Introduction
Patient decision-making aids (PDAs) are instruments that facilitate shared decision making and enable patients to reach informed, individual decisions regarding health care. The objective of this study was to assess the efficacy of a PDA compared with traditional information provision for adolescent patients considering fixed appliance orthodontic treatment.
Methods
Before treatment, orthodontic patients were randomly allocated into 2 groups: the intervention group received the PDA and standard information regarding fixed appliances, and the control group received the standard information only. Decisional conflict was measured using the Decisional Conflict Scale, and the levels of decisional conflict were compared between the 2 groups.
Results
Seventy-two patients were recruited and randomized in a ratio of 1:1 to the PDA and control groups. Seventy-one patients completed the trial (control group, 36; PDA group, 35); this satisfied the sample size calculation. The median total Decisional Conflict Scale score in the PDA group was lower than in the control group (15.63 and 19.53, respectively). However, this difference was not statistically significant (difference between groups, 3.90; 95% confidence interval of the difference, −4.30 to 12.11). Sex, ethnicity, age, and the time point at which patients were recruited did not have significant effects on Decisional Conflict Scale scores. No harm was observed or reported for any participant in the study.
Conclusions
The results of this study showed that the provision of a PDA to adolescents before they consented for fixed appliances did not significantly reduce decisional conflict. There may be a benefit in providing a PDA for some patients, but it is not yet possible to say how these patients could be identified.
Registration
This trial was registered with the Harrow National Research Ethics Committee (reference 12/LO/0279).
Protocol
The protocol was not published before trial commencement.
Highlights
- •
Giving adolescent patients PDAs before treatment did not reduce decisional conflict.
- •
Sex, ethnicity, age, and time of recruitment did not significantly affect DCS scores.
- •
A PDA might benefit some patients, but identifying them remains a challenge.
Patient-centered care, whereby patients are involved in decisions about their own health care and treatment options, is a core principle of modern health care. It is both an ethical and a legal obligation that is underpinned by legislation. Providing patient-centered care commonly involves shared decision making, where clinicians and patients work together to ensure that patients are involved in decisions about their health care.
Patient decision-making aids (PDAs) are evidence-based tools that facilitate shared decision making and are used to help patients make personal decisions about their health care. They are designed to engage patients in the decision-making process, assist them in considering the risks and benefits of different treatment options in line with their own values and help them to put information into a personal context. PDAs encourage patients to establish their preferences and to make decisions that are appropriate for them.
A large body of evidence exists on the benefits of PDAs, and they have been shown to help patients make choices in keeping with their own values as well as to improve the quality of decision making. PDAs have also been shown to improve patient knowledge, increase appreciation of risks, reduce decisional conflict, reduce passive decision making, reduce the number of patients who are undecided about their treatment options, and improve adherence to treatment because patients have taken a more active role in the decision-making process. Additionally, using PDAs commonly results in patients choosing less invasive treatment options, and their use has often been shown to be more cost effective than when they are not used.
Although PDAs are increasingly used in medicine, they are not commonly used in dentistry. This is despite research showing the need to consider patient values and preferences when making decisions about dental treatment, and there is evidence that dental patients prefer a collaborative style of decision making.
Orthodontics often involves decisions where there are several viable treatment options; therefore, it lends itself well to shared decision making and the use of PDAs. Currently, there are no published PDAs for orthodontics. Thus, the use of PDAs in orthodontics requires further research to assess whether the benefits of PDAs that have been shown to exist in medicine also exist in orthodontics.
Specific objectives and hypothesis
The aim of this study was to test the efficacy of a PDA in adolescent patients considering fixed appliances. This was through the measurement of decisional conflict, using the Decisional Conflict Scale (DCS). The null hypothesis was that there was no difference in the average level of decisional conflict in patients who received the PDA (intervention group) compared with those who did not (control group).
Materials and methods
Trial design and any changes after trial commencement
This study was a single-center, prospective, randomized controlled trial with a 1:1 allocation ratio. Ethical approval was granted from Harrow National Research Ethics Committee (reference 12/LO/0279). There were no changes to the trial after its commencement.
Participants, eligibility criteria, and settings
The study was carried out in the orthodontic department of a National Health Service teaching hospital, Eastman Dental Hospital, London, United Kingdom. Patients were eligible for inclusion if they were 10 to 16 years old, had not undergone previous orthodontic treatment, and had no craniofacial abnormalities. Patients were recruited from new patient clinic appointments and appointments to take initial records for treatment planning; both time points were before a full discussion of treatment options and before the patients and parents consented to treatment. Information leaflets were given to potential subjects and their parents, and consent and assent were obtained before recruitment to the study. Patients were recruited between July 2015 and February 2016.
Interventions
Participants in both the intervention and control groups received verbal information and patient information leaflets, according to their clinicians’ normal practices. To reduce the risk of bias because of different information being provided, all patients also received standardized verbal information from a researcher (K.P.) regarding the benefits and risks of fixed appliances. Participants in the intervention group additionally received the PDA, which the researcher discussed in detail with the patient.
