|
|
||||||||
Milieu in Dental School and Practice |
ar, M.D.
ener Büyüköztürk, Ph.D.
Key words: dental hygiene fear, anxiety, DHFS, reliability, validity
Submitted for publication 06/12/07; accepted 12/01/07
| Abstract |
|---|
|
|
|---|
Up to the present, anxiety has almost always been studied in relation to various dental treatment modalities.7–11 However, little research has been done in the area of fear as it relates to dental hygiene treatment. The anxiety-provoking character of this kind of treatment seems to have been systematically underestimated. There are a number of indications suggesting that anxiety and anticipation of pain are also relevant factors in dental hygiene care.1,12–14 Nevertheless, there have been only a few studies investigating the stimuli that contribute to anxiety for dental hygiene care. In a study in which anxiety and its relationship to regular dental treatment and dental hygiene treatment were investigated, 85 percent of the participants reported feelings of anxiety during dental hygiene treatment.1 Feelings of anxiety were greater in response to dental hygiene treatment than to dental treatment. The authors concluded that prevalence of anxiety for dental hygiene treatment is at least comparable to that found in patients receiving other dental treatments.
In our study, the differences between cultures and the specified effect of the procedure itself (not the effect of the dental hygienist versus the dentist) were investigated. Unlike in the United States and many European countries, where oral health care is provided by dental hygienists as well as dentists,1,13 the health system in Turkey does not involve any professionals equivalent to dental hygienists, and dental hygiene procedures are performed by dentists only. However, it would be logical to hypothesize that dental hygiene treatment, regardless of whether performed by a dental hygienist or a dentist, may possibly evoke a different fear response than for dental procedures.
Although it might be relevant to find out the exact role of anxiety as it relates to dental hygiene care, a thorough review of the literature reveals that there are only a few scales that have specific items about dental hygiene treatment. In the original version of the Corah Dental Anxiety Scale (DAS) the last (fourth) question refers to the dentist using cleaning instruments.15 Although the Ronis revised version of the DAS, the DAS-R,16 incorporates a more contemporary view of oral health care practice, aiming to acknowledge the roles of dental hygienists and female dentists by changing the third question to be gender-neutral and the last question to refer to either the dentist or the dental hygienist, it does not capture dental hygiene fear specifically. To our knowledge the Dental Hygiene Fear Survey (DHFS) is the only scale that specifically addresses the issues related to patients reactions to dental hygiene treatment procedures.
The DHFS was developed based on the theoretical model of dental fear that has been suggested by researchers at the University of Washington (UW) Dental Fears Clinic.11 The UW model includes four subtypes of fear: fear of specific stimuli, distrust of dental personnel, generalized anxiety, and fear of catastrophe. In the original study, the DHFS was administered to a convenience sample (n=300) of dental hygiene patients. It was demonstrated that the instrument was a reliable measure having a high internal consistency with an overall alpha coefficient of 0.92.13 The alpha coefficients of the specific dental hygiene stimuli, generalized anxiety, distrust, and fear of catastrophe subscales were 0.85, 0.79, 0.75, and 0.68, respectively. It was further determined that the DHFS behaves in a similar manner as dental fear surveys in that relationships among variables such as age, gender, and avoidance of dental care are consistent.13
The original study was limited to testing the criterion validity of the instrument by investigating the relationship of the scale scores with age, gender, and the avoidance of dental care. Results of the study showed higher DHFS scores in females, younger patients, and individuals who avoid dental care, which is consistent with the dental fears literature. These results can be accepted as the initial validation of the scale; however, further studies are needed to more comprehensively examine the validity of the instrument in subsequent trials involving different settings and types of patients.
