
with OPC.
Key Words. Oral and pharyngeal cancer;
risk factors; mouth neoplasm; early detection.

indexes. Many dentists indicated an interest in
receiving continuing education on this topic. Since
this dentist survey was administered, adjunctive
techniques have become available to enhance the
early diagnosis of oral cancer5 and have been an
impetus behind public educational campaigns and
a renewed professional interest in the early detec-
tion of this disease.6
With an age-adjusted oral cancer mortality
rate that is higher than the national average,7
North Carolina needs to improve its efforts at
controlling this often disfiguring disease. High
mortality rates need to be addressed in a collabo-
rative, community-based approach that involves
the general public, policy-makers, community
leaders and health care providers. Our study is
one component of a North Carolina needs assess-
ment project aimed at developing a state-focused
model for improved oral cancer control. It
addresses dental care providers’ readiness to
improve early detection efforts.
The purpose of this survey was to assess the
level of knowledge and factors associated with
knowledge levels (for example, a respondent’s
background) among North Carolina dentists
regarding oral cancer risk factors and diagnostic
concepts, such as signs and symptoms and high-
risk anatomical locations of oral cancer.
SUBJECTS, MATERIALS AND METHODS
In May 2002, we mailed a 38-item pretested
survey, cover letter and business reply envelope
to a random sample of 1,115 of 3,303 licensed
dentists practicing in North Carolina. The North
Carolina State Board of Dental Examiners sup-
plied mailing addresses for all currently licensed
dentists. Six weeks after the initial mailing, we
sent a reminder postcard. Six weeks later, we
sent a second complete mailing to nonrespon-
dents. Entry of respondents into a drawing for
one of five gift certificates served as an incentive
to participate. After two mailings, we received
584 completed surveys, for an effective response
rate of 52 percent.
The Committee on Research Involving Human
Subjects of the University of North Carolina
School of Dentistry, Chapel Hill, approved this
voluntary confidential survey. Data were entered
into a computer database (Epi Info 2002, Centers
for Disease Control and Prevention, Atlanta) and
analyzed with a statistical software package (SAS
8.2, SAS Institute, Cary, N.C.).
The survey asked respondents about their
Copyright ©2005 American Dental Association. All rights reserved.
C O V E R S T O R Y
background, dental education, dental practice
patterns, use of medical histories to assess
patients’ oral cancer risk, knowledge of oral
cancer risk factors and diagnostic concepts, and
qualifications and training with regard to oral
cancer prevention and early detection. We created
three-level (low, medium, high) composite index
scores for knowledge of 16 oral cancer risk factors
and 14 oral cancer diagnostic concepts using pre-
viously developed scales.8 Respondents were
asked specifically if they had ever heard of—and
if they had used in the past 12 months—each of
the following aids for the diagnosis of OPC:
dcomputer-assisted brush biopsy (OralCDx, CDx
Laboratories, Suffern, N.Y.);
dtoluidine blue vital dye;
ddisposable light for illumination of abnormal
tissue (ViziLite, Zila, Phoenix).
We conducted separate bivariate and logistic
regression analyses to assess the relationship of
the three-level outcome variables (risk factor
knowledge index and diagnostic knowledge index)
with the following independent variables:
dsex;
dtype of practice (solo versus other);
dgeneralist versus specialist;
dyear of graduation (before versus after 1982);
dbiopsy/referral pattern (those who performed
five or more biopsies and/or referred five or more
patients with suspicious lesions versus those who
referred fewer than five patients with suspicious
lesions in the previous 12 months);
dhaving heard of or used the newer oral cancer
diagnostic aids (that is, computer-assisted brush
biopsy, toluidine blue vital dye or disposable illu-
mination light).
We used Cochran-Mantel-Haenszel statistics to
test the two knowledge indexes for association.
The statistical significance of the coefficients in
the logistic regression models was tested using
the Wald statistic at the .05 level to determine
which variables to include in the regression
model. We calculated odds ratios and 95 percent
confidence intervals from the regression coeffi-
cients and standard errors. We used P < .05 as
the criterion for retaining variables in the final
logistic regression models.
RESULTS
Of the 584 respondents, 479 (82 percent) were
male, 365 (63 percent) were in solo practice and
469 (80 percent) were general dentists. Of the 469
general dentists, 90 (19 percent) also had com-
JADA, Vol. 136 www.ada.org/goto/jada May 2005 603
C O V E R S T O R Y

