, Introduction
When chronic kidney disease progresses to end-stage renal disease, renal replacement therapies such as hemodialysis (HD) or peritoneal dialysis (PD) become necessary. Both modalities aim to manage kidney failure but differ significantly in patient experience, clinical setting, and lifestyle implications. PD offers greater flexibility, as it can be performed at home, allowing patients more control over their treatment schedules and daily activities [1]. In contrast, HD is typically carried out in a clinical setting, such as dialysis centers, which requires multiple weekly visits [2].
Given the frequent healthcare contact and structured nature of HD treatment, patient satisfaction becomes a critical factor influencing treatment adherence and overall quality of life [3,4]. The clinical environment, interpersonal relationships with healthcare staff, availability of resources, and the physical and emotional demands of HD treatment all play pivotal roles in shaping the patient’s experience [5]. Assessing satisfaction among HD patients is vital to understanding their perspectives, addressing concerns, and ultimately improving the delivery of dialysis services.
Although several studies have explored patient satisfaction in various healthcare settings, there is growing recognition that HD patients have unique needs and face distinct challenges [6]. HD patients often experience significantly reduced quality of life due to the time-consuming nature of treatment, physical strain associated with repeated vascular access, and the emotional toll of dependence on life-sustaining therapy [7,8]. This rigid clinic schedule significantly limits patients’ autonomy and flexibility compared to PD patients, who can manage their treatments at home [9]. Additionally, fatigue, dietary restrictions, and limitations in social activities are more pronounced in HD patients, further exacerbating their challenges [10]. These specific stressors make HD patients particularly vulnerable to treatment dissatisfaction, which can affect adherence to therapy and health outcomes [11].
Thus, there is a pressing need to develop a satisfaction scale tailored to the HD population due to their distinct physical, emotional, and logistical challenges. A previous standard satisfaction scale designed for general patient populations or patients undergoing dialysis may not fully capture the unique aspects of HD care that contribute to satisfaction or dissatisfaction. Existing HD patient satisfaction scales, primarily developed in the United States and Europe, have predominantly focused on healthcare provider relationships, dialysis procedures, communication, and environmental factors. However, these instruments have limited applicability across different cultural contexts due to varying healthcare systems, accessibility, costs, and quality of life priorities between countries [12–14]. Additionally, the medical environment in Korea differs from that in the United States and Europe, where the previous HD scales were developed [15–17]. Korea has good healthcare accessibility, and the medical system primarily focuses on primary clinics, fostering high-quality medical care through competition. Patients also have the advantage of freely choosing their hospitals. Given these unique characteristics, developing satisfaction scales specifically tailored to Korea’s healthcare context is essential. Developing a specific instrument allows the identification of areas needing improvement in the HD care setting and ensures that the patient’s voice is adequately represented. Such a scale can offer valuable insights that drive improvements in clinical practices, enhance patient-provider communication, and ultimately lead to better patient-centered care. This manuscript outlines the development and validation of a satisfaction scale specifically designed to address key dimensions of the experience of HD patients in Korea. Although numerous patient satisfaction measurement scales exist, this tool was developed to ensure cultural relevance, reliability, and sensitivity to the unique challenges faced by Korean HD patients.
Methods
A formative research approach was employed to guide the development of the Hemodialysis Management Satisfaction Scale (HMSS). The formative research approach was chosen because it serves as a critical foundation for both the initial design of behavioral interventions and their evaluation and refinement during implementation. This approach allows for process evaluation, providing insights that can enhance the delivery of interventions or guide necessary adjustments throughout the development phase. Formative research approaches utilize both qualitative and quantitative mixed methods to triangulate findings across different participant types and data sources [18,19]. In this study, in-depth interviews were conducted to explore culturally specific dimensions of satisfaction that are inherently Korean, while a subsequent survey was used to quantitatively validate the identified constructs [20]. This combination ensured a thorough understanding of the factors influencing the satisfaction of HD patients and enhanced the robustness of the newly developed scale. A flow chart of our study is shown in Fig. 1.
In-depth interviews
A total of five in-depth interview sessions were conducted with 12 patients from Seoul National University Bundang Hospital, Severance Hospital, Hallym University Dongtan Sacred Heart Hospital, and Nuri Internal Medicine, a local HD center, between April 28 and June 21, 2023.
In the first phase of the study, these in-depth interviews were designed to capture culturally relevant aspects of patient satisfaction and uncover dimensions unique to the HD experience in Korea. The interview questionnaire consisted of open-ended questions that addressed a range of health and lifestyle factors and patients’ perspectives on various facets of their treatment. Specific areas explored included perceptions of the benefits and barriers associated with HD, dilemmas faced during the treatment process, and satisfaction with communication with attending doctors and medical staff. The insights gained from these interviews were instrumental in identifying key dimensions of patient satisfaction, which were subsequently incorporated into the development of the scale.
