|
|
||||||||
|
J Am Dent Assoc, Vol 131, No 10, 1449-1457.
© 2000 American Dental Association | ![]() |
CLINICAL PRACTICE |
| ABSTRACT |
|---|
|
|
|---|
Type of Studies Reviewed. Three computerized databases of published scientific literature were searched over a 10-year period. Only studies that included measures of anxiety or pain were included.
Results. Information on the reliability and validity of 15 measures of dental care anxiety and three measures of pain and pain-related behaviors is provided. Reliability and validity data for most measures are good. Corahs Dental Anxiety Scale is the most widely used measure of anxiety, although it may not be as sensitive as other measures. The McGill Pain Questionnaire is the measure of choice for the assessment of pain.
Clinical Implications. The authors have summarized properties of the scales for clinicians and researchers planning to use measures of anxiety, measures of pain, or both.
The behavioral sciences have become an increasingly important component of dental education and research.1,2 One component of this has been the application of psychological methods to the study of behavior and attitudes relevant to health, illness and health carein particular, fear of dentists and dentistry as well as of dental pain. This has included a wide range of methodological approaches and techniques, especially the use of questionnaires and behavioral measures. Several authors have emphasized the importance of ensuring that such measures are reliable, valid and applicable to the population toward which they are aimed.3,4 We seek to provide an overview of measures of anxiety and pain in dental research during a 10-year period and an appraisal of the psychometric properties of the measures used. We hope that this will serve two functions: to help researchers choose suitable measures when undertaking studies of behavior in dentistry, and to identify areas in which researchers might work to develop new psychometrically robust measures.
Fear of dentists and dentistry is a common and potentially distressing problem, both for the public and for dental practitioners. The most recent adult dental health survey conducted in the United Kingdom found that about 25 percent of adults are sufficiently fearful of dentistry to avoid or delay attendance.5 To date, there has been no comprehensive published review of the measures of anxiety used in dental research. Although Schuurs and Hoogstraten6 compared six measures of dental anxiety, they excluded measures that have been used with children, and made no attempt to determine the extent to which scales had been adopted for use by researchers. Locker and colleagues7 did make a direct comparison of three questionnaire measures of adult dental anxiety in a Canadian population. However, theirs was not a comprehensive review of available measures. Similarly, there is little published evidence on which to compare the utility of measures of pain in dental settings.
Search strategy.
A complete description of the search mechanism used is found elsewhere.8 We searched the MED-LINE, EMBASE and SSCI databases for a 10-year period from July 1988 through June 1998. For the purposes of this article, only studies that included measures of anxiety or pain will be revieweda total of 43 reports. We identified a total of 15 questionnaires in 38 articles that measured dental anxiety and three questionnaires in eight articles that measured dental pain and pain behaviors.
Data gathered.
We gathered specific information on each scaleinsofar as it was possibleusually from sources referenced in the articles. Articles, books and other materials containing this information that were published before our review period were included to obtain a complete picture of the qualities of the scale. Furthermore, cited articles were not restricted to the dental literature and included research in medicine and psychology.
The information we collected on each scale was as follows:
For some scales, we were unable to obtain complete information; in some cases, the authors had not assessed a particular property of the scale (for example, alternate forms of the questionnaire) or the evidence was published in a source that was not readily available (for example, a doctoral thesis). In the descriptions of scale reliability and validity, the terms "satisfactory" and "high" are used to refer to the following criteria. For internal consistency, measured by Cronbachs
![]()
MATERIALS AND METHODS
TOP
ABSTRACT
MATERIALS AND METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
We performed computerized searches of the dental literature published in a 10-year period to identify all studies in which questionnaire measures of psychological constructs had been included. We reviewed the articles to identify the scales that had been used. We then examined the primary sources to identify the psychometric properties of the measures identified.
