Effective communication of risk is a requisite for valid consent, shared decision-making, and the provision of person-centered care. No agreed standard for the content of discussions with patients about the risks of orthodontic treatment exists. This study aimed to produce a professional consensus recommendation about the risks that should be discussed with patients as part of consent for orthodontic treatment.
A serial cross-sectional survey design using a modified electronic Delphi technique was used. Two survey rounds were conducted nationally in the United Kingdom using a custom-made online system. The risks used as the prespecified items scored in the Delphi exercise were identified through a structured literature review. Orthodontists scored treatment risks on a 1-9 scale (1 = not important, 9 = critical to discuss with patients). The consensus that a risk should be discussed as part of consent was predefined as ≥70% orthodontists scoring risk as 7-9 and <15% scoring 1-3.
The electronic Delphi was completed by 237 orthodontists who reached a professional consensus that 10 risks should be discussed as part of consent for orthodontic treatment; demineralization, relapse, resorption, pain, gingivitis, ulceration, appliances breaking, failed tooth movements, treatment duration, and consequences of no treatment.
A professional orthodontic consensus has been reached that 10 key risks should be discussed with patients as part of consent for orthodontic treatment. The information in this evidence base should be tailored to patients’ individual needs and delivered as part of a continuing risk communication process.
Orthodontists agree that 10 risks should be discussed with patients for consent.
The risks deemed important to discuss include demineralization, relapse, and resorption.
This evidence base should be tailored to patients’ individual needs and values.
Risk communication involves giving patients information about potential risks they may encounter as a result of a disease, a clinical procedure, or a particular behavior. An orthodontist may be liable to legal action by the patient and disciplinary proceedings if a patient is not given sufficient, meaningful, and balanced information about the risks of treatment. Effective communication of risk is a requisite for valid consent, shared decision-making, and the provision of person-centered care.
The risks of orthodontic treatment have been defined broadly as any of the deleterious or iatrogenic effects of orthodontic treatment, or any potential adverse outcomes or consequences. The communication of risk is particularly difficult in orthodontics as care is often elective, takes place over an extended period and is delivered as part of a triad (professional, patient, and primary carer). Because of the considerable investments of time and resources, the potential harms must be carefully weighed against the anticipated benefits.
Landmark court rulings in the United States, Canada, United Kingdom, and Australia have shifted the way in which health care risks are communicated. This shift means that health practitioners are expected to provide patients with a reasonable amount of risk information in a patient-focused manner (which is likely to equate to a professional standard). In addition, the wants and needs of the particular patient must be identified and further information given relative to the material risks relevant to that subject elicited by their circumstances and response. Although paternalism has no place within health care, neither does the abandonment of patients by health care professionals failing to contribute to the decision-making process. The principles of shared decision-making encourage health care professionals to use their expert opinion for the benefit of patients as part of the consent process. In addition, because of heuristic strategies to make quick and effortless decisions, patients often do not seek new information but rely heavily on health care professionals’ advice about treatment. ,
Laws in many countries have now formalized that consent is not simply a process of giving all information, regardless of relevance. However, no agreed standard for the content of discussions with patients about the risks of orthodontic treatment exists, and the development of orthodontic risk communication tools have rarely been guided by an evidence base. Knowledge of a reasonable professional community standard pertaining to risk disclosure in orthodontics will allow clinicians to focus on and save energy for the additional risk information needs of the specific, individual patient. As such, this study aimed to gain a professional consensus on the risks that should be discussed as part of consent for orthodontic treatment.
Material and methods
Ethical approval was granted by the Cardiff University Dental School Research Ethics Committee (Ref no. 1507). A serial cross-sectional survey design using a modified electronic Delphi technique was used. Two survey rounds were conducted nationally in the United Kingdom.
