http://jdr.sagepub.com
Journal of Dental Research
DOI: 10.1177/00220345990780031201 1999; 78; 790 J DENT RES
D. Locker, A. Liddell, L. Dempster and D. Shapiro
Age of Onset of Dental Anxiety
http://jdr.sagepub.com/cgi/content/abstract/78/3/790 The online version of this article can be found at:
Published by:
http://www.sagepublications.com
On behalf of:
International and American Associations for Dental Research
can be found at:
Journal of Dental Research
Additional services and information for
http://jdr.sagepub.com/cgi/alerts
Email Alerts:
http://jdr.sagepub.com/subscriptions
Subscriptions:
http://www.sagepub.com/journalsReprints.nav
Reprints:
http://www.sagepub.com/journalsPermissions.nav
Permissions:
http://jdr.sagepub.com/cgi/content/refs/78/3/790
Age
of Onset of Dental Anxiety
D.
Lockerl*,
A.
Liddell2,
L.
Dempster',
and D.
Shapirol
'Facultyof Dentistry, Universityof Toronto, 124Edward Street,Toronto,Ontario, Canada M5G 1G6; and 2DepartmentofPsychology, Memorial University, StJohn's,Newfoundland,Canada;*correspondingauthor
Abstract. Little attentionhas been given tothe issue of the
age ofonsetof dental anxiety, even thoughit may have a bearingonthe origins of thistypeoffear. This
study
aimedto
identify
theageofonsetofdental anxiety andtoidentify
differencesbyageofonsetwithrespect topotential
etiologi-cal factors, such asnegative dental experiences, familyhis-tory of dentalanxiety, and general
psychological
states. Data werecollected bymeansoftwomailsurveys ofaran-dom sample of the adult
population.
Of 1420 subjects returningquestionnaires,16.4% weredentallyanxious.Half, 50.9%,reported
onset in childhood, 22.0% in adolescence, and 27.1% inadulthood. Logistic regressionanalyses
indi-cated that negative dental experiences were
predictive
of dental fearregardless ofage of onset.A family history of dental anxiety was predictive of child onset only. Adolescent-onset subjects werecharacterized
by trait anxi-ety and adult-onset subjects by multiple severe fears and symptoms indicative of psychiatric problems. The threegroups were similarin terms of their
physiological,
cogni-tive, and behavioral responses to dental treatment. However, adolescent- and adult-onset subjects weremorehostile toward and lesstrusting of dentists. These results indicate that child-onset subjects were morelikely tofall intothe exogenousetiological category suggested byWeiner
and Sheehan (1990), whileadult-onset subjects were more
likelytofallintotheendogenous category.
Key words: dental anxiety, etiology, age, psychological factors.
Received October7, 1997;Last Revision May 27, 1998; Accepted June 5, 1998
Introduction
Dentally anxiousindividuals are not a
homogenous
groupbut differ in terms of the origins
and/or
manifestations oftheir fears of dental treatment
(Milgrom
etal., 1985).
Forexample, Weiner and Sheehan
(1990)
suggested
thatthey
could be classifiedinto twogroups, exogenousand endoge-nous,withrespect to thesourceof theiranxiety. Inthefor-mer, dental anxiety is the result of conditioning via traumaticdental experiences orvicariouslearning, whilein
thelatter,ithasitsorigins in a
constitutional
vulnerabilityto anxiety disorders, as evidenced bygeneral
anxiety states,multipleseverefears, and disordersof mood. This
classifica-tion suggests thatnot all dentallyanxious
subjects
becomefearful
asaresult ofconditioning.
Evidence for the role of conditioning in dental anxiety,
through either aversive experiences or family influences, has beenprovided by Shoben and Borland (1954), Lautch (1971), Kleinknecht et al. (1973), Berggren and Meynert
(1984),
Ost
andHugdahl (1985),Davey(1989),
Milgrometal.
(1995), Locker etal.
(1996a),
and Poulton et al. (1997). Evidence for therole of additional severe fears andpsychi-atric problems has been provided by Fiset et al. (1989),
Mooreetal.
(1991),
and Roy-Byrneetal. (1994).However, todate,no study has addressed the relative contributions of
negativeexperiences,familial attitudes, and general
psycho-logicalstates todentalanxiety.