Of the 584 respondents,

339 (58 percent) reported
PERCENTAGE OF RESPONDENTS

100%
99%
100 95%
having performed biopsies
84%
80 76%
74%
76% and/or referred five or more
60% 60% patients with suspicious
60
44%
lesions in the previous year,
42%
40 39% 36% 456 (78 percent) had heard
29%
18%
of one or more oral cancer
20
8% diagnostic aids and 123 (21
0 percent) had used one or
o

on

ol

ge

60

ty
es

es

ng

es

er
more oral cancer diagnostic
s

le

od
ur

en
cc

ru
oh

si

nc
ur
tt

ag
A
si

ri
ab
ba

>
s

Fo

i
be
i

re

te
Le

yg
po

Ca
av
lc

er

er
ge

en
et
To

ga
A

us

lH
om
ld

y
ev
aids. Respondents reported
Ex
r

eg

of
A

D
ic
ce

of

Cl
Ci
O
of

ra
Sp
V

at

ng
ill
n

ry
an

al
n
se

s/

d
Su

O
se

ap

ha

to
ti
ed

an

ili
ng
lC

t
U

the following in regard to

ui

t
U

is
to

m
T

oo

Fi
os
ra

ti
Fr

H
an

Fa
od
y

Ea

y
gn
ed
O

ily
rl
of
um

is

Fo
s

at

ia

oo
having specific knowledge of

m
ou

R
e

D
el

ot

Fa
ss
vi

ta

H
it
re

Le
er

In

or

of
P

and using the diagnostic

nc

o
w

aj

cc

n
Ca

Lo

io
ba

pt
p

To

aids within the previous

Li

um
s

ns
es
el

Co

12 months:
ok
Sm

RISK FACTORS NONRISK FACTORS dsixty-nine percent had

heard of the computer-
Figure 1. Dentists responding correctly to the 16 risk factor knowledge items. assisted brush biopsy and
20 percent had used it;
dforty-six percent had
OF RESPONDENTS

99% 98%
100
84% 83%
heard of toluidine blue
82% 79% 79% 78%
80 77%
73%
70%
vital dye and 2 percent
60
60%
53%
had used it;
45%
dthirty-six percent had
40
heard of the disposable light
PERCENTAGE

