Four‐dimensional oral health‐related quality of life impact in children: A systematic review

Abstract Oral health‐related quality of life (OHRQoL) is an important dental patient‐reported outcome which is commonly based on 4 dimensions, namely Oral Function, Orofacial Pain, Orofacial Appearance and Psychosocial Impact. The Oral Health Impact Profile (OHIP) is the most used OHRQoL instrument designed for adults; nevertheless, it is used off‐label for children as well. Our aim was to describe the OHRQoL impact on children measured by OHIP and map the information to the 4‐dimensions framework of OHRQoL. A systematic literature review following the PRISMA statement was conducted to include studies assessing OHRQoL of children ≤ 18 years using OHIP. The OHIP seven‐domain information was converted to the OHRQoL 4‐dimension scores accompanied by their means and 95% confidence interval. Risk of bias was assessed using a six‐item modified version of quality assessment tool for prevalence studies. We identified 647 articles, after abstracts screening, 111 articles were reviewed in full text. Twelve articles were included, and their information was mapped to the 4‐dimensional OHRQoL. Most included studies had low risk of bias. OHRQoL highest impact was observed for Oral Function, Orofacial Pain, and Orofacial Appearance for children with: Decayed‐Missing‐Filled‐Surface (DMFS) of ≥10, anterior tooth extraction without replacement and untreated fractured anterior teeth, respectively. Across all oral health conditions, Psychosocial Impact was less affected than the other three dimensions. OHIP has been applied to a considerable number of children and adolescents within the literature. One instrument and a standardised set of 4‐OHRQoL dimensions across the entire lifespan seem to be a promising measurement approach in dental and oral medicine.

capture concepts that are only known to the patient, such as the impact of the disease in daily life, patient´s sufferings, including mental and social health aspects as well as the influence of contextual factors. 6 In dental and oral medicine, the patient perspective is mainly captured by dental patient-reported outcomes (dPROs) and its corresponding measures, the dental PROMs. 7 Evidence about their importance in research as well as clinical practice is increasing, 8 including their essential role in value-based oral health care for implementing simultaneously economic efficiency and the optimum quality of care and value for patients. 9,10 One of the most important dPROs is oral health-related quality of life (OHRQoL), which is commonly defined as how patients rate their well-being and satisfaction with the current state of oral health and its psychosocial consequences. 11 In addition, it has the potential to evaluate dental interventions from the perspective of patients in clinical practice and research. 8 Children and adolescents can be affected by numerous oral and orofacial disorders, which impact on physical functioning and psychosocial well-being. 11,12 Specific issues could arise when measuring OHRQoL in these age groups due to their phase of physical, cognitive, emotional, social and language development, as oral health and health cognition are considered age-dependent. 13,14 Several instruments have been specifically developed for children and adolescents. The most often used ones includes the Child Perceptions Questionnaire (CPQ), 11,15 the Child Oral Health Impact Profile (COHIP), 16 the Child Oral Impacts on Daily Performances (C-OIDP) 12

and the Early
Childhood Oral Health Impact Scale (ECOHIS). 17 In adults, the Oral Health Impact Profile (OHIP) 18,19 is the most comprehensive and widely accepted OHRQoL instrument internationally. It has sound psychometric properties and was adapted to many cultural settings. 20 Interestingly, while OHIP was developed for adults, it is currently being applied to evaluate OHRQoL in children and adolescents in many countries. 21,22 Moreover, it was tested for its validity and reliability when applied to children and was found satisfactory. 23 Recent studies have demonstrated that OHRQoL has four main components, so called dimensions-Oral Function, Orofacial Pain, Orofacial Appearance and Psychosocial Impact, which could provide a standardised and an efficient approach to measure what matters to patients. 7,[24][25][26] Moreover, studies' findings indicate that dPROMs including OHIP with this 4-dimensional approach characterise and summarise how adults´ comprising patients and general population individuals are impacted by oral diseases. [26][27][28][29] Although instruments measuring OHRQoL in adults such as OHIP were also applied to measure oral health problems in children and adolescents, possibly under the assumption that capturing oral problems could be similar for both, 30,31 and were even tested for its psychometric properties and found to be suitable when applied to children, 23 to date, a 4-dimensional approach to describe and characterise impact across oral diseases for children has not been performed yet.
Advantages of the novel approach of characterising the oral impact for children and adults using a 4-dimensional impact would be to allow measuring oral health impact across the entire lifespan. Adults and children could have one measurement system and not two-the current situation. Therefore, the overall aim of this project was to perform a systematic review in order to identify publications with information about OHRQoL dimensions in children measured using OHIP and map the available seven-domain information to the 4-dimensional framework of OHRQoL (Oral Function, Orofacial Pain, Orofacial Appearance and Psychosocial Impact), which was characterised previously in adults.
Moreover, we aimed to describe the levels of OHRQoL of different oral conditions including their clinical relevance in paediatric patients and the general population of children/adolescents.

| Design
A systematic literature review with the subsequent identification of instruments and extraction of data was conducted. The protocol was published in the PROSPERO database (registration number: CRD42017064033), and the PRISMA statement for reporting systematic reviews was followed. 32 5. In case of mixed population of children and adults, the following criteria were adopted (mean/median age should be ≤18, or more than 50% are children ≤ 18 years, or the scores of OHIP domains have to be available for the children separately in case of mixedage population).
2. Responses to the instrument are not in the proper 0 to 4 OHIP item response format.  (Table S1).