One issue which may have abearing on the originsof dental anxiety is that of ageof onset (Marks and Gelder,
1966). Ina studyofsixdifferenttypesof phobia,
Ost
(1987) found significantdifferences in age of onset. Animal and bloodphobias were largely acquired inchildhood, socialphobias in adolescence, and agoraphobia and
claustropho-biain adulthood. Of some importance was thefact that
these different ages ofonset were associated with distinct modesof acquisition anddifferencesin psychological char-acteristics.
Onsetof DentalAnxiety
The age of onset with respect to dental anxiety has received relatively littleattention(Ost,1987),largely because dental anxietyis usually viewed as a fear originating in childhood which persists into later life. However, two stud-ies that haveaddressed thisissue donotsupport this point of view. Ost (1987) studied a groupof dental phobicsand found that almost 20% reported onset after the ageof 14 years.Similarly, apopulation-based study by Milgrometal.
(1988) foundthat 33.3% becameanxiousduring adolescence oradulthood.
Apartfrom these data, littleisknown aboutvariations in the age ofonset of dental anxiety and itsimplications in terms of etiology. This paper reportsthe results ofa study which aimedto: (1) describe the distribution of the dentally anxious populationaccording to ageofonset, and (2) iden-tify differencesby ageof onset with respect toetiological
factors such asnegative dental experiences, family history,
and general psychologicalstates. The main hypothesis we testedwasthatchild-onsetsubjectswould bemorelikelyto fallinto the exogenous categorydescribed byWeiner and
Sheehan (1990), having acquired theiranxietythrough con-ditioning experiences, while adult-onset subjectswould be morelikelytofall into the endogenouscategoryand demon-strate a heightened constitutional vulnerability to anxiety
disorders.
Methods
Thedesign of the study,itsdatacollection methods, and proce-duresforobtaininginformedconsentandensuring confidential-ity were approved by the UniversityofToronto's Human SubjectsCertificationprocess.
Surveyprocedures
The datawere collected during the baselinephase ofa
longitu-dinal, population-basedstudy of the epidemiology of dental anxiety. The target population for the studywas all persons aged 18 years and overlivinginthe City ofEtobicoke,oneof five municipalities whichcomprise MetropolitanToronto. The sampling framewasthe listofvoterscovering thiscommunity. Since this iscompiledbyahouseholdenumerationprocedure,it containsthenamesofapproximately97% ofpersonseligibleto vote. Atwo-stage random-startsystematic samplingprocedure wasused,withsampling fractionsdesignedtogiveasampleof 6360subjects.
Data werecollectedbymeansoftwomailsurveysbasedon the Total Design Methodrecommended by Dillman
(1978).
Initially, all 6360 subjects were sent abasicquestionnaire ondentalanxiety. To maximizethe response rate,wesentthree fol-low-upmailings.Subsequently,a60% subsampleof those com-pleting the firstquestionnairewere sent asecondquestionnaire containinganumber ofpsychologicalmeasures.
Measures
Dental anxiety. Dental anxietywas measured with Corah's Dental Anxiety Scale (DAS) (Corah, 1969), theGatchel Fear Scale(Gatchel, 1989),and thesingleitemusedby
Milgrom
etal.(1988). Any subjects who scored 12 and above on the DAS, 8 or above on the Gatchel FS, or reported being very afraid or terri-fied of dentaltreatment were considered to be dentally anxious. This approachwasadopted since each of these measures alone fails to identify somedentally anxious subjects (Locker et al., 1996b). The Dental Fear Survey (DFS) (Kleinknecht and Bernstein, 1978) was used to assess the severity of dental anxi-ety. Subjects rated the amount of anxiety evoked by nine inva-siveand sixnon-invasive stimuli in the dental setting, using a six-pointscalerangingfrom 1 (not at all) to 6 (very much).
Age of onset. Subjects were asked toindicate whether they became dentally anxious during childhood (12 years of age or less), adolescence (age 13 to 17 years), or adulthood (18 years and over).
Conditioning experiences. Questions were asked about negative experiences in the dentalsetting, such as events whichwere painful, frightening,orembarrassing, andthe age at whichthey firstoccurred. The familyhistory withregardto dental anxiety wasassessed byitemspertaining tofear ofdentaltreatment on the part ofeach subject's mother,father,and siblings.