The PDA was developed using the International Patient Decision Aids Standards Collaboration criteria. The material in the PDA included information from evidence-based literature and in-depth interviews with patients undergoing, or having recently completed, fixed appliance treatment. During the interviews, patients were asked about their understanding of the risks and benefits of fixed appliances and also which risks and benefits mattered most to them. This information was then incorporated into the PDA with that from the literature; this ensured that the PDA included topics that were considered important by both patients and clinicians. The decision aid comprised 4 size A4 pages. The first page gave general information on fixed appliances including what they are, what they are used for, and what level of commitment is required from the patient. The second page detailed the overall benefits and risks of fixed appliance treatment, the third page contained a decision tree to guide patient decision making, and the final page contained questions to help patients in the final decision-making process.
All participants then completed the DCS unaided. This is the standard methodology used in most PDA research, where participants receive all the information, and the intervention group receives the PDA, and then the participants complete a DCS questionnaire only after all information has been given. The DCS is a validated, multidimensional questionnaire that indirectly measures shared decision making by measuring the degree of uncertainty in a patient’s decision making. It is the most popular tool used to measure shared decision making and the efficacy of the PDAs; it consists of 16 statements, each with 5 possible responses from “strongly agree” to “strongly disagree.” The DCS is completed by the patient and scored by the clinician or researcher to give a total score from 0 (no decisional conflict) to 100 (extremely high decisional conflict). The questionnaire has good psychometric properties and high reliability, with a test-retest correlation of 0.81 and a Cronbach alpha from 0.78 to 0.92. Subsections of the DCS can also be scored separately to illustrate different elements of decisional conflict (uncertainty, informed, values clarity, support, effective decision).
Outcomes (primary and secondary) and any changes after trial commencement
The primary outcome measure was the level of decisional conflict, as shown by the total DCS score, regarding whether to proceed with fixed appliances. Demographic information and the time point at which patients were recruited (new patient appointment or subsequent records appointment) were also recorded so that this information could be used in the statistical analysis to assess whether any of these factors affected decisional conflict. There were no outcome changes after trial commencement.
Sample size calculation
Using the data from the first 18 participants who completed the study as an internal pilot (common standard deviation, 11.5), it was calculated that a total of 54 participants (27 in each group) were required to demonstrate a clinically relevant difference between the 2 groups of 9 points on the DCS, with 80% power and a significance level of 0.01. This was increased by 15% due to using the Mann Whitney U test and increased a further 15% to allow for any dropouts and incomplete data. This gave a total sample size of 72 participants (36 in each group).
Interim analyses and stopping guidelines
Not applicable.
Randomization (random number generation, allocation concealment, implementation)
A random number sequence was generated using a random number table. Patients were randomized in blocks of 6 to ensure that there were equal numbers of participants in the intervention and control groups throughout the study. Allocation concealment was achieved using sequentially numbered, sealed, opaque envelopes containing the group to which participants were allocated. These envelopes were prepared before the study and were only opened once patients were recruited and consented to the study.
Blinding
Patients were recruited only after they had received the verbal information and patient information leaflets from their clinicians; thus, clinicians did not know whether patients had been recruited to the study and to which arm of the study they had been allocated. The verbal information given to both groups by the researcher had been standardized and practiced before recruitment to ensure that all patients were given the same information. It was not possible to blind the researcher and the participants to the group allocation because the researcher worked through the PDA with those allocated to this group, and patients knew whether they had received the PDA.
Statistical analysis
The total and subscale DCS scores for the control and intervention groups were compared to assess the efficacy of the PDA. Since the scores were not normally distributed, the nonparametric Mann-Whitney U test was used to compare the 2 groups. Univariable linear regression analyses were undertaken to assess whether any demographic factors or the time point at which patients were recruited affected the total DCS scores. The assumptions of the regression analysis were verified by a study of the residuals. All statistical analyses were completed using SPSS software for Windows (version 22.0; IBM, Armonk, NY). A significance level of 0.01 was used for the Mann-Whitney U tests instead of the conventional 0.05 to prevent spuriously significant results arising from multiple testing. A significance level of 0.10 was used for the univariable regression analyses to determine which, if any, variables to include in a multivariable regression analysis as potential confounders when comparing the scores in the 2 groups.
Materials and methods
Trial design and any changes after trial commencement
This study was a single-center, prospective, randomized controlled trial with a 1:1 allocation ratio. Ethical approval was granted from Harrow National Research Ethics Committee (reference 12/LO/0279). There were no changes to the trial after its commencement.
Participants, eligibility criteria, and settings
The study was carried out in the orthodontic department of a National Health Service teaching hospital, Eastman Dental Hospital, London, United Kingdom. Patients were eligible for inclusion if they were 10 to 16 years old, had not undergone previous orthodontic treatment, and had no craniofacial abnormalities. Patients were recruited from new patient clinic appointments and appointments to take initial records for treatment planning; both time points were before a full discussion of treatment options and before the patients and parents consented to treatment. Information leaflets were given to potential subjects and their parents, and consent and assent were obtained before recruitment to the study. Patients were recruited between July 2015 and February 2016.
