Development and validation of healthy dietary decision-making competence scale for inflammatory bowel disease: an instrument design study Authors: Yiting Li, Yongqi Liang, Murong Wang, Tingting Yin, Tianxu Chen, Ling Ji, Ping Zhang, Qiong Li, Wenjing Tu, Guihua Xu Published: 2025 ABSTRACT AbstractBackgroundDietary decision-making plays a crucial role in the management of patients with inflammatory bowel disease (IBD). However, it is complex and involves multiple challenges. Decision-making competence may assist healthcare professionals (HCPs) in evaluating patients’ competence to make healthy dietary decisions and to design targeted dietary support programs. Assessment of healthy dietary decision-making competence is essential for guiding IBD patients in daily dietary management. Currently, no valid instrument is available to measure this construct. This study aimed to develop an instrument to evaluate healthy dietary decision-making competence in patients with IBD.MethodsThe development and test of the healthy dietary decision-making competence scale (HDMCS) proceeded according to two phases: (i) instrument development: identifying preliminary items through literature reviews, interviews with 24 patients, an expert panel with 24 experts and cognitive interviews with 8 patients. (ii) scale verification: surveying 354 patients with IBD from five tertiary general hospitals in Nanjing and Shenzhen city, China. Item analysis, exploratory factor analysis, confirmatory factor analysis, and reliability rest were performed to evaluate the reliability and validity of this scale.ResultsThe exploratory factor analysis and confirmatory factor analysis yield a four-factor scale (χ2/df = 1.972, RMSEA = 0.068, GFI = 0.848, IFI = 0.929 and CFI = 0.928). The four factors were ‘dietary management perception’, ‘resource utilization competence’, ‘decisional balance competence’ and ‘post-decision reflection competence’. The reliability was demonstrated to be acceptable, with the Cronbach’s alpha coefficient of 0.941, and the 4 factors were from 0.818 to 0.912. Test–retest reliability (by ICC) of the total scale was 0.874.ConclusionsThe HDMCS demonstrated satisfactory reliability and validity to warrant further development as a measure of dietary decision-competence of patients with IBD. By evaluating dietary decision-making competence across multiple domains, HCPs could gain a comprehensive understanding of patients’ beliefs, preferences, and values related to dietary decisions, thereby tailoring interventions accordingly.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12876-025-04368-y.Keywords: Patient competence, Decision-making, Instrument development, Inflammatory bowel disease, Psychometric testing Background Inflammatory bowel disease (IBD) is an abnormal immune-mediated chronic gastrointestinal inflammation with periods of relapsing inflammation, leading to increased symptoms and a decreased quality of life. It mainly includes ulcerative colitis (UC) and Crohn’s disease (CD). Worldwide, the prevalence of IBD has continued to increase steadily in early industrialized regions and has risen rapidly in newly industrialized countries [1], making it a global public health challenge. The cause of IBD remains incompletely understood, and it has been widely believed that diet plays a significant role in the onset and progression of IBD [2]. The European Society for Clinical Nutrition and Metabolism (ESPEN) guideline has reported that diet and enteral nutrition can effectively reduce disease flares, alleviate intestinal symptoms, and decrease the need for surgery [3]. Thus, healthy dietary behavior has become an essential component of disease management among patients with IBD. However, accumulating evidence has highlighted multiple challenges in dietary behaviors within this population, including food avoidance [4], limited knowledge regarding dietary management [5], and poor adherence to dietary recommendations [6]. In addition, previous studies have reported that patients with IBD experienced a range of negative psychosocial experiences during their efforts to manage diet, such as uncertainty and frustration [7, 8]. These challenges substantially increase the risk of adverse health outcomes, such as malnutrition and gastrointestinal symptoms [9]. To effectively address these challenges, it is necessary to investigate the factors influencing healthy dietary behavior among patients with IBD. Currently, studies on dietary behaviors in patients with IBD have primarily focused on their dietary intake patterns, while neglecting a critical aspect: the behavioral decision-making processes involved in selecting and adhering to healthy dietary options, as well as the multifaceted factors influencing these decisions [10]. It is crucial to recognize that healthy dietary behaviors are closely linked to patients’ own decision-making [11]. According to the adaptive decision-making framework [12], dietary decision-making among patients with IBD can be understood at two levels. At the action level, it involves decisions regarding what, how much, how, and why to eat. At the reflection level, dietary decision-making entails considering one’s health status, beliefs, goals, and values to make choices that provide guidance for future dietary behaviors. Previous studies [13–15] have also demonstrated that dietary decision-making among patients with IBD is influenced not only by individual knowledge and specific skills, but also by factors such as beliefs, goals, and preferences. In addition, external influences (such as financial burdens, and social support) may further affect dietary decisions [16]. Thus, dietary decision-making is a complex process shaped by the mutual influence of cognitive, emotional, and social factors. Approaches that focus on a single dimension, such as knowledge, nutrition literacy, or dietary beliefs, are insufficient to fully account for individual differences and cannot provide adequate direction for interventions targeting long-term dietary management. Therefore, this study introduced the concept of decision-making competence to examine the process from a comprehensive perspective. Decision-making competence refers to an individual’s integrated competence to navigate the decision-making process, including problem identification, information integration, planning, and evaluative feedback, while being influenced by multiple factors (e.g., cognition, social support, and emotions) [17]. Initially, most studies focused on patients’ competence in participating in medical decision-making to safeguard their autonomy and right to informed consent [18]. Currently, research has increasingly focused on the decision-making competence of patients with chronic diseases in self-management, namely their competence to make appropriate decisions regarding disease-related aspects such as symptoms, medications, and diet [19, 20]. Within this context, healthy dietary decision-making competence can be defined as a multi-dimensional and integrative competence that that enables individuals to manage different decision-making processes and make dietary decision that support their health. Further clarification is required between the healthy dietary decision-making competence and related constructs. Self-efficacy refers to an individual’s belief or confidence in their ability to successfully perform specific behaviors in a given context [21]. Although it is an important determinant of health behavior, it does not encompass the decision-making process which is a key enabler for HCPs to develop individualized healthy behaviors based on patients’ needs, preferences, and values [22]. Patient activation describes the degree to which individuals are ready, willing, and able to engage in disease management [23]. Although closely related to decision-making competence, this construct primarily emphasizes proactive engagement and does not fully account for the stages of decision balance or post-decision reflection. Nutrition literacy, derived from the broader construct of health literacy, is defined as the basic literacy necessary to understanding and obtaining information about nutrition [24]. Compared with healthy dietary decision-making competence, nutrition literacy is focused on the literacy to access and utilize information, without addressing other essential components of the decision-making process (e.g. post-decision reflection). Taken together, these distinctions highlight that healthy dietary decision-making competence is a broader, integrative construct encompassing, but not limited to, self-efficacy, patient activation, and nutrition literacy. Through our literature review, we found that most existing instruments assessed dietary decision-making in patients with IBD only indirectly, focusing on aspects such as dietary intake, nutritional knowledge, or dietary behaviors. For example, the Groningen IBD Nutritional Questionnaire [25] was developed to evaluate food intake among patients with IBD, but it focuses mainly on the frequency and type of food intake rather than the broader construct of dietary decision-making