Factors influencing patient involvement in treatment decision-making for diabetic retinopathy: a mixed methods study protocol

Abstract

Introduction

The diversity of treatment options for diabetic retinopathy (DR) and the high uncertainty about the benefits and risks of different treatment modalities necessitate shared decision-making between patients and healthcare professionals. However, little is known about the involvement of individuals with DR in treatment decision-making in China. This study aims to gain insight into the current status and factors associated with involvement in treatment decision-making in patients with DR. Furthermore, we will explore the experiences and perceptions of patients with DR regarding their involvement in treatment decision-making.

Methods and analysis

We will conduct a mixed-method study using an explanatory sequential design. In the quantitative research (n=350), participants’ actual decision-making roles, sociodemographic data, disease-related data, health literacy, need for involvement in decision-making, decision-making self-efficacy, social support and ophthalmologist facilitation of patient involvement will be investigated to analyse the current state of patient involvement in treatment decision-making and the factors influencing it. Descriptive statistics, one-way analysis of variance and multinomial logistic regression will be performed. During this period, individual semistructured interviews will be conducted with a subset of these participants to understand the perceptions and experiences of people with DR regarding their involvement in treatment decision-making, and thematic analysis will be used to analyse the interview data. Finally, the joint display will be used to integrate quantitative and qualitative data.

Ethics and dissemination

Ethical approval for this study has been obtained from the Ethical Review Committee for Human Trials of Shanghai General Hospital, China (number: 2024–098). Written informed consent will be obtained from all participants after they have been fully informed about the study, prior to any data collection. The study’s findings will be disseminated through peer-reviewed publications and conference reports.

Trial registration number

ChiCTR2400087906.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study will employ an explanatory sequential mixed-methods design to integrate quantitative and qualitative data.

• The use of the Capability, Opportunity and Motivation-Behaviour model will provide a robust theoretical framework to guide the investigation.

• As the study participants are primarily from Eastern China, the findings may not be generalisable to populations from other regions with different characteristics.

• Self-reported data in the quantitative phase will be subject to potential social desirability bias.

Introduction

Diabetic retinopathy (DR) is a prevalent microvascular complication of diabetes mellitus, frequently resulting in significant visual impairment.¹ It is a leading cause of visual impairment and blindness among working-age individuals globally, necessitating complex and long-term management to prevent irreversible blindness.² Modern ophthalmology offers a diverse range of treatment modalities, including laser photocoagulation, intravitreal injections of anti-vascular endothelial growth factor (anti-VEGF), steroids and vitrectomy.³ Furthermore, combination therapies involving various treatment methods have also been explored in recent years to further improve clinical outcomes.⁴ While these various treatments and combinations offer a range of options, they also complicate the decision-making process. For example, although panretinal photocoagulation effectively maintains long-term retinal stability, it may result in permanent ocular complications such as constricted visual fields and impaired scotopic vision.⁴ Similarly, while intravitreal injections of anti-VEGF are considered first-line therapy for the treatment of DR, they impose substantial burdens. Beyond the high costs and the clinical complexity of selecting among various agents, these treatments require frequent clinic visits and carry serious potential risks, such as elevated intraocular pressure, iatrogenic lens injury or endophthalmitis.⁵ Because these options involve significant variations in treatment burden, cost and potential complications, patients with DR must navigate a high degree of clinical uncertainty.⁶

