Shared Decision-Making for Determining Treatment Strategies in Low-Risk Thyroid Cancer: Protocol of a Multicenter Cluster-Randomized Trial (MAeSTro-SDM)

Abstract

Active surveillance (AS) is recommended for low-risk papillary thyroid microcarcinoma (PTMC) in many guidelines. However, while its clinical application requires incorporation of patient values, implementing shared decision-making (SDM) in practice remains challenging. To generate reliable evidence, facilitate the integration of SDM into routine PTMC managements and improve patient satisfaction, this study developed a PTMC-specific SDM model (SM group) and aims to evaluate whether it improves patient-reported outcomes (PROs) compared to usual care (UC group) in patients with low-risk PTMC. This multicenter, parallel-group, cluster-randomized controlled trial will enroll 310 patients with low-risk PTMC across seven academic hospitals in Korea. Participants will be assigned to either the SM group (model) or the UC group (control) through cluster randomization of 26 clinicians, stratified by specialty and AS experience. The SM group will receive structured counseling using a newly developed PTMC-specific SDM model, supported by decision aids such as educational videos, web-based card news, and illustrated leaflets regarding disease-information and patients’ values. The UC group will receive standard counseling. The primary outcome is the Decisional Conflict Scale score. Secondary outcomes include satisfaction with decision-making process, decision regret, anxiety, and thyroid-specific quality of life. Data will be collected via the iCReaT v2.0 electronic Case Report Form, supplemented by electronic and paper-based PRO surveys. Assessments will be conducted at baseline, 1–4 weeks, and 6 months after the treatment decision.

Trial Registration

ClinicalTrials.gov Identifier: NCT06730893

Graphical Abstract

INTRODUCTION

Thyroid cancer is the most frequently diagnosed malignancy in South Korea, accounting for 12.7% of all cancer cases in 2021, with 35,303 new cases reported according to the Korea National Cancer Incidence Database.1 Owing to its high prevalence and the excellent prognosis of low-risk papillary thyroid microcarcinoma (PTMC), active surveillance (AS) has garnered increasing attention as a viable management strategy and is now recommended in several national and international guidelines.2 In Korea, the adoption of AS into clinical practice has been supported by a growing body of evidence, including data from multicenter prospective cohort studies.3,4,5 In line with this progress, the Korean Society of Thyroid Radiology and the Korean Thyroid Association (KTA) released dedicated imaging and clinical protocols for the implementation of AS in 2024 and 2025, respectively.6,7 Correspondingly, recent analyses of National Health Insurance Service claims data reveal a rising trend in patients choosing to defer immediate surgery within the first year after diagnosis,8 reflecting increased acceptance of AS in routine thyroid cancer care in Korea.

Despite these advances, several important challenges remain. Results from the Korean prospective MAeSTro study demonstrated that a notable proportion of patients who initially opted for AS ultimately underwent surgery without evidence of disease progression and continued to report impaired quality of life.9 Conversely, regret was also reported among those who underwent immediate surgery as well as among patients who later transitioned from AS to surgery, often citing the indolent nature of their disease.10,11 Patient surveys identified a lack of disease-specific information and insufficient communication with healthcare providers as key barriers to informed and confident decision-making11 (presented in 2024 KTA meeting). A nationwide survey of KTA members further highlighted practical barriers to implementing AS, including limited consultation time, challenges in delivering up-to-date information, uncertainty in selecting appropriate candidates, and concerns about disease progression and potential medicolegal risks.12

Shared decision-making (SDM) has been proposed as a fundamental component of value-based healthcare and has been legally recognized in the United States since the 1980s as a core element of patient autonomy.13,14 Over the past decade, efforts to implement SDM in clinical practice have increased. However, widespread adoption remains challenging due to various barriers inherent to the SDM process. To overcome these challenges and facilitate the integration of SDM into routine clinical care, various SDM models have been developed and tailored to specific diseases and clinical environments.15,16,17 These models have been shown to enhance patient satisfaction, clinician engagement, and the personalization of care. However, the widespread implementation of SDM remains limited due to practical barriers, including inadequate consultation time, lack of multidisciplinary support, and insufficient evidence regarding cost-effectiveness. Currently, integration of SDM into national healthcare systems has only been started in a few countries, such as Germany and Denmark.18,19,20 In Korea, awareness and application of SDM model are still in their early stages, with only a few pilot initiatives - for example, in renal replacement therapy - recently being introduced.21 Notably, no validated SDM models currently exist for thyroid diseases, either domestically or internationally. Given Korea’s exceptionally high prevalence of low-risk thyroid cancer, there is a clear need to develop a PTMC-specific SDM model (PTMC-SDM) that supports individualized decision-making between AS and immediate surgery. This model should be compatible with Korea’s national healthcare system and, importantly, culturally and emotionally aligned with the values and preferences of Korean patients. Importantly, its implementation should be supported by evidence of clinical effectiveness as well as cost-effectiveness to facilitate integration into the national healthcare framework.