competence. Similarly, the Inflammatory Bowel Disease-related Behaviors Questionnaire [26] includes the Avoiding Food and Activities dimension, which reflects IBD-specific dietary behaviors and indirectly reflects dietary decision-making competence through the outcomes of dietary decisions. In addition, a Nutritional Literacy Scale for patients with IBD was developed by Wang et al. in China [27], which evaluates nutrition literacy across five dimensions: nutritional knowledge, nutritional attitudes, nutritional practices, information interaction skills, and information appraisal skills. Nevertheless, since dietary decision-making includes weighing the pros and cons between different dietary goals or plans and actively adjusting behaviors in response to outcomes, this tool does not adequately capture the full spectrum of patients’ dietary decision-making competence. Furthermore, the limitations of existing instruments hinder the appropriate evaluation of interventions aimed at promoting healthy eating and dietary management among patients with IBD. Outcome measures in such studies often include symptoms [28], quality of life [29], fatigue [30], or adherence [31]; however, these indicators cannot directly reflect improvements in dietary decision-making competence or dietary management. Therefore, there is a critical need to develop and validate a novel instrument that specifically focuses on the dietary decision-making process and its multidimensional characteristics, in order to measure the proposed construct of healthy dietary decision-making competence in patients with IBD. Methods The HDMCS for patients with IBD was developed and validated through an instrument development phase and an instrument validation phase, in accordance with the scale development standards suggested by Boateng [32] and following the Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) Risk of Bias Checklist [33]. This study was conducted in two phases: (i) instrument development and (ii)scale validation (see Fig. 1). Instrument development and psychometric testing process. Summary of each step of the study, illustrating whether items were added, deleted, merged, or modified at each stage, and showing the changes in the number of items across the different steps Phase 1: instrument development Item generation To generate the initial items of the instrument, we constructed an item pool drawing upon two main sources: (i) a comprehensive literature review on dietary decision-making competence and (ii) findings from in-depth interviews with 24 patients diagnosed with IBD. Additionally, a five-member research group, comprising three experts familiar with IBD research area, one doctoral student and one master’s student, was established to design the original questionnaire. For the comprehensive literature review, the research group searched articles published from the establishment of the database to February 2024 in five international databases, including PubMed, Web of Science, Embase, Cochrane Library and CINAHL, and three Chinese databases, including CNKI, SinoMed and Wanfang Database, by combing keywords such as “decision making/decision”, “competence/ability”, “diet/eating/food/nutrition” and “inflammatory bowel disease/ulcerative colitis/Crohn’s disease/chronic disease”. The research group conducted a comprehensive review of the theories, professional knowledge, assessment tools, and scale development methodologies relevant to the study topic. For the in-depth interviews, 24 patients with IBD were recruited from four tertiary hospitals in Nanjing, China. The purposive sampling with maximum variation was used until data saturation was reached. The major interview questions were: “Has your diet changed since your diagnosis?”, “Why would you choose to change your diet?”, and “What information do you primarily rely on when deciding on your diet?” The collected interview data were analyzed using the qualitative thematic analysis method guided by Braun and Clarke’s [34]. Through a synthesis of findings from both the literature and interview data, the tentative attributes of the concept of healthy dietary decision-making competence in patients with IBD were identified. To systematically organize these attributes and ensure comprehensive coverage of the dietary decision-making competence, we applied a holistic competence framework to guide the item generation [35]. This framework delineates competence into four dimensions: cognitive competence (knowledge and understanding), functional competence (skills), social competence (behaviors and attitudes), and meta competence (promoting the development of other competences). Based on these attributes and guided by this framework, a preliminary draft of the scale was developed. Then, the research team reviewed all items, evaluating content relevance, face validity, and practical usefulness for dietary decision-making, deleting inappropriate items and integrating overlapping ones. Delphi survey A two-round Delphi survey was conducted to obtain the expert reviews of the preliminary items. Experts with at least five years of experience in working with patients with inflammatory bowel disease or in the field of scale development were invited to participate in the evaluation. The item rating was based on a 5-point scale: 1 (least important) to 5 (most important), to evaluate whether the preliminary items were suitable for measuring healthy dietary decision-making competence. Items with an importance assignment of ≤ 3.50 points and a coefficient of variation of > 0.25 were eliminated. Once consensus was reached among the panel, data collection process was concluded. Content and face validity test To assess the content validity of the instrument draft, ten experts in the healthcare of patients with IBD were invited to assess the relevance of each item using a 4-point Likert scale: 1 (not at all), 2 (somewhat), 3 (moderately so), and 4 (very much so). Then, the item-level content validity index (I-CVI: the ratio of the number of experts rating 3 or 4) and the scale-level content validity index (S-CVI: the average of I-CVI scores) were calculated. Items with I-CVI of more than 0.78 were retained, and the rest were deleted [36]. Then, cognitive interviews were conducted with eight patients with IBD to estimate the mean completion time, and examine the intelligibility of the items. The “think-aloud” method was encouraged when the participants were filling out the questionnaire. Meanwhile, the interviewer observed the participants’ response processes and asked probing questions regarding any pauses or modifications made during completion. To elicit more information, three types of verbal probing questions were employed: (1) comprehensiveness/interpretation probes (e.g., why do you think…), (2) paraphrasing probes (e.g., please repeat that statement in your own words), and (3) general probes (e.g., how did you arrive at that answer) [37]. Interviews were conducted face-to-face by the first author in a relatively quiet and private room within the hospital. Phase 2: scale verification Study setting and sampling A formal investigation was carried out to validate the HDMCS. The inclusion criteria were as follows: (1) adults (aged 18 or above) diagnosed with IBD (confirmed by a combination of endoscopic, radiological, biochemical and histological examinations); (2) patients who understood this study and agreed to participate. Patients with severe psychiatric illness or physical disabilities, a concomitant diagnosis of diabetes, severe kidney failure, or other chronic diseases which may require a special therapeutic diet, or those who did not provide informed consent were excluded. For exploratory factor analysis (EFA), the appropriate sample size was calculated to be five to ten participants per item of the scale. For confirmatory factor analysis (CFA), a sample of 200 or more was considered appropriate. Considering a non-response rate of 10%, the predefined sample size for this formal investigation was determined to be 367. A total of 382 participants were recruited, thus the sample size in our study was deemed acceptable. Data collection Data were collected between September 2024 and January 2025 from outpatient and inpatient wards of five tertiary general hospitals and patient groups at IBD centers in Nanjing (Jiangsu province) and Shenzhen (Guangdong province). These cities serve as major medical centers in eastern and southern China, attracting patients from a wide catchment area that extends beyond the cities themselves to surrounding provinces. Consequently, the participant sample likely included individuals from a broader geographic area beyond the two cities. Both paper-based questionnaires and the Questionnaire Star platform, an online system for surveys and assessments in China, were used for data collection. All data collection and processing were conducted anonymously. No financial or material incentives were provided to participants for their involvement in this study. Measures Socio-demographic and clinical informationParticipants were asked to self-report several