Experiences, perceptions and preferences regarding medical decision-making in South Korea: a nationwide cross-sectional survey of the general public

Abstract

Objectives

Active patient involvement is a core principle of patient-centred care, yet public experiences of medical decision-making in non-Western settings remain underexamined. In South Korea, nationwide evidence on how adults experience, perceive and prefer medical decision-making is limited. This study, therefore, examined Korean adults’ experiences, perceptions and preferences regarding medical decision-making.

Design

A cross-sectional study was conducted between March and April 2025 using online data collection for adults aged 19–59 years and face-to-face interviews for those aged 60 years or older.

Setting

This nationwide survey was conducted in South Korea.

Participants

A total of 1081 Korean adults were recruited using proportional quotas for sex, age group and region. After excluding withdrawals and invalid responses, 1000 were included (response rate 92.5%).

Results

Overall, 70.4% of respondents reported at least one significant health-related decision in the past 2 years. Although 34.1% reported making their most recent decision independently, a larger proportion preferred collaborative decision-making involving clinicians and/or family members. The clinician’s explanation was the most influential factor (77.4%). Preferences for primary decision-maker varied by clinical context: patient-led decisions were favoured for low-risk interventions such as vaccination (78.5%), whereas physician involvement was preferred for life-threatening illness (86.2%). Communication ratings were highest for presentation of treatment options (mean score 3.56±0.79 on a 5-point scale) and lowest for explanation of potential treatment risks (mean score 3.20±0.89). Participants satisfied with decision outcomes reported higher communication quality (p<0.001), while those reporting neutral satisfaction (37.1%) resembled dissatisfied participants (4.1%).

Conclusions

Medical decision-making was common, but respondents’ experiences did not always match their preference for collaborative involvement. The findings suggest that strengthening patient-centred care in South Korea will require not only improvements in patient–clinician communication, but also attention to family involvement and structural conditions such as limited consultation time and current reimbursement arrangements.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This nationwide survey used proportional quotas for sex, age group and region to enhance geographic and demographic coverage of Korean adults.

• The questionnaire was developed through literature review, expert review and pilot testing before field administration.

• Online forced-response validation and interviewer checks during face-to-face interviews resulted in a complete analytic dataset with no missing values.

• Different recruitment sources and survey modes were used for younger and older adults, and older adults were recruited in outpatient settings, which may have introduced selection and mode effects.

• Several items permitted multiple responses, which captured the breadth of influences on medical decision-making but limited estimation of the relative salience of individual options.

Introduction

Modern medicine has moved towards patient-centred care (PCC), which emphasises respect for and responsiveness to individual patients’ values and needs in all clinical decisions.¹ PCC emerged in the 1980s based on the recognition that improving patients’ experiences of care requires placing them at the centre of care.¹ Since then, shared decision-making (SDM) has been promoted as a practical route to achieve PCC—captured by the 2011 Salzburg Statement urging clinicians ‘to do things with patients, not to patients’.² In this framework, PCC describes why patients should be involved, whereas SDM describes how that involvement can occur in practice.³

Research from Western countries has examined how the public engages in medical decision-making (MDM). For example, a Canadian telephone survey of 635 adults identified common decisional needs among people facing complex health decisions.⁴ In the USA, a survey of 3010 adults aged 40 years and older described the prevalence of several clinical decisions and how they were made.⁵ A Danish national survey of more than 6000 respondents assessed public perceptions of clinical decisions and the factors influencing them.⁶ Collectively, these studies reported high interest in involvement and identified unmet needs for information and support during decision-making.

However, evidence from South Korea remains limited, creating a barrier to developing culturally appropriate PCC models. To our knowledge, no previous study has examined the MDM experience using a nationwide sample of Korean adults. A web-based survey in 2015 offered initial insights into Koreans’ views; however, it included only 500 adults in a metropolitan area and used convenience sampling.⁷ The generalisability is therefore limited, and more contemporary data are needed, given the rapid changes in healthcare over the past decade. Another study analysed secondary data from the Korea National Health and Nutrition Examination Survey, but focused on factors associated with patient participation rather than on people’s own experiences or attitudes.⁸ In summary, an important gap remains in understanding how adults in South Korea currently experience and perceive MDM. The absence of nationwide data comparable with large-scale studies from Western countries^{4 5} hampers efforts to promote patient-centred decision making in South Korea.

Addressing this gap is important, both locally and globally. Most models of patient involvement have been developed in Western contexts.⁹ In South Korea, family members often act as primary caregivers and are frequently involved in clinical communication and care decisions.¹⁰ Such involvement may provide practical support and help align decisions with patients’ day-to-day circumstances.^{10 11} In addition, Korean clinical culture has historically been characterised by relatively high trust in physicians, which may make clinician-led decisions more acceptable to some patients, especially in complex or high-risk situations.^{7 12} It remains unclear to what extent Western SDM models apply in settings that place greater value on family involvement and clinician authority. The Korean healthcare system also operates under distinctive constraints. It has one of the highest outpatient visit frequencies and the shortest average consultation times in the Organisation for Economic Co-operation and Development.¹³ In the 2025 Survey on the Experience with Healthcare Service in South Korea, the mean outpatient consultation time was 9 min, and 70.2% of respondents reported consultations lasting 4–10 min.¹⁴ Because time pressure is a frequently reported barrier to patient-centred communication,^{15 16} these conditions may hinder the communication and deliberation needed for high-quality MDM. Examining current MDM practices in South Korea may therefore offer useful lessons for other health systems attempting to implement PCC under high-volume and time-pressured conditions.

The present study aimed to fill this knowledge gap and inform clinicians, researchers and policymakers. Clinicians may better understand how potential patients perceive their medical encounters. Researchers can identify areas to develop tailored PCC implementation strategies, and policymakers may use this evidence to design supportive policies. Using a nationwide sample of Korean adults, we examined recent experiences, perceptions and preferences regarding MDM in South Korea. We also explored how these experiences and preferences varied across clinical contexts and according to satisfaction levels with previous decisions.

Methods

Study design

This study was designed as a nationwide cross-sectional survey to assess the public’s experiences with and attitudes towards MDM in South Korea.

Study population and sampling strategy

This study targeted adults aged 19 years or older. We used proportional quotas based on sex, age group and residential region (17 regions: nine provinces and eight metropolitan cities) to approximate the national population distribution. Eligible participants were adults residing in South Korea who were able to communicate in Korean and provide informed consent. Individuals whose cognitive impairment precluded survey completion were excluded. In the absence of a prior nationwide estimate for the main survey proportions, the minimum sample size was calculated using the standard single-proportion formula.¹⁷ We assumed p=0.50 because this provides the most conservative sample size when the expected proportion is unknown. With a 95% confidence level and a margin of error of approximately 3.1%, the minimum required sample size was 999, which was rounded to 1000.¹⁸

Questionnaire development

We developed a structured questionnaire to assess the participants’ experiences, perceptions and preferences regarding MDM. Because no single existing instrument covered all study objectives, questionnaire development proceeded in three stages. First, six researchers—two physicians, a registered nurse, two medical students and a public health expert—reviewed previous surveys of patients’ and clinicians’ experiences and attitudes towards MDM4,719 and Korean health service survey structures1420,22 to identify relevant domains and candidate items. Items were adapted from existing surveys where appropriate, and additional items were developed de novo for the Korean context, particularly to capture family involvement in decision-making. Second, an expert committee comprising a physician, two registered nurses, a sociologist, an ethicist and an independent public survey expert iteratively reviewed the draft for content and face validity. Third, the revised questionnaire was pilot tested in 10 adults who were not involved in questionnaire development; their results were not included in the final analysis. Minor corrections were made before finalisation.

The final questionnaire comprised 56 items in four sections: (1) sociodemographic characteristics (sex, age, education level, marital and cohabitation status and household income), (2) health status and medical history, (3) experiences, perceptions and preferences regarding medical communication, including decision-making and (4) others. Internal consistency of the 12-item medical communication section was assessed using Cronbach’s alpha in the final analytic sample. An English translation of the questionnaire items analysed in this study is provided in online supplemental file 1.

Data collection

Data were collected between 25 March and 24 April 2025 by Hyundai Research,²³ a professional survey agency with national fieldwork capacity. The agency implemented the prespecified quota scheme and administered the survey, but had no role in study design, questionnaire development, data analysis, interpretation or manuscript writing. The survey was conducted in two ways. Adults aged 19–59 years were recruited from the agency’s pre-existing online panel according to the quota targets.²⁴ Adults aged 60 years or older were recruited through face-to-face interviews in outpatient departments across the quota-defined regions because in-person administration was considered more feasible for respondents who might need assistance completing the questionnaire. Overall, 1081 individuals entered the survey process.

Statistical analyses

Data cleaning was completed before the main analyses. The online survey platform required responses to all items, and trained interviewers checked questionnaire completeness during face-to-face administration; consequently, the analytic dataset contained no missing values. Potentially careless responses, including straight-lining and implausibly short completion times, were identified during quality control and excluded before analysis.

Age was categorised as young (<40 years), middle-aged (40–64 years) and older (≥65 years). Average monthly household income was divided into high (≥£2550), middle (≥£1550 and <£2550) and low (<£1550) categories based on the Korea National Health and Nutrition Examination Survey,¹⁹ using an exchange rate of GBP 1=KRW 1950. Satisfaction with the most recent decision was categorised as dissatisfied (including ‘strongly dissatisfied’), neutral and satisfied (including ‘strongly satisfied’). Continuous variables are presented as means and SD, and categorical variables as frequencies and percentages. Group differences were assessed using the Kruskal-Wallis test, with the Mann-Whitney U test for post hoc pairwise comparisons. Bonferroni correction was applied to adjust for multiple testing. Statistical significance was set at p<0.05. All analyses were performed using IBM SPSS Statistics V.27.0 (IBM).

Reporting guideline

We used the Checklist for Reporting Of Survey Studies to enhance reporting quality (online supplemental file 2).²⁵

Patient and public involvement

Patients, caregivers and members of the public reviewed the near-final questionnaire and provided feedback on its relevance and clarity. They were not involved in participant recruitment, data collection, data analysis, interpretation or manuscript drafting.

Results

A total of 1081 individuals entered the survey process. Among 842 adults aged 19–59 years who started the online survey, 81 were excluded due to withdrawals (n=17), quota overfills (n=60) or careless responses (n=4), leaving 761 valid online responses. In the face-to-face survey of adults aged 60 years or older, 239 interviews were completed and retained. The final analytic sample comprised 1000 participants, corresponding to an overall response rate of 92.5% (figure 1).

Figure 1. Flow chart of the data collection process.

General characteristics

Table 1 summarises participant characteristics. Of the 1000 participants, 506 (50.6%) were female, and 494 (49.4%) were male. Middle-aged adults (40–64 years) comprised 469 (46.9%) of the sample, while 292 (29.2%) were younger than 40 years and 239 (23.9%) were aged 65 years or older. Most participants had an undergraduate degree or higher (n=625; 62.5%), and the most common marital status was married (582; 58.2%). A high average monthly household income was reported by 383 participants (38.3%).

Table 1. General characteristics (n=1000).

Variables and categories	N (%)
Sex
Female	506 (50.6)
Male	494 (49.4)
Age group (years)
Young (<40)	292 (29.2)
Middle-aged (≥40, <65)	469 (46.9)
Older (≥65)	239 (23.9)
Education level
Undergraduate or above	625 (62.5)
High school or below	375 (37.5)
Marital status
Married	582 (58.2)
Single	301 (30.1)
Bereaved	76 (7.6)
Divorced	33 (3.3)
Cohabitating	5 (0.5)
Separated	3 (0.3)
Average monthly household income*
High (≥£2550)	383 (38.3)
Middle (≥£1550, <£2550)	268 (26.8)
Low (<£1550)	349 (34.9)

^*GBP 1=KRW 1950.

Perceived health status and medical history

Table 2 summarises self-rated health, medical history and satisfaction with previous treatment experiences. Three hundred and four participants (30.4%) rated their health as good or very good, whereas 486 (48.6%) rated it as neither good nor poor. More than half were currently receiving treatment or had been diagnosed with a health condition (514; 51.4%) and were taking medication (513; 51.3%). Nearly half reported having a family member who was currently receiving treatment or living with a diagnosis (477; 47.7%). In the previous 2 years, 598 participants (59.8%) had undergone surgery, a procedure or a diagnostic test. Among the 925 participants who reported any previous medical treatment, 579 (62.6%) were satisfied with their most recent treatment experience.

Table 2. Perceived health status and medical history (n=1000).

Variables and categories	N (%)
Self-rated health
(Very) good	304 (30.4)
Neutral	486 (48.6)
(Very) poor	210 (21.0)
Currently being treated or diagnosed	514 (51.4)
Currently taking medications	513 (51.3)
Having family members who are currently being treated or diagnosed	477 (47.7)
Experience of hospitalisation in the past 2 years	190 (19.0)
Experience of surgery/procedure, or diagnostic tests in the past 2 years	598 (59.8)
Level of satisfaction with the most recent experience of medical treatments (n=925)
(Highly) satisfied	579 (62.6)
Neutral	294 (31.8)
(Highly) dissatisfied	52 (5.6)

Experiences and perceptions regarding MDM

Table 3 presents participants’ experiences and perceptions regarding MDM during the previous 2 years. Of the 1000 respondents, 704 (70.4%) reported at least one significant health-related decision during that period. The most common decision type was whether to undergo a diagnostic test (n=332; 47.2%), followed by medication therapy (n=208; 29.5%). Decisions were most often made in secondary-care settings (n=273; 38.8%). In their most recent decision, 240 participants (34.1%) reported deciding alone, whereas 211 (30.0%) decided with clinicians. When asked about their preferred decision-making arrangement, 285 of the 1000 respondents (28.5%) preferred making decisions with clinicians and 251 (25.1%) preferred decisions involving both family members and clinicians (figure 2). A total of 413 participants (58.7%) reported that clinicians had explained the medical options in detail during their most recent MDM. When multiple responses were allowed, the clinician’s explanation was the most influential factor (77.4%), followed by family opinion (25.1%) and respondents’ own values, beliefs or preferences (16.1%). Among all 1000 respondents, the most common non-clinician information source was internet searching, such as Google (n=358; 35.8%), followed by healthcare organisations (n=235, 23.5%). Among respondents with decision-making experience, 414 of 704 (58.8%) were satisfied with their most recent decision.

Table 3. Experiences and perceptions regarding medical decision-making (n=704).

Variables and categories	N (%)
Whether to make any significant health-related decisions in the past 2 years (n=1000)
Yes	704 (70.4)
No	296 (29.6)
Type of significant health-related decisions, based on the most recent experience
Diagnostic test	332 (47.2)
Medication therapy	208 (29.5)
Surgery/procedure	115 (16.3)
Admission	37 (5.3)
Pregnancy/contraception	12 (1.7)
Type of healthcare organisation where the decision was made, based on the most recent experience
Secondary level	273 (38.8)
Primary level	223 (31.7)
Tertiary level	200 (28.4)
Others	8 (1.1)
Whether clinicians explained medical options in detail
(Strongly) Yes	413 (58.7)
Neutral	226 (32.1)
(Strongly) No	65 (9.2)
People involved in the decision
Alone	240 (34.1)
With clinicians	211 (30.0)
With family	179 (25.4)
With clinicians and family	74 (10.5)
Influencing factors in the decision*
Explanation of clinicians	(77.4)
Opinion of the family	(25.1)
My values, beliefs or preferences, including religion	(16.1)
Information acquired from the internet search	(12.6)
Advice of significant others	(6.8)
Financial situations	(6.0)
Information acquired from support groups	(1.3)
Others	(0.1)
Health information sources, excluding clinicians (n=1000)
Searching the internet (eg, Google)	358 (35.8)
Healthcare organisations	235 (23.5)
Family or significant others	212 (21.2)
Television	99 (9.9)
Social media (eg, YouTube)	93 (9.3)
Others	3 (0.3)
Level of satisfaction with the decision
(Strongly) Satisfied	414 (58.8)
Neutral	261 (37.1)
(Strongly) Dissatisfied	29 (4.1)

^*Multiple choices were allowed.

Figure 2. Preferred type of health-related decision-making.

Preferred stakeholders involved in the decision-making by clinical situation

We examined preferred stakeholder involvement across eight clinical situations. Across all situations, respondents consistently endorsed patient involvement, ranging from 54.3% for emergency care decisions to 78.5% for vaccination decisions. Patients were most often selected as the primary stakeholders for vaccination (78.5%), whereas physician involvement was most prominent for life-threatening illness (86.2%) and emergency care (79.8%). Family involvement was most often preferred for emergency care (66.6%) and for life-sustaining treatment or hospice decisions (65.5%). In end-of-life care, preferences for the involvement of patients (68.2%), family members (65.5%) and physicians (63.9%) were relatively similar (figure 3).

Figure 3. Preferred involvement of patients, caregivers and clinicians across clinical decision-making contexts. Multiple choices were allowed. This figure presents the cumulative distribution of each question.

Medical communication during clinical encounters

Figure 4 illustrates participants’ ratings of 12 aspects of medical communication during their most recent clinical encounter. The highest mean score was for clinicians’ presentation of the available options (3.56±0.79 on a 5-point scale), followed by explanation of the medical condition (3.54±0.75). The lowest mean score was for explanation of possible treatment risks (3.20±0.89), followed by having sufficient time to decide among the available options (3.31±0.88). Overall satisfaction with clinician communication was 3.44±0.82. The 12 items assessing experiences of medical communication showed high internal consistency (Cronbach’s alpha=0.91).²⁶

Figure 4. Mean agreement scores for 12 clinician communication behaviours during medical decision-making. The 5-point Likert scale (1=strongly disagree to 5=strongly agree) was used.

Table 4 compares medical communication experiences across three levels of satisfaction with the most recent MDM: dissatisfied, neutral and satisfied. All 12 communication items differed significantly across the three groups (all p<0.001). The satisfied group consistently had the highest mean rank, followed by the neutral group. In post-hoc pairwise analyses, the satisfied group scored significantly higher than both the neutral group (all p<0.001) and the dissatisfied group (all p<0.001) on all items. By contrast, differences between the neutral and dissatisfied groups were not significant for most items, including explanations of benefits and risks (p=0.77).

Table 4. Comparison of medical communication experiences and perceptions by level of decisional satisfaction.

	Decisional satisfaction	N	Mean rank	Kruskal-Wallis	Mann-Whitney U test*
					D vs N	N vs S	S vs D
Clinicians provided a detailed explanation of my medical condition.	D	29	223.74	<0.001	1.00	<0.001	<0.001
	N	261	253.17
	S	414	424.14
Clinicians presented the available options for me to choose from.	D	29	220.38	<0.001	0.31	<0.001	<0.001
	N	261	279.41
	S	414	407.83
Clinicians explained the benefits and risks of available options.	D	29	235.24	<0.001	0.77	<0.001	<0.001
	N	261	277.85
	S	414	407.78
When necessary, clinicians asked for my values and preferences regarding available options.	D	29	218.86	<0.001	0.23	<0.001	<0.001
	N	261	284.94
	S	414	404.46
During my medical consultations, I asked clinicians questions until full comprehension of the information.	D	29	237.74	<0.001	0.55	<0.001	<0.001
	N	261	288.01
	S	414	401.19
I was given opportunities to ask questions to clinicians during the consultation.	D	29	162.64	<0.001	0.02	<0.001	<0.001
	N	261	266.53
	S	414	420.00
I was given sufficient time to decide on the available options.	D	29	205.12	<0.001	0.28	<0.001	<0.001
	N	261	268.21
	S	414	415.96
My values and preferences were reflected in the final clinical decisions.	D	29	232.81	<0.001	0.82	<0.001	<0.001
	N	261	273.71
	S	414	410.55
I was informed about the expected outcomes of the tests or treatments.	D	29	256.03	<0.001	1.00	<0.001	<0.001
	N	261	279.59
	S	414	405.22
I was informed about the possible risks of the tests or treatments.	D	29	260.45	<0.001	1.00	<0.001	<0.001
	N	261	286.90
	S	414	400.30
After tests or treatments, I was able to consult with clinicians if I experienced any discomfort.	D	29	187.83	<0.001	0.18	<0.001	<0.001
	N	261	257.87
	S	414	423.69
Overall, I was satisfied with the communication I had with clinicians.	D	29	151.14	<0.001	0.01	<0.001	<0.001
	N	261	258.79
	S	414	425.68

^*Adjusted by Bonferroni correction.

D, dissatisfied; N, neutral; S, satisfied.

Discussion

This cross-sectional survey provides, to our knowledge, the first nationwide overview of how Korean adults experience, perceive and prefer MDM. Most respondents had encountered at least one significant health-related decision during the previous 2 years. Although deciding alone was the single most common reported experience, most respondents preferred collaborative decision-making involving clinicians and/or family members. Taken together, these findings suggest a gap between preferred and experienced participation. This interpretation is reinforced by the prominence of clinicians’ explanations as a determinant of decisions, alongside comparatively lower ratings for explanations of risks and for having sufficient time to decide. Overall, the findings suggest that many respondents wanted more collaborative and better supported decision-making than they had experienced in practice.

Comparison with other countries suggests both similarities and differences. The prevalence of recent decision-making experience in South Korea (70%) lay between the figures reported in the USA (82% among adults aged 40 years or older)⁵ and Canada (65% among adults).⁴ The higher US estimate may partly reflect the older age profile of that study, which excluded adults aged 19–39 years, who generally have fewer health problems.²⁷ South Korea’s rate may also be related to high healthcare utilisation: among Organisation for Economic Co-operation and Development countries, South Korea has the highest number of outpatient visits per capita.¹³ However, frequent contact with healthcare services does not necessarily translate into more participatory decision-making. In our study, only 40.5% of respondents reported that clinicians were involved in their most recent decision. Although direct cross-national comparisons should be interpreted cautiously because measures and study populations differ, international experience suggests that policy and legal frameworks can influence the extent to which SDM is embedded in routine care.28,30 Compared with settings where SDM has been supported by explicit legal or institutional drivers, South Korea may have fewer formal levers that encourage deliberative decision-making in everyday practice.

Our findings on context-dependent preferences for stakeholder involvement support a flexible, triadic decision-making model in the Korean context.¹¹ Respondents did not endorse a rigid ‘patient always decides’ or ‘doctor knows best’ position. Instead, preferences varied according to the perceived risk and urgency of the scenario. Patients were most often seen as the main decision-makers for lower-risk and more routine choices such as vaccination,³¹ whereas physicians were preferred in higher-risk or emergency situations, presumably reflecting trust in professional expertise when the stakes are high. At the same time, many respondents wanted family members to be involved in stressful or serious situations. Two-thirds preferred family involvement in emergency care, and in end-of-life decisions the roles of patients, families and physicians were valued almost equally. This pattern is consistent with relational autonomy,³² in which decision-making is shaped by social relationships as well as individual choice. Western SDM models, which commonly focus on the patient–clinician dyad, may therefore require adaptation in contexts where patient–family–clinician triads are important.¹¹ Our findings support the view that family involvement should be explored rather than assumed.³³ Clinicians should therefore ask patients how they want decisions to be made and who they want involved, rather than relying on cultural stereotypes alone.

One notable finding was the non-linear relationship between communication quality and decisional satisfaction. On many items, the neutral satisfaction group was statistically indistinguishable from the dissatisfied group, suggesting that communication which is merely adequate may not be sufficient to produce a positive decisional experience. This pattern is consistent with a threshold-like interpretation³⁴: may need to go beyond one-way information transfer and achieve genuine two-way engagement before patients feel positively about a decision. In this respect, there is an important distinction between informed consent as disclosure of information and SDM as a deliberative exchange.³⁵ Our findings suggest that providing standard information alone may satisfy minimum procedural requirements, yet still leave many patients feeling unconvinced that their values were meaningfully incorporated. By contrast, when respondents perceived more active engagement from clinicians, decisional satisfaction was markedly higher.

However, the current Korean healthcare system presents institutional obstacles to achieving this higher communication standard. Very short consultation times, high patient volumes and fee-for-service incentives may discourage extended discussion.^{13 14 36} Clinicians may therefore prioritise efficient information delivery over the more time-intensive tasks of eliciting patient values and supporting deliberation.³⁶ Even well-intentioned clinicians may find it difficult to provide the depth of communication associated with positive decisional experiences. Strengthening PCC in South Korea may therefore require not only clinician-level communication strategies but also system-level changes such as payment reform, organisational support and longer consultation time for complex decisions.

This study has several limitations. First, the findings relied on self-reported experiences over the previous 2 years and may therefore be affected by recall bias. Future studies could combine self-reports with more objective sources, such as medical records or recorded clinical encounters.³⁷ Second, different recruitment sources and survey modes were used by age group: adults aged 19–59 years were recruited from an online panel, whereas adults aged 60 years or older were interviewed face to face in outpatient departments. These design features may have introduced both mode effects and selection bias, particularly in the older group, and may have affected generalisability. Future research could use population-based mixed-mode sampling with post-stratification adjustment.³⁸ Third, although the questionnaire underwent expert review, pilot testing and internal consistency assessment, full psychometric validation of all domains was beyond the scope of this study. Finally, several items allowed multiple responses to capture the complexity of MDM experiences; while this broadened descriptive coverage, it limited estimation of the relative salience of individual response options.³⁹

Despite these limitations, this study provides, to our knowledge, the first nationwide overview of public experiences, perceptions and preferences regarding MDM in South Korea. The findings suggest that efforts to advance PCC in South Korea should complement the system’s quantitative strengths with greater attention to communication quality and meaningful involvement in decisions. More broadly, the results support expanding conventional dyadic SDM frameworks to accommodate contexts in which family members often play an important role in care.^{11 33 40} This may be relevant not only in South Korea and other Asian settings but also in multicultural health systems in which family-centred values shape healthcare decisions.

Conclusions

MDM was common, and many respondents preferred collaborative models involving clinicians and, in some situations, family members. However, several aspects of communication—particularly discussion of risks and time for deliberation—received comparatively low ratings. Strengthening PCC in South Korea may therefore require attention not only to clinicians’ communication practices but also to family involvement and system-level conditions such as consultation time and reimbursement arrangements.

Abbreviations

MDM

medical decision-making

PCC

patient-centred care

SDM

shared decision-making

Data availability statement

Data are available on reasonable request.