Therefore, the aims of our study were to 1) investigate patient anxiety as it relates to dental hygiene procedures and 2) conduct validity and reliability testing of the Turkish version of the DHFS to test its psychometric properties in the Turkish culture. This study serves as the first investigation to examine the factorial construct of the DHFS and provides an example of how to test the reliability and validity of a survey instrument by demonstrating the types of statistical analyses performed.
| Methodology |
|---|
|
|
|---|
|
For purposes of our study, the DHFS was translated into Turkish. Translation and back translation of the DHFS were done by two independent researchers fluent in both Turkish and English. The Turkish version of the scale was re-examined by the authors (a periodontist, a psychiatrist, and a statistician) and administered to a pilot group for final corrections to ensure the clarity of the items in the questionnaire.20 It should be noted that "the dental hygienist" phrase in the original version of the scale is interchangeable with "the dentist" because the dental hygiene procedure is carried out by dentists and the oral health care team does not include any dental hygienists in Turkey. Although the original form of the DHFS was developed to explore whether patients reactions and fear levels were different when they were seeking services from a dental hygienist versus a dentist,13 the aim of our study was to explore patients reactions and fear levels when seeking preventive services from a dentist. The modified Turkish DHFS questions appear in the appendix.
The survey utilizes a Likert rating scale that ranges from 1 to 5 for each item with 1="not at all" to 5="very much."13,14 Questions 3, 5, 14, and 15 relate to fear of specific dental hygiene stimuli such as scaling instruments. Questions 4, 8, 9, and 11 address distrust of the dentist provider. Generalized anxiety related to the appointment and treatment is elicited through questions 7, 10, 12, and 16. Finally, questions 1, 2, 6, and 13 relate to fear of catastrophe. Scores on the DHFS can range from a minimum of 16 to a maximum score of 80.
The Modified Dental Anxiety Scale (MDAS), first introduced by Humphris et al.,21 is a self-rating instrument with questions being scored from not anxious to extremely anxious in an ascending order from 1 to 5. Each question thus carries a possible maximum score of 5 with a total possible maximum score of 25 for the entire scale. The psychometric properties of the Turkish version of the scale, which we used in our study, were investigated by Tunc et al.22 and Ilguy et al.,23 and the Turkish version of the MDAS was demonstrated to be an internally consistent measure in both studies with an overall Cronbach alpha coefficient of 0.91 and 0.96, respectively. In Tunc et al.s study, a cut-off point of
15 in distinguishing patients with significant dental anxiety was considered to be favorable. This cut-off point reveals a sensitivity and specificity of 0.80 and 0.74, respectively, which can be considered as acceptable, although the specificity is rather low.22
Personal characteristics of study participants were determined with questions about age, gender, education level, and self-reported oral health status (Likert type scoring from 1 to 5, stated as "very bad" to "very good"). The patients were also asked to score their subjective dental fear and subjective dental hygiene fear on a visual analog scale in an ascending order from 1 to 10.
All participants were asked to fill in three different questionnaires in the following order: DHFS, MDAS, and the sociodemographic, dental, and medical history form designed by the authors. Additionally, for further analysis of the results, we used a cut-off MDAS score
15, which was reported to be a favorable value by Tunc et al. to distribute the patients into two groups as having low and high dental anxiety.22
Statistical Analyses
Statistical analyses were conducted by using the software Statistical Package for the Social Sciences (SPSS), Version 11.0. To examine the factorial construct of the scale in the Turkish culture, exploratory factor analysis (EFA) was performed by using principal components analysis with direct oblimin rotation.
Factor analysis is a technique used to identify factors that statistically explain the variation and covariation among measures (for purposes of this study the sixteen items in the DHFS). Factor analysis can be viewed as a data-reduction technique whereby the number of factors is smaller than the number of measures. Ideally, the factors will correspond to constructs (i.e., Milgroms hypothesized types of fear) of a theory that helps us understand behavior. Factor analysis requires two stages: factor extraction and factor rotation.