demographic and illness-related information, such as gender, age, education level and disease types. Socio-demographic and clinical informationParticipants were asked to self-report several demographic and illness-related information, such as gender, age, education level and disease types. Participants were asked to self-report several demographic and illness-related information, such as gender, age, education level and disease types. Participants were asked to self-report several demographic and illness-related information, such as gender, age, education level and disease types. The Healthy Dietary Decision-Making Competence ScaleThe preliminary version of the HDMCS was a disease-specific self-report instrument designed to assess the level of healthy dietary decision-making competence. It consisted of 26 items rated on a five-point Likert scale. A total score was calculated by summing all items scores, with higher scores indicating greater competence in making healthy dietary decisions. The Healthy Dietary Decision-Making Competence ScaleThe preliminary version of the HDMCS was a disease-specific self-report instrument designed to assess the level of healthy dietary decision-making competence. It consisted of 26 items rated on a five-point Likert scale. A total score was calculated by summing all items scores, with higher scores indicating greater competence in making healthy dietary decisions. The preliminary version of the HDMCS was a disease-specific self-report instrument designed to assess the level of healthy dietary decision-making competence. It consisted of 26 items rated on a five-point Likert scale. A total score was calculated by summing all items scores, with higher scores indicating greater competence in making healthy dietary decisions. The preliminary version of the HDMCS was a disease-specific self-report instrument designed to assess the level of healthy dietary decision-making competence. It consisted of 26 items rated on a five-point Likert scale. A total score was calculated by summing all items scores, with higher scores indicating greater competence in making healthy dietary decisions. The 12-item Short-form Nutrition Literacy ScaleThe NL-SF12, developed by Mo et al. [38], is a 12-item self-report instrument rated on a five-point Likert scale. Higher scores reflect stronger beliefs in the necessity of dietary management and greater competence in making appropriate nutritional decisions. Considering that this scale measures a concept similar to dietary decision-making competence, it was hypothesized that the HDMCS scores would be positively associated with the NL-SF12 scores. The 12-item Short-form Nutrition Literacy ScaleThe NL-SF12, developed by Mo et al. [38], is a 12-item self-report instrument rated on a five-point Likert scale. Higher scores reflect stronger beliefs in the necessity of dietary management and greater competence in making appropriate nutritional decisions. Considering that this scale measures a concept similar to dietary decision-making competence, it was hypothesized that the HDMCS scores would be positively associated with the NL-SF12 scores. The NL-SF12, developed by Mo et al. [38], is a 12-item self-report instrument rated on a five-point Likert scale. Higher scores reflect stronger beliefs in the necessity of dietary management and greater competence in making appropriate nutritional decisions. Considering that this scale measures a concept similar to dietary decision-making competence, it was hypothesized that the HDMCS scores would be positively associated with the NL-SF12 scores. The NL-SF12, developed by Mo et al. [38], is a 12-item self-report instrument rated on a five-point Likert scale. Higher scores reflect stronger beliefs in the necessity of dietary management and greater competence in making appropriate nutritional decisions. Considering that this scale measures a concept similar to dietary decision-making competence, it was hypothesized that the HDMCS scores would be positively associated with the NL-SF12 scores. The Nine-Item Avoidant/Restrictive Food Intake Disorder Screen (NIAS)The Nine-Item Avoidant/Restrictive Food Intake Disorder Screen (NIAS) is a self-reported questionnaire consisting of nine items that evaluate eating behaviors associated with Avoidant/Restrictive Food Intake Disorder (ARFID) [39]. It comprises three subscales: the picky eating subscale, which assesses sensory-based food aversion; the poor appetite subscale, which captures lack of interest in food or eating; and the fear of negative consequences subscale, which measures fear of aversive outcomes related to eating. The NIAS was translated into Chinese by Jinbo He [40] and employs a 6-point Likert scale ranging from 0 (“strongly disagree”) to 5 (“strongly agree”). In accordance with existing literature, a total score of 28 or higher on the NIAS indicates a positive screen for ARFID. Previous studies [41–43] have indicated that ARFID was prevalent among individuals with IBD and was related to restrictive dietary attitudes, lower quality of life, and poorer psychosocial functioning. Therefore, the NIAS was employed in this study to measure dietary behaviors of patients with IBD and to examine the discriminant ability of the HDMCS. The Nine-Item Avoidant/Restrictive Food Intake Disorder Screen (NIAS)The Nine-Item Avoidant/Restrictive Food Intake Disorder Screen (NIAS) is a self-reported questionnaire consisting of nine items that evaluate eating behaviors associated with Avoidant/Restrictive Food Intake Disorder (ARFID) [39]. It comprises three subscales: the picky eating subscale, which assesses sensory-based food aversion; the poor appetite subscale, which captures lack of interest in food or eating; and the fear of negative consequences subscale, which measures fear of aversive outcomes related to eating. The NIAS was translated into Chinese by Jinbo He [40] and employs a 6-point Likert scale ranging from 0 (“strongly disagree”) to 5 (“strongly agree”). In accordance with existing literature, a total score of 28 or higher on the NIAS indicates a positive screen for ARFID. Previous studies [41–43] have indicated that ARFID was prevalent among individuals with IBD and was related to restrictive dietary attitudes, lower quality of life, and poorer psychosocial functioning. Therefore, the NIAS was employed in this study to measure dietary behaviors of patients with IBD and to examine the discriminant ability of the HDMCS. The Nine-Item Avoidant/Restrictive Food Intake Disorder Screen (NIAS) is a self-reported questionnaire consisting of nine items that evaluate eating behaviors associated with Avoidant/Restrictive Food Intake Disorder (ARFID) [39]. It comprises three subscales: the picky eating subscale, which assesses sensory-based food aversion; the poor appetite subscale, which captures lack of interest in food or eating; and the fear of negative consequences subscale, which measures fear of aversive outcomes related to eating. The NIAS was translated into Chinese by Jinbo He [40] and employs a 6-point Likert scale ranging from 0 (“strongly disagree”) to 5 (“strongly agree”). In accordance with existing literature, a total score of 28 or higher on the NIAS indicates a positive screen for ARFID. Previous studies [41–43] have indicated that ARFID was prevalent among individuals with IBD and was related to restrictive dietary attitudes, lower quality of life, and poorer psychosocial functioning. Therefore, the NIAS was employed in this study to measure dietary behaviors of patients with IBD and to examine the discriminant ability of the HDMCS. The Nine-Item Avoidant/Restrictive Food Intake Disorder Screen (NIAS) is a self-reported questionnaire consisting of nine items that evaluate eating behaviors associated with Avoidant/Restrictive Food Intake Disorder (ARFID) [39]. It comprises three subscales: the picky eating subscale, which assesses sensory-based food aversion; the poor appetite subscale, which captures lack of interest in food or eating; and the fear of negative consequences subscale, which measures fear of aversive outcomes related to eating. The NIAS was translated into Chinese by Jinbo He [40] and employs a 6-point Likert scale ranging from 0 (“strongly disagree”) to 5 (“strongly agree”). In accordance with existing literature, a total score of 28 or higher on the NIAS indicates a positive screen for ARFID. Previous studies [41–43] have indicated that ARFID was prevalent among individuals with IBD and was related to restrictive dietary attitudes, lower quality of life, and poorer psychosocial functioning. Therefore, the NIAS was employed in this study to measure dietary behaviors of patients with IBD and to examine the discriminant ability of the HDMCS. Data analysis The data were analyzed using SPSS 26.0 and AMOS 23.0. Socio-demographic characteristics and scale scores were summarized using descriptive statistics. Continuous variables (e.g., age and item scores) were presented as means and standard deviations (SDs), whereas categorical variables (e.g., educational level and occupation) were expressed as frequencies and percentages. A p-value < 0.05 was considered statistically significant. The specific methods for validity and reliability testing were detailed below. Item analysis Item analysis was performed to ensure variability in item difficulty and distinguishability among respondents. (1) Item distribution: Descriptive analysis was performed to analyze the concentration of answers to each item. If the selection rate of any option for an item exceeded 80% or the sum of the selected frequency of any two options was < 10%, indicating that the answers had a significantly skewed distribution and the item should be considered for deletion; (2) Item dispersion degree analysis: Any item with a standard deviation < 0.75 was considered for deletion; (3) Critical ratio (CR) analysis method: Participants were divided into a high group (the top 27%) and a low group (the bottom 27%) based on total scale scores [44]. An independent-samples t-test was performed to compare the difference between the two groups. Items were considered for deletion if P > 0.05 or CR value was less than 3.5. (4) Correlation coefficient method: Items with a low correlation coefficient (< 0.3) and/or P > 0.05 were considered for deletion. (5) Cronbach’s α coefficient method: Items were considered for deletion if their removal increased the overall Cronbach’s α. Following these methods, items meeting more than three criteria were deleted. If an item met less than three criteria, it was discussed for modification, merging or deletion with the research group based on theoretical relevance and content validity. Construct validity EFA and CFA were conducted to evaluate the construct validity. The data were deemed suitable for EFA based on a Kaiser–Meyer–Olkin (KMO) value of ≥ 0.80 and a statistically significant Bartlett’s test of sphericity (P < 0.05) [45]. EFA was performed using principal component analysis and varimax rotation. The factor structure of the HDMCS was identified by considering eigenvalues (≥ 1.00), the scree plot, the percentage of variance explained, and item factor loadings (>0.40). Items were deleted if their factor loadings were below 0.40, or belonging to a factor with fewer than three items [46]. EFA was repeated after deletion conducted until no further item deletion was needed. Items grouped together were carefully reviewed and compared with the provisional attributes and elements found through the literature review and interviews. CFA was subsequently performed to confirm the factor structure obtained from EFA, using covariance matrices and maximum likelihood estimations. Model fit was assessed using standard indices, including Root Mean Square Error of Approximation (RMSEA), Standardized Root Mean Square Residual (SRMR), Goodness-of-Fit Index (GFI), Comparative Fit Index (CFI), and Tucker-Lewis Index (TLI) [47]. The thresholds for acceptable and good fit are shown in Table 1. If most fit indices of the initial model did not meet acceptable thresholds, minor model modifications were made based on modification indices and theoretical plausibility. Table 1Criteria for evaluating model fit indicesFit criteriaχ²/dfRMSEASRMRGFINFITLICFIAcceptable fit<3≤ 0.08≤ 0.10≥ 0.90≥ 0.90≥ 0.90≥ 0.90Good fit<2≤ 0.05≤ 0.05≥ 0.95≥ 0.95≥ 0.95≥ 0.95 Criteria for evaluating model fit indices Discriminant validity The discriminant validity was analyzed by using the square root of the Average Variance Extracted (AVE) and Heterotrait-Monotrait (HTMT) ratio. First, First, The Average Variance Extracted (AVE) was calculated from the CFA sample, which indicates how closely items within each domain are related. An AVE greater than 0.5 was considered acceptable [48]. According to the Fornell-Larcker criterion, the square root of the AVE for each latent variable should be greater than its correlations with any other latent variables. Additionally, the HTMT test was conducted, which was proposed by Henseler et al. [49] as a sensitive and specific indicator of discriminant validity. When HTMT values are high, they suggest potential issues with discriminant validity. The threshold for the HTMT analysis was set at 0.85 [48]. In this study, discriminant validity was considered acceptable if either of the two criteria was met. Convergent validity Convergent validity, the extent to which dietary decision-making competence scores from the HDMCS demonstrated a strong relationship with scores on similar measures, was assessed by comparing the correlation between the HDMCS and a validated and widely used measure of nutrition literacy, the NL-SF12. The strength of the correlation coefficient was judged as follows [50]: <0.1 was set to negligible; a weak correlation was denoted by values between 0.1 and 0.3; a moderate correlation was indicated by ≥ 0.3 and < 0.5; and a value of ≥ 0.5 indicated a strong correlation. Reliability test Internal consistency was assessed using Cronbach’s α coefficient and split-half reliability, with a Cronbach’s α above 0.7 considered desirable. Moreover, the stability of the scale was examined by test-retest reliability. Twenty-five participants from the initial investigation, whom disease activity had no significant change over the two-week interval, were invited to complete the same questionnaire again. To further assess the temporal stability, an additional 28 participants who met the same inclusion and exclusion criteria were recruited in September 2025 from a tertiary general hospital in Nanjing; 3 of them dropped out during the 2-week interval. Ultimately, 50 paired responses were included in the test–retest reliability analysis. A two-way random-effect model based on absolute agreement was used to calculate the intraclass correlation coefficient (ICC). The ICC is considered poor at less than 0.40, fair at 0.4–0.6, good at 0.6–0.75, and excellent at 0.75–1.00 [51]. Discriminant ability Discriminant validity refers to the extent to which an instrument can differentiate groups that should theoretically differ on the construct of interest; in this study, it was applied to evaluate whether the HDMCS could distinguish between groups with potentially different levels of dietary decision-making competence. To enhance the robustness of the analysis, a data-driven exploratory strategy was implemented to identify relevant clinical and demographic variables. Based on preliminary analyses, variables showing significant associations with dietary decision-making competence were selected for discriminant validity assessment. For normally distributed data, independent-samples t-tests were conducted for dichotomous variables, whereas one-way analysis of variance (ANOVA) was applied for variables with more than two categories, followed by Tukey post-hoc comparisons to assess intergroup differences. The results of t-tests were presented with the test statistic (t), degrees of freedom (df), 95% confidence interval, and p-value. The ANOVA results were reported with the F statistic (F), df1, df2, and p-value, and Tukey’s tests were summarized with mean differences (Mdiff), 95% confidence intervals, and adjusted P values. Percentile analysis Percentile analysis is a data-driven approach that accounts for the distribution of scores within the sample, rather than relying on arbitrary or subjective criteria. To preliminarily establish cut-off scores for the HDMCS, quartile analysis was employed to identify specific points representing different levels of healthy decision-making competence. Analysis of variance (ANOVA) was then conducted to explore potential differences in mean HDMCS scores across the four groups, ranging from low to high competence. Subsequently, Tukey’s test was applied to identify pairwise differences between the competence groups. Ethical considerations This study was approved by the Medical Ethics Committee of the Nanjing Hospital of Chinese Medicine, affiliated with Nanjing University of Chinese Medicine (approval number KY2022029). Preliminary and main investigation data were collected after obtaining written or electronic consent from participants who understood the purpose of the study and voluntarily agreed to participate. Results Instrument development Item generation Based on a comparative analysis of findings from the literature review and interview data, seven attributes and eleven elements of healthy dietary decision-making competence in patients with IBD were identified. For each element, 4 to 8 items were developed, yielding an initial pool of 62 items (Supplementary Material 1). Specifically, the comprehensive literature review generated 45 preliminary items. Subsequently, interview transcripts were coded and thematically analyzed. Themes and statements repeatedly mentioned by patients were transformed into concrete questionnaire items, contributing an additional 17 items derived from the interview data. Subsequently, the research group reviewed all items in the pool, focusing on their theoretical relevance, readability, and potential redundancy. Following this discussion, 21 items were removed due to redundancy or limited relevance, resulting in a preliminary instrument comprising 41 items. Delphi survey Thirty experts were