This study aims to apply a newly developed PTMC-SDM model, specifically adapted to the Korean healthcare context, to support treatment decision-making in patients with low-risk PTMC, and to evaluate its impact on patient-reported outcomes (PROs) compared to usual care. This trial is a substudy of the MAeSTro-EXP cohort study and is designed to test the hypothesis that a structured SDM model enhances patient satisfaction with treatment decisions.22 The secondary objectives are: 1) to evaluate whether the SDM model influences patients’ choice between AS and surgery; 2) to assess the frequency of treatment decision changes over time; and 3) to compare PROs between the intervention and control groups.

METHODS

Study design and setting

This study is a multicenter prospective trial (MAeSTro-SDM), to facilitate shared decision-making for determining treatment strategies in low-risk thyroid cancer patients, conducted in Korea. A parallel, two-group cluster-randomized design will be used to test whether the PROs of SDM-model group (SM; model arm) is different from those of the usual care group (UC; control arm). Patients who are diagnosed with low-risk PTMC will be recruited from Seoul National University Hospital (Seoul, Korea), the National Cancer Center (Goyang, Korea), Seoul National University Bundang Hospital (Seongnam, Korea), Seoul Metropolitan Government-Seoul National University Boramae Medical Center (Seoul, Korea), Hanyang University Hospital (Seoul, Korea), Nowon Eulji Medical Center (Seoul, Korea), and Chungbuk National University Hospital (Cheongju, Korea). Twenty-six clinicians with various experiences of AS will participate in the study. The enrollment for study participants commenced in May 2025.

Eligibility criteria

Patients diagnosed with low-risk PTMC who are faced with choosing their treatment strategy comprise the study population.

The inclusion criteria for the MAeSTro-SDM trial are as follows:

• 1) Adults aged 18 years or older with a thyroid nodule of 1.0 cm or less confirmed as Bethesda category V (suspicious for papillary thyroid cancer) or VI (papillary thyroid cancer) based on cytopathological examination, or Bethesda category III (atypia of undetermined significance) with a confirmed BRAF^V600E mutation.

• 2) Patients with no evidence of distant metastasis, cervical lymph node metastasis, recurrent laryngeal nerve invasion, or tracheal invasion. Additionally, there should be no evidence of gross extrathyroidal extension, and high-risk subtypes of PTC (e.g., diffuse sclerosing, columnar cell, or solid subtypes).

The exclusion criteria are as follows:

• 1) Patients who cannot undergo regular follow-up or are expected to have difficulties with follow-up.

• 2) Patients with indeterminate or benign findings on thyroid biopsy.

• 3) Patients with anterior subcapsular tumors with strap muscle replacement, paratracheal tumors abutting ≥ 90° to the trachea, posteromedial subcapsular tumors without intervening normal thyroid parenchyma, and posterolateral subcapsular tumors with obvious protrusion.

Patients who cannot continue regular follow-up or withdraw consent will be withdrawn from the study. However, if a patient returns for follow-up after the designated study period, they will not be considered as dropout cases if their data still contributes to the long-term prognosis analysis of low-risk PTMC. If a participant discontinues follow-up voluntarily but has consented to data collection and utilization in secondary data analysis, their data will be used for interim or final analysis.

These eligibility criteria for this study are identical to those of the MAeSTro-EXP cohort, except for tumor size.22 While the MAeSTro-EXP study permitted enrollment of patients with thyroid tumors up to 1.5 cm, the MAeSTro-SDM study limits inclusion to patients with tumors measuring 1.0 cm or smaller. This restriction reflects the current lack of sufficient clinical evidence, particularly evidence that can be generalized and incorporated into decision aids used in routine practice, regarding the safety and efficacy of AS for low-risk PTC larger than 1.0 cm.

Interventions

We developed an SDM model, PTMC-SDM, specifically tailored to support treatment decision-making in patients with low-risk PTMC. This model was developed by comprehensive literature reviews and by incorporating the preferences and values of patients and clinicians, as identified through focus group interviews.12 The PTMC-SDM model (Fig. 1) is based on the six-step framework and is designed to incorporate patients’ values and preferences throughout the decision-making process.15 To support implementation of this model, a PTMC-specific decision aid, Thyroid-NAVI PTMC (Navigator for Patients with Low-Risk PTMC), was developed. Thyroid-NAVI PTMC is intended to provide essential information, support value clarification, and guide both patients and clinicians through each step of the SDM process, thereby enhancing usability and facilitating integration into clinical settings.

Fig. 1. Shared-decision making model for low-risk papillary thyroid microcarcinoma (PTMC-SDM).

KSThR = Korean Society of Thyroid Radiology, KTA = Korean Thyroid Association, PTMC = papillary thyroid microcarcinoma, SDM = shared decision-making, Thyroid-NAVI PTMC = Navigator for Patients with Low-Risk PTMC.

It consists of three educational videos, 15 web-based card news modules, informational leaflets incorporating diagrams or charts to summarize key aspects of the disease and treatment options, and self-administered questionnaires to help patients clarify their personal values. For healthcare providers, training is provided through instructional videos and a guidebook that includes protocols and step-by-step procedures. These components are structured to be delivered in a stepwise manner, aligned with each phase of the 6-step SDM process.

In the SM group, structured counseling is conducted using these aids in accordance with the PTMC-SDM framework.

In contrast, the UC group receives counseling based on usual clinical practice and is provided with a text-only informational document on treatment decisions - thus same document is also provided to the SM group.

Outcomes

The primary outcome is Decisional Conflict Scale (DCS), which assesses an individual’s perception and understanding of decision-making by 16 items.23 The scale is 5-point Likert scale from 0 (strongly agree) to 4 (strongly disagree), and scores are calculated through a) summed, b) divided by 16, and c) multiplied by 25, and range from 0 (no decisional conflict) to 100 (extremely high decisional conflict).

Secondary outcomes include various PROs, including satisfaction with SDM, Decision Regret Scale (DRS), thyroid-specific quality of life (QoL), and anxiety.

Satisfaction with SDM is measured by three questionnaires. SDM-Q-9 is a nine-item questionnaire assessing the quality of SDM from both patient and clinician perspectives.24 While previously applied in thyroid dysfunction (Graves’ disease), it has not yet been studied in thyroid cancer patients.

Patients’ Perceived Involvement in Care Scale (PICS) is a 13-item questionnaire evaluating patient engagement in primary care.25 It measures Clinician’s information provision, patient’s proactive information-seeking, and patient participation in treatment decisions (5-point Likert scale; 1 = strongly disagree, 5 = strongly agree). Higher scores indicate lower clinician control and greater patient involvement in decision-making.

Mappin’SDM is a 15-item tool adapted from the 12-item OPTION scale to evaluate observed patient involvement in decision-making.26 Each item consists of behavioral and outcome components (a total of 30 items), rated by patients, clinicians, and observers (through real-time or via video review). The Korean version consists of 11 validated items and has been used alongside DCS for indeterminate thyroid nodules.

DRS measures distress and remorse after making a healthcare decision.27 Scale ranges from 1 (strongly agree) to 5 (strongly disagree). The final score is an average of all items, ranging from 0 (no regret) to 100 (high regret).

Thyroid-specific QoL is assessed by a Korean version of a 30-item thyroid-specific QoL questionnaire originally developed by Dow et al.28 and further refined and validated by Ryu et al.29

Anxiety and Depression will be assessed by Hospital Anxiety and Depression Scale (HADS).30 HADS is a 14-item tool (7 for anxiety, 7 for depression), developed to measure hospital patients' levels of anxiety and depression. Each item is rated on a 0–3 scale, with total scores ranging from 0 to 21 per subscale. A cutoff score of 8 or higher indicates clinically significant symptoms, based on Korean validation studies.

In addition, another secondary outcome examined the proportion of patient selecting each treatment option (immediate surgery vs. AS) and the rate of changes in treatment decisions during follow-up.

Assessment and follow-up schedule

Patients will be assigned to either the SM or UC group based on the random allocation of clinicians, stratified by specialty and clinical experience in AS. Patients will engage in treatment decision-making with their clinicians according to the modality of their assigned group (SM or UC group). Regardless of group allocation, the final decision between immediate surgery and AS will be made by the patient. Following the treatment decision, patients are assessed for DCS and satisfaction with SDM.

In both groups, the follow-up schedule is identical (Table 1). However, slight variations may occur based on whether patients select immediate surgery or AS. Regarding patients who choose immediate surgery, the extent of surgery (lobectomy or total thyroidectomy) will be determined based on tumor location. At 6 months after treatment decision, secondary and exploratory outcome measures will be reassessed. The patients who choose AS will return to the hospital 6 months later for neck ultrasound and blood tests. At 6 months after treatment decision, secondary and exploratory outcome measures will be reassessed.

Table 1. MAeSTro-SDM study flow.

Period		Screening	Intervention	Assessment period
Visit		0	1	2	3
Study month		−3 to 0	0	1	6
Informed consent		X
Clinician Randomization and Allocation		X
Eligibility screen		X
Medical history/demographic data		(X)	X
Physical examination		(X)	X		X
Thyroid function test		(Xa)	(X)		X
Neck ultrasound with pathologic diagnosis		(Xa)	(X)		X
Intervention
	NAVI (SM group)		X
	Usual consulting (UC group)		X
Co-medication			X
Video recording (duration of consulting)			X	X	X
Choice of treatment (AS or surgery)			(X)	X	(X)b
Patient-reported outcomes
	1) Knowledge evaluation		X	X	X
	2) NAVI (patients’ value and preference, stage of decision)		X
	3) QoL questionnaire		X		X
	4) Hospital Anxiety and Depression Scale		X	X	X
	5) Decision Conflict Scale		X	X
	6) Decision Regret Scale				Xc
	7) SDM-specific questionnaire (Mappin’SDM, SDM-Q-9, PICS)		X	X	X

SDM = shared decision-making, NAVI = navigator for patients with low-risk papillary thyroid microcarcinoma, SM = shared decision-making model, UC = usual care, AS = active surveillance, QoL = quality of life, PICS = Patients’ Perceived Intervention in Care Scale.

^aTest results obtained within 3 months prior to screening may be used.

^bThe choice of treatment may be changed at any visit.

^cDecisional Conflict Scale, instead of Decision Regret Scale, will be assessed until a decision is made.

A flowchart will illustrate the study process (Fig. 2). All assessments and visit schedules may be adjusted based on clinical judgment if deemed unnecessary or inappropriate for individual participants.

Fig. 2. MAeSTro-SDM study flow.

DCS = Decisional Conflict Scale, HADS = Hospital Anxiety and Depression Scale, NAVI = navigator for patients with low-risk papillary thyroid microcarcinoma, PICS = Perceived Involvement in Care Scale, QoL = quality of life, SDM = shared decision making, SDM-Q-9 = nine-item questionnaire assessing the quality of SDM, SM = shared decision making model, UC = usual care.

Upon obtaining informed consent, the date of consent will be designated as Visit 1, marking the initiation of the study. Patients will be randomly assigned to either the SM or UC group based on the clustered randomization of their clinician, stratified by the clinicians’ specialty with AS. Patients will participate in treatment decision-making with their clinician according to the protocol of their assigned group. The only procedural difference between the two groups is whether the newly developed PTMC-SDM model is used (SM group) or the consultation follows usual standard practice without the model (UC group) at Visit 1.

At Visit 1, following the consultation (i.e., the SDM process), patients will complete the following questionnaires: DCS, QoL, and HADS. Visit 2 will be scheduled 1 to 4 weeks later to allow patients sufficient time to consider their treatment options. During Visit 2, patients will make a final treatment decision in consultation with their clinician. Regardless of group allocation, the choice between immediate surgery and AS will be made solely by the patient. After the decision is made, patients will be assessed for decisional conflict using the DCS and for satisfaction with the SDM process using the SDM-Q-9, PICS, and Mappin’SDM questionnaire.

All patients will undergo a follow-up visit 6 months after Visit 1, which will include a neck ultrasound and blood tests. During the period, participants are allowed to change the treatment decision made at Visit 2, either from AS to surgery, or from surgery to AS, based on their preferences or request.

The follow-up schedule, including Visit 2 and Visit 3, is identical for both the SM and UC groups. However, minor variations in the schedule may occur depending on the patient’s treatment choice (immediate surgery vs. AS) or due to medical or non-medical circumstances unrelated to the study protocol. Nevertheless, Visit 3 will be conducted at 6 months (± 3-month window) after Visit 1, ensuring that an equivalent amount of time has passed since the SDM process, regardless of treatment progression or changes. At this visit, secondary and exploratory outcome measures will be re-evaluated (Table 1, Fig. 2).

Initial evaluation (Visit 1, baseline; Visit 2, 1-4 weeks)

Demographic and clinical data will be collected in Visit 1 or 2 as follows:

• - Demographic and Physical variables: Gender, age, body weight, blood pressure, pulse rate

• - Imaging assessment: Neck ultrasound

• - Thyroid Function Tests: Free T4, Total T3 or free T3, TSH

• - Thyroid Cancer Markers: Thyroglobulin (Tg), Thyroglobulin Antibody (TgAb)

PROs (DCS, QoL, and HADS at Visit 1; DCS, SDM-Q-9, PICS, Mappin’SDM, and HADS at Visit 2) will be collected using paper forms or tablets via Google Forms.

Follow-up assessments (Visit 3, 6 months ± 3 months after treatment decision)

Demographic and clinical data will be collected as follows:

• - Physical variables: Body weight, blood pressure, pulse rate

• - Imaging Assessments: Neck ultrasound

• - Thyroid Function Tests: Free T4, Total T3 or free T3, TSH

• - Thyroid Cancer Markers: Tg, TgAb

PROs (DRS, SDM-Q-9, PICS, Mappin’SDM, QoL, and HADS) will be collected via paper forms or tablets using Google Forms.

Sample size

A stratified cluster-randomized design with 26 strata will be used to test whether the SM group mean (μ1) is different from the UC group mean (μ2) (H0: δ = 0 versus H1: δ ≠ 0, δ = μ1 − μ2). The comparison will be made using a generalized linear mixed effect model (GLMM), with a Type I error rate (α) of 0.05. The standard deviation of the individual-level residual error in the GLMM is assumed to be 1. With this assumption, 264 subjects are required to achieve 90% power to detect a mean difference (μ₁ − μ₂) of 0.4 at a two-sided significance level of 0.05.

For the cluster-randomized design, the required number of participants per cluster is calculated using an inflation factor to account for intracluster correlation. Assuming an intracluster correlation coefficient (ρ) of 0.01 and participation of 26 clinicians across 7 hospitals, the following calculation was applied:

Equation 1:

RCT Sample Size × (1−ρ)/(Number of Clusters − ρ × RCT Sample Size)

264 × (1 − 0.01)/(26 − 0.01 × 264) = 11.2

Thus, each of the 26 clinicians is required to enroll an average of 11.2 participants, resulting in a total of 290.9 participants (145.5 per group). To account for additional intracluster heterogeneity, 37 participants were added. Considering the short study duration and the fact that the intervention is not directly applied to patients, a dropout rate of 6% is assumed. Therefore, a total of 310 patients (155 per group) will be required. The sample size calculations were performed using PASS 2024 (version 24.0.2).

Allocation

To prevent information sharing between the SM and UC groups, cluster randomization was implemented at the level of recruiting clinicians. The clinicians involved will be assigned to study groups using a randomization table, stratified by specialty (surgeon vs. non-surgeon). A previous national survey demonstrated that strong adherence to surgery was driven by surgeons, irrespective of their clinical experience.12 A cluster randomization method will be applied at the clinician level, meaning that each clinician will consistently apply either the PTMC-SDM model or usual care. The allocation result is blinded for patients, but not for clinicians.

SM group (PTMC-SDM model group; model arm):

• - Patients will receive detailed information, including an information leaflet about treatment options.

• - Clinicians will guide patients through a structured PTMC-SDM model protocol before finalizing the treatment decision.

UC group (usual care group; control arm):

• - Patients will receive usual medical care following current clinical practices without adopting any structured SDM protocol.

