International Dental Journal (2000) 50, 367-370

Further evidence for the reliability

Aim: To gain further evidence of the psychometric properties of the Dental anxiety is recognised as one
Modified Dental Anxiety Scale. Setting: Dental admission clinics. of the major barriers to attendance
Design: Consecutive sampling, cross-sectional survey. Participants: in those who require a dental visit’.
Patients (n=800) in four cities (Belfast, Northern Ireland; Helsinki, The assessment of dental anxiety has
Finland; Jyvaskyla, Finland and Dubai, UAE). Methods: Questionnaire encouraged the development of a
booklet handed to patients, attending clinics, for completion following an variety of measure^^.^. The majority
invitation by the researcher to be included in the study. Measures: of these are self report question-
Modified Dental Anxiety Scale (MDAS), together with further questions naires. The direct approach of
concerning dental attendance and nervousness about dental procedures. inviting individuals to state their own
Results: Overall 9.3 per cent of patients indicated high dental anxiety. rating has merit4. The flexibility and
MDAS showed high levels of internal consistency, and good construct ease of administration of self report
validity. The relationship of dental anxiety with age was similar to previous measures has resulted in a number
reports and showed lowered anxiety levels in older patients. of scales for measuring various
Conclusion: Data from three countries has supported the psychometric dental aspects of the anxiety
properties of this modified and brief dental anxiety scale. construct5-’. Measures such as
Klienknecht’s Dental Fear Survey
Key words: Dental anxiety, reliability, validity, cross-cultural studies have been advocated5.However, for
brief assessment Corah’s DAS has
often been employed. The Modified
Dental Anxiety Scale briefly extends
Corah’s scale by a single question.
Small but important changes to the
answering scheme were made. That
is, consistent wording for the
response categories across the five
questions was introduced. These
improvements have shown the
MDAS as a quick and efficient
instrument for dental researchers and
clinicians8. The scale invites the
respondent to indicate their anxiety
level to five dental situations includ-
ing: sitting in the waiting room,
having a scale and polish, a tooth
drilled and a local anaesthetic
Correspondence to: Dr. G.M. Humphris, Department of Clinical Psychology, Whelan
Building, The University of Liverpool, L69 3G6,UK. Email: [email protected] injection. The scale has shown

0 2000 FDlMlorld Dental Press

Table 1 Means and standard deviations of items and total scores for the Modified Dental Anxiety Scale with reliability estimates
(Cronbach alpha) from Belfast, Dubai, Helsinki and JyvBskyla samples including percent of samples above MDAS cut-off and those
who were “extremelyanxious”for each item

Site (n)
Belfast (200) Dubai (200) Helsinki (200) Jyvaskyla (194) All samples (794) per cent
“extremely
mean sd mean sd mean sd mean sd mean sd anxious”
visit tomorrow 2.43 1.42 1.66 0.82 1.72 0.91 1.81 1.00 1.91 1.11 5.3
waiting room 2.44 1.35 1.81 1.01 1.80 0.93 1.94 1.02 1.99 1.12 5.3
drill 2.92 1.48 2.29 1.28 2.25 1.12 2.51 1.23 2.49 1.31 12.7
scale and polish 1.90 1.35 2.27 1.08 1.87 1.04 1.84 0.99 1.96 1.09 3.0
injection 2.76 1.45 2.83 1.18 1.81 0.65 2.44 1.21 2.45 1.23 10.5
Modified DAS 12.40 5.98 10.90 4.28 9.44 3.91 10.54 4.65 11.27 5.07
Percent scoring 19 or > 19.50 6.00 3.00 8.80 9.30
Cronbach’s alpha 0.90 0.86 0.89 0.88 0.89