| Screening and selection procedure
Two researchers (MO, MTJ, see acknowledgment, or KB) independently screened all titles and abstracts for inclusion based on the inclusion criteria mentioned above (criteria 1, 2 and 6 were used for abstracts and titles only). The two reviewers then discussed points of disagreements in inclusion of articles and agreed to the final decision. All potentially eligible articles underwent a full-text review to ensure that they fulfilled the inclusion criteria. Again, disagreements within this step of the review were discussed until consensus was achieved between the two reviewers.

| Assessment of risk of bias in included studies
Two researchers (MO and MTJ) assessed the risk of bias for the eligible articles using a modified version of the quality assessment for prevalence studies tool developed by Munn et al 35 Only six of the ten items (item number 1, 2, 4, 5, 6 and 7) in the appraisal tool were applicable, details about the items of the assessment tool and how they were used are depicted in Table S2. Each of these six questions could be answered with 'yes', representing a low risk of bias, 'unclear' representing unknown risk or 'no', representing a high risk of bias.
Any disagreements were resolved by arbitration.

| Data extraction
Data extraction of the included articles was carried out by one researcher (MO), and 10% of the extracted data was checked by an-

| Data analysis
The study determined the number of publications, the number of clinically relevant patient groups, (referred to as 'populations'), and the number of patient samples per population with 4-dimensional OHRQoL information.
The study also included mean values derived from any version of the OHIP questionnaire (14-item and 49-item). The data from OHIP-49 were converted to OHIP-14 means together with 95% confidence interval values restricted to fit on a 0 to 8 scale. Mean and standard deviation values were derived directly from the articles.

| RE SULTS
The literature search identified 647 articles. After screening of abstracts, 111 articles were reviewed in full text (Table S3) (Table 1).

| Assessment of study quality
The risk of bias assessment showed a high risk in the representativeness of the target population (does the source of population adequately represent the target population?), followed by the proper recruitment of the study participants (sampling technique, eg probability vs nonprobability samples). Another high risk and a large number of unclear risk of bias were shown by the coverage domain, which is a comparison of actual subjects/patients with the intended or planned subjects/patients (with the response rate being an important measure to assess the numerical discrepancy between both numbers). Nevertheless, because the included studies used OHIP which is validated in many languages, and well tested for its psychometric properties, the reliability and standard domains presented only low risk of bias. In addition, characterisation (if the study subjects and setting were sufficiently described) showed a low risk of bias as well. In general, a high proportion of studies (n = 8) and individual samples (n = 19) was deemed to have a low risk of bias ( Figure 2 and Table S4).

| Functional, pain, aesthetical and psychosocial OHRQoL impact
Overall results of the 0 to 8 converted OHIP scale showed that the most affected dimensions of the examined population were

| D ISCUSS I ON
Oral health conditions in children are in principle similar to adults, especially in terms of how they affect their lives and quality of life.
Similar to using a pharmaceutical drug off-label for an unapproved indication or in an unapproved age group, the OHIP was used for different age groups for which it was originally developed in a con-  We investigated the non-follow-up studies and found the highest levels of functional impairment in the general population of adolescents with DMFS of ≥10. This is consistent with adults and even elderly population studies. 36 Here, both caries and tooth loss were strongly associated with OHRQoL, providing evidence that these conditions affect directly the oral function of individuals regardless of their age. For all 4 dimensions, when DMFS increased ranging from low to high, the impact on OHRQoL increased as well. In concordance, such an association exists also in adult populations. 37 Regarding Orofacial Pain, interestingly, the highest pain value was found in the sample with the condition of anterior tooth extraction without replacement. The descriptive nature of our study cannot generate a reason for that, however, extraction of anterior teeth in school children could be related to possible traumatic injuries or dental caries. 38 Moreover, the potential high levels of anxiety and psychological distress of such condition among school children could be a reason for oral psychosomatic problems. 39 In general,

Both instruments reported a significant improvement in OHRQoL
at the end of orthodontic treatment in children. 50  OHRQoL items function differently for children and adults could be studied by examining the differential item functioning (DIF). Such techniques are essential to provide psychometrically solid dental patient-reported outcome measures.

| Limitation
Through OHIP being used 'off-label', that is applying the instrument to an age groups it was not designed for, standardised 4-dimensional information became available and the represented oral health conditions was low. Interestingly, researchers used OHIP in children and found it suitable for measuring several conditions including such important dental patient populations such as patients with malocclusion. Although many OHIP papers exist in children, we could not use all of them for the current project because we needed domain values (for the dimensions). Therefore, more OHRQoL information is actually available.

| CON CLUS ION
A considerable number of international researchers have applied OHIP, an OHRQoL instrument for adults, to children and adolescents.
Moreover, using OHIP for paediatric patients and the general popu-

ACK N OWLED G M ENT
The authors would like to thank Mike T. John (MTJ) for his assistance in screening the abstracts and full-text studies. And Nicole Theis-Mahon (NTM) for assisting in the electronic literature search.

CO N FLI C T O F I NTE R E S T
The authors declare that no conflicts of interest exist.

AUTH O R S CO NTR I B UTI O N S
MO, KB and TS conceived the ideas. MO and KB collected the data.
MO, KB and TS analysed the data. MO, KB and TS wrote the manuscript. MO, KB and TS revised and approved the manuscript.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/joor.13066.