invited via email to complete the Delphi survey. In the first round, 24 experts responded: eight clinical nursing specialists in IBD, three clinical specialists in IBD, one registered dietitian and two nursing experts with a specialization in nutrition, six chronic disease management specialists, two patient decision-making specialists, and two scale development specialists. The average length of experience was 19.92 years. In the second round, 23 experts completed the second round. The coefficient of expert authority was 0.89 in round 1 and 0.91 in round 2. Additionally, the Kendall’s concordance coefficient was 0.161 and 0.126 in round 1, and 0.138 and 0.120 in round 2. After the two-round Delphi survey, 14 items were excluded, 11 items were integrated into 4 items, and 6 new items were added. Moreover, 20 items were reworded based on the experts’ suggestions. Consequently, the draft of HDMCS comprised 26 items. Content and face validity test The I-CVI for the 26 items ranged from 0.80 to 1.00; the S-CVI/UA was 0.81 and the S-CVI/Ave was 0.98, indicating that HDMCS had acceptable content validity. Five adults (2 males and 3 females), aged 33 to 58, completed the first-round cognitive interviews, with an average interview duration of 17.6 min. A summary of the modifications was as follows: (i) Two participants found it difficult to use the original response scale ranging from “strongly agree” to “strongly disagree,” noting that a frequency-based format would better capture their experiences. Accordingly, for 21 items (items 7–26), the anchors on the 5-point scale were revised to “never” and “always.” (ii) Two participants reported difficulty understanding the phrase “emotion regulation skills.” After discussion, the item was revised to: “I employ strategies to relieve excessive anxiety, fear, or other emotional distress.” (iii) One participant expressed uncertainty about the meaning of “long-term impact.” To improve clarity, a note was added in parentheses. The revised item read: “When developing a dietary plan, I take the long-term effects (beyond six months) of such a diet into consideration rather than short-term benefits.” Following the second-round cognitive interviews with three adults (1 male and 2 females), aged 27 to 38, with an average duration of 11.5 min, positive feedback was received, and no further adjustments required. Scale verification Characteristics of the sample A total of 382 participants with IBD were enrolled, of whom 354 (92.67%) submitted valid questionnaires. The main reasons for non-participation or invalid questionnaires were absence during the survey, incomplete responses, and lack of interest in the study objectives. The mean age of 354 participants was 38.02 (12.68) years, with 205 (57.91%) being female and 245 (69.21%) diagnosed with CD. Among the 354 valid questionnaires, a sample of 144 was randomly chosen to carry out EFA, and the remainder (n = 220) were used to carry out CFA. Participants’ characteristics are shown in Table 2. Table 2Demographic characteristics of patients with IBD (n = 354)VariablesTotal sample (n = 354, %)EFA sample (n = 144, %)CFA sample (n = 210, %) Gender Male205(57.9)70(48.6)135(64.3) Female149(42.1)74(51.4)75(35.7) Age 18~221(62.4)97(67.4)124(59.0) 41~112(31.6)41(28.5)71(33.8) ≥ 6021(5.9)6(4.2)15(7.1) Education Junior high school or less53(15.0)16(11.1)37(17.6) High school65(18.4)25(17.4)40(19.0) College/Bachelors’ degree206(58.2)90(62.5)116(55.2) Master’s degree or above30(8.5)13(9.0)17(8.1) Marital status Unmarried122(34.5)55(38.2)67(31.9) Married216(61.0)83(57.6)133(63.3) Divorced16(4.5)6(4.2)10(4.8) Living place Rural33(9.3)15(10.4)18(8.6) Suburban45(12.7)19(13.2)26(12.4) Urban276(78.0)110(76.4)166(79.0) Occupation In school43(12.1)16(11.1)27(12.9) Employed214(60.5)89(61.8)125(59.5) Absence/unemployed due to illness60(16.9)28(19.4)32(15.2) Retirement37(10.5)11(7.6)26(12.4) Monthly family income (yuan/person) <300041(11.6)15(10.4)26(12.4) 3001–5000115(32.5)45(31.3)70(33.3) 5001–10,000118(33.3)54(37.5)64(30.5) >10,00080(22.6)30(20.8)50(23.8) Disease type UC163(46.0)62(43.1)101(48.1) CD184(52.0)79(54.9)105(50.0) IBDU7(2.0)3(2.1)4(1.9) Disease duration(years) <154(15.3)20(13.9)34(16.2) 1-5140(39.5)49(34.0)91(43.3) 6-1073(20.6)38(26.4)35(16.7) >1087(24.6)37(25.7)50(23.8) Disease activity Remission266(75.1)105(72.9)161(76.7) Active88(24.9)39(27.1)49(23.3) Treatment Drugs for IBD Yes340(96.0)137(95.1)203(96.7) No14(4.0)7(4.9)7(3.3) ARFIID Yes49(13.8)21(14.6)28(13.3) No305(86.2)123(85.4)182(86.7) Demographic characteristics of patients with IBD (n = 354) Item analysis The mean score for each HDMCS item ranged from 2.47 to 4.62, with SDs ranging from 0.71 to 1.45. The CR for each item ranged from 3.696 to 20.389 (P < 0.05). The Cronbach’s α coefficients ranged from 0.933 to 0.939, except for items 1 to 8 (Supplementary Material 2). The item analysis results showed that item 4 did not meet the criteria for item distribution, critical ratio analysis, correlation coefficient method, and Cronbach’s α coefficient method, and was therefore removed. It is worth noting that although 12 items showed suboptimal performance on the Cronbach’s α coefficient method, the research group decided to retain them temporarily due to the overall stability of the scale and the theoretical relevance of these items. These items would be removed if they demonstrated poor performance in later analyses. Construct validity Exploratory factor analysis After the item analysis, PCA was performed using the 25 items, excluding Item 4. In the first round, the KMO value was 0.895, and the Bartlett’s sphericity test was significant (χ2 = 2336.588, p < 0.001), indicating that the data were suitable for the EFA. Five factors with eigenvalues greater than 1 were extracted, and the cumulative variance explained by these factors was 68.71%. Item 19 was simultaneously loaded on Factor 1 (factor loading = 0.404) and Factor 2 (factor loading = 0.617), but the difference between the loadings was more than 0.2. After research group discussion, the research group determined that the competence to adjust dietary decision-making goals (e.g., nutritional supplementation, achieving symptom relief) in response to one’s health condition was a critical aspect of dietary decision-making competence in patients with IBD. Therefore, Item 19 was temporarily retained. Factor 5 contained only Items 5 and 6, a structure that could compromise the stability of the dimension. Moreover, both items showed suboptimal results in item analysis. Given these considerations, the research group ultimately decided to delete Items 5 and 6. In the second round of EFA, PCA was performed on the 23 remaining items, with a KMO value of 0.901 and a significant result on Bartlett’s sphericity test (χ2 = 2193.494, P < 0.001). Four factors with eigenvalues greater than one were extracted, explaining 20.12%, 18.94%, 17.59%, and 10.29% of the total variance, respectively, for a cumulative total of 66.93%. In this round, no items met the deletion criteria (Supplementary Material 3). The research group identified the characteristics of the four factors based on their loadings and assigned each factor a name that encompassed all the items within the factor. Factor 1 was named “dietary management perception”, as it reflected the attitudes and beliefs of patients with IBD toward healthy eating and provided an initial measure of their values of healthy dietary behaviors, with factor loadings ranging from 0.783 to 0.898. Factor 2 was named “resource utilization competence”, as it reflected patients’ competence to actively access, understand, and utilize informational, social, and instrumental resources to support healthy dietary decision-making, with factor loadings ranging from 0.606 to 0.752. Factor 3 was named “decisional balance competence”, as it reflected patients’ ability to weigh the pros and cons and make decisions based on their personal values, with factor loadings between 0.696 and 0.783. Factor 4 was named “post-decision reflection competence”, as it reflected patients’ competence to cope with various specific situations after making decisions, to summarize decision-making experiences, and to make necessary adjustments, with factor loadings between 0.574 and 0.785. Finally, the items were re-ordered according to their corresponding dimensions (supplementary Material 3). Confirmatory factor analysis Using the four-factor model with the 23 items, CFA was conducted to cross-validate the factor structure obtained through EFA. While some indices indicated an acceptable fit to the data (χ²/df = 2.463, P<0.01, SRMR = 0.0566), several others indicated poor model fit (RMSEA = 0.084, GFI = 0.808, NFI = 0.828, IFI = 0.806, TLI = 0.890, CFI = 0.875). Therefore, minor modifications were made to the model based on modification indices and theoretical justification. The analysis showed six pairs of items with high modification indices, along with similarities in wording and slightly overlapping in content, which might have displayed a non-random measurement error [52]. To address these issues, without artificially