Questionnaire development and Delphi validation
Following the interviews, a draft questionnaire was developed based on insights from the literature review and in-depth interviews. This questionnaire incorporated both validated items from existing scales and new items that captured dimensions identified as culturally specific during the qualitative phase. To refine and validate the culturally specific items derived from the qualitative phase, we applied the Delphi technique, a structured method used to achieve expert consensus through iterative feedback [21]. Eight experts—including three nephrologists, two hemodialysis nurses, and three health communication specialists—participated in two rounds of rating. Items were retained if they achieved a mean agreement score of ≥4.0 on a 5-point scale and at least 80% consensus. Feedback from the first round informed minor revisions, which were then confirmed in the second round. Through this iterative process, consensus was reached on 10 new questionnaire items specific to the Korean cultural context.
The survey aimed to evaluate the reliability and validity of the newly developed scale, including variables such as health efficacy and patient-provider interaction, assessing its psychometric properties. All items were rated on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). A total of 121 participants were recruited from two university hospitals (n = 61) and two local HD centers (n = 60). This sample size was guided by established psychometric research recommendations. A sample size of 100 is considered ‘fair’ for factor analysis, and a participant-to-item ratio of at least 3:1 is often deemed acceptable for both exploratory and confirmatory factor analysis (CFA) [22]. In our case, the ratio exceeded 4:1. Moreover, given the scale’s relatively simple factor structure, high item communalities, and strong factor loadings, the sample size was sufficient to conduct exploratory factor analysis (EFA), CFA, and structural equation modeling (SEM) [23]. Eligibility criteria included patients aged 19 years or older undergoing HD for at least 3 months. Patients with cognitive impairments or psychological conditions that prevented their participation were excluded from the study.
Ethical consideration
The study adhered to the principles of the Declaration of Helsinki. The Institutional Review Board (IRB) and Ethics Committee of the Seoul National University Bundang Hospital (SNUBH) approved the study protocol. We provided informed consent forms to the patients and obtained their consent prior to conducting the research (No. B-2303-816-301). The IRBs for the two local HD clinics was included in the SNUBH IRB review process and received approval. Consent was obtained using consent forms bearing the corresponding IRB number. To ensure confidentiality and anonymity, participants were informed that all data would be anonymized, and no identifying information would be included in transcripts. Pseudonyms were used during transcription and analysis.
Statistical analysis
Data analysis was conducted in three stages. In the first stage, descriptive statistics were used to analyze the demographic characteristics of the participants, providing an overview of the sample.
In the second stage, the psychometric properties of the newly developed scale were assessed through reliability and validity testing. EFA was performed using IBM SPSS version 26.0 (IBM Corp.) to identify the underlying factor structure. Items with low factor loadings (<0.40) or cross-loadings were removed to improve construct clarity. CFA was then conducted using AMOS version 21.0 (IBM Corp.) to validate the factor structure identified in the EFA.
To evaluate internal consistency, Cronbach’s alpha was calculated, with values ≥0.70 considered acceptable, and values ≥0.80 regarded as good [24]. Test-retest reliability was assessed using Pearson correlation coefficients, evaluating the temporal stability of the scale.
In the third stage, nomological validity was tested using SEM. The hypothesized relationships between patient satisfaction, health self-efficacy, and perceived health and life satisfaction were examined. The goodness-of-fit of the model was assessed using multiple indices: the chi-square test, comparative fit index (CFI), incremental fit index (IFI), root mean square error of approximation (RMSEA), and root mean square residual (RMR). A CFI or IFI ≥0.85 and RMR <0.05 were considered indicative of acceptable model fit, consistent with established recommendations [23,25].
Results
A total of 121 patients completed the survey questionnaires. The mean age of participants was 58.3 ± 13.8 years. The demographics of the study participants are summarized in Table 1.
Identification of hemodialysis patient satisfaction attributes extracted from the in-depth interviews
This study extracted 10 items representing patients’ perceptions of doctors, medical staff, and overall dialysis experience from the in-depth interviews. These items are as follows: (1) My attending physician always respects me; (2) My attending physician provides clear explanations about HD; (3) My attending physician provides clear explanations about dietary management for weight control; (4) My attending physician does their best to respond to questions from me and my caregiver; (5) I am satisfied with the way my attending physician explains things, including the use of visual aids; (6) The HD nurses are skilled; (7) The HD nurses do not discuss patients’ private matters among themselves; (8) The number of patients assigned to each HD nurse is appropriate; (9) The dialysis center where I receive treatment uses the latest and best equipment; (10) The dialysis machines at my treatment center are efficient.