20 for illumination of abnormal

0 tissue and 1 percent had
used it.
C

es

te

on

es

C
al

te
O

re

O
io

xe
ng
iv

Si
t

ag
ti
Si

Si
at

A
f

h
Fi
rv

di

ua
o

rl

ue

it

it
St
in

ed

on

on
Su

Ea

W
rm

As shown in Figure 1, the

or

in
ng
am
Fi

d
R

nt

ce
d

d
Fo
es

le
in

To
e

te

te
Ex

co
v

ob

an
es
ov

Co

Co
ti

on

ia

ia
y
ti

is
nl
e

dv
M
ga

percentage distribution of
pr

el

oc

oc
a

D
u
m

t
ai

om

k
ng
Im

A
os

os
e

s,
ss

ss
m

h
P

Li
N

es

it

in
M

A
To
Co

pt
as
on

W
ee

er

nl
ia

ly
ed
ym

correct responses to the 16

h

rd
t
ti

hr

ai
s

ve
Tw

Tw

on
ak
os

ug

ar
ec

os
P
s

o
T

ol
M

m
pl
B
e
et

ro

gn
d,
of

of
r

pp

es

m
te

ro
Is

al
D

ll

ho

ar

ia

risk factor knowledge items

R
ne

ne

Co
Ce
f

er

th
nt
y

D
H
T

ns
rl

at
O
ly

ry
ie
s

t
ns

en
e
ed
Ea

os
ou

al

io
ol

od
th

ue
at
io

ft
b

su

M
tr

d
m

ou

ng

Le
ri

ranged from 8 percent to 100

at

O
an
en
U
ua

ia
M

ph
in

To

t
es

o
V
C

os

k
ia
Sq
am

cc
O

of
D

ym

la
k

M
ba
y

op
la
st
Ex

percent. As Figure 2 shows,

r

L
rl

C
oo

op
ti

To

hr
Ea

O
en
C

Fl

uk

yt
O

s
D

es

Er
ue

Le

the percentage distribution of

el
in

ok
nt

Sm
co

correct responses to the 14

is
D
ot

DIAGNOSTIC KNOWLEDGE ITEM

nn

diagnostic knowledge items

Ca

Figure 2. Dentists responding correctly to the 14 diagnostic knowledge items.
OC: Oral cancer.
pleted either a general practice residency (GPR)
or an advanced education in general dentistry
(AEGD) program. Specialists included 27 oral and
maxillofacial surgeons (5 percent), 21 periodon-
tists (4 percent), 19 orthodontists (3 percent), 17
pediatric dentists (3 percent), 12 endodontists (2
percent), 10 public health dentists (2 percent) and
nine prosthodontists (1 percent), but no oral
pathologists or oral radiologists. Year of gradua-
tion from dental school ranged from 1944 to 2001,
with a median year of 1982.
ranged from 45 to 99 percent.
The mean correct risk factor
knowledge score was 8.7
(range, 0 to 15), and the mean correct diagnostic
knowledge score was 8.6 (range, 0 to 12). Knowl-
edge levels were significantly associated with each
other (Mantel-Haenszel χ2, P < .0001).
Table 1 shows the distribution of dentists
according to patterns of knowledge about risk fac-
tors and the clinical diagnosis of oral cancer. Only
181 (31 percent) of 584 respondents had consis-
tent medium-to-high levels of knowledge on both
indexes, and 171 respondents (29 percent) scored
low on both indexes.

604 JADA, Vol. 136 www.ada.org/goto/jada May 2005

Copyright ©2005 American Dental Association. All rights reserved.
 C O V E R S T O R Y

In examining for bivariate TABLE 1

associations with risk factor
and diagnostic knowledge
scores, we found that gradu-
ation from dental school
within the previous 20
years, performing biopsies or
referring five or more
DISTRIBUTION OF DENTISTS BY PATTERNS OF
KNOWLEDGE ABOUT RISK FACTORS AND CLINICAL
DIAGNOSIS OF ORAL CANCER.*
KNOWLEDGE OF RISK KNOWLEDGE OF CLINICAL DIAGNOSTIC CONCEPTS‡ (N = 584)
FACTORS†

patients for biopsies of suspi- Low Score Medium Score High Score All Dentists
(0 to 9 Items) (10 or 11 (12 to 14
cious lesions within the pre- Items) Items)
vious 12 months, and having
heard of oral cancer diag- Low Score 171 (29.3) 46 (7.9) 10 (1.7) 227 (38.9)
(0 to 8 Items)
nostic aids were significantly
associated with higher Medium Score 102 (17.5) 71 (12.2) 17 (2.9) 190 (32.5)
(9 or 10 Items)
knowledge scores (Table 2).
Women also were more High Score 74 (12.7) 75 (12.8) 18 (3.1) 167 (28.6)
(11 to 16 Items)
knowledgeable about diag-
nostic concepts than were ALL DENTISTS 347 (59.4) 192 (32.9) 45 (7.7) 584 (100.0)

men, yet this association * Data are number (percentage) of respondents.

was not maintained in the † Total of 16 risk factor items.
‡ Total of 14 diagnostic items.
logistic regression model.

TABLE 2

BIVARIATE ASSOCIATIONS WITH RISK FACTOR KNOWLEDGE AND

DIAGNOSTIC KNOWLEDGE.
VARIABLE PERCENTAGE OF RESPONDENTS (N = 584)

Risk Factor Knowledge Diagnostic Knowledge

Low Medium High P Value Low Score Medium High P Value

Score Score Score Score Score

Sex
Male 39.1 31.9 29.0 .650 61.8 30.8 7.4 .022
Female 35.6 36.5 27.9 47.1 43.3 9.6

Practice Type
Solo 40.3 32.3 27.4 .608 60.0 32.3 7.7 .930
Other 36.5 32.9 30.6 58.4 33.8 7.8

Practitioner
Generalist 37.5 33.6 29.0 0.371 57.4 34.2 8.4 .116
Specialist 44.3 28.7 27.0 67.7 27.9 4.9

Recency of Dental
School Graduation
Last 20 years 32.5 32.5 35.1 < .001 51.7 40.1 8.3 < .001
More than 20 years 45.7 32.6 21.6 67.7 25.2 7.1
ago