Data collection methods, management plan and monitoring

Data required by the clinical research protocol including primary, secondary, and exploratory variables will be collected at the study sites using the electronic Case Report Form (eCRF) within iCReaT v2.0, a clinical research management system operated by the Korea National Institute of Health (https://icreat2.nih.go.kr/). Clinical data entry, data management, and central monitoring will be performed using eCRF. PROs will be collected via paper forms or tablets using Google Forms.

Data access will be restricted to the principal investigator (PI) and designated research staff via a secure Google account, and the site PI, sub-investigator, and study collaborators will pay maximum attention to security management.

All data must be accurately transcribed into the eCRF and should be verifiable through source documents. Examples of source documents include hospital electronic medical records, laboratory records, and clinical charts. All eCRFs must be completed and signed by research personnel delegated by the PI. The PI is responsible for ensuring the accuracy of all data reported in the eCRF. Before finalizing the eCRF, researchers must check for any blank fields to avoid queries. Regular data audits and quality control procedures will be conducted to address discrepancies and maintain data integrity.

The data manager will transfer the final data set to the PI after statistical analysis and the data will be stored in the electronic format. Study-related data will be stored at participating institutions for 5 years after the end-of-study report or 3 years after all study-related publication, but desirably for as long as possible. After the retention period, any study-related sample or information will be discarded only after being anonymized. Auditing will be carried out by each institution’s audit system.

Statistical methods

The primary objective of this clinical trial is to evaluate whether the model group (SM group), supported by a decision-making model (PTMC-SDM) for treatment planning, demonstrates greater satisfaction with the decision-making process compared to the control (UC group) receiving usual care. Descriptive statistics will be used to summarize the general characteristics of the study participants; continuous variables will be reported as means and standard deviations, while categorical variables will be presented as frequencies and percentages.

For efficacy analysis, all randomized clinical trial participants will be included in the analysis, provided they have at least one efficacy evaluation after randomization (intent-to-treat full analysis set; FAS). Participants who violate the study protocol without any meaningful evaluation will be excluded from the FAS analysis. The per-protocol analysis set (PP) will include participants from the FAS who have completed the questionnaires at least twice.

The primary efficacy evaluation will be conducted in both the FAS and PP analysis sets, and a two-sided test with a significance level of 0.05 will be performed. The general characteristics of study participants and the secondary efficacy outcomes will be analyzed using the FAS analysis set, with a two-sided test at a significance level of 0.05.

Statistical analyses will be conducted at the individual participant level under the cluster-randomized design. Outcomes will be analyzed using GLMM to account for both clustering and repeated measurements. Treatment group will be included as a fixed effect, with clinician (cluster) and participant specified as random effects to model within-cluster and within-subject correlation, respectively. Appropriate distributions and link functions will be selected according to the outcome type. All tests will be two-sided with a Type I error rate of 0.05.

Ethics statement

The ethics of study protocol was reviewed and approved by the Institutional Review Board (IRB No. H-2501-118-1609 in Seoul National University Hospital, NCC2025-0008 in National Cancer Center, B-2506-976-405 in Seoul National University Bundang Hospital, EMCS IRB 2025-03-019 in Nowon Eulji Medical Center, and CBNUH IRB 2025-03-005-001 in Chungbuk National University Hospital). Participants will receive a detailed explanation of the study, and informed consent will be obtained before enrollment. The study protocol has been registered on ClinicalTrials.gov (Trial registration number: NCT06730893). Any protocol amendment will be reviewed by the Institutional Review Board and updated on the trial registration site.

DISCUSSION

SDM has become a fundamental component of patient-centered care, particularly in oncology, where treatment choices often involve complex trade-offs between risks and benefits. This study protocol outlines a cluster-RCT designed to evaluate the effectiveness of a disease-specific SDM model in improving decision quality and patient satisfaction among PTMC patients.

The primary goal of this trial is to determine whether a structured SDM model (PTMC-SDM) tailored for patients with low-risk PTMC, can enhance understanding of treatment options, align decisions with patients’ values and preferences, and reduce decisional conflict. Although prior studies suggest that SDM can improve patient engagement and satisfaction, evidence regarding its influence on actual treatment choices in low-risk thyroid cancer remains limited. This may reflect both limited awareness of SDM among patients and clinicians, and more critically, the practical difficulties of implementing SDM effectively in real-world clinical settings. To address this gap, this trial integrates a structured SDM process—guided by the PTMC-SDM model using Thyroid-NAVI PTMC decision aids and provider training—into routine clinical care for PTMC patients.

By comparing this intervention with usual care, this study will provide valuable insights into strategies for enhancing the applicability, acceptability, and real-world implementation of SDM in diverse clinical settings. Furthermore, evaluating its impact on PROs, such as decisional regret and anxiety, will help to clarify the broader psychosocial benefits of SDM in the management of PTMC treatment.

A key strength of this trial is its randomized design, which minimizes selection bias and strengthens causal inferences. The use of cluster randomization further accounts for clinician-related factors that may influence treatment decisions. Moreover, the integration of both qualitative and quantitative measures enables a comprehensive evaluation of the effectiveness of PTMC-SDM model. The use of validated instruments for assessing decision quality and PRO measures enhances the methodological robustness and reliability of the findings. In addition, comparing consultation time and clinician satisfaction between SM and UC groups will offer new insights into the practical aspects of PTMC management and valuable evidence regarding the feasibility of implementing SDM in routine clinical practice.

However, several challenges may arise in the implementation and evaluation of this trial. Ensuring adherence to the PTMC-SDM process may require additional training and periodic reinforcement for participating investigators. Variability in patient health literacy and external influences, such as advice from acquaintances, could also affect the effectiveness of the intervention. Clinical factors, such as the difference in extent of surgery (lobectomy vs. total thyroidectomy), which are determined based on tumor size and location, may influence surgical outcomes, including complication rates. In addition, among patients who choose immediate surgery or later transition to surgery after initially choosing AS, the timing of surgery may vary. Even, some patients may not undergo surgery until after Visit 3 due to unforeseen clinical or logistical issues. Particularly in those who switch from AS to surgery, the timing of the decision change may affect whether surgery is performed before or after Visit 3, potentially influencing survey-based outcome measures and interpretation. If these clinical events are not evenly distributed across study arms, they may introduce confounding effects on the trial outcome. Additionally, variability in follow-up adherence and potential participant attrition may impact on the internal validity of the study results.

To address these challenges, we have incorporated several mitigation strategies. First, to address potential bias from unblinded clinicians and variability in communication, we developed SOPs for physicians and nurses and provided structured training using a standardized information script and explanatory video. All participating investigators will undergo standardized training sessions and workshops to ensure consistent delivery of the PTMC-SDM model and to reinforce adherence to the structured decision-making process. Refresher sessions will be provided periodically throughout the study period. To account for variability in patient health literacy, decision aids, Thyroid-NAVI PTMC, have been developed with simplified language and visual support to facilitate understanding across a wide range of educational backgrounds.

Additional explanatory support from clinicians will be encouraged when needed. To minimize variability in surgical timing and its influence on outcome measures, detailed documentation of decision change dates, surgical dates, and reasons for delay will be collected. Sensitivity analyses will be conducted to assess the potential impact of delayed or unevenly distributed events between groups. To enhance follow-up adherence and reduce attrition, participants will receive reminders for scheduled visits, and flexible scheduling options will be offered. Lost-to-follow-up cases will be carefully monitored and reported to assess their potential impact on study results. All patient–clinician consultations will be audio-recorded and assessed using the Mappin’SDM observer-rated tool to ensure consistency in the SDM process.

This study was designed to minimize physicians’ subjective influence and provide neutral information via a decision aid, safeguarding patients’ autonomy. If successful, it will offer robust evidence for integrating SDM into routine care for PTMC patients and support its wider adoption in clinical practice. If successful, this trial will provide robust evidence supporting the integration of SDM into routine clinical care for PTMC patients and will facilitate its broader adoption in real-world practice. By developing a PTMC-specific SDM model tailored to the Korean healthcare context, the study has the potential to improve satisfaction with the decision-making process for both patients and clinicians when choosing between treatment options.

The findings may inform future clinical guidelines by highlighting the importance of structured decision aids, not only in PTMC but also in the management of other thyroid diseases and preference-sensitive clinical decisions. Furthermore, identifying key barriers and facilitators to SDM implementation will yield practical insights to enhance patient-centered communication and support shared decision-making in everyday clinical practice. Beyond thyroid cancer, this study has the potential to contribute to the broader field of oncology by providing empirical evidence on both the benefits and challenges of applying a structured SDM model. Ultimately, the results are expected to support ongoing efforts to enhance patient autonomy, improve decision quality, and promote truly personalized cancer care.