Translation and psychometric evaluation of the Chinese version of the King’s Stool Chart in patients receiving enteral nutrition Authors: Guandong Wang, Hao Dou, Haibo Wang, Jingjing Hu, Liang Wang, Wenjuan Li, Xueying Wang, Hongmei Zhang Published: 2026 ABSTRACT AbstractObjectivesThis study aims to translate, culturally adapt, and validate the Chinese version of the King’s Stool Chart among patients receiving enteral nutrition for use in nursing practice.DesignThis is a descriptive, cross-sectional study.SettingThis study was conducted in the intensive care unit (ICU) of a tertiary hospital in Henan Province, China.ParticipantsA total of 144 patients receiving enteral nutrition were included.MethodsThis study was conducted in two phases. Phase I involved the translation and cultural adaptation of the King’s Stool Chart using established methodologies, including forward translation, synthesis, back-translation and expert review. Phase II comprised psychometric evaluation with 144 patients receiving enteral nutrition. The Chinese version of the King’s Stool Chart was used to assess stool frequency, consistency, weight and the daily total score. Validity was tested through content, construct and concurrent validity, while inter-rater reliability was assessed using the kappa coefficient.ResultsThe Chinese version of the King’s Stool Chart demonstrated excellent content validity, with item-level indices ranging from 0.95 to 1.00. Construct validity was supported by the ability of the Chinese version of the King’s Stool Chart to differentiate between clinical subgroups with varying stool characteristics. Sensitivity rates for stool weight categorisation were above 89%, and substantial inter-rater reliability (kappa=0.735) was observed. The daily total score was effective in identifying patients at risk for diarrhoea, with significant differences observed among clinical subgroups. Diarrhoea classification, using a threshold of ≥15 points, showed strong construct validity.ConclusionsWithin the scope of this single-centre sample, the Chinese version of the King’s Stool Chart demonstrates acceptable validity and substantial to excellent inter-rater reliability for assessing stool frequency, consistency, weight and diarrhoea classification in enterally fed ICU patients in China. These psychometric properties provide preliminary support for its use in routine nursing practice for gastrointestinal function monitoring in ICU settings. Further multicentre and large-sample studies are required to verify its external validity and generalisability to broader Chinese-speaking populations and non-ICU clinical settings. STRENGTHS AND LIMITATIONS OF THIS STUDY Employed a standardised cross-cultural adaptation process (forward-back translation+expert review) for the King’s Stool Chart to ensure cultural and linguistic appropriateness. Employed a standardised cross-cultural adaptation process (forward-back translation+expert review) for the King’s Stool Chart to ensure cultural and linguistic appropriateness. Evaluated multiple psychometric properties (content, construct, concurrent validity, inter-rater reliability) for comprehensive tool validation. Evaluated multiple psychometric properties (content, construct, concurrent validity, inter-rater reliability) for comprehensive tool validation. Adopted a standardised stool weighing procedure to minimise measurement bias in concurrent validity assessment. Adopted a standardised stool weighing procedure to minimise measurement bias in concurrent validity assessment. Used convenience sampling for participant recruitment, which may limit the generalisability of study findings. Used convenience sampling for participant recruitment, which may limit the generalisability of study findings. Conducted as a single-centre study, with no validation across different regions of China to account for dietary or cultural variations. Conducted as a single-centre study, with no validation across different regions of China to account for dietary or cultural variations. Introduction Enteral nutrition (EN) is an essential component of nutritional support for patients unable to meet their nutritional needs orally, particularly in critically ill and hospitalised populations, and international guidelines recommend initiating EN within 24–48 hours of critical illness when clinically feasible.1 However, global audits reveal a gap between guidelines and practice; approximately half of ICU patients receive EN within the first 24 hours of admission, with this figure rising to around 75% by day 5 of ICU care.2 Outside the critical care setting, many malnourished inpatients also require EN, although the utilisation