inflating model complexity, we sequentially added six error covariances, thereby improving model fit while maintaining theoretical coherence. Thus, the final model showed a good fit to the data. Specifically, it allowed correlations among unique variances between three item pairs from the resource utilization competence factor (items 4–6, 4–9, and 5–9) and three item pairs from the post-decision reflection competence (items 17–19, 21–22, and 22–23) (see Table 3; Fig. 2 for details). Although the GFI and NFI were slightly below the recommended threshold of 0.90, all other indices exceeded the acceptable criteria. Taken together, the findings supported the adequacy of the four-factor model and confirm the factorial validity of the HDMCS. Table 3Goodness-of-fit indices for CFA of the HDMCS (n = 210)Fit criteriaχ²/dfRMSEASRMRGFINFIIFITLICFIInitial model2.4630.0840.05660.8080.8280.8060.890.875Modified model1.9720.0680.05060.8480.8660.9290.9170.928Acceptable fit< 3≤ 0.08≤ 0.10≥ 0.90≥ 0.90≥ 0.90≥ 0.90≥ 0.90 Goodness-of-fit indices for CFA of the HDMCS (n = 210) Modified CFA model for four factors model of scale (n = 210). The adjusted CFA model revealed a four-factor structure for the measurement construct of healthy dietary self-decision-making competence. The values displayed to the left of each item indicate the standardized factor loadings. Guided by modification indices and theoretical considerations, error covariances were specified among items 4–6, 4–9, and 5–9 within Factor 2, and among items 17–19, 21–22, and 22–23 within Factor 4 Discriminant validity The AVE values of the factors ranged from 0.501 to 0.620, with square roots ranged from 0.707 to 0.787. According to the Fornell-Larcker criterion, the square roots of Factor 2 and Factor 4 did not fully meet this criterion, as their AVE square roots were slightly lower than their correlations with Factor 3 (Table 4). Given the slight discrepancies, discriminant validity was further evaluated using the HTMT ratio, which showed values ranged from 0.274 to 0.831, all below 0.85, indicating acceptable discriminant validity (Table 5). Table 4Discriminant validity analyses: Fornell-Larcker criterion resultsFactor/scaleCorrelation between factors (n = 354)AVEFactor 1Factor 2Factor 3Factor 4Factor 1 0.772 0.596Factor 20.320** 0.707 0.501Factor 30.318**0.713** 0.787 0.620Factor 40.335**0.618**0.728** 0.721 0.520Total scale0.481**0.832**0.909**0.893**-The diagonal values (bold) of the matrix represent the square root of AVE, AVE average variance extracted; ** means P< 0.01. Discriminant validity analyses: Fornell-Larcker criterion results The diagonal values (bold) of the matrix represent the square root of AVE, AVE average variance extracted; ** means P< 0.01. Table 5Discriminant validity analyses: HTMT criterion resultsFactorFactor 1Factor 2Factor 3Factor 4Factor 11Factor 20.2741Factor 30.3120.7891Factor 40.3780.7150.8311 Discriminant validity analyses: HTMT criterion results Convergent validity Significant and positive correlations were found between the overall score of HDMCS and scores on NL-SL12 (r = 0.457, P < 0.01, details are provided in Table 6). Table 6Convergent validity and reliability test of the HDMCSFactor/scaleCorrelation between NL-SF12Cronbach’s α (n = 354)ICC (n = 50)ICC 95% CIFactor 10.335**0.8180.831(0.721, 0.900)Factor 20.409**0.8600.804(0.669, 0.886)Factor 30.373**0.9120.697(0.519, 0.817)Factor 40.361**0.9000.898(0.826, 0.941)Total scale0.457**0.9410.874(0.787, 0.927) Convergent validity and reliability test of the HDMCS Reliability test The Cronbach’s α coefficient for the HDMCS was 0.941, and the coefficients for the four factors ranged from 0.818 to 0.912 (Table 6). The result of inter-item correlations revealed that the highest correlation coefficient was 0.721 (P < 0.01). Furthermore, the correlations between items from different factors ranged from 0.105 to 0.582 (P < 0.05). The split-half reliability of the scale was 0.870. Additionally, the test–retest reliability (by ICC) was ranged from 0.697 to 0.898, indicating acceptable reliability for this scale (Table 6). Discriminant ability To evaluate the discriminant ability of the HDMCS, exploratory analyses were conducted to identify significant clinical and demographic variables, followed by examinations of whether the scale could distinguish between groups differing on these factors. The ANOVA results indicated a significant effect of disease type [F(2, 351) = 4.018, P = 0.019]. Tukey’s post-hoc comparisons revealed a marginally significant difference between the UC and CD groups [Mdiff = 4.179, 95% CI (−0.09, 8.45), Padj = 0.056]. As the data on monthly family income per capita were not normally distributed, the Kruskal–Wallis test was applied to assess group differences in HDMCS scores (χ²(3) = 12.045, P = 0.007). Dunn’s post-hoc analyses with Bonferroni correction indicated that, compared with participants earning below 3,000 yuan, those with incomes of 5,001–10,000 yuan (Z = −2.470, Padj = 0.037) scored significantly higher, and those with incomes above 10,000 yuan (Z = −2.603, Padj = 0.056) showed a marginally significant difference. Independent-samples t-test results indicated that individuals without ARFID had significantly higher mean HDMCS scores compared with those with ARFID [t(352) = 3.176, Mdiff = 7.722, 95%CI (2.941, 12.502), P = 0.002]. Percentile analysis The results of the percentile analysis were as follows: the 25th percentile (Q1) = 66, the 50th percentile (Q2) = 78, and the 75th percentile (Q3) = 88. Based on these values, decision-making competence scores were categorized into four groups: scores from 0 to 66 were classified as “Low Competence,” scores from 67 to 78 as “Mild Competence,” scores from 79 to 88 as “Moderate Competence,” and scores from 89 and above as “High Competence.” Results of Tukey’s pairwise comparisons showed that, when the “Lowest Competence” group was used as the reference, all mean differences in HDMCS scores were statistically significant. The final HDMCS The final HDMCS is a 5-point scale containing four dimensions with 23 items (Table 7.). The four dimensions are dietary management perception (three items), resource utilization competence (six items), decisional balance competence (six items) and post-decision reflection competence (eight items). All items are positively scored, yielding a total score from 23 to 115; higher scores indicating greater competence in healthy dietary decision-making. The HDMCS can be utilized as an assessment tool in any organization, requiring only 8 min to complete. Table 7Final version of the scaleDimensionItemDietary management perception1. I think that healthy diet is an important component of disease management for inflammatory bowel disease.2. I think that dietary management strategies should be adjusted according to different disease phase (eg, remission or active phase).3. I pay attention to my daily diet based on healthcare professionals’ advice.Resource utilization competence4. I can seek reliable dietary and nutritional information for disease through various sources, including healthcare professionals, the internet, newspapers and books.5. I can maintain detailed diet diary, recording aspects such as meal timing, location, type, quantity and gastrointestinal symptoms.6. I can use nutritional assessment tools (eg, applications) or consult healthcare professionals to evaluate daily energy and nutrient intake.7. I can identify foods that trigger gastrointestinal symptoms or physical discomfort by reviewing dietary diaries or food intolerance test results.8. I can break long-term dietary management goals into smaller, achievable short-term goals (eg, consistently recording dietary intake for a week).9. I can actively seek assistance from others (eg, physicians, dietitians or peers with same disease), when encountering difficulties in dietary management.Decisional balance competence10. When developing a dietary plan (including food intake, types and cooking methods), I carefully weigh the potential benefits and risks associated with each dietary plan.11. When developing a dietary plan, I take my personal food preferences into consideration and evaluate their impact on disease management.12. When developing a dietary plan, I take the long-term effects (beyond six months) of such diet into consideration rather than short-term benefits.13. When developing a dietary plan, I take my healthy goals into consideration.14. I make informed judgments when faced with conflicting dietary or nutritional related information.15. I regularly set and evaluate my dietary management goals based on disease status and physical needs.Post-decision reflection competence16. I adjust my dietary management goals promptly in response to changes in healthy conditions.17. I proactively modify my dietary plan by considering disease status and suggestions from others.18. When encountering unexpected situations (eg, business trips, social dining), I flexibly adjust my dietary plan to accommodate changing circumstances.19. I reflect on past dietary management experience and use them to improve my dietary plan.20. I