Psychological characteristics. General fearfulness was measured by a short form of theFearSurveySchedule II (FSSII) (Geer, 1966). Thedegreeof fearoranxietycreatedbyeach of20objects orevents was rated on a scaleranging from 0 (none)to 6 (ter-ror). Fearsrated5 or 6 wereconsideredtobe severe. Trait anxi-ety wasassessedby the20-itemSpeilbergerTraitAnxiety Index (STAI) (Speilberger etal., 1983). Inresponse to each item, sub-jects rate howthey generally feel using a four-point scale in which 1 indicates 'almost never' and 4 indicates 'almost always'. The 12-item version of the General Health Questionnaire (GHQ) (Goldberg and Williams, 1988) wasused tomeasurethefrequency ofsymptomspotentiallyindicativeof psychiatric disturbance. This also uses afour-point response scale whichvariesaccordingtoeachitem. Foreachscale,scores were generated from the sum of the responses toeach item. Subsequently, acut-off ofthe mean score plus one standard deviation wasused to differentiatebetween subjectswho had 'high' and thosewhohad 'low'scores.
Other characteristics.Physiologicalresponsesduringdental treat-ment wereassessedbysix itemsfromthe DFS. These included events suchasincreases inbreathingrate,increased heart rate, and feelingnauseous. Cognitiveresponses were measured
by
meansof a 13-item scaleconcerning thenegativethoughts sub-jects had prior to and during dental treatment (Kent and Gibbons, 1987). Behavioralresponses todental anxiety weremeasuredby meansofquestions concerningthe use of dental services,avoidanceofdentalcare,cancelingappointments,and failingto showupbecause ofanxiety about dentaltreatment. The Dental Beliefs Survey (Milgromet al.,1985) was used to
assess subjects' perceptions offour components of the dentist-patientrelationship that have a bearing on dental anxiety;
namely, communication,belittlement,lack ofcontrol,andtrust.
Dataanalysis
Subjectswereclassifiedintooneof fourgroups:(1)not
dentally
anxious, (2) dentally anxious-childhood onset,(3) dentally anxious-adolescent onset, and(4) dentally anxious-adult
onset. Two sets of analyses wereundertaken, the first compar-ing the threedentally anxious groups with eachother, andthe second comparing each of thesegroups with thenon-anxious. Inthisrespect, those who were not dentally anxious were used as a normative orreference group.Chi-squaretests wereused to assess thesignificance ofdifferences inproportions, and odds
ratiosand their 95% confidence intervals wereused to indicate andcompare the strengths of associations. The significance of
differences in means was assessedwith one-way analysis of
variance. Logistic regression analysiswas used to assess the
independenteffects ofpotential etiologicalfactors on the
proba-bility of dental anxiety.
Results
Response
The initial questionnairewas completed by 3055 subjects.
This represents 60.4% of the5061 individualspresumed to
be alive and livingatthe listed address. Of these,1420
com-pleted thesecondpsychological questionnaire. Census data indicated that, when compared with thetarget population,
subjects completing
both questionnaires were somewhat older and better educated.Ageofonset
Usingthe definition cited above, we classified 16.4% (n = 233)ofsubjects as beingdentallyanxious. One half, 50.9% (n
= 111), reported becoming fearful of dental treatment in
childhood,22.0% (n=48) sufferedonset inadolescence, and
27.1%(n=59) becameanxious asadults.There were no dif-ferences amongthe three groups in terms of gender. The
mean(SD)agesof the threegroups atthetimeof data
collec-tionwere 45.1 (13.7)years, 43.8 (13.7) years, and 49.6 (14.9)
years, respectively
(ns:
one-wayanalysis ofvariance). Themean(SD)age of the non-anxious referencegroup was 48.7
(16.6) years.
Themean (SD) DASscores of thethree groups were 14.1
Table1.Ageatfirsttraumaticdentalexperiencebydental anxiety status(%)
AgeatFirstTraumatic DentalExperience DentalAnxiety Status Child Adolescent Adult
Notdentallyanxious 39.4 22.3 38.3 (n= 1187)
Child-onset 67.9 18.3 13.8
(n= 111)
Adolescent-onset 31.1 46.7 22.2 (n=48)
Adult-onset 27.3 10.9 61.8
(n=59)
Differencesinproportions between dentally anxious groups: p < 0.0001:Chi-squaretest.
(2.7),13.7
(2.8),
and 13.5(2.7), respectively;
thesewere notsignificantly
different. However, those with child-onset den-talanxietyreported
significantly
moredistress thanadoles-cent-andadult-onset
subjects
withrespecttosixof thenine DFSitems concerned withinvasiveprocedures
and two of the six DFS items referring to non-invasiveprocedures.
