We found that the improvements in OHRQoL detected by PIDAQ, were much larger than those detected by CPQ11-14-ISF-16. The correlation between the change in CPQ11-14-ISF-16 scores and change in IOTN-AC assessments, were smaller in comparison to the correlations with PIDAQ. However, for the CPQ11-14-ISF-16 subdomains, there were higher, but relatively weak associations between the improvement in the SWB subdomain and improvement in IOTN-AC (Table 5).
In the present study, the improvement in OHRQoL, as shown by CPQ11-14-ISF-16 is considered to be lower than a minimally important difference.17 Norman et al.,18 found that half the standard deviation of baseline scores, represents the most reported minimally important difference. When the CPQ11-14-ISF-16 subdomain scores were taken into consideration, the mean change was highest for the OS and EWB subdomains. When the size and direction of treatment effect in reducing the total CPQ11-14-ISF-16 scores was investigated, this resulted in the effect size to be “medium”; furthermore for the CPQ11-14-ISF-16 FL and SWB subdomains, the effect sizes were“small”; however they were found to be“medium”for both EWB and OS subdomains. These findings are in contrast to the findings of Agou et al. 19 in a sample of 74 Canadian children, who observed a slightly higher mean change in CPQ11-14-ISF-16 total scores of 4.9 (SD=14.57) after treatment, this was in comparison to an untreated control group; however when dimensional improvements were accounted for, the changes were higher and significant for the EWB subdomain. The findings of Benson et al. 20 in the United Kingdom, were almost similar to the current study, they observed a mean change in the total CPQ11-14-ISF-16 scores of 3.2 (SD=6.9; p=0.009), this was in comparison to a group of participants that had no history of receiving orthodontic treatment and the reported mean change in this group was 2.4 (SD=8.8; p < 0.001); however the difference between the two groups was not of statistical significance (p=0.584). It is worth noting that the findings of Benson et al. 20 were based on a community sample; consequently, such findings are only comparable to this study to a limited extent. On the contrary to these findings, Healey et al. found among a clinical sample of children in New Zealand, that the mean changes in total CPQ11-14-ISF-16 scores after orthodontic treatment (at debond) were small -0.3 (SD=14.4) with a “small” effect size of 0; however at the end of their study (mean=21 months after treatment) the CPQ11-14-ISF-16 total scores showed significant reduction with a mean change of 10.4 (SD=7.7) and a “large” effect size of 0.9; therefore reporting an overall mean change in CPQ11-14-ISF-16 total scores of 9.8 (SD=14.6) and a “large” effect size of 0.7 (this was “medium” for all the CPQ11-14-ISF-16 subdomains).21 It is not surprising that an improvement in OHRQoL was not detected among their sample immediately following orthodontic treatment, as young people tend to require and need time to adapt to the new arrangement of their teeth.
The “large” improvements in OHRQoL when PIDAQ was used as a measure, are similar to the results of a study conducted on 93 patients in India, where the authors reported “large” (Effect size=1.47) significant (P<0.001) reductions in mean PIDAQ scores after treatment.22 When age was accounted for in relation to the mean change in PIDAQ total scores, a statistical significance (p=0.04) was observed for two age groups (Adolescents 10-19 years and young adults 20-35 years); however, this was not found to be significant when gender differences (males and females) were considered (p=0.31); nonetheless such findings can only be confirmed with larger samples, that would help and be able to account for a variety of individual characteristics.22
The discrepancy between examiner IOTN-AC assessments and self-perceived or lay IOTN-AC assessments for the need of orthodontic treatment has been reported in other studies,22-24 and this was evident in this study. We found that the aesthetic changes detected by laypeople and orthodontists were variable and the improvements detected in the aesthetic appearance by the laypeople group were slightly larger (Table 4).
The results indicated that the correlation between the change in Total CPQ11-14-ISF-16 scores and change in laypeople and orthodontists IOTN-AC scores were small and statistically non-significant. However, when the CPQ11-14-ISF-16 subdomains were accounted for (Table 5), there were better associations for the SWB subdomain when assessed by both groups; however they were only statistically significant when assessed by the laypeople group. In one study, although total CPQ11-14-ISF-16 scores were associated with the IOTN-AC scores, these changes were mainly influenced by the EWB and SWB subdomains.25 This demonstrates the detrimental effects of dental aesthetics on the SWB and EWB of children.
The correlations between the changes in OHRQoL measured by PIDAQ and the IOTN-AC aesthetic changes, were higher and significant in comparison to the associations with CPQ11-14-ISF-16 (Table 5). The findings of the present study are consistent with the findings of Garg et al. 22 who found statistically significant associations (P<0.001) between the PIDAQ changes and self-perceived IOTN-AC changes after treatment among their sample. Overall, the findings direct research to employ more condition-specific measures in evaluating patient-reported outcomes in orthodontic treatment, and they also support the perception that orthodontic treatment and improved dental aesthetics have a positive impact on individuals OHRQoL.
Considering the limitations of this study is deemed appropriate at this stage. Lacking an untreated control group among this prospective cohort is considered to be a limitation, as one large community-based longitudinal study among school children, found improvements in OHRQoL (CPQ11-14-ISF-16) among those children with no history of receiving orthodontic treatment i.e. in the control group;19 however It could be argued that such changes in OHRQoL among younger age groups are due to natural fluctuations in OHRQoL. Another important weakness of our study is that we did not explore other likely influences on OHRQoL such as age and gender that could have resulted in the observed improvements in CPQ11-14-ISF-16 and PIDAQ scores. The small sample of this study was based on a clinical sample rather than a population based sample; therefore the introduction of sampling bias can be considered a limitation.
The study’s strengths include recruitment of patients in both specialist orthodontic practices and hospitals settings, and this implies that this was a “real world” study. Furthermore, using validated generic and condition-specific measures of OHRQoL. Also, alongside the OHRQoL measures, the inclusion of IOTN-AC allowed for the associations between dental aesthetics and OHRQoL to be examined. Whilst the longitudinal construct validity of the CPQ11-14-ISF-16 was examined, it was not possible to test the longitudinal construct validity of PIDAQ, because no global questions were included in the measure.