Principal components analysis is a factor extraction technique used with the primary objective of making an initial decision about the number of factors underlying a set of measures. To determine the number of factors to extract, the researchers considered factors with eigenvalues >1. Eigenvalue is defined as the sum of squares of the factor loadings of each factor. It indicates the proportion of variance explained by that factor. The larger the eigenvalue, the more variance is calculated. The second stage, factor rotation, is carried out to make the factors more meaningful as unrotated factors are typically not very interpretable. The rotated factors may be uncorrelated (orthogonal) or correlated (oblique). Direct oblimin rotation is a type of oblique rotation method that was carried out. This method simplified the factors by minimizing cross products of loading.24,25
The significance of differences in the scale and subscale scores with respect to gender and between the low and high anxiety groups was determined by independent samples t-test. The correlations between the scale and subscale scores and the other study variables such as age, self-reported oral health, subjective dental fear, etc. were determined by the Pearsons correlation coefficient (p<0.05). Test-retest reliability was calculated using intraclass correlation coefficient, and Cronbach alpha value was estimated for evaluating the internal reliability of the scale.
| Results |
|---|
|
|
|---|
value when item 11 was deleted increased from 0.9076 to 0.9104.
Therefore, we reanalyzed the oblimin rotation by omitting item 11. The exploratory factor analysis excluding item 11 revealed a model with three factors having eigenvalues >1, explaining 45.1, 10.9, and 6.8 of the variance, respectively, which together could explain 62.8 of the total variance. Seven items comprising the fear of specific dental hygiene stimuli and distrust subscales of the original DHFS except item 11 (3, 4, 5, 8, 9, 14, and 15) loaded on the first factor; items 1, 2, 6, and 13 (items of the fear of catastrophe subscale) loaded on the second factor; and items 7, 10, 12, and 16 (items of the generalized anxiety subscale) loaded on the third factor with no cross-loadings in any of the items (Table 2
). This three-factor model seems to be consistent with the original four subscale model with the exception of the two subscales (the fear of specific dental hygiene stimuli and distrust) behaving as a single factor, whereas the fear of catastrophe and generalized anxiety of dental hygiene treatment subscales continued to exist as separate factors, just like the original DHFS.
|
Criterion Validity
The correlations of the DHFS with subjective dental fear, subjective dental hygiene fear, and MDAS scores were investigated to test the criterion validity. We also investigated the relations of the DHFS total score and subscale scores with age, gender, and self-reported oral health status, which are factors reported to be related to dental fear in previous studies.4–6
When the subjects were separated into two groups in terms of the MDAS score, it was revealed that 149 (74.5 percent; eighty-two females, sixty-seven males) of the subjects were in the low anxiety group, while fifty-one (25.5 percent; forty females, eleven males) of the subjects were in the high anxiety group. The scores of the subjects on different anxiety measures with respect to gender and the anxiety level (group) are presented in Table 3
. It should be noted that, to be able to compare the DHFS scores with previous studies carried out with the original version of DHFS, both the raw scores and the standardized scores obtained by dividing the sum of the item scores by the number of the items in the scale were used.
|
15, all of the scores of the subscales and the total DHFS scores were significantly different between the two groups (p<0.001), the scores of the high anxiety group being higher (Table 3
The correlation coefficients of the DHFS and its subscale scores with age, education, subjective dental fear, subjective dental hygiene fear, and MDAS score appear in Table 4
. There were significant negative correlations between age and DHFS total and MDAS scores. Subjective dental fear and subjective dental hygiene fear scores revealed significant positive correlations with DHFS total and MDAS scores (p<0.05). There were significant negative correlations between self-reported oral health scores and DHFS and its subscale scores (p<0.05).
|
|
| Discussion |
|---|
|
|
|---|
As in other dental treatment modalities, the dental anxiety level of the patient and the predisposing factors for the dental anxiety related to dental hygiene treatment procedure are prerequisites for development of management strategies to deal with dental anxiety effectively. Milgrom et al.11 propose that the ability to identify specific kinds of fear allows the practitioner to customize management strategies. For example, a patient who has a specific fear such as fear of injections might be taught basic relaxation skills. Once the patient has mastered these skills, he or she can gradually be exposed to the anesthetic needle through a series of increasingly realistic and repeated rehearsals. During each rehearsal the dental provider can coach the patient to practice breathing and muscle relaxation skills. Thus, an objective instrument for measuring the anxiety about the dental hygiene procedure is necessitated to further investigate the etiology of the anxiety and possible modes of management of the fearful dental patient.