Widely advocated for decades, patient involvement in treatment decision-making is recognised for its capacity to improve clinician–patient concordance, enhance satisfaction and optimise resource utilisation, ultimately leading to better patient outcomes and fewer medical disputes.7,10 As mainland China’s healthcare model shifts toward patient-centredness, the importance of involving patients in treatment decisions has gained significant traction and increased recognition among medical professionals.¹¹ However, the clinical reality of patient involvement in DR treatment decisions may be suboptimal, and evidence regarding their actual roles is currently insufficient. For example, a cross-sectional study in Germany involving 810 adults at secondary diabetes care centres found that as high as 74.3% of patients preferred shared decision-making before treatment of DR, whereas only about 17.4% wanted to delegate the decision-making process to their ophthalmologists.¹² In contrast, research in Vietnam indicated that patients frequently rely heavily on ophthalmologists’ expertise rather than actively selecting the treatment options that best suit their individual preferences.¹³ These contrasting findings suggest that the nature of patient involvement may be highly context-dependent. Furthermore, the actual decision-making status of patients with DR has not yet been systematically and comprehensively explored, especially as previous ophthalmic research has predominantly focused on glaucoma and cataracts.¹⁴ Identifying the specific factors associated with involvement is the foundation for providing effective support and promoting active participation among patients with DR. Therefore, there is a clear need for a systematic investigation using integrated methodologies to fill these gaps in the literature and offer evidence-based insights for clinical practice.

Patient involvement in treatment decision-making is a complex behaviour influenced by a multifaceted array of individual factors, including age,¹⁵ gender,¹⁶ income,¹⁷ education level,¹⁸ disease severity,¹⁹ health literacy²⁰ and self-efficacy,²¹ as well as environmental influences such as social support²² and physician support.²³ For example, older adults often adopt more passive roles in treatment decisions compared with younger individuals,²⁴ while patients with higher education levels are typically more inclined toward shared involvement.²⁵ Despite the identification of these general influences, it remains currently unclear whether such factors effectively apply to the unique treatment-related decisions encountered by patients with DR. Furthermore, existing research exhibits an over-reliance on single methodologies. While purely quantitative studies can identify trends and correlations, they often lack the depth required to uncover the underlying reasons behind the data.²⁶ Although qualitative approaches offer rich experiential insights, they face challenges in quantifying prevalence or statistically identifying predictive factors, thereby limiting the generalisability of the findings.²⁷ This methodological disconnection may hinder the systematic identification of the factors influencing patients with DR, leaving the underlying mechanisms and real-world processes through which these clinical decisions are navigated largely unexplored.

Theoretical framework

A theoretical framework is crucial for identifying relevant factors and gaining a better understanding of the mechanisms behind patient involvement in decision-making. The Capability, Opportunity and Motivation-Behaviour (COM-B) model has been applied in health-related behaviour research within healthcare settings to provide a systematic perspective.28,30 Schöne et al employed the COM-B model to identify facilitators and barriers to participation in a prehabilitation programme among older people with frailty syndrome prior to elective surgery.³¹ Likewise, Gogoi et al applied the COM-B model to investigate how migrants in the UK perceive participation in health research, identifying 16 distinct factors and suggesting corresponding intervention strategies.³² These studies collectively underscore the applicability of the framework in the field of patient engagement. As the core of the Behaviour Change Wheel, the COM-B model provides a systematic theoretical approach to identifying the cognitive, affective, social and environmental influences on patient involvement. Furthermore, its constructs can be mapped to identify a range of potential targets for future evidence-based interventions, such as tailored decision aids, to address the unique physiological and clinical constraints faced by patients with DR.

The COM-B model posits that Capability, Opportunity and Motivation constitute the three essential prerequisites for behaviour to occur.³³ In this model, Capability refers to the psychological and physical capacity required for an individual to execute a specific behaviour.³³ Health literacy reflects a patient’s fundamental ability to access, comprehend and use health information, playing a crucial role in patient participation in medical care.³⁴ Consequently, in this study, health literacy is incorporated into the Capability dimension. Motivation encompasses all conscious and unconscious internal driving forces.³³ Need for involvement in decision-making refers to a patient’s intrinsic desire to participate in the treatment decision-making process, where patients with a stronger need tend to engage more actively in medical care and exhibit greater autonomy.³⁵ Self-efficacy is defined as an individual’s belief in their capacity to execute specific behaviours to achieve desired outcomes, which can significantly influence their perception of the behaviour.³⁶ Therefore, these two concepts are mapped to the Motivation dimension in this study. Opportunity encompasses all factors lying outside the individual that facilitate or prompt the behaviour. Prior research has shown that social support and physician support serve as environmental conditions that enhance patient engagement in healthcare.^{37 38} Accordingly, social support and ophthalmologist facilitation of patient involvement are incorporated into the Opportunity dimension in this study.