can implement my dietary plan independently or with assistance from others.21. Despite challenges (eg, opposition from others, low mood), I adhere to my dietary plan by making healthy dietary decisions.22. If recurrence or prodromal symptoms occur after eating certain foods, I take steps to mitigate the symptoms.23. I employ strategies to relieve excessive anxiety, fear, or other emotional distress. Final version of the scale Discussion The HDMCS is a self-report instrument developed to evaluate healthy dietary decision-making competence among patients with IBD. The seven attributes identified during the development phase were validated and refined into four final factors. Notably, the HDMCS measures the cognitive, functional, social, and meta competence within the complex context of dietary decision-making, which faithfully reflects the framework of competence proposed by Le Deist and Winteron [35]. The factor structure of the HDMCS The HDMCS comprises four factors of dietary decision-making competence in IBD patients. Factor 1 (dietary management perception) refers to cognitive competence and aims to measure whether patients with IBD have established an appropriate dietary management perception. The higher awareness of the importance of dietary management, the better engagement in disease management [53]. Moreover, when they recognize the relationship between IBD and diet, they may actively engage in dietary management to maintain health [54]. Therefore, the HCPs should help patients with IBD to develop a reasonable perception of dietary management and dietary decision-making. Factor 2 (resource utilization competence) refers to functional and social competence, which involves one’s competence in actively seeking and making full use of dietary-related resources and in obtaining assistance from others. Previous study [55] indicated that some tools can help patients initially identify potential symptom-triggering foods and calculate their nutrient intake. Therefore, Factor 2 highlights not only patients’ competence to actively seek and apply relevant information, but also their proficiency in utilizing appropriate tools to facilitate informed dietary decision-making. Factor 3 (decisional balance competence) is derived from the Transtheoretical Model [56], which describes the process of weighing the pros and cons of behavioral change, with action more likely when perceived benefits outweigh the costs. In this scale, it refers to the cognitive and functional competence, which reflects how patients evaluate the advantages and disadvantages of dietary-related content to guide decisions. Based on previous studies [13, 57] and the qualitative findings, patients with IBD faced different conflicts, such as dietary benefits vs. drawbacks, enjoyment vs. disease control. Items in this factor assess dietary decision-making preferences among IBD patients, thereby aiding HCPs in developing targeted value clarification interventions. Factor 4 (post-decision reflection competence) refers to meta and functional competence, measuring patients’ competence in critically reflecting on their dietary decision-making processes and physiological responses, adjusting their decisions accordingly, and deriving insights from these experiences. This inference is consistent with previous studies [8, 58], which note that patients with IBD can develop the competence to quickly judge whether foods are suitable for them through accumulated experience over time. The items in this factor include the implementation of their dietary decisions, the adjustment of patients’ decision goals and plans based on reflection, and dealing with the consequences of decisions. The psychometric properties of the HDMCS After a rigorous instrument development process with various methods, a 23-item HDMCS with four dimensions was developed and validated. Item analysis and EFA were conducted based on established criteria, leading to the removal of three items. The resulting factors were relatively consistent with the initial attributes. However, items originally assigned to goal-setting competence and flare management competence were merged into other factors. The adaptive decision-making framework suggested that goal setting could be regarded as a form of decision-making, with both the behavioral and reflective dimensions involving the setting and adjustment of goals [12]. For example, during decisional balance, personal goals shape perceived value, whereas in reflection, individuals consider potential improvements and adjust their goals accordingly. This theoretical perspective aligns with the finding that items under “goal-setting competence” were redistributed into decisional balance competence and post-decision reflection competence. In the scale development phase, items within the attribute of “flare management competence” emphasized patients’ competence to identify potential flare-up symptoms and foods that might trigger gastrointestinal symptoms. However, EFA results reassigned these items to the post-decision reflection competence dimension. After discussion, the research team agreed that flare management competence is developed through continuous reflection and exploration [59]. Finally, the model generated by the EFA results was accepted. Most CFA indicators in this study met the evaluation criteria, although a few did not. While six error covariances were adjusted based on modification indices and theoretical considerations, some of the goodness-of-fit indices in the adjusted model remained slightly below the reference thresholds. However, several researchers have cautioned that using the cut-off value of model fit as a golden rule may cause the false rejection of acceptable models and that model fit should not be an absolute [60]. Therefore, we considered the fit indices from the CFA to be acceptable, supporting the robustness of the four-factor model identified through EFA. It may be useful to re-verify the results by performing a larger population study. With regard to discriminant validity between factors, some values were slightly higher or close to the threshold (factor 2 and factor 3, factor 4 and factor 3). Given that the instrument was designed to assess a single overarching construct, it was difficult to assume complete independence among elements within the construct [61]. Taken together, we considered the discriminant validity indicated by the CFA results to be acceptable. Consistent with our initial hypothesis, the NL-SF12 showed a moderate positive correlation with the HDMCS, indicating good convergent validity of the scale. We confirmed the expected relationships between the HDMCS and related constructs, demonstrating the effectiveness of the HDMCS in assessing patients’ competence in acquiring, understanding, and utilizing nutrition- or dietary-related information and services to make appropriate dietary decisions. Furthermore, the scale’s reliability was tested with item-total correlation coefficients, Cronbach’s α coefficient and the test-retest reliability coefficients. The correlation coefficients between each item and the total scale exceeded 0.30 (P<0.01). Moreover, the total Cronbach’s α coefficient of the HDMCS was 0.941, indicating a high level of internal consistency, with each factor’s Cronbach’s α coefficient ranging from 0.818 to 0.912, suggesting that all items contributed effectively to the measurement of both the overall construct and its individual factors. However, a high value of Cronbach’s α coefficient may be a risk of overly strong inter-item correlations, leading to redundancy among the items [62]. Since all inter-item correlations were below 0.75, the items were deemed to have acceptable correlations, and no deletions were made at this stage. Moreover, with an ICC of 0.874 and 95% CI (0.787, 0.927) of the total scale, the findings indicated that the instrument maintained stability over time. Therefore, the HDMCS is a valid and reliable measurement instrument. The results of discriminant ability demonstrated variations in mean HDMCS scores for healthy dietary decision-making competence across various demographic and clinical factors. Regarding disease type, patients with UC had higher mean scores than those with CD. This finding was consistent with clinical observations, as CD patients experience gastrointestinal symptoms more frequently and tend to adopt stricter dietary restrictions to avoid relapse due to concerns about symptom exacerbation and disease severity [63, 64], which may limit their capacity for rational decision-making. In terms of socioeconomic status, patients from families with a per capita monthly income exceeding 5000 yuan had higher mean scores than those from families with a per capita income below 3,000 yuan, reflecting potential socioeconomic disparities that may influence dietary decision-making competence in IBD patients [16]. In addition, a preliminary finding of this study indicated that participants with positive ARFID status had lower HDMCS scores compared to those with negative status. These results suggest that the HDMCS can, to some extent, capture