rates vary significantly. For example, a survey from the USA and Europe reported that only approximately 5% of malnourished hospitalised patients received tube feeding.3 While EN remains a cornerstone of therapeutic nutrition, its implementation often results in gastrointestinal complications, with diarrhoea being the most prevalent adverse outcome.4 The incidence of diarrhoea among patients receiving EN varies widely due to inconsistent definitions, ranging from 2% to 70% in different studies.4 In critical care settings, approximately 15%–38% of EN-fed patients develop diarrhoea,5 while similar high incidence rates have been observed in general wards (eg, a 22% incidence in a tertiary hospital audit of enterally fed inpatients).4 6 This complication can have significant clinical consequences, as prolonged diarrhoea is associated with fluid and electrolyte imbalances, skin breakdown, nutrient malabsorption and worsened patient outcomes. For example, critically ill patients who develop diarrhoea experience significantly longer ICU stays and nearly double the mortality rate compared with those without diarrhoea.7 As a result, careful monitoring of bowel movements is a key aspect of care for patients on EN, with clinical practice standards highlighting the need for routine assessment of gastrointestinal tolerance, including stool output and consistency, to enable early identification of feeding intolerance and timely intervention.4 8 Despite the clinical significance of diarrhoea, the definition and documentation of this condition in patients receiving EN have lacked standardisation. To address this gap, the King’s Stool Chart (KSC), as shown in online supplemental file 1, was developed at King’s College London as a standardised tool for objectively characterising faecal output in patients receiving EN.9 The KSC enables the systematic documentation of stool consistency, approximate stool mass/volume and frequency over a 24-hour period, using both descriptive categories and pictorial illustrations to enhance accuracy.10 By assigning numerical scores to each stool based on these attributes, the KSC generates a daily cumulative score; a 24-hour stool score ≥15 points typically indicates diarrhoea according to this system.4 10 Initial evaluations of the KSC in enterally fed patient populations demonstrated strong measurement reliability and validity.9 11 Furthermore, the utility of the chart has been confirmed across different language versions, such as the Turkish adaptation (KSC-Tr), which showed robust reliability and validity when applied to ICU patients receiving EN.9 However, a validated Chinese-language version of the KSC has not yet been developed, presenting a gap in supporting clinical practice in Chinese-speaking regions. Given the widespread use of EN and the pressing need for standardised stool monitoring in these settings, there is a clear demand to translate and culturally adapt the KSC for use in Chinese-speaking populations, while also evaluating its psychometric properties, including validity and reliability. Therefore, the objective of this study was to translate and culturally adapt the KSC into Chinese and assess its psychometric performance in patients receiving EN. Methods Study design This descriptive, cross-sectional study was conducted in two phases: (1) translation and cultural adaptation of the KSC into Chinese following the methods of Sousa et al12 and (2) psychometric evaluation of the translated version in a clinical setting to confirm the validity and reliability. Informed consent was obtained from all patients or their legal guardians. Phase I: cultural and linguistic adaptation of the chart Forward translation Initially, two native Chinese speakers independently translated the English version of KSC (source language chart, SL chart) into Chinese. The translators were selected based on the following criteria: had at least a master’s degree, proficiency in both English and Chinese language and possessed comprehensive understanding of the local Chinese language and culture. Among them, the first translator possessed expertise in questionnaire design and medical terminology, and the second translator, in addition to bilingual proficiency, had extensive knowledge of colloquial expressions, Chinese idioms, customary phrases, medical jargon and commonly used emotional terminology. Through this process, two forward translation versions were generated, namely translation 1 (TL1) and translation 2 (TL2). Synthesis of the forward translation by comparing TL1 and TL2 A third bilingual expert translator was invited to conduct a comparative analysis of TL1 and TL2 while simultaneously