For the effective management of DR, context-specific data are needed to understand how patients participate in treatment decision-making. This protocol provides novel insights by focusing on the Chinese population, where shared decision-making practices are still developing and specific barriers remain under-explored.³⁹ It emphasises a systematic understanding of the behavioural determinants using the COM-B theoretical framework. Therefore, the proposed study aims to investigate the current status of involvement in treatment decision-making among Chinese patients with DR, identify the influencing factors and explore their experiences related to the decision-making process.

Aims and objectives

The primary aim of this study is to investigate the current status of decision-making roles among Chinese patients with DR and to evaluate the relationships among personal characteristics, COM-B factors and these decision-making roles. Furthermore, the study aims to explore patients’ subjective experiences to provide an in-depth explanation of the quantitative results. The study will evaluate the following specific objectives:

1. To assess the distribution of decision-making roles among Chinese patients with DR.

2. To analyse the relationship between patients’ personal characteristics (eg, age, gender, disease severity) and their decision-making roles.

3. To identify the association between COM-B dimensions (eg, social support, ophthalmologist facilitation, self-efficacy) and the decision-making roles adopted by patients.

4. To explore patients’ subjective experiences and attitudes towards the decision-making process to explain the quantitative findings.

Methods and analysis

We will conduct a mixed-methods study in China. This research will employ an explanatory sequential design consisting of two consecutive phases. An initial quantitative phase will be followed by a qualitative phase, which aims to explain, contextualise and elaborate on the quantitative results.⁴⁰

The quantitative research will be a cross-sectional descriptive design using self-reported questionnaires to measure the actual types of patient involvement in treatment decision-making, sociodemographics, clinical characteristics, health literacy, need for involvement in decision-making, decision-making self-efficacy, social support and ophthalmologist facilitation of patient involvement. Qualitative research will be conducted through semistructured interviews with patients with DR to explore their perceptions and experiences of involvement in treatment decision-making. Finally, both quantitative and qualitative data will be integrated using a joint display technique to enhance our understanding of the treatment decision-making process among patients with DR.⁴¹

The Good Reporting of a Mixed Methods Study checklist will be used to guide the reporting of this study.⁴² The COM-B model will serve as the overarching theoretical framework to guide both data collection and analysis. Specifically, it will inform the development of the quantitative survey items and the qualitative interview guide. Furthermore, the COM-B constructs will provide the primary structure for coding and thematically analysing the qualitative data, ensuring a systematic exploration of the determinants of involvement in treatment decision-making. As outlined in the Introduction, the study variables have been theoretically mapped to the COM-B model domains (figure 1).

Figure 1. Theoretical framework of factors influencing involvement in treatment decision-making among patients with DR based on the COM-B model. This figure illustrates the hypothesised relationships between the three domains of the COM-B model, including Capability, Opportunity and Motivation, and their influence on the behaviour of patient involvement in treatment decision-making. Capability is operationalised as health literacy. Motivation includes decision-making self-efficacy and the need for involvement. Opportunity encompasses social support and ophthalmologist facilitation. The model also accounts for the influence of personal characteristics on behaviour, which ultimately manifests as actual decision-making roles. COM-B, Capability, Opportunity, Motivation-Behaviour; DR, diabetic retinopathy.

Quantitative study

Phase one is a cross-sectional questionnaire survey addressing the first, second and third specific objectives. It aims to assess the distribution of decision-making roles, analyse their relationships with personal characteristics and identify the associations with COM-B dimensions among patients with DR.