Interventions
Participants in both the intervention and control groups received verbal information and patient information leaflets, according to their clinicians’ normal practices. To reduce the risk of bias because of different information being provided, all patients also received standardized verbal information from a researcher (K.P.) regarding the benefits and risks of fixed appliances. Participants in the intervention group additionally received the PDA, which the researcher discussed in detail with the patient.
The PDA was developed using the International Patient Decision Aids Standards Collaboration criteria. The material in the PDA included information from evidence-based literature and in-depth interviews with patients undergoing, or having recently completed, fixed appliance treatment. During the interviews, patients were asked about their understanding of the risks and benefits of fixed appliances and also which risks and benefits mattered most to them. This information was then incorporated into the PDA with that from the literature; this ensured that the PDA included topics that were considered important by both patients and clinicians. The decision aid comprised 4 size A4 pages. The first page gave general information on fixed appliances including what they are, what they are used for, and what level of commitment is required from the patient. The second page detailed the overall benefits and risks of fixed appliance treatment, the third page contained a decision tree to guide patient decision making, and the final page contained questions to help patients in the final decision-making process.
All participants then completed the DCS unaided. This is the standard methodology used in most PDA research, where participants receive all the information, and the intervention group receives the PDA, and then the participants complete a DCS questionnaire only after all information has been given. The DCS is a validated, multidimensional questionnaire that indirectly measures shared decision making by measuring the degree of uncertainty in a patient’s decision making. It is the most popular tool used to measure shared decision making and the efficacy of the PDAs; it consists of 16 statements, each with 5 possible responses from “strongly agree” to “strongly disagree.” The DCS is completed by the patient and scored by the clinician or researcher to give a total score from 0 (no decisional conflict) to 100 (extremely high decisional conflict). The questionnaire has good psychometric properties and high reliability, with a test-retest correlation of 0.81 and a Cronbach alpha from 0.78 to 0.92. Subsections of the DCS can also be scored separately to illustrate different elements of decisional conflict (uncertainty, informed, values clarity, support, effective decision).
Outcomes (primary and secondary) and any changes after trial commencement
The primary outcome measure was the level of decisional conflict, as shown by the total DCS score, regarding whether to proceed with fixed appliances. Demographic information and the time point at which patients were recruited (new patient appointment or subsequent records appointment) were also recorded so that this information could be used in the statistical analysis to assess whether any of these factors affected decisional conflict. There were no outcome changes after trial commencement.
Sample size calculation
Using the data from the first 18 participants who completed the study as an internal pilot (common standard deviation, 11.5), it was calculated that a total of 54 participants (27 in each group) were required to demonstrate a clinically relevant difference between the 2 groups of 9 points on the DCS, with 80% power and a significance level of 0.01. This was increased by 15% due to using the Mann Whitney U test and increased a further 15% to allow for any dropouts and incomplete data. This gave a total sample size of 72 participants (36 in each group).
Interim analyses and stopping guidelines
Not applicable.
Randomization (random number generation, allocation concealment, implementation)
A random number sequence was generated using a random number table. Patients were randomized in blocks of 6 to ensure that there were equal numbers of participants in the intervention and control groups throughout the study. Allocation concealment was achieved using sequentially numbered, sealed, opaque envelopes containing the group to which participants were allocated. These envelopes were prepared before the study and were only opened once patients were recruited and consented to the study.
Blinding
Patients were recruited only after they had received the verbal information and patient information leaflets from their clinicians; thus, clinicians did not know whether patients had been recruited to the study and to which arm of the study they had been allocated. The verbal information given to both groups by the researcher had been standardized and practiced before recruitment to ensure that all patients were given the same information. It was not possible to blind the researcher and the participants to the group allocation because the researcher worked through the PDA with those allocated to this group, and patients knew whether they had received the PDA.
Statistical analysis
The total and subscale DCS scores for the control and intervention groups were compared to assess the efficacy of the PDA. Since the scores were not normally distributed, the nonparametric Mann-Whitney U test was used to compare the 2 groups. Univariable linear regression analyses were undertaken to assess whether any demographic factors or the time point at which patients were recruited affected the total DCS scores. The assumptions of the regression analysis were verified by a study of the residuals. All statistical analyses were completed using SPSS software for Windows (version 22.0; IBM, Armonk, NY). A significance level of 0.01 was used for the Mann-Whitney U tests instead of the conventional 0.05 to prevent spuriously significant results arising from multiple testing. A significance level of 0.10 was used for the univariable regression analyses to determine which, if any, variables to include in a multivariable regression analysis as potential confounders when comparing the scores in the 2 groups.
Results
Participant flow
Seventy-two patients were randomized in total. One patient failed to fully complete the DCS; therefore, data for 71 patients were analyzed. A CONSORT diagram showing the flow of participants through the study is given in the Figure .
Baseline data
The baseline characteristics for age, sex, ethnicity, and the time point at which patients were recruited were similar in both groups ( Table I ).