Consequently,
those with childonsethadsignificantly
higher
scoresthan theothersonascaleconstructed
by summing
theresponsecodesto all15DFSitems
(means
of54.5,49.7,
and48.0,respectively;p<0.01,one-way
analysis
ofvariance).
Conditioning
experiences
andfamily history
Overall, 74.8%
reported
painful
dentalexperiences,
30.7%experiences whichwere
frightening,
and 13.3%experiences
whichwereembarrassing. Therewas anassociationbetweendental anxiety status and theage atwhich negative
experi-ences werefirst encountered
(Table 1).
Those withchild-onset anxiety werethemost
likely
to have had their first experienceinchildhood,
those with adolescent onset were the mostlikelyto have had their first experienceinadoles-cence, and those with adultonset werethe most
likely
to have had their first experience asadults. Forthe non-anx-ious,thefirstnegativedentalexperiences were moreevenly distributed withrespect to age.There were no differences among the three groups of dentallyanxioussubjectsinthe proportionsreporting nega-tivedental experiences
(Table
2). However, all threegroups were significantly different from the non-anxious. Oddsratios, obtained by comparing each dentally anxious group
with the non-anxious reference group, indicated that the
associationbetweenaversive eventsand dentalanxiety was strongestfor the child-onset group(OR = 9.2) and weakest for the adult-onsetgroup
(OR
=3.7).
Ageofonsetofdentalanxietyand family history of den-talanxiety were associated (Table 2). Half (55.9%) of those with childhood onset had a mother, father, or sibling who
was anxiousabout dental treatment compared with
one-third
(35.6%)
of those with adultonset(p<0.05). Theassoci-ated odds ratios and their
95%
confidence intervals indicated that afamily history of dentalanxiety was apre-dictorof child-onsetanxietyonly.
Psychological characteristics
Table 3 shows the proportions in each group with high scores on the three scales measuring general psychological
states.The oddsratiosand95% confidence intervals indicate
no differences between child-onset subjects and the non-anxiousreference group. Adolescent-onset subjects were
significantly
different on two ofthe three measures, andadult-onsetsubjectsweresignificantly differentonall three. The odds ratios indicated that the associations were strongest foradult-onsetsubjects.
Independent predictorsofdental anxiety by age of onset
Threelogisticregressionanalyses wereundertaken.Thefirst
sub-OnsetofDentalAnxiety
jects withthe reference group,and the third Table 2.Perc( compared the adult-onset subjects with the tory of dental reference group. The three types of negative ious
reference
dental experiences (painful, frightening,embarrassing) were entered asseparatevari- Dental
ables. Gender was included inthemodels as Anxiety
a control variable. All dependent and inde- Status pendent variables were in a binaryformat
coded0and 1. The resultsaresummarized in
Table 4. Not dentallya
Inthemodelpredictingchild-onset dental Child- t
anxiety, only a family history of dental
anxi-ety and the three variables describing aver- Adolescent-oi
sive experiences entered the regression Adult-onset
equation. Foradolescent-onset anxiety, nega- aDifferences tive experiences and the score on the Trait anxious refe Anxiety Index were significant predictors, bDifferences:
while for adult-onset anxiety, frightening anxious refe experiences and thescores ontheFSS II and
the GHQ emerged as significantpredictors.
Theseresults tend to confirmthe hypothesis beingtested. Direct andvicarious conditioning experiences are important with respect to child-onset dental anxiety, while general
psychological states play a more prominent role in adult-onsetanxiety.
Other characteristics
Therewere nodifferences among the three groups in terms
ofphysiological,cognitive, orbehavioralresponses todental
treatment. However, the Dental Belief Scale scores ofthe
three groups (4.5, 5.9, and 5.9,respectively; p < 0.05, one-way analysis of variance) indicated that adolescent- and adult-onset subjects were significantly morenegative
con-cerning dentists and their behavior toward patients than
werechild-onsetsubjects.
Discussion
Although theresponse tothe study
was acceptable, there were differ-
Table
3. Proportions wi ences between study subjects and Comparisonofeachder
the population from which theyweredrawn. Theformerwereolder and somewhat bettereducated than
the latter. Since the
magnitude
and DentalAnxiety
Status direction ofbias inducedby
thesedifferences are difficult to
judge,
some caution needsto be exercised Not dentallyanxious when the results of the study are Child-onset
generalized.