Setting

Participants will be recruited using a convenience sampling approach from the ophthalmology wards and outpatient clinics of Shanghai General Hospital. As a National Clinical Research Center for Eye Diseases, the hospital serves a vast patient population drawn primarily from Shanghai and the neighbouring provinces of Jiangsu, Zhejiang and Anhui, which collectively form the Yangtze River Delta region. This geographical diversity ensures that the sample will likely represent the target population in Eastern China. To minimise selection bias and potential coercion, recruitment will be facilitated by researchers (LX and HZ) who are not directly involved in the patients’ clinical care. Specifically, potential participants will first be identified by the researchers through a preliminary screening of clinical registration lists and medical records based on the inclusion and exclusion criteria. Information sheets regarding this study will be distributed in the waiting areas to ensure maximum reach. Following this initial identification, the researchers will approach eligible patients to provide detailed information and invite participation. Participants will be offered the option to complete the questionnaires via face-to-face interviews to accommodate those with visual impairment, or through self-administration under researcher supervision.

Sample size

The study included 16 sociodemographic and clinical characteristics and 12 dimensions corresponding to six scales, resulting in 28 variables. According to Kendall’s logistic regression sample estimation method, the sample size should be 10–15 times the number of independent variables.⁴³ Based on this calculation, a raw sample size of 280–420 cases is required. In consideration of a 10% attrition rate due to lost follow-up or invalid questionnaires, we determined to recruit approximately 350 participants to ensure an adequate sample size for robust statistical analysis.

Inclusion and exclusion criteria

Eligible patients for this survey must be adults aged 18 years or above, diagnosed with DR in stages III–VI and willing to volunteer to participate and provide informed consent. Patients who have a past or present history of mental illness, intellectual disabilities or impaired verbal communication will be excluded to ensure the validity of data and informed consent. Additionally, those diagnosed with severe systemic conditions, such as severe cardiac, hepatic or renal dysfunction, respiratory failure or critical illness, will also be excluded to avoid imposing undue physical or psychological burden on their health.

Measures

Data collection will be performed by using the following measures:

1. General information questionnaire: Based on the literature review, the researchers designed their questionnaire that included information about the participants’ gender, age, ethnicity, marital status, education level, occupation, per capita monthly household income, healthcare payment method, residency, disease duration, stage and comorbidities.

2. Control Preference Scale: This scale was initially developed by Degner and subsequently adapted and revised by Nolan et al.⁴⁴ It is a unidimensional scale with five options to describe patients’ decision-making involvement in clinical treatment decisions. Options 1 and 2 represent the active type, option 3 represents the shared decision type and options 4 and 5 represent the passive type. The Cronbach’s α coefficient for the Chinese version of the scale was 0.899.

3. All Aspects of Health Literacy Scale (AAHLS): The scale was developed by Deborah et al⁴⁵ and translated into Chinese by Wu Qing.³⁸ We will use AAHLS to assess participants’ health literacy. The scale consists of three dimensions: reading skills and understanding health information, communication with health professionals and evaluation and application of health information. There are 11 items, and a Likert 3-point scale is adopted. Higher scores indicate better health literacy levels. The Cronbach’s α coefficient for the Chinese version of the scale was 0.811.

4. Patient Expectation for Participation in Medical Decision‐making Scale: The scale was developed by Xiaolin Xu et al to assess patients’ need for involvement in treatment decision-making.⁴⁶ The scale includes three dimensions: information needs (including three items), communication needs (including six items) and decision-making needs (including three items). Each item is scored on a 5-point Likert scale, with higher scores indicating a higher need for patient involvement in treatment decision-making. The overall Cronbach’s α coefficient of the scale is 0.961.

5. Decision Self Efficacy Scale: The scale was developed by O’Conner et al to assess patients’ self-confidence in their ability to make treatment decisions.⁴⁷ The scale is a unidimensional scale comprising 11 items, each rated on a Likert scale from 0 (not at all confident) to 4 (very confident). Higher scores represent higher decision-making self-efficacy in patients. The Cronbach’s α coefficient for this scale was 0.92.