To our knowledge, the DHFS is the only scale developed specifically to look at the patients reactions to dental hygiene treatment procedures.13,14 It has been demonstrated that the DHFS is a highly reliable measure with an overall internal consistency of 0.92 in measuring the fear of dental hygiene treatment as performed by dental hygienists. In the original study investigating the psychometric properties of the scale, the initial validation investigations revealed that the scale scores were higher in females, younger subjects, and patients avoiding dental care. These findings pointed to the fact that the scale was behaving in a similar manner with dental fear surveys, but it was mentioned that further studies were needed to examine the validity of this instrument.
One of the aims of our study was to test the psychometric properties of the DHFS in the Turkish culture. While doing this, we also aimed to perform further validation of the scale by investigating its factor structure by EFA. The peculiarity of our study is that the adapted Turkish form of the scale refers to the dentists performing the dental hygiene treatment rather than the dental hygienists, since there are no dental hygienists in Turkey. Therefore, what we investigated was the psychometric property of the scale in measuring patient fear about the dental hygiene treatment procedure itself, not the effect of the professional performing the treatment.
The findings of our study confirm that the Turkish form of the DHFS acts as a reliable instrument with a high overall Cronbach alpha coefficient of 0.91. This value is quite similar to that of the original scale (0.92) for which the Cronbach alpha coefficients of the subscales ranged between 0.79 and 0.86. When the Cronbach alpha coefficients of the subscales of the original and the Turkish versions are compared, the Cronbach alpha coefficients of the Turkish generalized anxiety subscale and the catastrophe subscales (0.83 and 0.79 respectively) were higher than the Cronbach alpha coefficients of the mentioned subscales of the original DHFS (0.79 and 0.68, respectively). Regarding the results of the exploratory factor analysis (EFA), we preferred to treat the fear of specific dental hygiene stimuli and distrust subscales of the original DHFS by omitting item 11 from the scale as a single factor comprising seven items named dental hygiene specific anxiety. The Cronbach alpha coefficient of the seven-item subscale was determined to be 0.86, which is higher than those of either the specific fear and the distrust subscales of the original version (0.85 and 0.75, respectively).
We performed EFA for further validation of the Turkish version of the DHFS. As revealed by the EFA with a direct oblimin rotation, it was observed that item 11 separately loaded on a third factor by itself. The item-total correlation value of item 11 in the scale was the lowest of all the items being 0.30. Therefore we preferred to omit item 11, which was freestanding, and rerun the analysis with the scale retaining fifteen items as recommended by Costello and Osborne.34 Following omission of item 11, the EFA with direct oblimin rotation revealed three subscales, which we named as general anxiety, dental hygiene specific anxiety, and catastrophe. The pattern matrix of the scale with fifteen items resembled the original DHFS much better in that two of the subscales appeared as different factorial constructs. The remaining two subscales constituted a single factor. In this three-factor model, the differentiation of the separate factors was quite satisfactory since there were no cross-loadings in any of the items. The three factors accounted for 62.8 percent of the variance. We preferred to name the factor consisting of seven items with three items of the distrust subscale and four items of the fear of specific dental hygiene stimuli of the original DHFS as the dental hygiene specific anxiety. This subscale addresses the anxiety related to the dental hygiene treatment experience, either concerning the interaction with the dentist or the specific characteristics of the dental hygiene procedure such as instruments, needles, etc.
The measures we investigated for criterion validity were gender, age, and self-reported oral health. In our study group, DHFS and MDAS scores were found significantly higher in females than males, which are in line with the findings of previous studies of Berggren and Meynert,10 Ekanayake and Dharmawardena,35 and Erten et al.36 The increased levels of anxiety may arise from women subjects greater willingness to express fear.37 Males generally may not express their fears as openly as do females.