The risks used as the prespecified items scored in the Delphi exercise were identified through a structured literature review. Search strategies focused on identifying articles reporting on the probability and nature of the risks of orthodontic treatment. As stated in the literature, , orthodontic treatment risks were defined broadly as any deleterious or iatrogenic effects of treatment, or any potential adverse outcomes or consequences. Risks associated with specific treatment modalities, such as headgear, miniscrew implants, and orthognathic surgery, were deemed to be outside the scope of this study and not included. Search strategies were developed using a combination of free-text terms, based on keywords and phrases, and controlled vocabulary in the form of appropriate subject headings. The databases Ovid MEDLINE (1946 to November 1, 2016), EMBASE (1947 to November 1, 2016), and PsycINFO (1806 to November 1, 2016) were searched, and search engines, such as Google (Google, LLC, Mountain View, Calif) and Google Scholar (Google, LLC), were also used. Key international orthodontic journals and the bibliographies of articles were used to identify additional studies and further search terms. Literature searches were kept up to date using e-mail notifications from Ovid MEDLINE (Wolters Kluwer Health 2016). Relevant risks were extracted from the studies using a reference table system, and 2 authors (J.P and H.P) generated a final list of risks by combining similar risk categories and resolving conflicts by discussion.
Custom-made surveys using Key Survey (WorldAPP, Braintree, Mass) were developed for the Delphi exercise and refined during steering group meetings of the research team. The surveys were based on previously reported Delphi methodology. Pilot surveys were conducted with 23 orthodontic clinicians practicing in a range of sectors (hospital, public, and private practice) in South Wales (100% response rate). These subjects were chosen as a representative sample of professionals similar to those who would complete the Delphi exercise correctly. Feedback was obtained, and subsequent amendments to the survey layout and wording were made.
The risks identified in the structured literature review formed a template for the survey used in round 1 of the Delphi ( Fig 1 ). To avoid weighting, we listed risks randomly in each round using a random number generator (Microsoft Office Excel; Microsoft, Redmond, Wash).
People with an e-mail address registered on the British Orthodontic Society (BOS) membership database were deemed eligible to participate. Subjects registered as retired, international, or core trainee members were excluded. Participant consent to be involved in the study was implicit on completing the surveys, and entry to a prize draw was offered to participants for completing the Delphi exercise.
The BOS disseminated the survey link directly to members. Two reminder e-mails were sent to participants, 1 and 2 weeks after initial contact. The survey was closed after an additional week. E-mail addresses were collected for participation in round 2. It took participants approximately 10-15 minutes to complete round 1.
Participants were asked to score the importance of discussing each risk with patients as part of the consent process for orthodontic treatment. Risks were scored on an ordinal scale, from 1 to 9, with 1 being “not important at all” and 9 being “completely critical.” Extra information to explicitly describe risks and avoid ambiguity was provided. If participants felt a risk only applied in specific circumstances, instead of scoring the risk, they could provide details in a free text box ( Fig 1 ). A function was provided for participants to add extra risks they thought were relevant, which had not already been listed.
Data were exported from Key Survey into Microsoft Office Excel and SPSS Statistics (version 20; IBM, Armonk, NY) for analysis. The risk scores were reviewed against a predefined definition of consensus ( Table I ). Risks classified as consensus in/out were not assessed in round 2.
|Consensus in||The consensus that risk should be discussed with patients as part of the consent process for orthodontic treatment||≥70% participants scoring as 7-9 and <15% scoring 1-3|
|Consensus out||The consensus that risk is not normally important to discuss with patients as part of the consent process for orthodontic treatment (but clinicians should use their discretion)||≥70% participants scoring as 1-3 and <15% scoring 7-9|
|No consensus||Uncertainty about the importance of discussing risk as part of the consent process for orthodontic treatment||Anything else|
Risks stated by the majority of participants (>50%) as applying only in specific circumstances were forwarded for assessment in round 2. The free-text responses for these risks were thematically analyzed and coded by 2 authors (J.P and H.P), generating a list of specific circumstances for when each risk might apply.
The free-text responses describing additional risks were analyzed similarly but coded according to the original risk list. Risks not already represented were included in the list of risks forwarded for assessment in round 2.
Those participants who responded in round 1 and provided a valid e-mail address were contacted and asked to complete the survey for round 2. Similar to Round 1, reminder e-mails were sent, and the survey was closed after 3 weeks. It took participants approximately 5-10 minutes to complete round 2.
Participants were provided with the following results from round 1 for each risk carried forward: (1) overall quartiles for the response scores from all participants; and (2) a reminder of their score (if they scored the risk).
After considering the results of round 1, participants were asked to review the risks listed and rescore them. They were informed that for each risk, they could change their score from round 1 or keep it the same ( Fig 2 ).
Participants were also asked to score the risks that had previously been identified as applying only in specific circumstances according to the list of circumstances defined in round 1.
The definition of consensus was applied again, including only the responses from round 2. Risks classified as consensus in, after either round (and not identified as applying only in specific circumstances), were included in a core set of risks.
To identify whether attrition in round 2 would introduce bias, we calculated the median score across risks from round 1 for each participant. These scores were compared for those completing both rounds and those completing round 1 only.
The structured literature review identified 30 risks, which were included in round 1 of the Delphi exercise ( Table II ).
|Risk highlighted by study (+/− specific circumstances when risk might apply)||Orthodontist opinion (% of participants scoring risk 1-3, 7-9)||Evidence in the literature|
|Demineralization||Consensus in (0, 99)||
|Relapse||Consensus in (1, 98)||
|Length of treatment||Consensus in (1, 95)||
|Root resorption||Consensus in (2, 93)||
|Pain/discomfort||Consensus in (3, 89)||
|Consequences of doing nothing||Consensus in (5, 86)||
|Appliances breaking||Consensus in (4, 85)||
|Failure to achieve desired tooth movement(s)||Consensus in (9, 76)||
|Gingivitis||Consensus in (7, 76)||
|Cuts and ulcers||Consensus in (4, 75)||
|Gingival recession and/or crestal alveolar bone loss|
|With patients with a preexisting periodontal condition||Consensus in (0, 99)||
|If there are specific anatomic considerations||Consensus in (1, 90)|
|With adult patients||Consensus in (5, 72)|
|If using certain treatment modalities||No consensus (4, 67)|
|With specific skeletal patterns/malocclusions||Consensus in (0, 96)||
|Development or worsening of black triangles between teeth|
|With patients with preexisting periodontal conditions/black triangles||Consensus in (0, 96)||
|With patients with specific tooth anatomy||Consensus in (1, 87)|
|With adult patients||Consensus in (4, 74)|
|With patients with anterior crowding||No consensus (18, 45)|
|With patients whose physicians recommend antibiotic prophylaxis||Consensus in (3, 92)||
|With patients with a history of cardiac disease||No consensus (18, 57)|
|Negative effect on playing wind/brass instrument|
|With patients who are wind/brass instrumentalists||Consensus in (5, 79)||
|Tooth wear caused by opposing brackets|
|If using certain appliance types||Consensus in (5, 78)||
|With patients with specific occlusal features||Consensus in (6, 76)|
|With patients with bruxism||No consensus (8, 64)|
|Problems eating||No consensus (7, 67)||
|Periodontitis||No consensus (10, 61)||
|Devitalization of teeth||No consensus (8, 61)||
|Problems speaking||No consensus (11, 55)||
|Missing school lessons/time off work||No consensus (11, 48)||
|Damage to teeth or restorations on debonding||No consensus (11, 26)||
|Flattening of the facial profile||No consensus (35, 12)||
|Risks associated with tooth extraction(s)||No consensus (69, 7)||
|Teasing, embarrassment, impact of the appliance on interpersonal relationships||No consensus (6, 7)||
|Temporomandibular dysfunction||No consensus (48, 7)||
|Soft tissue injury during placement or manipulation of the appliance by the clinician||Consensus out (70, 10)||
|The negative effect of the appliance on sleeping patterns||Consensus out (74, 9)||
|Radiation exposure||Consensus out (70, 9)||
|Airway or ingestion risks||Consensus out (72, 8)||
|Allergies to orthodontic materials||Consensus out (83, 5)||
|Cytotoxic effects and mutagenic potential of orthodontic materials||Consensus out (91, 2)||