verifying them against the original SL chart. The third translator was required to meet the following criteria: (1) bilingual and bicultural competence and (2) a minimum of 1 year’s full-time work experience in a hospital setting. Then, after group discussions among the research team, the first synthesised Chinese version of the KSC (KSC-Chinese Synthesis I) was developed. Blind back-translation In this step, KSC-Chinese Synthesis I was independently back-translated into English by two translators who had not participated in the previous steps. The first translator was a bilingual nursing expert with ten years of nursing faculty experience in Australia, and the second was a bilingual professional with 8 years of university-level English teaching experience. Each translator independently produced a back-translated version, resulting in back translation 1 (B-TL1) and back translation 2 (B-TL2). Review and finalisation by comparing B-TL1 and B-TL2 All bilingual translators and researchers discussed and compared B-TL1 and B-TL2 and cross-checked both against the original SL chart. Through iterative discussion and consensus-building, the final Chinese version of KSC (KSC-Cn), as shown in online supplemental file 2, was established. Content validity evaluation We invited 20 experts to evaluate the content validity of the KSC-Cn. The experts rated the importance of each item in KSC-Cn using a four-point scale ranging from 1 (not important) to 4 (very important). An item-level content validity index (I-CVI) greater than 0.78 and a scale-level CVI (S-CVI) greater than 0.90 were considered acceptable according to established standards.13 Phase II: psychometric evaluation Clinical setting and samples This study employed a convenience sampling method to recruit patients receiving EN in the intensive care unit of a tertiary hospital in Henan Province, China.14 The stool samples were collected during the first day of defecation (within 24 hours, from 8:00 am to 8:00 am the following day) from these patients, resulting in a total of 221 stool samples for analysis. The study inclusion criteria were: (1) patients receiving EN and (2) did not use devices for stool collection, such as faecal drainage systems. Exclusion criteria included: (1) patients who died, were discharged, transferred to other clinical units or had EN discontinued during the data collection period and (2) patients with severe complications (eg, gastrointestinal bleeding) that prevented stool sample collection and assessment. Finally, a total of 144 patients receiving EN were included. The total sample of 144 patients exceeds the recommended threshold of ≥100 patients considered adequate for hypothesis-testing construct validity in measurement-property studies.15 This sample size is also larger than those used in the original development9 11 and Turkish validation of the KSC,10 supporting its adequacy for comprehensive psychometric evaluation. Data collection A colour-printed paper version of the KSC-Cn was placed beside each patient’s bed, and the responsible bedside nurses were given detailed instructions regarding its use and the required data collection procedures. The nurses were instructed to mark the corresponding letter code on the KSC-Cn that best matched the stool sample after each defecation event. Defecation data for the first 24 hours after enrolment were recorded for all 144 patients receiving EN. Each stool sample was weighed using a high-precision electronic scale, following a standardised procedure: for conscious patients, stool was collected in a bedpan and weighed with the weight of the empty bedpan subtracted; for unconscious or sedated patients, stool was collected on nursing pads, and the weight of the empty pad was subtracted. All weighing was completed before any clinical handling to ensure the integrity of the samples. Validity testing Content validity was evaluated as described above. The validity of faecal frequency and faecal consistency characterisation was assessed by comparing the scores between different patient groups that were expected to show differences in stool characteristics. For faecal weight, validity was assessed by comparing the weight category assigned by the KSC-Cn with the actual measured weight of the stool samples. Sensitivity was calculated as the proportion of samples that were correctly categorised according to their actual weight. The daily total score was also validated by comparing differences between patient groups, and the threshold for diarrhoea classification (defined as a daily total score ≥15) was established through the same comparison approach. Specifically, the following patient groups were analysed: patients who received antibiotic prescriptions versus those who did not, patients who received potassium chloride prescriptions versus those who did not, patients with serum albumin levels ≤35 g/L versus those with serum albumin levels>35 g/L and patients who received prokinetic prescriptions versus those who did not. These comparisons were based on the expectation that patients with different clinical conditions would exhibit different frequencies of diarrhoea. Reliability testing To assess inter-rater reliability, 40 stool samples from the patients were independently evaluated by two intensive care unit nurses, each recording the results on separate forms. Given the clinical constraints of on-site sample generation, immediate assessment and subsequent destruction, a quota-controlled convenience sampling method was adopted for sample selection, with a judgement panel (core researchers + 2 certified wound, ostomy and continence experts) established to standardise KSC-Cn classification of samples. A predefined quota of two to four samples was set for each of the 12 KSC-Cn categories to ensure category representativeness. For each sample, the panel first determined its exact KSC-Cn category and checked the quota completion; only unfulfilled quota samples were included. The two nurses then conducted double-blind independent assessments of included samples. Sample collection was terminated only when the total 40 samples were reached, and all 12 categories met the two to four sample quota. Prior to the assessment, the two nurses received a 2-hour standardised training on the KSC-Cn, which included: (1) the theoretical design and assessment principles of the KSC; (2) detailed criteria for identifying the 12 stool types; (3) standardised procedures for stool consistency and weight classification; and (4) a proficiency test with 10 simulated stool samples (kappa ≥0.80 required for formal assessment). The two raters conducted independent assessments with strict blinding: all stool samples were labelled with anonymous numerical codes, and the raters did not communicate their assessment results with each other and were unaware of the sample’s clinical information and the other rater’s judgments, to avoid assessment bias. The reliability of faecal consistency was tested by examining the agreement between the two independent raters using the chart to assess the consistency of the same sample. Similarly, the reliability of faecal weight was determined by evaluating the agreement between the two raters using the chart to classify the weight category of the same faecal sample. Statistical analysis Data analysis was performed using IBM SPSS Statistics for Windows V.25.0. Armonk, NY: IBM Corp. Descriptive statistics were used to analyse the demographic characteristics of the patients and the distribution of stool samples according to the KSC-Cn. Content validity was evaluated using the I-CVI and the S-CVI. The I-CVI was calculated as the proportion of experts rating an item as 3 or 4 out of 4, while the S-CVI was the average of all I-CVIs. Differences in faecal frequency and total scores between different groups were analysed using the Mann-Whitney test, while differences in stool consistency and diarrhoea incidence were analysed using the χ2 test. The mean faecal weight across different weight categories was compared using analysis of variance. Inter-rater reliability for faecal consistency and weight was evaluated using the kappa coefficient. The sample size calculation was based on ensuring adequate reliability, which is a prerequisite for instrument validity. To detect an agreement rate of 80% with a statistical power of 80% at a 5% significance level, a minimum of 40 faecal samples assessed independently by two nurses was required.10 Results Participant characteristics To characterise the study population, demographic and clinical data of all enrolled patients were collected. A total of 144 patients receiving EN were included, with a mean age of 67.73±13.56 years; 52.78% of the participants were male. The median number of stool samples per patient was 1 (range: 1–4), with 87 patients providing 1 sample, 41 patients providing 2 samples, 12 patients providing 3 samples and 4 patients providing 4 samples. Most patients received whole-protein EN formulas (68.06%), and most were fed through nasogastric tubes (70.83%) using continuous feeding modes (75.69%). The primary diagnoses were respiratory failure (61.11%) and acute chronic obstructive pulmonary disease (22.22%). The detailed demographic information is presented in table 1. The stool samples were collected during the first day of defecation for each patient, resulting in a total of 221 faecal specimens (table 2). These samples provided the basis for subsequent validation analyses. Table 1Demographic and clinical characteristics of patients receiving enteral nutrition (n=144)DemographicsGroupsNumber of participants (N, %)SexFemale68 (47.22)Male76 (52.78)Age, mean (SD)67.73 (13.56)Enteral nutrition formulaWhole protein98 (68.06)Short peptide34 (23.61)amino-acid-based formula12 (8.33)Feeding tubeNasogastric tube102 (70.83)Nasointestinal tube42 (29.17)Enteral nutrition modeContinuous mode109 (75.69)Intermittent mode35 (24.31)DiagnosisRespiratory failure88 (61.11)ACOPD32 (22.22)Stroke10 (6.94)Renal failure8 (5.56)Others6 (4.17)ACOPD, acute chronic obstructive pulmonary disease. Demographic and clinical characteristics of patients receiving enteral nutrition (n=144) ACOPD, acute chronic obstructive pulmonary disease. Table 2Distribution of stool consistency and weight categories assessed by the KSC-Cn (n=221 samples)Consistency (N, %)Weight (N, %)<100 g100~200 g>200 gTotalHard and formed4 (1.8)10 (4.5)10 (4.5)24 (10.9)Soft and formed3 (1.4)8 (3.6)12 (5.4)23 (10.4)Loose and unformed7 (3.2)18 (8.1)33 (14.9)58 (26.2)Liquid35 (15.8)45 (20.4)36 (16.3)116 (52.5)Total49 (22.2)81 (36.7)91 (41.2)221(100) Distribution of stool consistency and weight categories assessed by the KSC-Cn (n=221 samples) Content validity This was assessed to determine whether the KSC-Cn items were representative and appropriate for evaluating stool characteristics in patients receiving EN. A panel of 20 experts, comprising 8 nursing educators, 10 clinical nurses and 2 intensive care physicians, participated in a Delphi survey. Our results showed that the S-CVI of the chart was 0.95, and the I-CVI for individual items ranged from 0.95 to 1.00, all exceeding the acceptable threshold of 0.78 (table 3). Overall, these results indicate excellent content validity of the KSC-Cn. Table 3Content validity indices for items and overall scale of the KSC-Cn based on expert evaluationVariablesGroupsVery important (N, %)Important (N, %)Slightly important (N, %)Not important (N, %)I-CVIS-CVIStool frequency20 (100)0001.000.95Stool consistencyHard and formed18 (90)1 (5)1 (5)00.95Soft and formed17 (85)2 (10)1 (5)00.95Loose and unformed19 (95)1 (5)000.95Liquid18 (90)1 (5)1 (5)00.95Stool weight<100 g16 (80)3 (15)1 (5)00.95100~200 g18 (90)1 (5)1 (5)00.95>200 g19 (95)01 (5)00.95I-CVI, item-content validity index; S-CVI, scale-content validity index. Content validity indices for items and overall scale of the KSC-Cn based on expert evaluation I-CVI, item-content validity index; S-CVI, scale-content validity index. Construct validity for stool frequency The construct validity of stool frequency assessment was examined to determine whether the KSC-Cn could differentiate between patients expected to have different stooling patterns. The median daily stool frequency across all patients was 1.00 (IQR: 1.00–2.00). Group comparisons showed that patients who were prescribed antibiotics, potassium chloride, prokinetics or had serum albumin levels ≤35 g/L had significantly higher daily stool frequencies compared with their respective control groups (all p <0.001, table 4). Collectively, these findings support the ability of the KSC-Cn to discriminate stool frequency differences related to clinical risk factors. Table 4Construct validity of the KSC-Cn for stool frequency, consistency, daily total score and diarrhoea classification across clinical subgroupsVariablesGroupsAntibiotic prescriptionPotassium chloride prescriptionSerum albuminProkinetics prescriptionNoneAt least oneYesNo≤35 g/L>35 g/LYesNoFaecal frequencyPatients (n, %)73 (50.69)71 (49.31)59 (40.97)85 (59.03)58 (40.28)86 (59.72)65 (45.14)79 (54.86)Frequency, M (Q1, Q3)1.00 (1.00, 1.00)2.00 (1.00, 2.00)2.00 (1.50, 2.50)1.00 (1.00, 1.00)2.00 (2.00, 2.75)1.00 (1.00, 1.00)2.00 (1.00, 2.00)1.00 (1.00, 1.00)Z (P value)−7.80 (<0.001)−7.31 (<0.001)−7.46 (<0.001)−6.73 (<0.001)Faecal consistency, n (%)Total faecal samples79 (35.75)142 (64.25)122 (55.20)99 (44.80)120 (54.30)101 (45.70)129 (58.37)92 (41.63)Hard and formed18 (75.00)6 (25.00)2 (8.33)22 (91.67)3 (12.50)21 (87.50)3 (12.50)21 (87.50)Soft and formed15 (65.22)8 (34.78)4 (17.39)19 (82.61)3 (13.04)20 (86.96)7 (30.43)16 (69.57)Loose and unformed26 (44.83)32 (55.17)28 (48.28)30 (51.72)25 (43.10)33 (56.90)33 (56.90)25 (43.10)Liquid20 (17.24)96 (82.76)88 (75.86)28 (24.14)89 (76.72)27 (23.28)86 (74.14)30 (25.86)χ² (P value)44.17 (<0.001)55.76 (<0.001)59.11 (<0.001)40.09 (<0.001)Daily scorePatients (n, %)73 (50.69)71 (49.31)59 (40.97)85 (59.03)58 (40.28)86 (59.72)65 (45.14)79 (54.86)Score, M (Q1, Q3)6.00 (3.00, 8.00)18.00 (10.00, 24.00)20.00 (12.00, 24.00)6.00 (3.00, 8.00)20.00 (14.00, 24.00)6.00 (3.00, 9.50)18.00 (10.00, 24.00)6.00 (3.00, 9.00)Z (P value)−7.01 (<0.001)−7.82 (<0.001)−7.62 (<0.001)−6.87 (<0.001)Diarrhoea, n (%)Total patients73 (50.69)71 (49.31)59 (40.97)85 (59.03)58 (40.28)86 (59.72)65 (45.14)79 (54.86)Yes2 (4.08)47 (95.92)40 (81.63)9 (18.37)42 (85.71)7 (14.29)41 (83.67)8 (16.33)No71 (74.74)24 (25.26)19 (20.00)76 (80.00)16 (16.84)79 (83.16)24 (25.26)71 (74.74)χ² (P value)64.56 (<0.001)50.77 (<0.001)63.74 (<0.001)44.54 (<0.001)Q₃, third quartile; Q₁, first quartile; Z, Mann-Whitney test. Construct validity of the KSC-Cn for stool frequency, consistency, daily total score and diarrhoea classification across clinical subgroups Q₃, third quartile; Q₁, first quartile; Z, Mann-Whitney test. Construct validity and inter-rater reliability for stool consistency Assessment of stool consistency validity and reliability was performed to confirm the KSC-Cn’s ability in accurately capturing stool form differences. Group analyses revealed significant differences in stool consistency among patients receiving antibiotics, potassium chloride, prokinetics or having low serum albumin compared with controls (all p <0.001, table 4). In addition, two independent intensive care nurses assessed the consistency of 40 stool samples using the KSC-Cn, yielding a kappa coefficient of 0.916, indicating excellent inter-rater reliability. Taken together, these results demonstrate that the KSC-Cn provides a reliable and valid assessment of stool consistency. Concurrent validity and inter-rater reliability for stool weight Concurrent validity for stool weight characterisation was evaluated to determine whether KSC-Cn-assigned weight categories corresponded with actual measured faecal weights. The mean faecal weight was 189.81±100.76 g, and statistically significant differences were observed among the mean weights of the <100 g, 100–200 g and >200 g categories (p<0.001, table 5). The corresponding sensitivities for correct weight assignment were 89.80%, 95.06% and 94.51% for the <100 g, 100–200 g and >200 g groups, respectively. Furthermore, the inter-rater reliability for stool weight categorisation based on independent nurse assessments yielded a kappa coefficient of 0.735, reflecting substantial agreement. These findings support both the concurrent validity and reliability of the KSC-Cn in assessing stool weight. Table 5Concurrent validity of stool weight categories and sensitivity of classification using the KSC-CnVariablesWeight categoryF (P value)<100 g100~200 g>200 gFaecal weight, M (95% CI)75.39 (68.69, 82.08)149.10 (141.95, 156.25)287.65 (272.49, 302.81)306.071(<0.001)Sensitivity* (%)89.80%95.06%94.51%*Percentage of samples from each weight category that were correctly assigned.F, Analysis of variance. Concurrent validity of stool weight categories and sensitivity of classification using the KSC-Cn Percentage of samples from each weight category that were correctly assigned. F, Analysis of variance. Construct validity for daily score The validity of the daily total score was assessed to determine whether it appropriately reflected the severity of stool changes associated with clinical conditions. The results indicated that the median daily total score for all patients was 8.00 (IQR: 4.00–18.50), and patients who received antibiotics, potassium chloride, prokinetics or had serum albumin ≤35 g/L had significantly higher daily scores compared with controls (all p <0.001, table 4). These results confirm that the daily total score could be used to effectively differentiate patients with higher risks of abnormal stool patterns. Construct validity for diarrhoea classification Construct validity for diarrhoea classification was evaluated to assess whether the established threshold (daily total score ≥15) could accurately identify patients with diarrhoea. Among the 144 patients examined, 49 (34.03%) were classified as having diarrhoea. Group comparisons showed that patients receiving antibiotics, potassium chloride, prokinetics or having serum albumin ≤35 g/L were significantly more likely to develop diarrhoea (all p <0.001, table 4). Collectively, these findings support the appropriateness of the KSC-Cn diarrhoea classification threshold and its clinical applicability. Discussion This study aimed to translate, culturally adapt and psychometrically validate the KSC-Cn for patients receiving EN. The results demonstrated that KSC-Cn possesses excellent content validity, good construct and concurrent validity, and substantial to excellent inter-rater reliability for assessing stool frequency, consistency, weight, daily scores and diarrhoea classification, indicating that our proposed KSC-Cn could be a reliable and valid instrument for evaluating faecal characteristics and diagnosing diarrhoea in Chinese patients receiving EN. The first major finding of this study was that the KSC-Cn exhibited excellent content validity, with an S-CVI of 0.95 and I-CVIs ranging from 0.95 to 1.00 for all items. These results are consistent with the original English version and the Turkish adaptation studies,911 suggesting that the translation and cultural adaptation processes successfully preserved the original conceptual framework of the KSC. The expert panel’s unanimous high ratings underscore the relevance and clarity of the KSC-Cn items for clinical evaluation in the Chinese context. Regarding the construct validity of stool frequency assessment, the KSC-Cn effectively differentiated between clinical groups based on antibiotic use, potassium chloride prescription, serum albumin levels and prokinetic use. Patients in these risk groups exhibited significantly higher stool frequencies, supporting the instrument’s ability to reflect clinically expected patterns of bowel movements. These results align with previous studies indicating that antibiotic therapy, electrolyte supplementation, hypoalbuminaemia and prokinetic administration are associated with altered gastrointestinal motility and increased stool output.11 16 17 The evaluation of stool consistency using the KSC has demonstrated robust construct validity and excellent inter-rater reliability, with a kappa coefficient of 0.916 indicating almost perfect agreement, which underscores the reliability of the KSC-Cn in detecting variations in stool form, which is essential for distinguishing between normal and pathological defecation patterns.18 Consistency patterns have been shown to differ significantly among clinical groups, with softer or liquid stools being more prevalent in patients with clinical risk factors such as hypoalbuminaemia, antibiotic use or Clostridium difficile infections.19 This aligns with prior research, highlighting the clinical relevance of stool consistency assessments in monitoring gastrointestinal health and tailoring interventions for patients receiving EN.20 The analysis of stool weight in this study further strengthens the clinical value of the KSC-Cn, confirming its concurrent validity through significant differences in mean measured faecal weights across the KSC-Cn categories. Sensitivity rates exceeding 89% for all categories demonstrate the accuracy of the tool in classifying stool weight without the need for direct measurement, facilitating efficient monitoring in clinical settings, and this capability is particularly crucial in ICU settings, where accurate faecal output monitoring is vital for guiding fluid and electrolyte management.21 The ability to estimate stool weight categories effectively supports timely interventions and enhances patient care,22 especially in critically ill populations who are at increased risk for gastrointestinal disturbances. The construct validity of the KSC-Cn was further supported by the analysis of daily total scores, which revealed significantly higher scores in patients with clinical risk factors, which suggests that the cumulative scoring system effectively captures the severity of stool changes, providing valuable insights into the patient’s gastrointestinal status. The median daily score of 8.00 across the cohort reflects typical bowel output in patients receiving EN, while higher scores in the clinical risk groups highlight the tool’s sensitivity to disease-related changes in bowel function. We note that the inter-rater reliability for stool weight (kappa=0.735) was lower than that for stool consistency (kappa=0.916), and this difference is attributable to two main methodological and clinical factors. First, stool weight assessment relies on subjective visual estimation of faecal volume/infiltration range, whereas stool consistency is assessed via direct visual matching to the KSC-Cn’s pictorial illustrations, which reduces subjective bias. Second, the categorical boundaries for stool weight are less visually distinct for loose/liquid stool samples—the most common type in this study—which increases assessment variability. Future studies could address these issues by incorporating standardised visual reference tools for faecal weight estimation. Lastly, the KSC-Cn demonstrated strong construct validity through its classification of diarrhoea based on a daily total score threshold of ≥15, whereby patients with higher clinical risk could be more likely to be classified as having diarrhoea, which aligns with the known pathophysiology of conditions such as