Among thedentallyanxioussub- Adolescent-onset jects identified in this population- Adult-onset
based study, only half
reported
a Difference between thbecoming
dentally anxious in child-cant;
p <0.05,
Chi-squ hood.One-fifth reportedadolescent b Differencebetween th onset, and almost one-third cant; p <0.01: Chi-squreported onset in adulthood. These c Differencebetween th results confirm the earlier observa- cant; p<0.001,Chi-sq
entreporting one or more negative dental experiences and a family his-anxiety:Comparisonof eachdentally anxious group with the
non-anx-egroup
PercentReporting Percent Reporting Negative Dental Family History of Experiences Dental Anxiety
Odds Odds
% Ratio 95%CI % Ratio 95% CI
anxious 74.4 1.0 35.4 1.0
96.4a 9.2 3.2-21.6 55.9a 2.3 1.5-3.5 nset 93.8b 5.2 1.5-15.8 47.9 1.7 0.9-3.1
91.5b 3.7 1.4-10.7 35.6 1.0 0.8-1.8
inproportionsbetween the dentallyanxious group andthe non-dentally zrencegroupsignificant: Chi-squaretest p < 0.0001.
inproportionsbetweenthedentallyanxious group andthe non-dentally Zrencegroupsignificant: Chi-squaretest p < 0.01.
tions ofMilgrom et al. (1988) and challenge the view that
dental anxiety isinvariably a fear which has its origins in
childhood. Although the DAS scores of the three groups were similar, scoresderived from the Dental Fear Survey
suggested that the anxiety ofchild-onset subjects wasmore severe than that of adolescent-onset and adult-onset
sub-jects. This was largely dueto the fact that the former reported significantly more anxiety withrespect to stimuli associated withinvasivedentalprocedures, suchas restora-tions and the extraction of teeth. Kleinknecht and Lenz
(1989) have suggested that since such treatments involve
blood and body injury types of stimuli, these anxiety responses may be linked to blood and body injury fears, whichtypicallyarise inchildhood(Ost, 1987). Such fearsare
characteristicofapproximately one-third of
dentally
anxiousadults(Lockeretal., 1997).
Rachman (1977) hassuggested that threetypesof condi-tioning may play a role inthe acquisition of fears. While many fears have their origins in direct experience, others
ithhighscoresonthe threemeasuresofgeneral anxiety and fearfulness: ntallyanxious groupwith thenon-anxiousreferencegroup
GeneralAnxiety and Fearfulness
FearSurvey Trait General Health ScheduleII AnxietyIndex Questionnaire
Odds Ratio OddsRatio OddsRatio % 95%CI % 95%CI % 95%CI
15.3 1.0 12.5 1.0 7.7 1.0 21.6 1.5/0.9-2.5 16.2 1.4/0.7-2.4 12.6 1.7/0.9-3.3 27.1a 2.1/1.1-4.1 27.1b 2.6/1.3-5.2 14.6 2.0/0.8-3.9 42.4c 4.0/2.3-7.2 28.8c 2.8/1.5-5.3 23.3c 3.7/1.9-7.4
edentallyanxiousgroup and the non-anxious reference group
signifi-iare test.
iedentallyanxiousgroup and the non-anxious reference group
signifi-iare test.
iedentallyanxiousgroup and the non-anxious reference group
signifi-juare
test.Table4.Resultsof thelogistic regressionanalyses
Model1 Model2 Independent Child-onsetvs. Adolescent-onsi
Variable: ReferenceGroup vs.ReferenceGro p Odds Ratio p Odds Rati
Familyhistory of dental anxiety
(Yes=1; No=0) <0.05 1.8 ns
Painful dental experiences
(Yes=1,No=0) <0.01 4.5 <0.05 3.1 Frightening dental
experiences
(Yes=1;No=0) <0.001 4.2 <0.05 2.4 Embarrassing dental
experiences
(Yes=1;No=0) <0.05 2.2 <0.05 2.8 FearSchedule
SurveyII score
(High=1;Low=0) ns ns TraitAnxiety
Indexscore
(High=1; Low=0) ns <0.05 2.6 General Health
Questionnairescore
(High=1;Low=0) ns ns
Sex(Male=1;Female=0) ns ns
Modelchi-square 87.4 38.1
df 8 8
p <0.0001 <0.0001
may arise viamodeling orexposure tothreatening
informa-tion. This
theory helps
explainwhy
notallfearful individu-alshave beenexposedto atraumaticconditioning episode.With respect to direct
conditioning
experiences, therewere nodifferences among the three groups of dentally anx-ioussubjects interms of theirreports of dentalexperiences
involving pain,
fear,
orembarrassment.