favourable psychometric properties tal approached the patients and tions about the patient’s age,
in comparison to the original Dental informed them of the research. The gender, past experience of the
Anxiety Scale of C0rah~2~~ and has questionnaire was handed to the dentist and frequency of dental
good specificity and sensitivity for patient for completion prior to visits was included. These ques-
dental phobia’. However the scale being seen by the dentist. All ques- tions were based upon the UK
has been limited to use in England, tionnaires were collected at each Adult Dental Health Survey.
Scotland and Wales. centre and keyed onto disc using The questionnaires used in
The aim of this study was to SPSS version 9.0 for Windows NT. Finland and Dubai were translated
determine the reliability and validity A template was constructed for this into Finnish and Arabic respectively.
of the Modified Dental Anxiety Scale purpose and supplied to each centre Back translations were performed at
in two other European countries and with coding instructions. Analysis the International Oral Health Unit,
the Middle East. was performed calculating Cronbach Helsinki to ensure accuracy and
alphas to inspect internal consistency comparability of translation.
and analysis of variance to examine
Methods
the influence of clinic centre on anxi-
Participants ety levels. Simple linear regression Results
Two hundred consecutive patients was applied to the total scale scores Some respondents refused to take
attending admission clinics in dental derived for each age group. Polyno- part. Reasons given were: insufficient
hospitals of four cities from three mial terms were fitted to model the time, attended clinic without specta-
different countries, were invited to relationship of dental anxiety to age. cles and dislike of completing forms.
complete the questionnaires. The The overall response rate was 79 per
four sites were: Belfast, Northern cent ranging from 69 to 98 per cent
Measures over the four centres. Reasons given
Ireland; Helsinlil and Jyvaskyla,
Finland; and Dubai, UAE. A brief Two questionnaireswere included in by the patients for attending the
description of the service at each the pack handed to the patient. These dentist did not vary across the c h -
hospital is detailed below. included: its (cht square = 4.19, 3df, P 0 . 2 ) .
Belfast - walk in, every day service Modified Dental Anxiety Scale Twenty eight per cent (range 23-31
and over-flow facility for the W A S ) * . This measure is simi- per cent) reported attending the
General Dental Service of the lar to the CDAS, but includes an dentist routinely, whereas the remain-
National Health Service. extra question about a local der went when they were in dental
anaesthetic injection, as well as a pain or trouble.
Helsinki - mainly patients who want The means and standard devia-
simplified and a consistent
to come for dental treatment. All
answering scheme across all five tions of the individual items and
come through the polyclinic and 25 the total scale scores (derived by
questions. Each question was
per cent are emergency patients. summing the individual item scores
answered by the patient indicat-
Jyvaskyla - a drop-in service for ing whether they were ‘not together) for the four samples are
emergency patients. Catchment area anxious’, ‘slightly anxious’, ‘fairly presented in Table 1. A previous
is mainly rural. anxious’, ‘very anxious’ or sensitivity/specificity analysis has set
Dubai - Patients attend for emer- ‘extremely anxious’. Simple Likert 19 or over as the cut-off score for
gency treatment on first come, first scoring (1-2-3-4-5) was assigned high dental anxiety. The prevalence
serve basis. Catchment area is mixed to these categories, with a high estimates for the four centres,
rural and urban. score denoting a high anxiety Belfast, Dubai, Jyvaskyla and
response. Items were summed to Helsinki were 19.5 per cent, 6.0 per
Administration of questionnaires derive the total score. cent, 8.8 per cent and 3.0 per cent
The researcher based at each hospi- A further sheet including ques- respectively (c2=36.6, df3, R.001).
International Dental Journal (2000) Vol. 50/No.6
 369