6. Social Support Rating Scale: The scale was developed by Chinese scholar Xiao Shuiyuan to assess social support.⁴⁸ It consists of three dimensions: objective support, subjective support and the utilisation of support, comprising 10 items. A higher score indicates a greater level of social support for the patient. The Cronbach’s α coefficient for this scale was 0.92.

7. Facilitation of Patient Involvement Scale: The scale was developed by Martin et al to assess patients’ perception of health professionals’ facilitation of patient engagement.⁴⁹ It is a unidimensional scale with nine items. Each item is scored on a 6-point Likert scale ranging from 1 (never) to 6 (always). The Cronbach’s α coefficient for this scale was 0.885.

Data collection

Before administering the questionnaire survey, all research team members will receive standardised training to ensure consistent terminology is used throughout the survey. When administering the questionnaire, the researchers will explain the study’s purpose, content and procedures to the participants who meet the inclusion criteria and obtain their informed consent. Following the consent process, the researchers (LX and HZ) who are independent of the patient’s direct clinical care team will access the hospital’s electronic medical record system to minimise potential coercion. They will extract specific clinical data relevant to the study objectives, such as disease duration, stage and comorbidities.

Questionnaires will be distributed and collected on-site. For participants with adequate visual acuity to read the text independently, self-administration will be encouraged with a researcher nearby to answer queries. For participants unable to complete the questionnaire independently due to visual impairments, a researcher-assisted interview method will be used with strict safeguards to ensure accuracy. Specifically, the researcher will read each question and option neutrally to avoid any intonation that might influence the choice. Furthermore, a read-back verification technique will be implemented where the researcher repeats the participant’s verbal response for confirmation before recording it. Prompt checks will be conducted immediately after collection to ensure completeness. The questionnaire’s average completion time is approximately 15 min. Data collection will take place from September 2024 to September 2026.

Data analysis

Quantitative data will be statistically analysed using SPSS V.26.0. Descriptive analyses will be performed for all included variables to summarise the characteristics of the participants. The χ² test will be used to examine differences in actual decision-making roles among categorical variables, while continuous variables will be analysed using one-way analysis of variance or the non-parametric Kruskal-Wallis H test. Before performing the regression, multicollinearity among independent variables will be assessed using the variance inflation factor (VIF), with a VIF of less than 10 indicating the absence of significant multicollinearity. Multinomial logistic regression analyses will then be conducted to identify predictors of actual decision-making roles, incorporating significant factors from the univariate analyses as independent variables. Sensitivity analyses will be performed to evaluate the robustness of the findings by assessing the impact of missing data on the primary outcomes. Two-tailed p values will be adopted, with p<0.05 considered statistically significant.

Qualitative study

A descriptive qualitative approach using semistructured personal interviews will be used to explore the perceptions and experiences of individuals with DR regarding their involvement in treatment decision-making. This study follows an explanatory sequential mixed-methods design, where the qualitative phase is conducted after the quantitative phase to provide an in-depth explanation of the survey results.

Sampling

Purposive sampling will be used to select participants from those who have completed the quantitative phase and expressed willingness to participate in semi-structured interviews. The principle of maximum variation will guide the selection process, considering factors such as age, education and disease stage, as well as the diverse decision-making roles identified in the quantitative phase to ensure a comprehensive explanation of the survey findings.⁵⁰ The final sample size will be determined by the criterion of data saturation, meaning that recruitment will continue until no new themes emerge.⁵¹ It is expected that 10–20 respondents will need to be recruited for this phase.

Data collection

Data will be collected using a face-to-face semistructured interview guide. To ensure consistency, the first author (LX) will conduct all interviews in person at the hospital. The first author is a master’s student with systematic training in qualitative research. We have developed an interview guide based on the study aims, relevant literature and the COM-B model. We also conducted pilot interviews with two patients to refine the formal interview outline based on their feedback (table 1). Before each formal interview, the interviewer will explain to the interviewee the purpose of the interview, its content, the reason for recording and the principle of data confidentiality. The interviewee will then be asked to sign an informed consent form indicating their understanding and agreement. Each interview will be conducted in a private and quiet setting at a time and place convenient for the interviewee and is expected to last approximately 30–45 min. All interviews will be audio-recorded with the participants’ permission and transcribed verbatim for analysis. The researcher will maintain a reflective diary immediately after each interview to capture contextual insights and personal reflections.