However, inboth bivariate and multivariateanalyses,
the associated oddsratiosindicate that such experiencesmadea greater
contri-bution to child-onset anxiety than adult-onset anxiety. The multivariate analyses also suggested that while all three
typesofexperiences were importantwith respecttothe
for-mer, only
frightening
experiences were influential withrespect to the latter. That child-onset subjects were more fearful of invasive procedures and adolescent- and adult-onset subjects morenegative concerning dentists' behavior
may also reflect differences in the types of conditioning
experiences giving rise to dental anxiety (Locker etal.,
1996a).
Although therewas anassociationbetween age of onset and age of first negativedental experience, there was a
degree
ofdiscordance ineach group of dentallyanxious subjectswithrespect tothetiming of traumaticexperiences.Almost one-third ofchild-onsetsubjects reported that their first traumaticexperiencedidnot occur untiladolescenceor
adulthood,
andalmost two-fifths of adult-onsetsubjects
reported
Model 3 such experiences prior tobecom-et Adult-onset vs.
up ReferenceGroup
ing
dentallyanxious.
Davey o p Odds Ratio(1989),
in exploring the latent inhibitionhypothesis,
haspro-vided evidence to demonstrate
that the relationship between
ns traumatic experiences and dental
anxiety
isnot asimple
one.Inhisstudies,
traumaticexperiences
ns were more
likely
togive
risetodental anxietyif theyoccurred <0.05 2.5
early
in anindividual's dentalcare
history
than ifthey
werepre-ceded
by
a series ofrelatively
ns -painless dental visits. Whilethis
explains
why
manynon-dentally
anxioussubjects reporttraumatic
<0.05 2.6 dental experiences, it does not
accountfor thepatternsobserved
in this study. Milgrom and
ns - Weinstein (1993) have suggested
that the consequences of trau-matic experiences aredependent <0.05 2.8 upon the context in which they
occur.Thatis, pain inflicted bya ns dentist who isperceived as caring
is likely to have less
psychologi-38.0 cal impact than pain inflicted bya
dentist who is cold and
control-8 ling.
<0.0001 These observations indicate that while direct conditioning experiences areimportant in the genesisof dental anxietyregardless ofage ofonset,factors other thantraumaappear tobe involved. Rachman's (1977) model of fearacquisition suggeststhat the
family history
of fears andphobias is important, while Weiner and Sheehan (1990) suggestthatgeneralpsychological
traits and statesplayarole.
In this study, a
family history
of dental anxiety was important with respect to child-onset anxiety only. Thisfinding
isconsistent withthe study by Ost(1987),
who reported thatphobias acquiredinadulthoodwerelesslikely thanthose acquiredinchildhoodtobe ascribedtomodeling and/orvicariouslearning. Itisalsoconsistentwith the find-ings ofMilgromet al.(1995),
who found that directcondi-tioning and modeling were bothimportant predictors of
dentalanxietyoriginatinginchildhood.
Bivariate andmultivariateanalyses ofourdata suggested
thatgeneralpsychologicalstatesand traitsmadea contribu-tionto adolescent- andadult-onset anxietybut not
child-onsetanxiety.Adolescent-onset subjectsweremorelikelyto
have high trait anxiety, while adult-onset subjects were morelikely to have multiple severe fearsand symptoms
potentially
indicativeofpsychiatric disorder. Thereis grow-ing evidence that, for some dentallyanxious adults, fearsconcerningdental treatment arelinkedto orsymptomatic of
Onsetof DentalAnxiety
with respect to children is less clear, perhaps because appro-priatemeasures have not been widely used in studies of
children. Milgrometal. (1995) found no association between
dentalanxiety and a measure of behavioral problems and
socialcompetency inchildren. However, studies have found a link between child temperament and anxiety regarding dental treatment (Liddell, 1990).
The role played by general psychological factors in the acquisition of dental anxiety appears to be a complex one. Davey (1997), in describing howcontemporaryconditioning
models help explain the acquisition and maintenance of phobias, has suggested that they are associated with a
ten-dency to focus onand rehearse the negative outcomes of encounters with feared objects and events. Since subjects with high traitanxiety selectively process threatening infor-mation, theyinflate the unpleasantness ofunconditioned
stimuli and experience conditioned responses of a higher
magnitude. Davey (1997) cites experimental evidence in supportof thischaracterization of theway inwhich fears are induced.
