Study design
A systematic review and, if possible, a meta-analysis will be conducted to map and screen the literature on adult patients’ motivations and expectations of orthodontic treatments and their effects on satisfaction of treatment outcome, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (16).
Study Eligibility
Type of studies
Papers written in English or Chinese with no study design restrictions will be included. Review articles, experimental studies (randomized or non-randomized controlled trials, controlled before-after studies, prospective or retrospective cohort studies), case-control studies and cross-sectional studies and qualitative explorations, that assessed motivations, expectations and satisfaction with orthodontic treatment outcome will be included.
Types of participants
The inclusion criteria are adult patients aged 18 years and above who seek orthodontic treatments. Orthodontic treatment uses different techniques, such as labial fix appliance, lingual appliance, clear aligner and removable appliance will be included. Adolescent patients 17 years and under, patients’ parents and patients with craniofacial hypoplasia, such as cleft lip and palate, and patients who receive orthodontic treatment combined with orthognathic surgery will be excluded from the systematic review.
Outcome measures
The primary outcome is to identify the adult patients’ motivations and expectations of orthodontic treatments. Motivation for treatment stems from several sources and can be defined as a concept that describes the conscious or unconscious stimulus for action toward a desired goal (4).Adult patients’ treatment expectation can be defined by Thompson and Sunol as a kind of realistic and practical belief that something will happen actually during process or after the treatment completed(17). As there is no consensus on measures to assess treatment expectation/motivation, adult patients’ treatment expectation and motivation can either be measured by surveys with a general question or a series of questions(4, 18, 19) or other measures(5) developed to assess expectation and motivation for orthodontic treatment. The instrument can be unipolar or bipolar scale. For example, continuous scale such as visual analogue scale (VAS) or five-point Likert scale.
The secondary outcome is mainly to investigate whether an association exists between adult patients’ treatment expectation/motivation and satisfaction with treatment outcomes. To what extent patient’s expectation/motivation affects their satisfaction with treatment outcomes. Again, there is no established systematic reviews to specifically explore this topic, we are therefore to look broadly to identify potential mechanism on the effect of patient’s treatment expectation/motivation on their satisfaction with treatment.
Search strategy
The sources to be searched are online databases including MEDLINE via PubMed, Scopus, Embase, Cochrane, Web of Science, PsychINFO and Assia. The grey literature will be searched using Google Scholar and OpenGrey (www.opengrey.eu/). The literature written in Chinese will be searched by Cnki (https://www.cnki.net) and Wanfang (https://www.wanfangdata.com.cn) online databases. In addition, a manual search will be conducted based on the reference lists of the selected studies. The search dates are from the year 1969 to present. Literatures written in English or Chinese will be included.
The search will be developed using medical subject headings (MeSH) and text words related to adult, young adult, orthodontics, malocclusion, motivation and expectation. A draft MEDLINE via PubMed search strategy is included in additional file 1(see Additional file 1). After the MEDLINE strategy is completed, the draft search protocol will be used in the other databases listed. When the electronic database search is finished, the references of the selected studies will be examined. Until the full search strategy is completed, a topic expert will be invited to check the list of the selected publications to identify any known studies that might have been missed. Software EndNote X9 will be used to manage the searched results.
Selection of studies
We will remove duplicates using EndNote’s duplicate identification tool, and then manually. At the first stage, all the title/abstract of records will be screened by YQ using Rayyan(20) (http://rayyan.qcri.org). Any studies that are not related to adult patients’ motivation and expectation will be excluded at this phase. At the second stage, the full text of records which match the selection criteria will be screened independently and in duplicate by YQ and SY, which will follow the exclusion criteria (see Additional file 2). If any discrepancies in the selection process occurs, it will be discussed first between the two peer reviewers. If the discrepancies will not be resolved after the discussion between the two reviewers, a third reviewer (RF) will be invited to discuss in detail and arbitrate to ensure a consensus is reached. Throughout the review, records and data will be managed according to the PRISMA protocol flow diagram(16). Included studies will be presented in a “Characteristics of included studies” table (Table 1), containing study reference, quality rating, sample and population characteristics, study design, measures, study characters, results and main findings.