Table 1. Semistructured interview guide.

Attributes	Topics
Involvement in decision-making experiences in treatment	Can you recall when the last treatment decision was made during your treatment? Do you remember the process of making that decision? Please describe it in detail.
Influences on involvement in treatment decision-making	Capability	What details do you recall from your ophthalmologist’s presentation about your condition and treatment? Were there any parts of the information that you found difficult to understand? Did you ask the ophthalmologist for further clarification? What factors do you think contributed to your difficulty in understanding the information? Besides the information given to you by your ophthalmologist, have you received information about your disease or treatment from other sources? How did you get this information? What information sources were most helpful in your treatment decisions? Did you compare all the information you received? What factors influenced your final decision?
	Opportunity	Did you think your family, friends or others should be involved in treatment decision-making? Why? How did you think their involvement in treatment decision-making affected your final decision? Do you feel that your ophthalmologist has provided you with adequate support in making treatment decisions? Do you feel that your ophthalmologist influenced your involvement in treatment decision-making?
	Motivation	How would you like to be involved in decision-making during treatment? Why? What do you think are the benefits and drawbacks of involvement in decision-making during treatment? What are the implications? How confident were you in your involvement in decision-making? What would affect confidence in involvement in treatment decision-making? Was there ever a discrepancy between your thoughts and the ophthalmologist’s recommendations when making a treatment decision? How was this resolved? Do you think healthcare professionals can help you with involvement in treatment decision-making?
	Other factors	What other factors do you think caused/hindered your involvement in treatment decision-making? Can you be more specific? Do you have any ideas/suggestions for patient involvement in treatment decision-making?

Trustworthiness

Qualitative rigour will be established via the criteria of credibility, dependability, confirmability and transferability.⁵² Credibility will be maintained by returning verbatim transcripts to participants for verification shortly after each session and inviting selected informants to evaluate preliminary thematic findings, ensuring their lived experiences with DR and treatment decision-making are accurately reflected. Dependability and confirmability will be ensured through a transparent audit trail and regular peer debriefing within the research team to reach a consensus on themes. Transferability will be established through comprehensive accounts of the research methodology, participant demographics, data processing and the inclusion of representative participant quotes.

Data analysis

The recorded interviews will be transcribed verbatim within 24 hours of their completion. The transcribed text will then be imported into MAXQDA V.24.0 software for management, coding and analysis. We will use a combined inductive and deductive qualitative analysis approach to ensure methodological integrity. First, two independent researchers will conduct an inductive analysis of the interview data using Braun and Clarke’s thematic analysis to generate main themes and subthemes.⁵³ Following this, they will apply deductive thematic analysis guided by the COM-B model to organise the extracted subthemes. Both researchers (LX and HZ) will transcribe and analyse the data simultaneously and independently. Throughout the analysis process, all identified themes and subthemes will be reviewed, revised and discussed within the research team to ensure consistency in the final results.

Integration study

The Pillar Integration Process will be used to integrate qualitative and quantitative data and present them through the joint display.⁵⁴ This process consists of four stages (listing, matching, checking and pillar building): (1) listing of the most relevant qualitative and quantitative data in the form of tables/charts/matrices; (2) matching of the raw data and categories derived from the quantitative analyses with those derived from the qualitative analyses; (3) cross-checking of the completeness of all the data to ensure appropriate matching; and (4) comparing and contrasting the results derived from the listing, matching and checking stages to define integrated themes. This analysis will help us better identify the factors that influence the involvement in treatment decision-making processes of patients with DR. We will use the COM-B model to guide the presentation and conceptualisation of the data.

Patient and public involvement

While patients were not involved in the initial design, several measures have been implemented to ensure their active participation. Two patients with DR were invited to provide feedback and suggestions on the recruitment materials and interview questions to ensure they are culturally sensitive and understandable. Furthermore, after the preliminary analysis of the quantitative data, a group of patients will be invited to discuss which findings they find most significant, which will help refine the focus of the qualitative phase. Finally, a plain-language summary of the findings will be shared with the participants to ensure the results are accessible to the target population.

Discussion

To our knowledge, this will be the first mixed-methods study in mainland China to investigate both the current status and experiences of patients with DR regarding their involvement in treatment decision-making. Unlike previous studies on patient decision-making preferences, this study aims to evaluate the actual decision-making roles and associated psychosocial factors among patients with DR within the Chinese clinical context. By using an explanatory sequential mixed-methods design, the protocol will integrate quantitative findings with qualitative insights to delve into the lived experiences of patients with DR throughout the decision-making process. This integration is expected to provide a comprehensive and holistic perspective on the mechanisms underlying patient involvement, offering insights that quantitative surveys alone cannot capture.

The application of the COM-B model as a theoretical framework will enable a systematic exploration of the barriers underlying the involvement of patients with DR. As DR progresses, declining visual function is expected to affect the decision-making capability of patients with DR, which in turn will influence their motivation to engage in complex treatment choices. Moreover, with advancing disease stages, patients with DR may rely more heavily on physician-led decision-making, yet the specific influence of ophthalmologist facilitation remains largely underexplored. By identifying whether barriers originate from capability, opportunity or motivation, this research will provide evidence-based insights for ophthalmologists to tailor communication strategies to the unique needs of individual patients with DR.

The findings are expected to have a direct impact on clinical practice by informing the development of patient-centred resources and professional training programmes. If potential barriers such as physician communication styles or patients’ treatment beliefs can be better understood, clinicians can move towards a more effective shared decision-making process. This has the potential to improve clinician–patient concordance and patient satisfaction within the ophthalmology department.

Certain limitations of this study should be recognised. The recruitment of patients with DR will be conducted at a single tertiary hospital in Shanghai, which may restrict the generalisability of the findings to patients with DR in primary care or rural settings. Additionally, the cross-sectional nature of the data collection will preclude the exploration of longitudinal changes in the decision-making roles of patients with DR as the disease progresses. Regarding the potential for social desirability bias in self-reported data, the research team will ensure strict anonymity and confidentiality for all patients with DR to promote the accuracy and honesty of the responses.

Several potential directions for future research can be proposed based on the findings of this study. First, subsequent research could replicate these results in more diverse clinical settings, such as primary care facilities or rural hospitals, to further validate the psychosocial barriers identified in this protocol. Second, as treatment decision-making is a process involving multiple stakeholders, future studies may be expanded to include the perspectives of ophthalmologists, nurses and family members to provide a more comprehensive and multifaceted understanding of shared decision-making. Finally, based on the identified barriers within the COM-B framework, future work will aim to develop and evaluate tailored supportive tools, such as patient-centred educational resources and decision aids, designed to enhance the active involvement of patients with DR.

Ethics and dissemination

Ethical considerations

The study protocol has been reviewed and approved by the Human Research Ethics Committee of the Shanghai General Hospital, Shanghai, China (number: 2024–098). Written informed consent will be obtained from all participants before the survey. Participants will have the option to consent to participate only in the questionnaire or in both the questionnaire and a semistructured interview. If participants agree to participate in the interview, they will be informed that what they say in the interview will be quoted verbatim in the report of the findings. All participants will be informed of their right to withdraw from this study at any time without affecting their healthcare services. All data will be entered and stored anonymously. Computer-stored data will be kept under password protection, and access to the data will be restricted to research team members.

Dissemination

The findings of this study will be disseminated through publication in academic conferences and peer-reviewed journals. Additionally, a report summarising the findings, written in everyday language, will be provided to participants to ensure that the results are accessible and meaningful to the patient community.