Table 2 Means and standard deviations of the Modified Dental of that reported previously (see
Anxiety Scale across self-reported dental attendance and levels of F@re 1). The older age groups (those
dental treatment nervousness (all samples)
______~ above 55 years) showed less anxiety.
mean sd n The polynomial curve fitted ( ~ 0 . 9 9 )
self-repoi3ed dental attendance
with less error than simple linear
“did go in the last year” 10.28 4.64 506 regression ( ~ 0 . 9 1 and
) described an
“did not go in the last year” 11.81 5.13 278 inverted ‘U’ relationship.
nervousness (I am nervous of some kinds of dental treatment)
“don’tfeel like that” 9.63 4.06 378
“to some extent“ 10.53 3.36 202 Discussion
“definitely feel like that” 13.26 6.43 206 This study provides further evidence
to support the reliability and validity
[7.2] [15.0] [15.6] [12.7] [4.8] [2.6] of the Modlfied Dental Anxiety Scale
using samples drawn from outside
131 T England, Scotland and Wales. The
internal consistency of the scale (a
12- = 0.89) is comparable to previous
reports (range of alphas from 0.84
Modified to 0.90)8. Of interest is the stabllity
Dental 11- of the size of the internal consist-
Anxiety ency coefficients across the four
Scale 10- samples. The reliability of the scales
appears to be independent of anxi-
ety level. The significant associations
9- with self reported attendance and
a single item assessing respondent’s
8- ‘nervousness’ about dental treat-
ment supports the scale’s construct
Validity.
A limitation of the study is the
16-25 26-35 36-45 46-55 56-65 65+ samplmg of the clinics that partici-
pated. Ideally, a random selection
Age in years procedure of all admission clinics
Figure I . MDAS mean scores (with 95% confidence intervals) and prevalence rates in w i h n the countries involved would
brackets across age groups of whole sample (n = 597) with polynomial curve fitted (y = assist in providing representative
10.73 + 0 . 9 3 ~- 0 . 2 2 :~r~= 0.99) estimates of dental anxiety. However,
the reasons given for attending the
dentist were similar across the
The rank order of anxiety for the self-reported dental attendance and clinics, hence some evidence was
five items in each of the centres was the individual’s rating of nervous- found for comparability of the
not identical. In Belfast,Jyvaskyla and ness about visiting the dentist in participants’ behaviour in this study.
Dubai the injection was found to be general. The predicted results of These results provide some of the
the most anxiety provoking item, lower anxiety levels with less frequent first direct comparisons across a
whereas in Helsinh the most fearful dental visiting (t[782] = 4.24, number of international boundaries
procedure was the use of the drill. A P=O.OOl) and being less nervous using an identical questionnaire.
visit to the dentist the next day was about some kinds of dental treat- The overall level of h g h dental
rated as the least anxious aspect of ment (F[2,783] = 41.29, P=O.OOl) anxiety (or dental phobia”) was 9.3
the five scenarios among the was confirmed (Table 2). per cent. Studies using the Corah
respondents from Jyvaskyla, Dubai The relationship of dental anxi- Dental Anxiety Scale give prevalence
and Helsinki. The Belfast sample ety with age was tested by analysis of rates as follows: 10.9 per cent in
rated a scale and polish as the least variance with age group (sample Ontario, Canada’, 10.2 per cent in
anxiety provohng procedure. divided into five, 10-year categories Seattle, USA”, 6.7 per cent in
The internal consistency coeffi- ranging from 16 to 65 years of age SwedenI3 and 4.2 per cent in
cients (Cronbach alphas) ranged from with a further category of patients N o r ~ a y ’ ~Caution
. is required in
0.86 to 0.90 for the four samples. aged 66 years or above) as the comparing these ‘prevalence rates’.
The overall reliability for the total between subjects factor. Dental anxi- The MDAS and CDAS are closely
sample was 0.89. Validity was ety was found to vary significantly related ( ~ 0 . 8 5 but
) ~ the cut-off for
checked by investigating the relation- with age (F[5,788] = 7.35, P=O.OOl). the CDAS was determined by clini-
ship of the dental anxiety ratings to The relationship was characteristic cal recommendation”’as opposed to

Hurnphris eta/.:The Modified Dental Anxiety Scale

the MDAS whch was scientifically by Corah of 13 or over to indicate Hoogstraten J, e t al. Appraisal of
determined” Moreover, the patients those who were dentally anxious. behavioural measurement techniques
who participated in this study were Furthermore, they reported those for assessing dental anxiety and fear in
attending admission c h c s for emer- scoring 15 or over as highly children: A review. J Psychopathol Behav
gency dental care in their respective anxious15. Our cut-off for the Assess 1996 18: 153-171.
3. Kleinknect R, McGlynn F D ,
countries. Reports of dental anxiety MDAS at 19 or over refers to dental
Thorndike R M, e t al. Factor analysis
levels for these participants have not phobic respondents. Some authors
of the dental fear survey with cross-
been previously published. It advocate the use of multiple ques- validation. ] Amer Dent Assoc 1984 108:
appeared that there were differences tionnaires and combined cut-off 59-61.
in dental anxiety prevalence between levels’. Locker’s recent study uses a 4. Corah N L. Methodological needs and
the three countries: Northern complex set of cut-offs”. For exam- behavioural research with adult dental
Ireland, United Arab Emirates and ple they specify a cut-off of 12 or patients. Anaestheia Pmg 1986 3 3 4 6 4 9 .
Finland. The two admission centres over for the CDAS which is lower 5. Schuurs A H, Hoogstraten J. Appraisal
in Finland (Jyvaskyla and Helsinki) than previous recommendations, as of dental anxiety and fear question-
have different dental anxiety levels. well as two further scales and their naires: a review. Comm Dent Oral Epide-
The patients from the centre in respective cut-off levels. It is recog- mioll993 21: 329-339.
Helsinlil were attending from an nised that multiple measures of 6. Kent G, Rubin G, Getz T, et al. Devel-
opment of a scale to measure the
urban setting, and the majority dental anxiety have a part to play in
social and psychological effects of
arrived in the facility already with studying what is a multifactorial severe dental anxiety: Social attributes
some experience of the services psychological condition, however of dental anxiety scale. Comm Dent
provided at the dental hospital. there may still be advantage (clinical Oral Epidemiol 1996 24: 394-397.
Comparisons with other reports based studies, where time with a 7 . Locker D , Shapiro D, Liddell A. Who
using the CDAS, mindful of the patient is at a premium) to advocate is dentally anxious? Concordance
cautions expressed already, suggests the use of a single measure. The between measures of dental anxiety.
that the Scandinavian respondents MDAS may function as an assess- Comm Dent Oral Epidemiol 1996 24:
report lower levels of dental anxiety ment to assist initial comparisons 346-350.
than English speaking participants. although further work is required to 8. Humphris G M, Morrison T, Lindsay
The relationshtp of dental anxi- validate the stated cut-off and deter- S J E. The Modified Dental Anxiety
Scale: UK norms and evidence for
ety and age is similar to a previous mine the relationships of the MDAS
validity. Communily Dent Hlth 1995 12:
report adopting the MDAS8. The with other available measures of 143-150.
pattern of h g h scores for younger dental anxiety. One possible advan- 9. Corah N L. Development of a dental
individuals in comparison to older tage of using the MDAS over Corah’s anxiety scale. J Dent Res 1969 48: 596.
individuals (55 years and above) is scale is the translation of the response 10 Corah N L, Gale E N, Illig S J. Assess-
consistent with a report from Canada categories. The MDAS requires ment of a dental anxiety scale. ] Amer
using Corah’s Dental Anxiety Scale’O. translating into the native language Dent Assoc 1978 97: 816-819.
They found 11 per cent of their various (ordinal) levels of anxiety, 11. Freeman R. A psychodynamic theory
50-64 year old respondents and 4 whereas Corah’s scale demands the for dental phobia. Brit Dent] 1998 1 8 4
per cent of their 65-74 year old searching of a greater variety of 170-176.
respondents were dentally anxious words (for example, tense, anxious) 12. Milgrom P, Fiset L, Melnick S,
(overall rate = 8.4 per cent). Our to match Corah’s original categori- Weinstein P. The prevalence and prac-
tice management consequences of
sample showed a similar but less sation. The comparison between
dental fear in a major US city. J Amer
strikingreduction in anxiety level with different translations may be Dent Assoc 1988 116: 641-647.
increasing age level as shown by the conducted with greater confidence 13. Hakeberg M, Berggren U, Carlsson S
56-65 and 66+ year age groups (4.3 using the MDAS. G. Prevalence of dental anxiety in an
per cent and 2.3 per cent highly The translations of the MDAS adult population in a major urban area
dentally anxious respectively). The are available in Arabic and Finnish. in Sweden. Comm Dent Oral Epidemiol
Canadian sample (n=580) was drawn The MDAS has also been translated 1992 20: 97-101.
using a telephone interview proce- into Hind, Russian, Estonian and 14. Vassend 0.Anxiety, pain and discom-
dure and a low response rate (28 per Swahili. New translations and fort associated with dental treatment.
cent) was achieved. Locker’s group normative data from dental research- Behav Res Therapy 1993 31: 659-666.
ers internationally will be gratefully 15. Locker D , Liddell A, Burman D.
has reported a further general
Dental fear and anxiety in an older
population sample survey of dental received.
adult population. Comm Dent Oral
anxiety from the Ontario area’. Epidemiol 1991 19: 120-124.
Prevalence rates for the 50-69, and 16. Schuurs A, Hoogstraten J. Appraisal
70+ year age groups were 9.5 and References
of dental anxiety and fear question-
7.5 respectively (overall rate = 10.9 1. Finch H, Keegan J, Ward K. Barriers to naires: a review. Commun Dent Oral
per cent). the receipt of dental care - a qualitative Epidemioll993 21: 329-339.
The issue of where to set a cut- research study. Social and Community 17. Locker D, Liddell A, Dempster L, e t al.
off point is controversial“. Locker et Planning Research: London; 1988. Age of onset of dental anxiety. J Dent
a l l 5 used the recommendation made 2. Aartman I, van-Everdingen T, Res 1999 78: 790-796.

International Dental Journal (2000) Vol. 501No.6