Biopsy/Referral
Pattern
≥ Five patients/year 33.2 33.8 32.9 .002 53.1 37.1 9.8 < .001
< Five patients/year 46.6 30.8 22.7 68.0 27.1 4.9

Heard of Diagnostic
Aids
Yes 33.2 34.1 32.8 < .001 54.2 36.7 9.2 < .001
No 59.5 27.0 13.5 78.6 19.1 2.4

Used Diagnostic Aids

Yes 33.9 33.1 33.1 .346 54.0 33.9 12.1 .093
No 40.2 32.4 27.4 60.9 32.6 6.5

JADA, Vol. 136 www.ada.org/goto/jada May 2005 605

Copyright ©2005 American Dental Association. All rights reserved.
C O V E R S T O R Y

TABLE 3 with GPR or AEGD

training. In addition, 13 (48
MULTIVARIABLE MODELS OF FACTORS percent) of the 27 oral and
SIGNIFICANTLY ASSOCIATED WITH RISK FACTOR maxillofacial surgeons
KNOWLEDGE AND DIAGNOSTIC KNOWLEDGE. scored high on the risk
factor knowledge index,
FACTOR RISK FACTOR KNOWLEDGE DIAGNOSTIC KNOWLEDGE while four (15 percent)
Odds Ratio 95% Odds Ratio 95%
scored high on the diag-
Confidence Confidence nostic knowledge index.
Interval Interval
Twenty (23 percent) of the
Having Heard of 2.7 1.8 to 4.0 2.7 1.7 to 4.4 88 other specialists com-
Diagnostic Aids bined scored high on the
Recency of Dental 1.8 1.3 to 2.5 1.8 1.3 to 2.5 risk factor knowledge index
School Graduation and 2 percent scored high
Biopsy/Referral 1.5 1.1 to 2.1 1.7 1.2 to 2.4 on the diagnostic knowledge
Frequency index.
Continuing education.
In constructing a multivariable model of fac- Once in practice, dentists should obtain the best
tors significantly associated with knowledge of continuing education in early detection of oral
risk factors and diagnostic concepts (Table 3), we cancer to meet their needs. Hands-on diagnostic
found that the strongest independent variable in evaluation of oral lesions and palpation of lym-
each model was having heard of diagnostic aids, phadenopathy in patients, as well as hands-on
followed by more recent dental school graduation activity in simulated clinical situations (including
and greater frequency of having performed biop- demonstration of, and practice in, biopsy tech-
sies or referred patients for biopsies. The fol- niques), may have the highest educational value
lowing variables did not have an impact on either and the most immediate applicability to clinical
of the knowledge categories: sex, practice type, practice. However, continuing education via video
generalist versus specialist and (digital video disk and video home
actual use of a diagnostic aid. system videotapes) and seminar/
The strongest lecture formats may be more
DISCUSSION accessible.9
independent variable
Being able to routinely detect oral Because of the asymptomatic
in each model was
cancer at an early stage and nature of premalignant and early
counsel patients in prevention is a having heard of malignant oral lesions, a primary
continuous challenge for the dental diagnostic aids. strategy of early detection lies in
profession. Dentists must be the health care provider’s review of
familiar with the risk factors and the patient’s medical and dental
clinical signs and symptoms of oral cancer if they history, as well as the extraoral and intraoral
are to be effective in identifying, referring and visual inspection of the head and neck (including
counseling high-risk patients. Several opportuni- manual palpation of related sites).10 Practice pat-
ties exist to educate dental professionals: the first terns including routine, systematic oral cancer
is in dental school (compulsory), the second is examinations and biopsy (or referral for biopsy) of
postgraduate education (self-selective) and the suspicious lesions must follow from improved
third occurs later, in continuing education (self- knowledge and a heightened index of suspicion.
selective). The association between higher knowl- Stahl and colleagues6 reported that 62.2 percent
edge scores and recency of dental school gradua- of a random sample of U.S. dentists surveyed
tion is important. about the impact of the ADA oral cancer cam-
Of the 379 general dentists without GPR or paign said that having a better understanding of
AEGD training, 102 (27 percent) scored high on what to look for would motivate them to test
the risk factor knowledge index and 30 (8 percent) small, benign-appearing lesions routinely for
scored high on the diagnostic knowledge index. early signs of oral cancer. The box provides a
This compares with 32 (36 percent) and 10 (11 simple, seven-step guide for improving the
percent), respectively, of the 90 general dentists chances of detecting oral cancer at an early stage.

606 JADA, Vol. 136 www.ada.org/goto/jada May 2005

Copyright ©2005 American Dental Association. All rights reserved.

In 2002, a statewide random-digit-dial tele-
phone survey of adults conducted in North Car-
olina revealed that only 29 percent of respondents
reported ever having had an oral cancer exami-
nation when this procedure was described to
them.11 Despite the fact that 85 percent of adults
stated that they had seen a dentist in the pre-
ceding three years, among those aged 40 years
and older, only 23 percent reported having
received an oral cancer examination in the pre-
vious year. It is troubling that smokers reported
significantly less oral cancer examination
experience than did nonsmokers.11
Dentists must be vigilant about oral mucosal
alterations.12 If suspicious erythroleukoplakic or
ulcerative lesions persist after possible traumatic
or infectious etiology has been ruled out, the clini-
cian should sample the tissue in a full-thickness
surgical or punch biopsy procedure and send the
tissue to a laboratory for histologic evaluation.
Alternatively, the clinician can refer the patient
to another practitioner for the surgical sampling
procedure. As mentioned above, several new early
detection technologies, including the brush
biopsy, toluidine blue vital dye and the chemilu-
minescent light, are available to assist dentists by
alerting them to the need for further analysis of
possibly diseased tissue.2
Toluidine blue. Use of toluidine blue has
improved the sensitivity and specificity of visual
examinations when used in selected cases in
which suspicious mucosal characteristics are
present.13-15 After years of research into its appli-
cation as an oral rinse or a topical agent, tolo-
nium chloride (toluidine blue) has been shown to
be effective in detecting oral precancerous and
malignant lesions.16 This metachromatic dye is a
nuclear stain that binds to DNA, thus showing
increased dye uptake in areas of tissue with high
nucleic acid content, such as those undergoing
dysplastic or malignant change. The product is
marketed in Europe, Asia and the Middle East as
OraTest rinse (Zila Europe, Salisbury, England),
and the parent company is testing a version of the
kit in Phase 3 clinical trials in the United States.
Toluidine blue and acetic acid can be obtained for
clinical use in the United States, but the sequence
of rinses with vital dye has not been used widely
owing to the dye’s staining properties. However, a
study in the United Kingdom did demonstrate
high levels of acceptance by patients and
dentists.17
Brush biopsy. The brush biopsy may have
Copyright ©2005 American Dental Association. All rights reserved.
C O V E R S T O R Y
BOX
SEVEN-STEP GUIDE FOR IMPROVING
THE CHANCES OF DETECTING ORAL
CANCER AT AN EARLY STAGE.
dDentists should update patients’ medical histories,
with an emphasis on tobacco- and alcohol-use history.
They should take the opportunity to encourage tobacco-
use cessation and moderation in alcohol use.
dDentists should let patients know they will receive a
thorough oral cancer examination because they deserve
it and the dentist cares about their well-being. Dentists
should tell patients that they will deliver the best pos-
sible oral health care.
dDentists should palpate the patient’s neck bilaterally
and refer him or her if swollen nodes are located that
are hard, painless and mobile or fixed to underlying
structures.
dDentists should visually examine and bimanually pal-
pate the oral mucosal surfaces for painless, red and/or
white lesions or ulcers. They should pay particular atten-
tion to the high-risk areas of the ventrolateral tongue and
floor of the mouth.
dDentists should inform patients about their findings.
They should seek a probable cause: is it traumatic/irrita-
tional, infectious, developmental, nutritional, caused by
systemic disease, unknown? If the dentist is not particu-
larly concerned, he or she should consider using adjunc-
tive techniques, such as toluidine blue or the brush
biopsy, encourage removal of all potential causes of the
lesion and schedule a recall appointment for re-
evaluation in two weeks. If the dentist is particularly
concerned by the lesion’s appearance and history of
growth behavior, he or she should perform a full-
thickness scalpel biopsy or refer the patient for a biopsy.
dIf the lesion has not healed after two weeks, the den-
tist should perform a scalpel biopsy or refer the patient
for a biopsy.
dIf the biopsy results are uncertain or indicate that the
lesion is benign, yet the dentist believes the lesion still
might be malignant, he or she should perform a second
scalpel biopsy or refer the patient for a second biopsy.
utility in evaluating oral mucosal lesions of
unknown significance. Using a “cytobrush,” the
dentist samples the altered oral mucosal surface,
places a full-thickness sample of mucosal cells on
a slide with fixative and mails the sample to a
central laboratory facility. At the laboratory, the
sample is analyzed by computer—along with
standard microscopic evaluation by a patholo-
gist—for abnormal cell characteristics indicating
dysplasia or malignancy. 2,5,18 Despite its high sen-
sitivity, specificity and positive predictive values
in the original U.S. clinical trial5 and in addi-
JADA, Vol. 136 www.ada.org/goto/jada May 2005 607
C O V E R S T O R Y

tional studies,19,20 considerable controversy has survey evaluating the impact of the ADA’s oral
arisen regarding the efficacy of the brush biopsy cancer campaign reported that they evaluated
technique in clinical practice. Because not all small lesions more frequently because of the
potentially malignant disease is detected with availability of the brush biopsy test.6
21-23
this noninvasive procedure, complete reliance Study limitations. Given the 52 percent
on this cytologic screening test probably should response rate, we did not calculate population-
be avoided. based estimates, and nonresponse bias may have
Disposable light for illumination of limited our study findings. Similar response rates
abnormal tissue. The newest marketed tech- have been reported for mail surveys of practicing
nique is modeled after a chemiluminescent light dentists about oral cancer conducted nationally6,26
that, when combined with Papanicolaous tests of and in Maryland.27 We believe that survey non-
cervical mucosa, has been shown in a multicenter respondents were more likely to have had lower
trial to improve the detection of cervical premalig- knowledge levels and less interest in the topic
nant lesions compared with Pap tests alone.24 In a than were those who completed and returned the
darkened room, the clinician holds the chemilu- survey. If nonresponse bias affected our results, it
minescent light wand over oral mucosal tissue would most likely have resulted in higher knowl-
that has been prerinsed with 1 percent acetic edge levels than truly exist among practitioners
acid; abnormal mucosa appear as an opaque in North Carolina.
acetowhite alteration, alerting the We conducted this survey shortly
clinician to the need for a scalpel after the ADA initiated a nation-
biopsy.2 One pilot study of 150 Dentists with greater wide effort to increase the public’s
25
patients demonstrated the feasi- knowledge of both and providers’ knowledge about
bility of its use intraorally; how- risk factors and early detection of oral cancer. In
ever, large-scale studies are diagnostic concepts late 2001, the ADA’s national oral
required to elucidate issues related were significantly cancer awareness campaign was
to the sensitivity, specificity and launched to increase public aware-
more likely to have
predictive value of this technology ness of oral cancer.6 Cities in North
in the detection of oral premalig- identified suspicious Carolina were not among the 11
nant and malignant changes. lesions and to have cities targeted for this media inter-
Dentists with greater knowledge performed biopsies vention, which included advertise-
of both risk factors and diagnostic or referred patients ments in a variety of outdoor
concepts were significantly more for biopsies. venues (such as billboards, bus-
likely to have identified five or shelter signs and taxi tops). Thus,
more suspicious lesions and to have we do not suggest that this public-
performed biopsies or referred focused media event had a direct
patients for biopsies in the previous 12 months impact on dentists’ knowledge levels, use of
than were less knowledgeable respondents. We adjunctive diagnostic aids or practice patterns in
did not ask respondents specifically how many North Carolina. However, the educational cam-
lesions had positive biopsy results. However, paign targeting dentists, which included adver-
these findings indicate that increased referral tisements in ADA publications, may have had an
and biopsy practices are associated with greater impact on dentists’ awareness in North Carolina
provider knowledge levels. at the time that our survey was conducted.
Because this study showed that having actu- The lessons learned from promoting early
ally used a diagnostic aid was not significantly detection of breast, cervical and colorectal cancer
associated with knowledge levels, we cannot con- (for which effective screening tests exist) suggest
clude that use of aids reinforced the dentist’s that a diverse set of strategies targeted to each
practice of referring patients or performing biop- group involved (public policy–makers, clinicians,
sies. It is likely, however, that exposure to addi- patients and organizational systems/practice set-
tional diagnostic techniques improves the den- tings) is most effective at improving cancer
tist’s confidence in regard to his or her decision to screening rates when used synergistically.28 An
perform a more thorough examination. This pos- example of this is the concurrent public and
sibility is supported by the finding that nearly provider educational approach taken in the 2001
two-thirds of dentists responding to a national ADA campaign.

608 JADA, Vol. 136 www.ada.org/goto/jada May 2005

Copyright ©2005 American Dental Association. All rights reserved.

More public education (such as training pro-
vided by health care professionals or lay health
care workers) in conducting oral self-
examinations may enhance awareness and
empower patients. Physicians would benefit from
training in early detection of oral cancer provided
by dental educators. To enhance the effectiveness
of this important dental public health effort, we
need public policy and professional organiza-
tional/insurance industry changes that facilitate
patient access to providers, result in consensus
recommendations regarding preventive care, pro-
vide appropriate insurance coverage and reim-
bursement, and result in consensus regarding the
efficacy and validity of the existing and emerging
adjunctive diagnostic technologies.
CONCLUSIONS
The results of our survey of dentists in North
Carolina show that higher levels of knowledge
regarding oral cancer risk factors and diagnostic
concepts were associated with more recent dental
school training, having heard of oral cancer diag-
nostic aids, and having performed more biopsies
or referred more patients for biopsies of suspi-
cious lesions. Educational strategies to enhance
clinicians’ knowledge of oral cancer risk factors
and diagnostic concepts, as well as to promote
early detection, are needed in North Carolina.
These programs should include information
about the availability and use of adjunctive
diagnostic techniques. Along with these pro-
grams, we need enhanced public awareness;
professional organizational support; and imple-
mentation of policies that facilitate continued
patient and provider participation in routine
oral cancer examinations, as well as dentist
education and training to ensure proficiency in
conducting these examinations. Combined,
these efforts have the potential to increase early
detection, diagnosis and referral of patients
with suspicious oral lesions and improve the
health of the public. ■
Dr. Patton is a professor, Department of Dental Ecology, and
director, General Practice Residency program, University of North Car-
olina School of Dentistry, CB #7450, Chapel Hill, N.C. 27599-7450,
e-mail “lauren_patton@dentistry.unc.edu”. Address reprint requests to
Dr. Patton.
Dr. Elter is a research assistant professor, Department of Dental
Ecology, University of North Carolina School of Dentistry, Chapel Hill.
Dr. Southerland is an assistant professor, Department of Dental
Ecology, and director of Hospital Dentistry, University of North Car-
olina School of Dentistry, Chapel Hill.
JADA, Vol. 136 www.ada.org/goto/jada
Copyright ©2005 American Dental Association. All rights reserved.
C O V E R S T O R Y
Dr. Strauss is a professor and chair, Department of Dental Ecology,
University of North Carolina School of Dentistry, and a professor of
Social Medicine, University of North Carolina School of Medicine,
Chapel Hill.
This study was supported by National Institutes of Health grant
DE 14413.
The authors acknowledge the thoughtful contributions of Dr. Rick
Mumford, chief, Division of Oral Health, North Carolina Department of
Health and Human Services, Raleigh, in the project initiation and the
valuable assistance of Jacqueline C. Garner for data entry.
1. Jemal A, Murray T, Ward E, et al. Cancer statistics 2005. CA
Cancer J Clin 2005;55(1):10-30.
2. Sciubba JJ. Oral cancer and its detection. History-taking and the
diagnostic phase of management. JADA 2001;132(supplement):12S-
18S.
3. Silverman S Jr. Demographics and occurrence of oral and pharyn-
geal cancers: the outcomes, the trends, the challenge. JADA
2001;132(supplement):7S-11S.
4. Yellowitz JA, Horowitz AM, Drury TF, Goodman HS. Survey of
U.S. dentists’ knowledge and opinions about oral pharyngeal cancer.
JADA 2000;131:653-61.
5. Sciubba JJ for the U.S. Collaborative OralCDx Study Group.
Improving detection of precancerous and cancerous oral lesions:
computer-assisted analysis of the oral brush biopsy. JADA 1999;130:
1445-57.
6. Stahl S, Meskin LH, Brown LJ. The American Dental Association’s
oral cancer campaign: the impact on consumers and dentists. JADA
2004;135:1261-7.
7. Ries LAG, Eisner MP, Kosary CL, et al, eds. SEER Cancer statis-
tics review, 1975-2000. Bethesda, Md.: National Cancer Institute.
Available at: “seer.cancer.gov/csr/1975_2000/”. Accessed March 16,
2005.
8. Clovis JB, Horowitz AM, Poel DH. Oral and pharyngeal cancers:
knowledge and opinions of dentists in British Columbia and Nova
Scotia. J Can Dent Assoc 2002;68:415-20.
9. Christensen GJ. Continuing education: the good, the bad and the
ugly. JADA 2004;135:921-4.
10. Perform a death-defying act: the 90-second oral cancer exami-
nation. JADA 2001;132(supplement):36S-40S.
11. Patton LL, Agans R, Elter JR, Southerland JH, Strauss RP,
Kalsbeek WD. Oral cancer knowledge and examination experiences
among North Carolina adults. J Public Health Dent 2004;64(3):173-80.
12. Neville BW, Day TA. Oral cancer and precancerous lesions. CA
Cancer J Clin 2002;52(4):195-215.
13. Martin IC, Kerawala CJ, Reed M. The application of toluidine
blue as a diagnostic adjunct in the detection of epithelial dysplasia.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998;85:444-6.
14. Rosenberg D, Cretin S. Use of meta-analysis to evaluate tolonium
chloride in oral cancer screening. Oral Surg Oral Med Oral Pathol
1989;67:621-7.
15. Warnakulasuriya KA, Johnson NW. Sensitivity and specificity of
OraScan (R) toluidine blue mouthrinse in the detection of oral cancer
and precancer. J Oral Pathol Med 1996;25(3):97-103.
16. Patton LL. The effectiveness of community-based visual screening
and utility of adjunctive diagnostic aids in the early detection of oral
cancer. Oral Oncol 2003;39:708-23.
17. Feaver GP, Morrison T, Humphris G. A study to determine the
acceptability in patients and dentists of toluidine blue in screening for
oral cancer. Prim Dent Care 1999;6(2):45-50.
18. Christian DC. Computer-assisted analysis of oral brush biopsies
at an oral cancer screening program. JADA 2002;133:357-62.
19. Svirsky JA, Burns JC, Carpenter WM, et al. Comparison of
computer-assisted brush biopsy results with follow up scalpel biopsy
and histology. Gen Dent 2002;50:500-3.
20. Scheifele C, Schmidt-Westhausen AM, Dietrich T, Reichart PA.
The sensitivity and specificity of the OralCDx technique: evaluation of
103 cases. Oral Oncol 2004;40:824-8.
21. Poate TW, Buchanan JA, Hodgson TA, et al. An audit of the effi-
cacy of the oral brush biopsy technique in a specialist oral medicine
unit. Oral Oncol 2004;40:829-34.
22. Slater LJ. Brush biopsy: false negative results. J Oral Maxillofac
Surg 2004;62:764.
23. Rick GM. Oral brush biopsy: the problem of false positives. Oral
Surg Oral Med Oral Pathol Oral Radiol Endod 2003;96:252.
24. Loiudice L, Abbiati R, Boselli F, et al. Improvement of Pap smear
May 2005 609
C O V E R S T O R Y

sensitivity using a visual adjunctive procedure: a co-operative Italian
study on speculoscopy (GISPE). Eur J Cancer Prev 1998;7:295-304.
25. Huber MA, Bsoul SA, Terezhalmy GT. Acetic acid wash and
chemiluminescent illumination as an adjunct to conventional oral soft
tissue examination for the detection of dysplasia: a pilot study.
Quintessence Int 2004;35:378-84.
26. Horowitz AM, Drury TF, Goodman HS, Yellowitz JA. Oral
pharyngeal cancer prevention and early detection: dentists’ opinions
and practices. JADA 2000;131:453-62.
27. Canto MT, Drury TF, Horowitz AM. Maryland dentists’ knowl-
edge of oral cancer risk factors and diagnostic procedures. Health Pro-
motion Practice 2001;2:255-62. Available at: “hpp.sagepub.com/cgi/
content/abstract/2/3/255”. Accessed March 29, 2005.
28. Zapka JG, Lemon SC. Interventions for patients, providers, and
health care organizations. Cancer 2004;101(5 supplement):1165-87.

610 JADA, Vol. 136 www.ada.org/goto/jada May 2005

Copyright ©2005 American Dental Association. All rights reserved.