Two samples were recruited independent for the current study: 201 participants from a local community recruited via advertisement and 201 dental patients from the outpatient clinic of Taipei Veterans General Hospital (Table 1). The participants provided written informed consent, approved by the Institutional Review Board of National Yang-Ming University (ID: YM106095E) and Taipei Veterans General Hospital (ID: 2018-12-003AC) before all the assessment started.
Dental Treatment Experience
The questionnaire for dental treatment experience was customized with descriptions about 12 conditions of common dental procedures (Table 2), based on our previous study  and an earlier study  that adopted a set of selected dental procedures. To ensure that the conditions are common to most patients, we excluded the procedures about complicated orofacial surgery or orthodontic therapy. Trait dental anxiety was assessed using the Chinese version  of the Index of Dental Anxiety and Fear (IDAF-4C+) .
For each condition of dental treatment, the participants were asked to indicate (a) whether they have experienced that condition of treatment in the past (i.e., history of treatment, HT), the degree of (b) fear of the condition (Fear), and (c) intention of avoidance of the condition (IA), respectively, according to the following instruction:
- ‘If you have previously experienced a condition about the treatment, please rate the degree of fear and intention of avoidance about that condition, by recalling your prior experience about it.’
- ‘If you have never experienced a condition about the treatment, please rate the degree of fear and intention of avoidance about that condition, by anticipating what you would feel about it, according to what you know the treatment.’
All the ratings were scored based on a 10-point numerical rating scale, ranging from 1 (the least degree of fear/IA) to 10 (the maximal degree of fear/IA). The scores of HT, fear of experienced treatment (ExpFear), IA of experienced treatment (ExpIA), fear of non-experienced treatment (NExpFear), and IA of non-experienced treatment (NExpIA), were calculated by including the ratings from all the 12 conditions, according to the following methods:
- The conditions that subjects have experienced were indexed by the value ‘1’ and those they have not experienced were indexed by the value ‘0’. The average HT was the mean of the 12 values, denoting the proportion of conditions that a subject has experienced.
- ExpFear and ExpIA were calculated as the mean of Fear and IA, respectively, across the conditions that subjects have experienced (i.e., the conditions valued as ‘1’ for HT).
- NExpFear and NExpIA were calculated as the mean of Fear and IA, respectively, across the conditions that subjects have not experienced (i.e., the conditions valued as ‘0’ for HT).
Analysis of descriptive statistics
The analysis of descriptive statistics was conducted for all the variables (age, sex, and the average of HT, IDAF-4C+, ExpFear, ExpIA, NExpFear, and NExpIA), across all subjects (n=402) and respectively for the community and the clinical groups (n=201 for each). Normality of the score distribution was assessed using the Shapiro-Wilk test, with p<0.1 indicating non-normality.
For each of the 12 conditions of treatment, the analysis descriptive statistics was also conducted for the following variables: IDAF-4C+, ExpFear, ExpIA, NExpFear, and NExpIA. For each condition, the prevalence of experiencing a condition was calculated as the mean HT averaged across all subjects. The magnification factor, i.e., the discrepancy in the fear/IA of non-experienced vs. experienced conditions, was calculated as follows:
(a) Magnification factor of fear = 100% X (NExpFear – ExpFear) / ExpFear
(b) Magnification factor of IA = 100% X (NExpIA – ExpIA) / ExpIA
Analysis 1: Comparison between experienced and non-experienced fear/IA
To test Hypothesis 1, we performed the Wilcoxon signed-rank test for comparing the scores between ExpFear and NExpFear as well as the scores between ExpIA and FearIA, across all subjects. The choice of non-parametric method is based on the non-normality of the distribution of the scores (Table 1).
Analysis 2: Association between dental anxiety, experienced and non-experienced fear/IA
To test Hypotheses 2A, we examined the strength of association between (a) ExpFear and ExpIA and (b) between ExpFear and NExpIA, using the Spearman’s rho coefficient. To test Hypotheses 2B, we examined the strength of association between (a) IDAF-4C+ and NExpFear and (b) between IDAF-4C+ and NExpIA.
Analysis 3: Association between the magnification factor of a dental condition and the prevalence of experiencing it
To test Hypothesis 3, we examined the strength of association between (a) the prevalence of experiencing a condition and its magnification factor of fear and (b) the association between the prevalence and its magnification factor of IA, using the Spearman’s rho coefficient.
All the statistical analyses were performed using IBM SPSS Statistics (v. 24). For all the statistical tests, the level of statistical significance (alpha) was 0.05.