Significant negative correlations were found between age and the dental hygiene-specific anxiety subscale, total DHFS, and MDAS scores. These findings are consistent with various studies that reported dental anxiety levels decrease with age.19,38 Furthermore, as mentioned in a longitudinal analysis, dental fear, like many other general and specific phobias, declines with age.39
There were significant negative correlations between total DHFS scores and self-reported oral health scores in our study. Doerr et al.40 have reported that having a low perception of ones oral health status is associated with higher dental anxiety. This finding is in agreement with previous studies that report that subjects who rated their oral health as poor had higher levels of dental anxiety than subjects who rated their oral health as fair, good, or excellent.19,41
When the study sample was divided into two groups as having high and low dental anxiety with a reference cut-off point of MDAS
15, it was seen that the high and low anxiety groups exhibited significantly different total DHFS scores and subscale scores, the scores for the high anxiety group being higher.
All of the above findings indicate that the Turkish version of the DHFS acts as a reliable and valid scale in measuring the dental anxiety in dental hygiene treatment procedure. A very important question that arises at this point is this: is the dental hygiene treatment anxiety a unique type of dental anxiety, or does it act as the component of a so-called dental trait anxiety, which could be expected to appear in similar levels regardless of the dental procedure applied? Regarding the above mentioned issue, if the dental hygiene treatment fear is a specific type of dental fear, then we should expect the correlation between DHFS and subjective dental hygiene fear to be stronger than the correlation between DHFS and subjective dental fear. Indeed, our results reveal a higher correlation coefficient for the former relation (0.499) than the latter (0.420). However the difference between the two correlation coefficients is rather small, and this may point to the fact that dental anxiety in different modes of dental treatment tends to be determined by baseline characteristics rather than situational factors such as the dental procedure applied.
There were significant positive correlations between all of the DHFS scores with MDAS, subjective dental fear, and subjective dental hygiene fear, though these correlation coefficients were of moderate strength, the correlation between MDAS and total DHFS score being 0.682. One could expect the correlation coefficient to be higher; however, regarding the fact that the MDAS does not have any specific items about the dental hygiene treatment process, this relatively moderate level of correlation seems reasonable.
When the average MDAS scores of our study and those of Tunc et al.s22 are compared, it is seen that in Tunc et al.s study the average MDAS score in regular dental patients was 11.3 ±4.7, while in our study it was 12.46 ±5.13. These findings suggest that, in the Turkish culture, dental hygiene treatment evokes anxiety that is even higher than that in regular dental treatment. However, this comparison should be interpreted with caution since the higher MDAS scores in our study may also be due to the higher female proportion of the subjects in our sample.
To compare the DHFS scores in Gadbury-Amyot and Williamss14 study and our study, we should take the standardized scores into consideration since the original DHFS consists of sixteen items while the Turkish version consists of fifteen items. The previous study reported that the total DHFS scores were 27.5 ±10.9 and 23.9 ±9.1 for females and males, respectively, which correspond to standardized scores of 1.72 ±0.68 and 1.49 ±0.57. The standardized total DHFS scores were higher in our study, being 2.13 ±0.80 for females and 1.74 ±0.70 for males. This difference may be attributed to cultural differences; however, more complex issues related to the income level, education, and dental hygiene habits should also be taken into consideration.
Our study has some limitations. First, the study sample consisted of patients who were already admitted to the periodontology clinic and thus may be representing a patient group with lower dental anxiety since patients with higher dental anxiety are expected to avoid dental treatment. It is possible that dental hygiene treatment anxiety may even be higher in the general Turkish population, including those who avoid dental treatment. Second, we did not utilize an objective measure to determine the periodontal health status of the subjects. Further studies with samples including the subjects from the general population and data on the objective periodontal health of the subjects would be informative in determining the causes and consequences of high DHFS scores.
| Conclusion |
|---|
|
|
|---|
| APPENDIX Dental Hygiene Fear Survey |
|---|
|
|
|---|
| Footnotes |
|---|
kent University, Faculty of Education, Ankara, Turkey; and Dr. Gadbury-Amyot is Professor, University of Missouri-Kansas City, School of Dentistry, USA. Direct correspondence and requests for reprints to Dr. Zuhal Yetkin Ay, Süleyman Demirel University, Faculty of Dentistry, Department of Periodontology, 32260-Çünür, Isparta, Turkey; 90–246–211–3327 phone; 90–246–237–0607 fax; zuhalyetkin{at}yahoo.com, zyetkin{at}med.sdu.edu.tr. | REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |