Dietary intake patterns and nutritional adequacy in older adults with predialysis chronic kidney disease: a comparison by diabetes status Authors: Jeong Eun Kim, Jisoo Kim, Yu-Jin Kwon, Jung Eun Lee, Jung Joo Lee, Ji-Won Lee, Yoo Kyoung Park Published: 2026 ABSTRACT AbstractObjectiveNutritional management is essential in caring for patients with chronic kidney disease (CKD), older adults at higher risk of malnutrition and comorbidities. However, data on actual dietary intake patterns in older adults with predialysis CKD, especially by diabetes mellitus (DM) status, remain limited.MethodsThis cross-sectional study included 106 patients aged ≥65 years with CKD stage G3a or higher, divided into DM (n=67) and non-DM (n=39) groups. Dietary intake was assessed using a single 24-hour recall. Nutrient and food-group intakes were compared with recommended levels.ResultsIn both groups, energy intake was lower than recommended levels. More than half of the participants exceeded sodium limits, and approximately half consumed excess protein. Patients with DM had significantly higher protein intake and blood urea nitrogen (BUN) levels than those without DM. Most food groups, except protein foods, were consumed below recommended levels.ConclusionDietary patterns in older adults with predialysis CKD showed low energy intake, high sodium intake, and relatively high protein intake. Those with DM had higher protein intake and BUN levels, suggesting dietary differences by diabetes status. These findings underscore the need for age-sensitive, individualized nutritional management strategies that consider kidney function and DM status.Keywords: Diabetes mellitus, Dietary patterns, Energy intake, Malnutrition, Nutrients INTRODUCTION Chronic kidney disease (CKD) is a prevalent and increasing global health issue, affecting approximately one in ten adults worldwide [1]. It is associated with substantial morbidity and mortality, including higher risks of death, cardiovascular events, and hospitalization [2]. Nutrition plays a key role in CKD management and is an important component of nephroprotection, as it can affect disease progression, metabolic status, and clinical outcomes [3]. In patients with predialysis CKD, dietary intake—especially energy and nutrient intake—is an important determinant of nutritional status and is associated with clinical outcomes [3-5]. Dietary assessment is therefore essential before nutritional intervention to identify nutritional problems, guide individualized counseling, and support ongoing intake monitoring [3,4]. Older adults are a rapidly growing CKD subgroup. Aging-related changes, including reduced appetite, altered taste, lower energy intake, and increased risk of protein-energy wasting (PEW) and sarcopenia, can negatively affect nutrition [5,6]. Older adults with CKD often have comorbidities, such as hypertension or diabetes mellitus (DM), which further complicate nutritional management and increase the risk of malnutrition and functional decline [5,7]. Among these comorbidities, DM is a leading cause of CKD worldwide and a major driver of disease progression and related complications [1]. It is also the leading cause of kidney failure requiring dialysis or transplantation [8,9]. The co-occurrence of predialysis CKD and DM complicates dietary management, requiring a balance of glycemic control and kidney-specific nutrition recommendations while maintaining adequate energy and protein intake to prevent malnutrition [9]. Despite its clinical importance, data on actual dietary intake patterns in older adults with predialysis CKD, particularly by DM status, remain limited [4,5]. Only a few observational studies have directly compared energy and nutrient intake between diabetic and nondiabetic patients with predialysis CKD, and evidence in older adults is limited [10]. To address this gap, the study compared dietary intake patterns—including energy and nutrient intake—between older adults with predialysis CKD with and without DM and assessed intake adequacy against established nutrition recommendations. By identifying differences according to DM status, this study aims to inform individualized nutrition education and intervention strategies aimed at improving clinical outcomes among older adults with CKD. METHODS Ethics statement This study protocol was approved by the Institutional Review Board of Yongin Severance Hospital (IRB No. 9-2021-0075), and all participants provided written informed consent. The trial was registered at the Clinical Research Information Service (No. KCT0006503). Study design and participants This observational cross-sectional study was conducted at the Department of Nephrology at Yongin Severance Hospital from May to October 2021. Participants were aged ≥65 years with CKD stage G3a or higher. Eligibility criteria included an estimated glomerular filtration rate <60 mL/min/1.73 m2, calculated using the Chronic Kidney Disease Epidemiology Collaboration equation, and no dialysis treatment. The final sample included 106 participants, comprising 67 with DM and 39 without DM. Participant characteristics and nutritional status General participant characteristics, including sex, age, educational level, drinking status, smoking status, ability to prepare meals, and dietary management, were assessed using a structured questionnaire. Information on treatment duration and comorbidities was obtained from electronic medical records. Nutritional status was assessed using the seven-point Subjective Global Assessment (SGA) [11], which includes medical history (weight loss in the past 6 months, dietary intake changes, gastrointestinal symptoms lasting >2 weeks, functional capacity, and disease-related metabolic stress) and physical examination findings (muscle wasting, loss of subcutaneous fat, and edema) [12]. Each component was scored on a seven-point scale, with detailed scoring criteria provided in Table S1. Based on total SGA scores, participants were classified as well nourished (6–7), mildly to moderately malnourished (3–5), or severely malnourished (1–2). Anthropometric and biochemical measurements Anthropometric measures included height, weight, systolic blood pressure, diastolic blood pressure, triceps skinfold thickness (TSF), and mid-arm circumference (MAC). Body mass index (BMI) was calculated as weight (kg) divided by height (m²). Mid-arm muscle circumference (MAMC) was calculated from TSF and MAC using the following equation [13]: MAMC (cm)=MAC (cm)−[TSF (cm)×0.314] Blood samples were collected by a registered phlebotomist after at least 8 hours of overnight fasting. Biochemical parameters, including glucose, albumin, blood urea nitrogen (BUN), and creatinine, were analyzed in the hospital laboratory. Dietary intake assessment Dietary intake was assessed using a single 24-hour dietary recall of all foods and beverages consumed during the previous day. Trained researchers conducted face-to-face interviews. Participants reported foods, ingredients, and portion sizes, aided by food models and measuring tools to improve recall accuracy. Daily nutrient intake was analyzed using CAN-Pro 5.0 (Computer Aided Nutritional Analysis Program, The Korean Nutrition Society, 2015). Food-group intake was assessed using the food exchange list for DM and the Korean Dietary Reference Intakes [14]. Dietary adequacy was evaluated according to the International Society of Renal Nutrition and Metabolism (ISRNM) recommendations [15]. Recommended protein intake for nondialysis CKD patients was 0.6 to 0.8 g/kg/day. Recommended energy intake was estimated from each participant’s height and weight and compared with actual intake. Daily exchange units for food groups were also calculated (Table S2). Statistical analysis Statistical analyses were performed using IBM SPSS ver. 26.0 (IBM Corp.). Continuous variables are presented as mean±standard deviation, and categorical variables are frequencies and percentages. Differences between the DM and non-DM groups were assessed using independent t-tests for continuous variables and chi-square tests for categorical variables. A two-tailed P-value of <0.05 was considered statistically significant. RESULTS Group comparisons of participant characteristics, measurements, and nutritional status Table 1 presents the general characteristics and nutritional status of 106 participants by DM status. The mean age was 76.6±6.2 years, and 60 participants (56.6%) were male. Nearly half (49.1%) were classified as CKD stage 3b. Compared with the non-DM group, the DM group had a higher prevalence of hypertension (P=0.022) and a lower prevalence of dyslipidemia (P=0.023). Only 24.5% of participants reported following dietary management, with a higher proportion in the non-DM group than in the DM group (P=0.038). Based on the SGA, 12 participants (11.3%) were mildly to moderately malnourished, with no significant difference between groups. Table 1.General characteristics of participants according to DM comorbidityCharacteristicDM comorbidityTotal (n=106)DM (n=67)Non-DM (n=39)P-valueAge (yr)76.6±6.276.4±6.476.9±5.80.671Male sex60 (56.6)42 (62.7)18 (46.2)0.098Periods of medical treatment (mo)31.2±35.232.2±36.529.5±33.10.704CKD stage Stage 3a22 (20.8)15 (22.4)7 (17.9)0.304 Stage 3b52 (49.1)29 (43.3)23 (59.0) Stage 429 (27.4)20 (29.9)9 (23.1) Stage 53 (2.8)3 (4.5)0 (0.0)Comorbidity Hypertension96 (90.6)64 (95.5)32 (82.1)0.022* Anemia19 (17.9)11 (16.4)8 (20.5)0.596 Dyslipidemia12 (11.3)4 (6.0)8 (20.5)0.023* Proteinuria63 (59.4)42 (62.7)21 (53.8)0.371 Hyperkalemia19 (17.9)12 (17.9)7 (17.9)0.996Education Uneducated8 (7.5)3 (4.5)5 (12.8)0.245 Elementary school28 (26.4)15 (22.4)13 (33.3) Middle school14 (13.2)11 (16.4)3 (7.7) High school39 (36.8)27 (40.3)12 (30.8) College or higher17 (16.0)11 (16.4)6 (15.4)Smoking status Never85 (80.2)53 (79.1)32 (82.1)0.888 Current7 (6.6)5 (7.5)2 (5.1) Former14 (13.2)9 (13.4)5 (12.8)Current drinking status Yes16 (15.1)12 (17.9)4 (10.3)0.288 No90 (84.9)55 (82.1)35 (89.7)Diet management Yes26 (24.5)12 (17.9)14 (35.9)0.038* No80 (75.5)55 (82.1)25 (64.1)Subjective Global Assessment Mildly to moderately malnourished12 (11.3)9 (13.4)3 (7.7)0.368 Well nourished94 (88.7)58 (86.6)36 (92.3)Values are presented as mean±standard deviation or number (%) and compared using the chi-square test or independent t-test.DM, diabetes mellitus; CKD, chronic kidney disease.*P<0.05 for chi-square test. General characteristics of participants according to DM comorbidity Values are presented as mean±standard deviation or number (%) and compared using the chi-square test or independent t-test. DM, diabetes mellitus; CKD, chronic kidney disease. P<0.05 for chi-square test. Table 2 shows anthropometric and biochemical parameters by DM status. Anthropometric measurements did not differ significantly between groups. However, BUN (P=0.035) and phosphorus levels (P=0.036) were significantly higher in the DM group than in the non-DM group. Table 2.Anthropometric and biochemical measurements of participants according to DM comorbidityParameterDM comorbidityTotal (n=106)DM (n=67)Non-DM (n=39)P-valueAnthropometric parameter Height (cm)157.7±12.0158.9±8.9157.9±7.20.548 Weight (kg)65.0±14.165.4±10.461.9±9.90.087 BMI (kg/m2)25.5±3.425.9±3.324.8±3.40.109 SBP (mmHg)134.1±18.3135.6±18.3131.5±18.30.262 DBP (mmHg)65.0±12.364.6±12.365.7±12.50.655 TSF (mm)16.0±6.015.7±5.716.6±6.50.432 MAMC (cm)22.1±3.022.3±3.121.8±2.80.385Biochemical parameter Glucose (mg/dL)111.7±27.8114.8±31.3106.5±20.00.098 Hemoglobin (g/dL)11.8±1.811.9±2.011.8±1.60.913 Total protein (g/dL)6.9±0.56.9±0.56.8±0.50.632 Albumin (g/dL)4.3±0.34.3±0.34.3±0.30.505 eGFR (mL/min/1.73 m2)35.4±11.135.0±12.236.2±9.00.596 BUN (mg/dL)31.2±11.732.8±13.428.5±7.30.035* Creatinine (mg/dL)1.8±0.71.9±0.81.6±0.40.035* Cystatin C (mg/L)2.0±0.62.1±0.71.9±0.40.099 PCR (mg/g)579.3±796.5689.7±873.8394.3±614.10.052 Uric acid (mg/dL)6.6±1.66.8±1.76.3±1.50.131 Calcium (mg/dL)9.2±0.49.2±0.49.2±0.40.982 Phosphorus (mg/dL)3.7±0.63.8±0.63.5±0.60.036* Sodium (mmol/L)140.7±2.2140.6±2.5140.8±1.70.553 Potassium (mmol/L)5.0±0.75.0±0.74.9±0.60.232 Chloride (mmol/L)106.5±3.0106.5±3.3106.5±2.30.983 Cholesterol (mg/dL)138.6±29.9134.2±29.5146.0±29.40.050Values are presented as mean±standard deviation and compared using the independent t-test.DM, diabetes mellitus; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; TSF, triceps skinfold thickness; MAMC, mid-upper arm muscle circumference; eGFR, estimated glomerular filtration rate; BUN, blood urea nitrogen; PCR, protein creatinine ratio.*P<0.05 was considered statistically significant. Anthropometric and biochemical measurements of participants according to DM comorbidity Values are presented as mean±standard deviation and compared using the independent t-test. DM, diabetes mellitus; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; TSF, triceps skinfold thickness; MAMC, mid-upper arm muscle circumference; eGFR, estimated glomerular filtration rate; BUN, blood urea nitrogen; PCR, protein creatinine ratio. P<0.05 was considered statistically significant. Comparison of dietary intake with recommended levels Table 3 shows the mean dietary intake by DM status. Energy and protein intake per kilogram body weight were calculated using current or adjusted body weight when BMI adequacy was <95% or >115%. Although energy intake was similar between groups, the DM group had significantly higher protein intake (P=0.017), protein per kg of body weight (P=0.049), folate (P=0.006), and niacin (P=0.038) than the non-DM group. Table 3.Nutrient intake of participants according to DM comorbidityNutrient intakeDM comorbidityTotal (n=106)DM (n=67)Non-DM (n=39)P-valueEnergy (kcal)1,388.7±435.01,429.1±472.41,319.2±357.00.211Energy (kcal/kg)24.5±7.024.9±7.423.9±6.30.462Macronutrient Carbohydrate (g)208.2±65.4209.8±69.3205.5±58.80.742 Protein (g/kg)0.9±0.41.0±0.40.8±0.20.049* Protein (g)52.3±21.055.6±23.546.7±14.20.017* Animal (% total protein)40.4±20.540.5±21.540.3±18.90.958 Plant (% total protein)59.6±20.559.5±21.559.7±18.90.958 Fat (g)32.3±18.133.0±20.631.1±13.00.564 Animal (% total fat)45.6±26.644.1±26.248.1±27.40.460 Plant (% total fat)54.4±26.655.9±26.251.9±27.40.460 Saturated fat (g)10.9±14.79.8±10.112.8±20.40.304 Cholesterol (mg)209.2±191.4235.8±209.1163.5±147.90.060 Fiber (g)21.4±13.122.2±14.620.1±10.00.434 Carbohydrate:protein:fat (%)63:16:2163:16:2164:15:21Mineral and vitamin Sodium (mg)2,659.7±1,407.52,707.1±1,529.82,578.3±1,182.80.652 Potassium (mg)2,052.3±993.52,193.9±1,090.41,809.1±752.50.054 Phosphorus (mg)804.0±318.1843.6±339.0736.1±269.30.093 Calcium (mg)420.4±235.5433.8±232.4397.6±242.10.448 Iron (mg)17.6±23.018.8±25.015.6±19.00.498 Vitamin A (µg RAE)254.1±222.0274.4±243.7219.3±176.40.219 Vitamin E (mg)11.6±8.012.2±8.310.6±7.40.326 Folate (µg)372.8±204.7408.6±233.0311.2±123.80.006* Vitamin C (mg)62.3±65.571.3±74.947.0±41.60.065 Thiamin (mg)1.3±0.61.3±0.71.2±0.50.172 Riboflavin (mg)1.0±0.61.1±0.60.9±0.60.259 Niacin (mg)8.6±4.39.2±4.87.6±3.10.038*Values are presented as mean±standard deviation and compared using the independent t-test.DM, diabetes mellitus; RAE, retinol activity equivalents.*P<0.05 was considered statistically significant. Nutrient intake of participants according to DM comorbidity Values are presented as mean±standard deviation and compared using the independent t-test. DM, diabetes mellitus; RAE, retinol activity equivalents. P<0.05 was considered statistically significant. Table 4 compares energy and nutrient intake with ISRNM recommendations (Table S3). In both groups, mean energy intake was below the recommended range of 30–35 kcal/kg/day. Over half of the participants exceeded the recommended sodium intake of 1,840–2,300 mg/day. Protein intake above the recommended range (0.6–0.8 g/kg/day) was observed in 58.2% of the DM group and 48.7% of the non-DM group, with no significant difference between groups. Table 4.Energy and nutrient intake of patients with predialysis chronic kidney disease (stages 3–5) compared with the recommended levela)VariableDM comorbidityTotal (n=106)DM (n=67)Non-DM (n=39)P-valueBelowOnAboveBelowOnAboveBelowOnAboveEnergy90 (84.9)6 (5.7)10 (9.4)56 (83.6)3 (4.5)8 (11.9)34 (87.2)3 (7.7)2 (5.1)0.427Protein16 (15.1)32 (30.2)58 (54.7)10 (14.9)18 (26.9)39 (58.2)6 (15.4)14 (35.9)19 (48.7)0.584Phosphorus58 (54.7)23 (21.7)25 (23.6)34 (50.7)14 (20.9)19 (28.4)24 (61.5)9 (23.1)6 (15.4)0.311Sodium30 (28.3)15 (14.2)61 (57.5)18 (26.9)11 (16.4)38 (56.7)12 (30.8)4 (10.3)23 (59.0)0.665Values are presented as number (%) and compared using the chi-square test.DM, diabetes mellitus.a)2013 International Society of Renal Nutrition and Metabolism guideline. Energy and nutrient intake of patients with predialysis chronic kidney disease (stages 3–5) compared with the recommended levela) Values are presented as number (%) and compared using the chi-square test. DM, diabetes mellitus. 2013 International Society of Renal Nutrition and Metabolism guideline. Fig. 1 shows a radial graph comparing food-group intake with recommended levels (set at 100%). In both groups, intake was below recommended levels for all food groups except protein foods, with no significant differences between the DM and non-DM groups. Major food-group intake of participants compared with recommendation. DM, diabetes mellitus. DISCUSSION This study examined dietary intake patterns in older adults with predialysis CKD and compared them based on DM status. Overall, participants had lower energy intake than recommended, more than half exceeded the recommended sodium intake, and approximately half consumed protein above the recommended range for predialysis CKD. Participants with DM had higher protein intake than those without DM. These findings indicate poor alignment with dietary patterns and support the need for individualized nutritional management strategies in this population. A key finding was insufficient energy intake in many participants. Although ISRNM recommends 30–35 kcal/kg/day for patients with nondialysis CKD to prevent PEW [15], the mean energy intake in both groups was below this range. These findings align with those of previous studies reporting that patients with CKD often consume less energy than recommended [10,16,17]. Reduced energy intake in CKD has been attributed to anorexia, uremic symptoms, and dietary restrictions [18,19]. Inadequate energy intake may increase protein catabolism and the risk of PEW, which is associated with poor clinical outcomes [15,20]. This is especially relevant in older patients with CKD, as aging-related declines in appetite and function further limit dietary intake [5,6]. These findings highlight the importance of maintaining adequate energy intake in the nutritional management of older adults with CKD. Despite the high prevalence of inadequate energy intake, few participants were classified as malnourished according to the SGA. This discrepancy may reflect that SGA evaluates changes in dietary intake rather than adequacy based on recommended energy requirements. In addition, the use of a single 24-hour recall may not fully reflect habitual intake. These factors may explain the observed mismatch between reported energy intake and nutritional status. For patients with CKD, protein restriction to approximately 0.6–0.8 g/kg/day is commonly recommended to reduce intraglomerular pressure and mitigate hyperfiltration, thereby slowing disease progression [15,21]. However, adherence to a protein-restricted diet is often difficult in real-world home settings. Consistent with our findings, previous studies report that actual protein intake in patients with CKD frequently exceeds recommendations [21,22]. In this study, protein intake was similar to the recommended Korean Society of Nephrology (0.8 g/kg/day), and notably, approximately half of the participants met or exceeded this level. Given the older age of the cohort, these findings highlight a critical clinical dilemma: balancing the need for adequate protein intake to prevent sarcopenia against the risk of additional burden on compromised renal function. These results support that evidence-based, age-specific nutritional recommendations are warranted for the management of elderly patients with CKD. Notably, patients with DM consumed significantly more protein than those without DM, consistent with previous studies [17,23]. In nondialysis CKD, patients without diabetes tended to initiate protein restriction earlier, whereas those with DM were less likely to do so [17]. Although total energy intake did not significantly differ between groups, higher protein intake in the DM group may reflect differences in overall dietary intake patterns, including macronutrient distribution. In particular, carbohydrate restriction for glycemic control may lead to compensatory increases in protein intake [24-26]. Higher protein intake in the DM group was accompanied by higher BUN levels. As BUN reflects both dietary protein intake and kidney function, higher levels in this group may partly indicate greater protein consumption. This pattern may relate to DM-related dietary modifications, particularly carbohydrate restriction for glycemic control, which can lead to compensatory increases in protein intake. In addition, increased intake of protein-rich foods, especially from animal sources, may contribute to higher phosphorus intake and may partly explain the elevated serum phosphorus levels observed in the DM group. Previous studies similarly reported that higher protein intake increases circulating urea levels, indicating a relationship between protein intake and BUN levels [27]. High protein intake has also been associated with increased intraglomerular pressure and glomerular hyperfiltration, potentially contributing to progressive kidney damage over time [27]. These findings therefore highlight the importance of appropriate protein intake management in CKD [3]. However, optimal protein intake in older adults with CKD remains unclear. While older adults generally require approximately 1.0 to 1.2 g/kg/day to maintain muscle mass and prevent sarcopenia [28], this exceeds the lower levels recommended for CKD. Therefore, nutritional management in this population requires balancing kidney protection with the need to maintain adequate nutritional status. Another key finding was that more than half of the participants exceeded the recommended sodium intake. Excessive sodium intake contributes to fluid retention and hypertension and is strongly associated with increased cardiovascular risk and CKD progression [29]. Previous studies likewise report high sodium intake in CKD, while sodium restriction has been shown to reduce blood pressure and proteinuria [29]. These findings highlight the importance of sodium restriction in CKD management. Analysis of food-group intake showed that all groups except protein were consumed below recommended levels, suggesting that dietary imbalance in CKD may extend beyond individual nutrients to broader patterns of food selection. Previous studies have also reported food-group consumption imbalance among patients with CKD [17,30]. In older adults, food-based dietary guidance may be more practical and easier to implement than nutrient-based recommendations, and translating nutrient intake targets into meal-based patterns may improve dietary adherence. These findings support the use of food-based nutritional education in dietary interventions for patients with CKD. This study has several strengths. First, it focused on older adults with CKD, a rapidly growing population for whom nutritional management is particularly important. Second, dietary intake was assessed in terms of nutrient amounts and adherence to recommended intakes and food-group patterns. Several limitations should be acknowledged. First, the cross-sectional design precludes causal inferences between dietary intake patterns and clinical outcomes. Longitudinal studies are needed to determine whether observed dietary patterns contribute to disease progression or reflect adaptations to declining kidney function. Second, dietary intake was assessed using a single 24-hour dietary recall, which may not adequately capture habitual dietary patterns due to day-to-day variability. Although widely used, multiple recalls or food frequency questionnaires may provide a more reliable estimate of usual intake. Third, the relatively small sample and single-center design limit statistical power and generalizability to the broader populations of older adults with predialysis CKD. In addition, reliance on participant memory introduces recall bias, particularly in older adults with CKD, where cognitive decline may result in under- or inaccurate-reporting. These limitations may have led to misestimation of nutritional intake and attenuated associations with clinical parameters. Therefore, these findings should be interpreted with caution, and further studies using multiple recalls or food frequency questionnaires are warranted for a more robust estimation. In conclusion, older adults with predialysis CKD had insufficient energy, excessive sodium, and protein intake above recommended levels. Protein intake and BUN levels were higher in patients with DM than in those without. These findings underscore the need for individualized nutritional management strategies that consider both kidney function and diabetes status in older adults with CKD. Footnotes Authors’ contributions Conceptualization: JEK, YJK, JEL, JJL, JWL, YKP. Formal analysis: JEK. Investigation: JEK, JK, YJK, JEL, JJL, JWL. Methodology: YJK, JEL, JJL, JWL. Supervision: YKP. Writing–original draft: JEK. Writing–review & editing: all authors. All authors read and approved the final manuscript. Conflicts of interest None. Funding This study was supported by a grant from the Korea Institute of Planning and Evaluation for Technology in Food, Agriculture and Forestry through the High Value-Added Food Technology Development Program funded by the Ministry of Agriculture, Food, and Rural Affairs (Grant no. 322010-5). Data availability Data of this research are available from the corresponding author upon reasonable request. Supplementary materials Supplementary materials are available from https://doi.org/10.7762/cnr.2026.0014. Table S1. Seven-point Subjective Global Assessment questionnaire parameters for nutritional assessment Table S2. Recommended exchange units of each food group at different calorie levels Table S3. Recommended energy and nutrient intake for patients with predialysis CKD (stages 3–5) REFERENCES Associated Data Supplementary Materials Table S1. Seven-point Subjective Global Assessment questionnaire parameters for nutritional assessment Table S2. Recommended exchange units of each food group at different calorie levels Table S3. Recommended energy and nutrient intake for patients with predialysis CKD (stages 3–5)