patients’ dietary decision-making competence and further predict their dietary behaviors. Clinicians may also consider enhancing IBD patients’ dietary decision-making competence as a strategy to promote healthier dietary behaviors. Practice implications The HDMCS was considered suitable for measuring the dietary decision-making competence among HCPs, supported by good to excellent content validity, construct validity, and reliability. By evaluating the level of dietary decision-making competence and the scores across different dimensions, HCPs can comprehensively evaluate the healthy dietary decision-making competence of patients with IBD and accordingly tailor interventions. For instance, patients with low scores in the diet management perception dimension may require targeted education on the importance, and characteristics of dietary management to correct misconceptions and facilitating gradual acceptance of dietary management as part of disease control. When patients score low in the resource utilization competence dimension, this indicates insufficient competence to obtain, understand, and apply health information; therefore, interventions should focus on strengthening basic health literacy, enhancing dietary management competence, and introducing reliable sources for accessing health information. Patients with low scores in the decisional balance competence dimension may benefit from decision-support strategies such as providing comparative analyses of the advantages and disadvantages of different dietary patterns, clarifying personal values [65]. Finally, when patients score low in the post-decision reflection competence dimension, it is important to guide them in reflecting on and summarizing prior dietary management experiences while also offering necessary positive feedback. In addition, the HDMCS could provide novel insights into the behavioral determinants of healthy eating in IBD. Future studies could apply the HDMCS to examine its associations with a range of clinically relevant outcomes in patients with IBD. For example, it may be used to explore whether higher levels of healthy dietary decision-making competence are related to healthier dietary behaviors, better dietary adherence, or improved diet-related quality of life. Beyond these patient-reported outcomes, the HDMCS may also serve as a valuable tool to investigate potential links with clinical indicators such as symptom management, disease activity, or nutritional status. Limitations Several limitations should be considered when interpreting these results. First, the relatively small sample size and recruitment from only two cities in China may limit the generalizability. Only 144 participants were included in the EFA and 210 in the CFA, which is near the lower end of recommended thresholds and may influence the stability of construct validity. In addition, the sample included a higher proportion of male and urban participants, broadly consistent with recent epidemiological data on IBD in China [66], but potentially limiting applicability to other demographic groups. Future studies should expand the sample size, recruit participants from multiple provinces in China, adopt stratified sampling, and include patients from both tertiary and community hospitals to improve representativeness and reliability. This will constitute a key focus of our future studies. Second, patients with comorbid diabetes or severe kidney failure were excluded to reduce confounding from disease-related dietary modifications. However, this excluded a substantial subgroup of IBD patients, limiting representativeness. Future studies are warranted to extend validation to patients with common comorbid conditions, which will be essential to enhance the scale’s clinical utility and generalizability. Third, validity testing in this study was insufficient, as it convergent, discriminant, and criterion validity of the HDMCS were not examined using multiple measures. Future studies could incorporate measures of dietary adherence or similar construct to evaluate convergent validity, as well as measures of unrelated constructs such as psychological resilience to assess discriminant validity. In addition, instruments for assessing food-related quality of life and dietary patterns in patients with IBD may help assess predictive validity. This will be a key focus in our future efforts to further validate the psychometric properties of the HDMCS. Fourth, although items were translated by bilingual researchers, the scale was developed and applied within a Chinese cultural context. Therefore, before international application, cross-cultural adaptation procedures remain necessary, including forward translation, synthesis, back-translation, and expert committee review. We are planning to collaborate with researchers from diverse cultural backgrounds to strengthen the international applicability of the HDMCS. Finally, as the HDMCS relies on self-reported data, recall bias and social desirability bias cannot be ruled out. Future studies may benefit from incorporating more objective measures or observations (e.g., dietary intake and nutritional status) to support and validate the findings. Conclusions The HDMCS, consisting of 23 items across four domains (dietary management perception, resource utilization competence, decisional balance competence, and post-decision reflection competence), may be a valid and reliable instrument for measuring dietary decision-making competence. The total score ranges from 23 to 115, with higher scores indicating greater competence in making healthy dietary decisions among patients with IBD in the Chinese context. We hope that the HDMCS will serve as a tool for evaluating the quality of dietary decisions in patients with IBD, indicate the effectiveness of interventions promoting healthy eating behaviors, and provide a foundation for developing such interventions. Supplementary Information Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Acknowledgements The authors would like to express their appreciation to the physicians, nurses, and staff at the participating hospitals for their support in facilitating this study. We also extend our heartfelt gratitude to the patients for their active participation and commitment to completing the surveys, which was essential for validating this innovative measure of dietary decision-making competence in IBD. Abbreviations Crohn’s disease confirmatory factor analysis Consensus-based Standards for the Selection of Health Measurement Instruments critical ratio exploratory factors analysis European Society for Clinical Nutrition and Metabolism healthcare professionals healthy dietary decision-making competence scale inflammatory bowel disease the item- level content validity index short-form nutrition literacy scale the scale-level content validity index ulcerative colitis Author contributions Yiting Li: Writing – original draft, Methodology, Investigation, Conceptualization, Data curation, Formal analysis, Writing – review & editing. Yongqi Liang: Writing review & editing, Conceptualization, Data curation, Investigation, Methodology. Murong Wang: Methodology, Investigation, Formal analysis. Tingting Yin: Validation, Data curation, Investigation, Conceptualization. Tianxu Chen: Data curation, Investigation. Ling Ji: Data curation, Investigation. Ping Zhang: Data curation, Investigation. Qiong Li: Writing – review & editing, Methodology, Formal analysis. Wenjing Tu: Writing review & editing, Conceptualization, Methodology, Supervision, Project administration. Guihua Xu: Writing – review & editing, Supervision, Conceptualization, Formal analysis, Methodology, Supervision. Funding This research was funded by the Young Scientists Fund of the National Natural Science Foundation of China (No. 72204124), Postgraduate Research & Practice Innovation Program of Jiangsu Province (SJCX24_0818) and the Qing Lan Project of Jiangsu Province. Data availability The datasets generated and/or analyzed during the current study are not publicly available due to ethical restrictions, but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was approved by the Medical Ethics Committee of the Nanjing Hospital of Chinese Medicine, affiliated with Nanjing University of Chinese Medicine (approval number KY2022029). The study was performed according to the ethical guidelines of the Declaration of Helsinki. Informed consent was obtained from all study participants. Consent for publication Patients signed informed consent regarding publishing their data. Informed consent Informed consent was obtained from all individual participants included in the study. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yiting Li and Yongqi Liang contributed equally to this work. Contributor Information Wenjing Tu, Email: 837017@njucm.edu.cn. Guihua Xu, Email: 7115@njucm.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Data Availability Statement The datasets generated and/or analyzed during the current study are not publicly available due to ethical restrictions, but are available from the corresponding author on reasonable request.