, satisfactory is used when the value of
is greater than .60; high is used when
is greater than .80. For test-retest correlations, a value of r > .80 is rated as satisfactory and a value of r > .90 as high. For interscale correlations used to determine the validity of a scale, r > .60 is considered satisfactory and r > .80 is considered high.
The use of questionnaire measures of anxiety in dentistry is well-established.
![]()
RESULTS
TOP
ABSTRACT
MATERIALS AND METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
Measures of dental anxiety.
We identified a total of 15 measures used to measure dental anxiety. These can be subdivided into the following categories: specific measures of dental care anxiety in adults, specific measures of dental anxiety in children and general measures of anxiety that have been used in the dental setting. The use of questionnaire measures of anxiety in dentistry is well-established, as evidenced by the number of articles identified. Table 1
provides a summary of the measures identified.
|
Corahs DAS has been modified by the addition of a fifth item that asks about responses to administration of local anesthetic and by a change in the response format.16 The modified scale has high levels of reliability and validity. Mean scores are provided for phobic and nonphobic patients. Corahs DAS is widely used, but has been criticized as exhibiting a range of total scores that is too narrow to be used effectively in clinical studies6; by increasing the number of possible responses from four to five and by introducing an additional item, the modified DAS helps to rectify this problem.
Kleinknechts Dental Fear Survey.17 Kleinknechts Dental Fear Survey asks respondents to rate their anxieties about 27 specific situationssuch as making an appointment or hearing the dental drillon a five-point Likert scale ranging from "none" to "great." Three dimensions of the questionnaire have been derived from factor analysis: avoidance of dental treatment, somatic symptoms of anxiety and anxiety caused by dental stimuli. These factors are reliable and stable across different groups of respondents.18 Data are available on the internal consistency of the scale, test-retest reliability and validity of the questionnaire, all of which are satisfactory.
Dental Anxiety Question.19 The Dental Anxiety Question, or DAQ, is a single-item construct: "Are you afraid of going to the dentist?" It has four possible responses: "no," "a little," "yes, quite," "yes, very." These responses are scored from 1 to 4 in the direction of increasing anxiety. This question also has been used with a five-point response scale.7,20 The DAQ correlates well with Corahs DAS (r > .71 in studies of adult and child populations7,14,19). Single-item inventories have been regarded with skepticism by scale developers because they do not provide opportunities to control for response-set bias (such as the tendency to give responses that the participant believes are "correct"), and because they do not allow for the isolation of components of multidimensional constructs. However, for some purposes, such as screening people who are likely to be highly anxious about dental treatment, it is a useful and brief tool,21 although it has a tendency to overestimate the prevalence of severe dental anxiety.7
Gatchels 10-Point Fear Scale.22 This is a single-item scale that asks respondents to rate their dental fear on a 10-point scale. A score of 1 indicates no dental fear; 5, moderate fear; and 10, extreme fear. A score of 8 or greater is considered to indicate a significant degree of anxiety; approximately 11 percent of a North American sample scored at this level.22 Agreement between the Gatchel scale and Corahs DAS in the identification of people with significant dental fear is only moderate, suggesting that the two measures are related but tap independent dimensions.7 The Gatchel scale reflects an overall fearfulness of dental treatment, whereas Corahs DAS measures a more specific fear of the dental situation (for example, sitting in the chair, having treatment performed).
Photo Anxiety Questionnaire.23 This questionnaire asks respondents to imagine they are to undergo dental treatment and consists of 10 items illustrating different moments before, during and after treatment. The response scale is nonverbal, consisting of five photographs with facial expressions of differing anxiety levels. There are different versions for men and women. Patients indicate which photograph best expresses their feelings. The scale is scored by summing the item ratings (1 = relaxed, 5 = very anxious).
The internal consistency of the scale is high, and it correlates well with Corahs DAS.23 There are potential advantages to a response format that does not rely on verbal markers: problems of differences in the interpretation of words and phrases do not arise, andprovided that the facial expressions depicted are universaldifficulties in translating responses do not arise. De Jongh and Stouthard24 used this scale to analyze the degree of anxiety experienced by patients receiving treatment from a dental hygienist.