The datapresented here support the classification of den-tally anxious subjects suggested by Weiner and Sheehan (1990) and indicate that child-onset subjects were more likely tobeexogenous, whileadult-onsetsubjects were more
likely to beendogenous. There was, however,a degree of overlapinthat allsubjectswerecharacterized byexogenous
factors. Consequently, this classification scheme should be
considered to bedimensional rather than categorical.
A majorlimitation of the study was its cross-sectional design. Eventhough theanalysis approximated a
case-con-trolstudy, temporal relationships cannotbe established. At
best,we have documented associations between potential etiological factors and dentalanxietyoriginatingatdifferent
points inthe life span. Even so, the associationsbetween direct conditioning experiences and dental anxiety were
basedonsubjects' retrospective reports,which maybe
sub-ject toproblems of recallandretrospectivere-interpretation.
For example, Kent (1985) found that
dentally
anxious patients reported morepain threemonths after treatment than was reported immediatelyfollowing
treatment. Thisreconstruction made their actual experiences more consis-tentwith theiranxieties andexpectationsabout dental treat-ment.
These caveats also apply to thefindings concerning the family history of dental anxiety.
Although
the dataclearly
suggest that negative attitudes on thepart of the mother and/or siblings, and vicariouslearning
processes, make a greater contributionto child- onset dentalanxiety than to adolescent-oradult-onset dentalanxiety,
problems
of recalland retrospective re-interpretation may also have had an influence on these
findings
(Kent,
1989).
Because of theseproblems,
conclusionsregarding
the role ofconditioning
events andvicarious
learning
in theonset of dentalanxiety
arebestframedashypotheses
rather thanasdefinitive find-ings. Consequently,although
the data are consistent with theviewthat thepsychological
characteristics ofasubset of thepopulationrender them vulnerabletodentalanxiety,
the precise role of exogenous andendogenous
factors,
and the sequence in whichthey
occur in groups definedby
age ofonset,canbeelucidated
only by
meansoflongitudinal
stud-iesandqualitative research. It is entirely plausible, for exam-ple, that the psychological states which characterize
adult-onsetsubjects are related more to the maintenance ofdental anxiety than to its origins. In this regard, Liddell andLocker (1994) reportedthat adults who carried their dental anxiety into old age were more generally fearful than subjectswho hadbeen dentally anxious but recovered withaging.
Another implication of the study for research into dental
anxietyis that greaterrecognition needs to be given to the factthat thedentally anxious population is nothomogenous butis comprised of groups which differ along several
psy-chological and other dimensions. If data are not analyzed specifictothesesubgroups, thenimportant associationsmay be masked or attenuated. For example, when the logistic regression analysis was repeated comparing all dentally anxious subjectswith the non-anxious reference group,a family history of dentalanxiety and scores on the FSS II and theMHQ didnotenterthemodel.
Themainimplication ofthis study for dental practice is
that the dental team needs to take as much care with adult patients as withchild patients and use appropriate commu-nicationtechniqueswhichenhance trust and feelings of
con-trol. This preventiveapproach mayreducetheincidenceof dentalanxiety inpsychologically vulnerable individuals. In
addition, although child-onset subjects have more severe anxiety concerning dentaltreatment, adult-onset patients,
because of theirpsychological characteristics and greater
hostilitytowarddentists,maybemoredifficult for the prac-titionerto manage.Informationondentalanxietyandageof
onset,
then,
shouldbe collected from dentalpatients sothat dentistsmayplananappropriate managementand/or treat-mentstrategy.Acknowledgments
Theresearchonwhich thispaper isbasedwassupportedby GrantNo. 5RO1 DE10622-02from the
NIDR/NIH.
References
Berggren U, Meynert G (1984). Dental fear andavoidance. Causes, symptoms and consequences. JAm DentAssoc 109:247-251.
Corah N(1969). Developmentofadentalanxiety scale. JDent Res 48:596.
DaveyG(1989).Dentalphobiasandanxieties:evidencefor
con-ditioningprocesses intheacquisition and modulation ofa
learnedfear. BehavResTher27:51-58.
DaveyG(1997).Aconditioningmodelofphobias.In:Phobias:a
handbook oftheory,researchandtreatment.Davey G, edi-tor.London:JohnWileyandSons,pp. 302-322.
Dillman D(1978). Mailandtelephonesurveys:the total
design
method. New York:JohnWileyand Sons.FisetL, Milgrom P, Weinstein P, Melnick S (1989). Common fears and theirrelationshiptodental fear and utilization of the dentist. AnesthesProg36:258-264.
Gatchel R(1989). Theprevalence of dental fear andavoidance:
expandedadult and recentadolescentsurveys.JAm Dent
Assoc118:591-593.
Geer R(1966). Development ofascaleto measurefear. Behav Res Ther3:45-53.
Goldberg D,Williams P (1988). Auser's guide tothegeneral
healthquestionnaire.Windsor,UK:NFER-Nelson. KentG(1985).Memoryof dental pain. Pain21:187-194.
KentG (1997). Dental phobias.In:Phobias: ahandbook of the-ory,researchandtreatment.Davey G, editor. London:John Wiley and Sons,pp. 107-127.
KentG, Gibbons R (1987).Self-efficacy and the control of anx-ious cognitions.JBehavTher ExpPsychol18:33-40.
Kleinknecht R, Bernstein D (1978). Theassessment of dental fear. Behav Ther9:626-634.
Kleinknecht R, LenzJ (1989). Blood/injury fears, fainting and avoidance ofmedically-related situation: a family corre-spondencestudy. BehavResTher27:537-547.
Kleinknecht R, Klepac R,Alexander LD(1973). Origins and characteristics of fear ofdentistry.JAm Dent Assoc 86:842-848.
LautchH(1971). Dental phobia.BrJPsychiat119:151-158. LiddellA(1990). Personality characteristicsversusmedical and
dental experiences of dentallyanxious children. JBehav Med13:183-194.
LiddellA, LockerD (1994).Disproportionate dental anxiety in the over-50s. JBehav TherExpPsychiat25:211-216.
Locker D, ShapiroD,LiddellA(1996a).Negativedental experi-ences andtheir relationship todental anxiety.Community DentHealth13:86-92.
LockerD, ShapiroD, Liddell A (1996b). Who is dentally anx-ious? Concordance among measures of dental anxiety. CommunityDentOral Epidemiol 24:346-350.
Locker D, ShapiroA,LiddellA(1997). Overlap betweendental anxietyandblood-injury fears: psychologicalcharacteristics andresponse todentaltreatment.BehavResTher35:583-590.
MarksIM,Gelder MG(1966). Differentagesofonsetinvarieties ofphobia.AmJPsychiat123:218-221.
MilgromP, WeinsteinP(1993). Dental fearsingeneral practice: new guidelinesfor assessment andtreatment. Int DentJ 43:288-293.
Milgrom P,WeinsteinP, KleinknechtR, Getz T(1985). Treating fearfuldentalpatients.Reston, VA: RestonPublishing Co. MilgromP, FisetL, MelnickS,Weinstein P (1988). The
preva-lence and practice management consequences of dental fearin amajorUScity.JAm Dent Assoc116:641-647. Milgrom P,Mancl L, King B, Weinstein P(1995). Origins of
childhood dental fear. BehavResTher33:313-319.
MooreR, Brodsgaard I,BirnH (1991). Manifestations, acquisi-tion and diagnostic categories of dental fear in a self-referred population. BehavResTher29:51-60.
OstL(1987). Age ofonsetof differentphobias.
J
AbnormPsychol 96:223-229.Ost L,HugdahlK(1985). Acquisition of blood and dental pho-bia andanxiety responsepatterns inclinicalpatients.Behav ResTher23:27-34.
Poulton R,Thomson WM, DaviesS, Kruger E, Brown RH,Silva P (1997). Good teeth, bad teeth and fear of the dentist. BehavResTher35:327-334.
Roy-Byrne P,MilgromP,Khoon-MeiT, Weinstein P, Katon W (1994). Psychopathology and psychiatric diagnosis in sub-jectswith dentalphobia.
J
AnxDisord8:19-31.Shoben E, BorlandA(1954).Anempirical study of the etiology of dentalfears.JClinPsychol10:171-174.
Speilberger C, GorsuchR, Luchene R (1983). Manual for the state-trait anxiety inventory. Rev. ed. Palo Alto, CA: Consulting PsychologistsPress.