This study found that the oral health-related quality of life of the participants is impacted by oral diseases. The proportion of participants with at least one OHIP-14 domain impacted (94%) is one of the highest ever reported. The next highest prevalence reported is in the homeless people and patients with the temporomandibular joint disorder, both at 91%. In these studies, however, the sample sizes are smaller and, the threshold that defines prevalence is less conventional, where it includes the responses occasional, often and very often (40, 41). Only the latter two responses are commonly used to define impact prevalence, for example in the reports on the impact prevalence of the general population of England (16%), and Australia (18%) (37) and, dentate (19%) and edentulous (31%) populations (37, 42); which are much lower compared to the present report. Correspondingly, the impact severity in the present study is also somewhat high (mean add-score = 37.6 unit) compared to patients with the gastrointestinal and hepatic disorder, temporomandibular joint disorder and; socially vulnerable and underserved populations (mean ranges from 8.8–27.8 unit) (40, 43, 44).
The results showed that the prevalence (> 74%) and severity of impacts (mean add-score > 2.7) are greater in the psychosocial items (items: 1, 5, 6, 9, 10, 11) compared to oral functions items (2, 4, 7, 8) (< 72%, < 2.7 respectively). The prevalence of psychosocial impact is much lower in earlier reports (< 7%) (37, 40, 42). Many participants in the sample have disfigured lips, exposed teeth, severe periodontal condition, gum recession, dry mouth and halitosis. They are uncomfortable being ‘stared’ at by the researcher and somewhat agitated during the oro-facial examination. The reactions are not uncommon and similar responses including embarrassment, tense, anxiety, and irritation have been reported (30, 45). Some participants seemed annoyed about having to repeat themselves when their pronunciation cannot be understood. These lower the self-esteem and increases the stress, anxiety and depression level during social interaction. and because they view the social impact of the oral condition as severe, thus they are more likely to report an impact with high frequency (often and very often). The consequence of poor dental condition in facial burn patients surpasses the dental pain and oral function limits to affect the psychological and social well-being. That reporting behaviour is in contrast to the denial attitude described in Slade et. al., (2005) as the willingness to report minor discomfort or unwilling to report a severe impact which leads to reporting an adverse impact with low frequency (never, seldom) (37). A less likely explanation for the high psychosocial impact is that the participants misinterpret the context of OHIP-14 instrument as that related to disfigurement instead of a dental condition. This is because the instruction is given by the investigator and specified in the instrument is very clear. Also, the instrument is shown to be valid and reliable for self-administration.
The analysis showed that the effect of oral health indices, either individually or combined, is greater compared to other non-dental factors. Oral health status is the main predictor of poor OHRQoL in the participants; the more severe caries, periodontal and oral hygiene status, the greater the impact. This is consistent with the general understanding of the relationship between oral health status and OHRQoL (12). Poor oral health behaviour, i.e. less frequent tooth brushing and episodic / no dental visit in the past year, is the most likely explanation for the oral health status in the participants (46, 47); but the reason behind it is not clear from this study. Nonetheless, it is possible that the psychosocial factors adversely influence the health behaviours, specifically the dental visit. Most of the participants avoided the dentist for fear of treatment and feeling embarrassed with their dental conditions. Those with social barriers have problems to interact with other people because they are shy, lack self-confidence or have low esteem (28); which could be due to the dental condition and/or disfigured appearance. Thus, they lose the opportunity to learn about good oral hygiene practice and get the treatment for their condition. Further investigations to better understand and overcome the issue is recommended.
The cost of dental treatment is shown to be associated with better OHRQoL. on further assessment, the data show that the majority of participants with a cost issue have significantly less severe injury and better oral health condition (p < 0.05). Because of that, they are less likely to have a major oral hygiene care problem compared to more severe conditions; hence, have a lower risk of having impacted OHRQoL. Caries development involves the interaction between time, plaque, acid from the bacteria and fermentable carbohydrates (48). Thus the longer the time of exposure to the risk factors after the burn incident, i.e. the oral health behaviours and psychosocial changes, the greater the risk of developing caries and having poor OHRQoL. The associations between OHRQoL and scalding, fire and chemical burns are likely to be chance observations as there is no logical explanation for the significant findings. The sex-education interaction is significantly associated with OHRQoL in all three models. In the participants with low education, the OHRQoL is better in females than the males but; it is worse in the females who are better educated. This could be because the women are generally more conscious about their appearance and have greater aesthetic concerns than men and more expressive in describing a depressive impact, particularly those who are educated (49, 50).
It has been recommended that different scoring formats should be used in the analysis of participant-based outcomes such as the OHIP-14 (51, 52). However, only the results of the add-score are presented in this report because the analysis of impact prevalence did not yield any significant multivariate model. This is partly due to multicollinearity between the oral indices, barriers to health care utilisation, cause of the burn, and burn-related measurements that affect both outcomes. Multicollinearity issues are not unexpected because of the apparent correlation between the parameters. For example, all of the severity based burn measures such as the clinical assessments (the type of burn, TBSA) and subjective assessment (disfigurement and SWAP) are expected to be correlated to each other. Thus, only fitting only one measure in a model would be adequate. Thus, it is possible to obtain variations of reasonable ‘final’ model depending on the choice of measures included in a model. Models 2 and 3 demonstrate that effect by examining the disfiguration and SWAP separately. Similarly, in Models 2 and 3, the clinical oral status resolves the collinearity issue between the oral indices and supports the hypothesis for the effect of oral health measures on OHRQoL.
The results of this study should be interpreted with caution. Several limitations relating to the limited inference from the cross-sectional study design, reference population from a single institution in Pakistan, lack of reliability study due to patients unwilling to return for clinical reassessment and recall bias, have been discussed earlier (28). A very limited inference can be drawn for facial burn patients who are not followed-up at a burn centre. The key strength of this study is in its originality as this is the first study that investigated the OHRQoL of facial burn injury victims; a small niche and underserved population with unknown need. Highlighting their problem might help in getting them the attention and open a path to provide services to meet the oral health needs. Hopefully, these will improve the oral health-related quality of life and assist the victims to reintegrate into society.