These items reflect key aspects of the patient experience, encompassing doctor-patient interactions, communication, nursing care, and perceptions of facility quality, highlighting culturally specific factors relevant to the satisfaction of HD patients in Korea.
Reliability
The reliability of the newly developed items derived from the in-depth interviews was assessed using Cronbach’s alpha, which yielded a value of 0.844, indicating good internal consistency. Cronbach’s alpha for the overall satisfaction scale for HD patients was 0.83, with the coefficients for the subdomains ranging from 0.80 to 0.93, demonstrating acceptable to excellent reliability [24,26]. The Spearman-Brown coefficient ranged from 0.824 to 0.829, and the Guttman split-half coefficient was 0.795, both meeting the criteria for split-half reliability and further supporting the consistency of the scale [27].
Validity
Construct validity
Construct validity of the newly developed scale was assessed using both EFA and CFA. The Kaiser-Meyer-Olkin measure of sampling adequacy was 0.881, and Bartlett’s test of sphericity was significant (p < 0.001), indicating that the data were appropriate for factor analysis. Initially, six principal components were extracted through principal component analysis, accounting for a cumulative variance of 71.34%. However, the final component, which contained only one item, was removed following established guidelines [28], resulting in five principal components with a refined cumulative variance of 68.20%. The five principal components were identified as nephrologist (11 indicators), medical staff (7 indicators), facility (4 indicators), environment (4 indicators), and proficiency (4 indicators) factors. The results of this analysis are summarized in Table 2, including factor labels, component items, factor loadings, and Cronbach’s alpha values for each principal component. Internal reliability, assessed using Cronbach’s alpha, was sufficiently high for all factors, with values exceeding the commonly accepted threshold of 0.70 [29]. The results of the EFA, including the identified factor structures, are illustrated in Fig. 2.
Construct validity of the factor structure was supported by CFA, which demonstrated overall acceptable model fit: chi-square (395), 703.121; p < 0.001; CFI, 0.881; RMR, 0.061. While most indices, including RMSEA and RMR, met acceptable thresholds, the CFI and IFI values were slightly below the conventional cutoff of 0.90. Convergent and discriminant validity were also assessed and provided additional support for the proposed factor structure.
Convergent validity, a subcategory of construct validity, refers to the degree to which two measures of theoretically related constructs are, in fact, related. In contrast, discriminant validity tests whether concepts or measurements that are theoretically unrelated are indeed unrelated [30]. The Fornell-Larcker criterion assessed convergent and discriminant validity by examining the average variance extracted (AVE) and composite reliability (CR) values.
The AVE measures the level of variance captured by a construct relative to the variance due to measurement error, with values above 0.70 considered very good and values of 0.50 considered acceptable [30]. The CR, a less biased reliability estimate compared to Cronbach’s alpha, is expected to have values of 0.70 or higher for acceptable reliability [30]. To conduct a more rigorous analysis of convergent and discriminant validity, the AVE and CR of each latent variable were calculated using the formula provided by Fornell and Larcker [31]. The AVE values for the constructs of nephrologist, medical staff, facility, environment, and proficiency factors were all greater than 0.50, indicating adequate convergence. Additionally, the CR values for each latent variable exceeded 0.80, demonstrating both convergent validity and internal consistency.
Discriminant validity was assessed by comparing the AVE with the squared correlation coefficients between the factors. According to Fornell and Larcker [31], discriminant validity is confirmed when the AVE of a construct is greater than the square of the correlation coefficient between that construct and other constructs. In this study, the AVE for each factor exceeded the squared correlation coefficients between each factor, indicating that discriminant validity was adequately established (Table 3).
Nomological validity
This study assessed nomological validity using SEM. Nomological validity refers to the degree to which predictions from a theoretical network (nomological network) are confirmed [29,30]. In simpler terms, it assesses how well a scale fits within a broader theoretical framework. Previous studies have supported the relationship between the satisfaction of HD patients and perceived well-being or health status, reinforcing the argument for the nomological validity of our model [32,33].
Accordingly, we hypothesized that patient satisfaction would positively influence the health self-efficacy of HD patients, which would enhance perceived health status and daily life satisfaction. The SEM results provide evidence supporting our hypothesized model. The model fit indices indicated an acceptable fit: chi-square, 215.389; degrees of freedom, 64; p < 0.001; IFI, 0.838; CFI, 0.836; and RMR, 0.053. Patient satisfaction was found to positively influence health self-efficacy (β = 0.55, p < 0.001), which subsequently improved patients’ perceived health status (β = 0.34, p < 0.001) and their satisfaction with daily life (β = 0.38, p < 0.001). These results demonstrate that the nomological validity of the newly developed satisfaction scale for HD patients was adequately established, as the theoretically derived relationships were confirmed empirically (Fig. 3). The final validated items are shown in Table 4.
Discussion
According to the prior research, the components of the satisfaction of HD patients are multifaceted. Provider-related determinants of patient satisfaction in HD centers include doctors, medical staff, facilities, services, and treatment, while patient-related characteristics associated with satisfaction include demographics and health status history [34]. The present study successfully developed and validated a Korean satisfaction scale for HD patients, identifying five distinct dimensions: nephrologists’ care, medical staff, facility, proficiency, and the overall dialysis environment. These factors represent the key domains of the experiences of HD patients and the specific aspects that significantly impact their satisfaction. The findings align with existing literature, indicating that higher satisfaction with medical care is positively correlated with better physical and mental health among patients undergoing dialysis. The demonstrated association between patient satisfaction and perceived well-being underscores the crucial role of patient-centered care in promoting positive health outcomes for HD patients.
Importantly, this study found that higher patient satisfaction was associated with enhanced health self-efficacy, which, in turn, led to improvements in patients’ perceived health status and their satisfaction with daily life. Health self-efficacy is a critical factor in managing chronic conditions, as it represents a patient’s belief in their ability to manage their own health and make informed health-related decisions. Previous studies have shown that higher health self-efficacy can lead to better adherence to treatment regimens, healthier lifestyle behaviors, and improved clinical outcomes among HD patients [35,36]. Thus, enhancing patient satisfaction contributes to their emotional well-being and positively impacts their confidence in managing their health, ultimately leading to better health outcomes.
The importance of patient-centered care cannot be overemphasized, particularly for vulnerable populations such as individuals undergoing dialysis. The data highlights that, when nephrologists and medical staff provide clear, empathetic communication, and when facilities and processes are perceived as efficient and proficient, patients tend to report better well-being. This improves their physical and mental health outcomes and enhances treatment adherence, contributing to more effective disease management and overall quality of life.
From an academic perspective, this study contributes to the growing body of knowledge surrounding patient satisfaction measurement tools, particularly in the context of culturally specific instruments tailored to the Korean healthcare setting. Although several patient satisfaction scales exist, the necessity of a tool specific to Korean HD patients is evident, given cultural nuances that influence healthcare perceptions. This scale provides a more nuanced understanding of how various elements of care impact the experiences of HD patients, thereby addressing an essential gap in healthcare evaluation research.
From a practical standpoint, this newly developed satisfaction scale provides healthcare practitioners and policymakers with a reliable instrument for assessing and improving the quality of care in dialysis settings. By utilizing this scale, dialysis centers can identify specific areas needing improvement, such as staff training, facility management, or enhanced communication between medical providers and patients. Additionally, healthcare providers can use insights from the HMSS to develop training programs focused on culturally sensitive care practices, directly contributing to improved patient satisfaction and adherence. Consequently, this instrument facilitates targeted quality improvement initiatives, ultimately enhancing patient outcomes.
One notable limitation of the present study is the modest survey sample size. Although the in-depth interview phase included a sufficient number of participants, only 121 HD patients completed the survey. While this sample size provided an adequate basis for preliminary psychometric validation, it may still be limited for ensuring robust statistical power and generalizability. This constraint could explain the marginal model fit indices observed—such as the CFI (0.881) and IFI (0.883)—which fell slightly below conventional thresholds. A larger sample would likely enhance the stability and fit of the measurement model, especially for assessing the predictive validity of the HMSS.
In addition, the cohort is somewhat limited in representativeness, as participants were recruited from four institutions: two university-affiliated hospitals and two local dialysis centers, all located in metropolitan areas. Although this provides some institutional diversity, community-based and rural perspectives remain underrepresented. Future studies should aim to include more geographically and demographically diverse samples to strengthen the cultural relevance, generalizability, and applicability of the scale across varied care environments.
In conclusion, this study highlights the significance of patient-centered care in improving the physical and mental health outcomes of HD patients and the importance of fostering health self-efficacy as an intermediary step toward better patient satisfaction and perceived well-being. The development and validation of this culturally tailored satisfaction scale are of great academic and practical value, providing a specific and reliable tool for evaluating the quality of HD care in the Korean context. The HMSS may also be applicable in other Asian countries with similar cultural and healthcare dynamics, further validating its utility. Future research should focus on increasing sample size and ensuring repeated studies to enhance predictive validity and achieve a more comprehensive understanding of the factors that affect the satisfaction of HD patients.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