Table 1
Characteristics of included studies
Study reference | Quality rating | Sample and population characteristics | Study design | Measures | Study characters | Results / main findings |
Data collection and management
Once the studies are selected for final analysis, data will be extraction by two reviewers independently and in duplicate using a standardised data collection form (see Additional file 3). Extraction information will include the tile, author, study population (number of participants, age, gender, ethnicity/race and baseline characteristics), study setting, intervention (types of orthodontic treatment, stage of treatment, types of interview and questionnaire), methodology (design of study, duration, recruitment, completion rates and data analysis) and outcomes. YQ and SY will examine the extracted data independently and if any objection or disagreement occurs, it will be resolved by consensus between them or calling the third reviewer (RF) to arbitrate. Any missing data will be requested from the study authors through emails.
Risk of bias assessment
The methods used to assess risk of bias will be conducted at the outcome and study level. At the study level selection bias such as ‘inappropriate sampling frame, exclusion criteria, observer bias’, and information bias such as missing data and variables will be assessed using the Joanna Briggs Institute for bias measurement for the selected study types.
At the outcome level, “motivations and expectations affect satisfaction with treatment outcome” will be assessed by reporting the level of reliability of the measures used to assess treatment satisfaction. In addition, GRADE will be used to assess the evidence at the outcome level as appropriate (21). The search of literature will be restricted to English-language and Chinese-language publications, which may introduce citation bias and jeopardize the evidence synthesis.
Quality assessment will be carried out using tools appropriate to study type. These will include AMSTAR 2 to assess systematic review quality (22); the Reporting of Observational Studies in Epidemiology (STROBE) for cohort, case–control and cross-sectional study quality (23). All eligible papers will be assessed independently by the reviewers using AMSTAR 2 or STROBE as appropriate.
Data synthesis
As the systematic review will consider both quantitative studies and qualitative studies that fulfil the inclusion criteria, for quantitative research meta-analysis will be conducted. Data will be extracted from each study and effect size will be assessed. The programme Comprehensive Meta-Analysis will be employed to assist with the conversion of the aggregate statistical details and entered into one of the 100 formats available to transform this information into effects sizes with weights. The z and standard error values will be copied into the STATA programme to produce editable ‘forest-plots’ and measures of heterogeneity, random overall effect sizes and meta-regression of moderators (e.g. year of publication and develop status of country from where study was conducted). Sensitivity analyses will be conducted to determine the robustness of the results and detect potential sources of bias.
For qualitative research articles, the data will be synthesised using thematic analysis(24) by identifying the important and prominent themes and the findings summarized accordingly. The following thematic headings may include: (1) study design; (2) quality of evidence; (2) Study Population (Number of participants/ Age/ Gender/ Country Ethnicity); (3) Study Setting; (4) Intervention (Types of orthodontic treatment/Stages of treatment/Types of interview/Time of interview/Questionnaire). NVivo software will be employed to organize, analyse and find insights in qualitative data. The data will be synthesised and presented in a narrative form.
Statistical analysis
If quantitative synthesis is possible, a meta-analysis will be conducted. Data will be extracted from each study and effect size will be assessed. Associations will include Pearson's and Spearman's correlation coefficient. Means and sample size of low/high value groups, odds ratio and sample size, beta coefficient and raw regression coefficient with standard error. A number of visual presentations from the analysis will be conducted including forest and funnel plots and meta-regression graphs. The heterogeneity statistics Q and I squared will also be reported and interpreted. A random error analysis will be appropriate for this study. Publication bias will be inspected through the conventional ‘File Drawer’ statistics in addition to the funnel plots.
Quality of evidence
Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework will be used to evaluate the quantitative data for quality of evidence. PRISMA-P of the protocol checklist was provided in the additional file (see Additional file 4).