antibiotic-associated diarrhoea, hypokalaemia and hypoalbuminaemia.23 24 This finding also underscores the clinical applicability of the KSC-Cn as a reliable tool for early identification and management of diarrhoea, a common and significant complication in EN therapy. By enabling healthcare providers to quickly detect and address this issue, the KSC-Cn can contribute to better patient outcomes and more effective management of EN in both ICU and general ward settings. To further clarify the clinical positioning of the KSC-Cn, we compared it with two mainstream tools: the Hart & Dobb Score25 and the Bristol Stool Form Scale.26 Compared with the Hart & Dobb Score, the KSC-Cn offers intuitive pictorial illustrations for all classification categories and a more refined stool consistency classification (four levels: hard/soft/loose/liquid vs three broad types in Hart & Dobb), enabling sensitive detection of subtle gastrointestinal changes, while both tools support quantitative diarrhoea diagnosis via cumulative scoring. In contrast to the globally prevalent Bristol Stool Form Scale, the KSC-Cn uniquely integrates stool weight assessment alongside consistency evaluation, whereas the Bristol Stool Form Scale excels in widespread validation across diverse populations and comprehensive psychometric testing—an area where the KSC-Cn requires further multicentre and multipopulation verification to enhance generalisability. The KSC-Cn has important clinical implications for the nursing care of enterally fed ICU patients and can be easily integrated into routine clinical practice. First, the KSC-Cn can be used as a standardised daily assessment item for enterally fed patients, with results recorded alongside vital signs and nutrition tolerance to form a comprehensive gastrointestinal function monitoring system. Second, the ≥15-point diarrhoea threshold provides an objective trigger for timely nursing intervention: nurses can initiate targeted interventions for patients with a score ≥15 and monitor the change in KSC-Cn scores to evaluate intervention effectiveness. Third, the standardised assessment results of the KSC-Cn improve inter-professional communication between nurses, physicians and dietitians, providing objective data for clinical decision-making. Finally, from a feasibility perspective, bedside nurses informally reported that completing the KSC-Cn entry for each defecation event required approximately 1–2 min, suggesting that the tool can be integrated into routine ICU documentation without excessive burden. There are some limitations to this study. First, the study employed convenience sampling to recruit participants from a single tertiary ICU in Henan Province, which may limit the external validity and generalisability of the findings to other Chinese-speaking populations, non-ICU settings or lower-level hospitals.13 Second, as a single-centre study, the KSC-Cn has not been validated across different regions of China, and dietary variations (eg, northern wheat-based vs southern rice-based diets) and dialectal differences in clinical terminology may affect the tool’s applicability; multicentre validation across eastern, southern, western and northern China is required to confirm its generalisability. Third, while inter-rater reliability for stool consistency was excellent, concurrent validity requires verification through laboratory quantification of faecal water content in samples categorised by the KSC-Cn. Future studies should employ gravimetric analysis (eg, lyophilisation) to establish objective moisture thresholds for each consistency type (hard/soft/loose/liquid), strengthening evidence for clinical utility. Finally, diagnostic accuracy metrics—including sensitivity, specificity, predictive values and the area under the ROC curve—have not been assessed for the ≥15 diarrhoea threshold. Prospective cohort studies tracking diarrhoea-related outcomes are needed to calculate these indices and further optimise the cut-off. Conclusion In summary, the KSC-Cn demonstrated excellent psychometric properties, including high content validity, robust construct and concurrent validity and substantial to excellent inter-rater reliability, suggesting that it could be used to effectively assess stool frequency, consistency, weight and diarrhoea classification in patients receiving EN. These findings support the use of the KSC-Cn as a practical and reliable tool in Chinese clinical settings to monitor bowel function and guide timely interventions for patients at risk of diarrhoea. However, future research is needed to explore its applicability in broader populations and in different healthcare environments. Supplementary material