Dental Anxiety Inventory.25
The Dental Anxiety Inventory, or DAI, is a 36-item scale. Answers are given on a five-point scale, and scores range from 36 (no anxiety) to 180 (high anxiety). The internal consistency of the scale is high (
= .95 for a student population, and .98 for a general population sample). The test-retest correlation during a two-week period also was satisfactory (r = .87). The scale correlates with Corahs DAS (r = .87), although the factorial validity of the scale is unsatisfactory. The data did not fit well with either a three- or four-factor solution. Schuurs and Hoogstraten6 concluded that the validity of the DAI must still be determined.
There are potential advantages to a response format that does not rely on verbal markers: problems of differences in the interpretation of words and phrases do not arise.
Scales for children and adolescents. Childrens Fear Survey Schedule.26 The Childrens Fear Survey Schedule, or CFSS, is designed to assess a range of general fears in children. A dental subscale has been devised (CFSS-DS27) that consists of 15 items rated on a five-point scale, ranging from 1 (not afraid) to 5 (very afraid). Scale scores are calculated by summing item scores; the total score can range from 15 to 75. Scores above 38 indicate significant dental fear.28 Both the full-scale and dental subscale are internally consistent.26,27 The CFSS-DS has been found to disciminate between children who do and do not display dental fear and behavioral problems during dental treatment.28
Venham Picture Scale.29 This scale consists of a series of eight paired drawings of a child. Each pair consists of a child in a nonfearful pose and a fearful pose (for example, running away). The respondent is asked to indicate, for each pair, which picture more accurately reflects his or her feelings at the time. Scores are determined by summing the number of instances in which the child selects the high-fear stimulus. To our knowledge, there is no published information about the reliability and validity of the scale.
Venham Anxiety and Behavior Rating Scales.30 These two scales assess the anxiety and uncooperative behavior of children in the dental setting. Both scales consist of five behaviorally defined categories ranging from 0 to 5, with higher scores indicating greater levels of anxiety or lack of cooperation. The criteria used to assign the scores are described by Venham and colleagues.30 Using the method of paired comparisons, they were able to ascertain that the points on the scale can be treated as interval data. A high degree of interrater reliability was found for both scales, even for untrained observers. Alwin and colleagues31 found that the scale was able to distinguish between children referred to a specialist treatment center for management of uncooperative behavior in dental treatment and a control group of children.
Adolescents Fear of Dental Treatment Cognitive Inventory.32
This scale measures the thoughts and ideas an adolescent may have during dental treatment. It is unique among the childrens scales in that it focuses solely on the cognitive manifestations of fear. It is a 23-item scale, with a five-point response format. Scores range from 23 (no fear) to 115 (high fear). The scale shows high internal consistency (
= .91) and moderate test-retest reliability over a one-week period (r = .84). Factor analysis of the scale revealed four components: fear of pain, negative perceptions of the dentist, avoidance of the dentist and a fourth scale that was not interpreted by the researchers. The scale has been cited only in its original development study. Despite the advantages of assessing the cognitive dimensions of anxiety, there is insufficient evidence to support the usefulness of this scale.
Behavior Profile Rating Scale.33 This scale consists of 27 uncooperative behaviors considered to be related to dental anxiety. The behavior of the child in the dental setting is observed for three-minute intervals throughout the 30-minute consultation, and each item is scored if the behavior occurs at the point of observation. An overall score is calculated on the basis of the frequency of each behavior, together with a weighting for the severity of the behavior (for example, kicking is perceived to be more severe than oral complaints). The scale has adequate interrater reliability, given adequate training of observers. The face validity of the scale is high, and it has been found to distinguish between children referred for behavioral management of uncooperative behavior in the dental setting and a control group of children.31 As with many behavioral measures, use of this scale is likely to be time-consuming.
Spielbergers State-Trait Anxiety Inventory distinguishes between anxiety as a general aspect of personality and anxiety as a response to a specific situation.
General adult scales. Spielbergers State-Trait Anxiety Inventory.34 Spielbergers State-Trait Anxiety Inventory, or STAI, distinguishes between anxiety as a general aspect of personality (trait anxiety) and anxiety as a response to a specific situation (state anxiety). It consists of 40 statements, 20 of which measure trait anxiety and 20 state anxiety. Items are scored on four-point scales, with response categories varying according to the nature of the question. This questionnaire has been tested extensively for reliability and validity.34 A six-item version of the state scale has been devised that is reliable and valid and yields scores that are similar to those of the full version.35
Fear Survey Schedule.36 The original Fear Survey Schedule is a 51-item scale consisting of a list of commonly encountered stimuli. Each stimulus is rated on a seven-point scale, ranging from 0 (no fear) to 6 (terrified). Scores indicate the extent of general fearfulness, which Geer,36 who developed the scale, suggests is a personality characteristic. A shortened form (18 items) has been developed and was used in three of the studies identified in the literature search.3739 Both the full-length and shortened scales have been subject to extensive psychometric analyses, and both display high levels of internal consistency and validity.36,37 Scale scores vary between men and women (mean scores for women are higher). The scale has been used in a number of European countries, including the United Kingdom, Norway, Sweden and other Scandinavian countries.
Weiner Fear Questionnaire.40 The Weiner Fear Questionnairre has two parts. Part A consists of 16 questions with a five-point answer format seeking to reveal respondents general and dental care fears. Part B consists of 18 questions about autonomic stress reactions, and has a five-point answer format; three of the questions in this part address severe anxiety attacks. Few data are available on the reliability and validity of the scale; the only published study using this scale was the one for which the scale was devised. People with significant levels of dental care anxiety were included in the sample used to develop the scale.
Measures of dental pain. We found three questionnaires (in eight articles) that have been used to assess dental pain. These are the McGill Pain Questionnaire, the West Haven-Yale Multidimensional Pain Inventory and the Pain Anxiety Symptoms Scale.
The McGill Pain Questionnaire.41 This questionnaire was designed to assess three components of reported pain: the sensory, affective and evaluative components. Respondents indicate the location of their current pain by marking an area on a drawing of a human figure. They then are asked to choose the words that best describe the pain from a list of 78 adjectives (such as flickering, sickening). The adjectives are grouped into 20 subclasses that describe different aspects or types of pain. A third part of the questionnaire assesses how the pain changes over time and what relieves it or increases it. The final part is a single measure of pain intensity. This scale has been used in a variety of settings for the assessment of pain from a variety of origins.
Grushka and Sessle42 reported that the McGill Pain Questionnaire could discriminate between pain originating from a reversibly inflamed tooth pulp and pain resulting from an irreversibly inflamed or necrotic tooth pulp. Similarly, Turp and colleagues43 found that the McGill Pain Questionnaire distinguished facial pain from other non-orodental pain. These authors suggested that it should be used as an adjunct to simpler clinical assessments of tooth pain (such as asking a patient about his or her pain). Using the McGill Pain Questionnaire, Beese and Morley44 found that memory in regard to dental pain was inaccurate over a two-week period.
The West Haven-Yale Multidimensional Pain Inventory.45
Reisine and colleagues46 used this scale as part of a battery of tests to compare groups of patients receiving dental treatment with those not receiving treatment. The scale exhibited high levels of internal consistency (
= .87) and was found to discriminate between these groups. We found no other dental research studies in which this scale has been used during the 10-year period studied.
The Pain Anxiety Symptoms Scale.47 This scale is not a simple measure of pain experience, but focuses on the cognitive aspects of pain. It consists of three subscales: avoidance of pain, acceptance of pain and fear of pain. Reliability analysis of all subscales shows high levels of internal consistency, and the scale correlates well with Corahs DAS. Scores within the general population are higher for women.48
| DISCUSSION |
|---|
|
|
|---|
|
Because children may not have a fully developed ability to recognize and interpret the physiological and cognitive manifestations of anxiety, measures of dental fear in children have tended to concentrate on the behavioral component of fear or have used non-verbal tools such as pictures. Behavioral measures rely heavily on the training of observers to ensure interrater reliability and are cumbersome to apply in practice. Pictorially based measures such as the Venham Picture Scale are rapidly administered, reliable and understandable to a broad age range. While these present a potentially useful approach to the measurement of anxiety in children, more work is necessary to establish their reliability and validity, as Table 2
illustrates.
General measures of anxiety in adults are useful in that they allow comparison of dental anxiety with other fears. The Spielberger measure34 has been used widely in both dental and nondental settings, and is available in a short form that could be useful in both research and clinical settings.
Pain routinely is measured by visual analogue scales. Although these provide a useful method of describing pain experience, they do not assess the multidimensional nature of pain. More sophisticated measures include analysis of the sensory, affective and cognitive components of pain. In our review, we found the McGill Pain Questionnaire to be the most widely used pain scale. In a review of the literature on pain measurement, Zakrzewska and Feinmann50 suggested that it is the measure of choice.
Any conclusions drawn from this study should be tempered by consideration of its limitations. It seems likely that some measures have been missed in the computerized literature search, despite the extensive search strategies adopted. Furthermore, there may have been developments in this area of research since 1998, the last year of our review. Still, this report provides a fairly extensive review of the scales used to measure anxiety, fear and pain in dental settings.
As our review demonstrates, there is a wide variety of measures available to assess levels of dental anxiety. Most of them exhibit adequate levels of internal consistency, and correlate with other measures of the same construct. Choice of a particular measure will depend, in part, on the purpose for which the measure is intended and on the particular aspects of dental fear that are being assessed. Longer measures (that is, those with more questions) tend to provide a wider range of scores and, therefore, would be expected to be more sensitive to change over time or to variation between groups (for example, the DAI or Kleinknechts Dental Fear Survey).6
Other scales (for example, the modified Corahs DAS) ask respondents to rate particular dental situations and so might be of value in planning interventions aimed at alleviating dental anxiety, such as systematic desensitization.51 In general, fear is expressed as changes in thoughts, physiological state and behavior. The scales reviewed above place different degrees of emphasis on these components; most scales for adults are concerned with thoughts, and many scales for children are concerned with behavior. Corahs DAS provides a useful and short measure of dental anxiety. It has been used widely and extensive comparison data are available.
There are fewer scales available to assess pain. The McGill Pain Questionnaire emerges as a psychometrically sound measure with the ability to discriminate between pains of different origin. It has been used widely in both dental and medical settings. Although it may be lengthy for use in the clinical setting, components such as the single-item measure of pain intensity could be used in isolation.
Researchers interested in the areas of dental anxiety and dental pain should consider carefully the measurement of these phenomena. Existing measures vary widely in their length and content. Both anxiety and pain are multidimensional constructs, and it is important to identify which dimensions are being assessed. Furthermore, the use of standardized instruments provides the opportunity for comparison of data across groups, experimental manipulations and treatment approaches.
| CONCLUSIONS |
|---|
|
|
|---|
Studies are needed to measure the impact of dental anxiety and dental pain on behavior in adults. In addition, more scales of anxiety in children are needed that assess the cognitive manifestations of anxiety. Finally, further research is required to determine the reliability and validity of measures of dental anxiety in children.
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G.M. Humphris, K. Milsom, M. Tickle, H. Holbrook, and A. Blinkhorn A new dental anxiety scale for 5-year-old children (DA5): Description and concurrent validity Health Education Journal, January 1, 2002; 61(1): 5 - 19. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |