NRS-2002–defined nutritional risk and associated factors in intensive care unit patients: a cross-sectional study Authors: Derya Bıçak Ayık, Derya Akça Doğan Published: 2026 ABSTRACT BackgroundMalnutrition is common in intensive care units (ICUs) and is associated with adverse clinical outcomes. This study aimed to evaluate NRS-2002-defined nutritional risk in adult ICU patients and to examine factors associated with high nutritional risk.MethodsThis cross-sectional study was conducted in a tertiary-level ICU of a training and research hospital in Türkiye between April 2024 and May 2025. A total of 169 adult patients were included. Nutritional risk was assessed using the Nutritional Risk Screening-2002 (NRS-2002), and patients were classified as low (< 3) or high (≥ 3) risk. Disease severity was evaluated using the Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Group comparisons were performed using the Mann–Whitney U and chi-square tests. Factors associated with high nutritional risk were examined using modified Poisson regression with robust variance, and adjusted prevalence ratios (aPRs) with 95% confidence intervals (CIs) were reported.ResultsThe median age of the patients was 74 years, and patients at high nutritional risk were significantly older than those at low nutritional risk (p = 0.018). Enteral or parenteral nutrition was more common in the high-risk group (p < 0.001). In the adjusted analysis, absence of sedative medication use (aPR: 0.800, 95% CI: 0.646–0.990) and higher hemoglobin level (aPR: 0.898, 95% CI: 0.852–0.948) were associated with a lower prevalence of high nutritional risk.ConclusionsSedative medication use and lower hemoglobin levels were identified as clinical correlates of NRS-2002-defined high nutritional risk in ICU patients. These findings support the importance of regular nutritional risk screening in ICU settings and suggest that nutritional risk may not always be fully reflected by disease severity scores alone. Further multicenter studies with larger samples are needed to better clarify factors associated with nutritional risk in ICU populations.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12871-026-04005-5. Background Intensive care units (ICUs) are essential components of healthcare systems, providing advanced monitoring and treatment for critically ill patients. In these settings, inadequate nutrition represents a significant health problem that has been associated with poorer clinical prognosis and higher morbidity and mortality rates [1, 2]. Malnutrition has multidimensional consequences, including loss of body mass and muscle tissue, impairment of respiratory and cardiac functions, weakening of the immune system, and increased susceptibility to infections [3]. In addition, it has been associated with longer hospital stays, higher complication rates, and increased healthcare costs [4, 5]. The etiology of malnutrition in the intensive care setting is complex and involves multiple factors, such as disease severity, metabolic stress, inflammation, medication use, reduced appetite, and limitations in nutritional intake routes [6, 7]. During critical illness, the development of a hypercatabolic state and rapid depletion of energy stores may be associated with protein–energy malnutrition and poorer clinical outcome [8]. This condition has been associated with prolonged ICU stay, increased complication rates, and higher mortality risk [5]. Therefore, early identification of malnutrition and timely initiation of appropriate nutritional support are of vital importance for patient prognosis [1]. Nutritional risk screening enables early identification of patients who are at risk of malnutrition or who are already malnourished [9]. The European Society for Clinical Nutrition and Metabolism (ESPEN) recommends that all patients staying in the ICU for more than 48 h should be considered at nutritional risk and undergo systematic nutritional assessment [6]. In this context, tools such as the Nutritional Risk Screening 2002 (NRS-2002), Malnutrition Universal Screening Tool (MUST), Mini Nutritional Assessment (MNA), and Subjective Global Assessment (SGA) are commonly used for nutritional evaluation [10]. Among these tools, the NRS-2002 was preferred in the present study because it is a practical and widely used screening tool for hospitalized adults and allows the structured evaluation of nutritional risk together with disease severity, which is particularly relevant in ICU patients. In addition, disease severity scores, including the Acute Physiology and Chronic Health Evaluation II (APACHE II), have been considered in studies evaluating nutritional risk in critically ill patients [11, 12]. This study was designed to determine the NRS-2002-defined nutritional risk in ICU patients and to examine demographic, clinical, and laboratory factors associated with NRS-2002-defined nutritional risk. The findings obtained from this study are expected to contribute to the development of nutritional management strategies in intensive care settings. Methods Study design and sample This cross-sectional, descriptive study was conducted in a tertiary-level intensive care unit of a training and research hospital in Türkiye to evaluate NRS-2002-defined nutritional risk in critically ill patients requiring advanced monitoring and treatment, as well as factors associated with nutritional risk. Adult patients aged ≥ 18 years who were followed in the ICU for more than 48 h between April 2024 and May 2025 were included. Patients in the terminal stage, those receiving palliative care, or those with incomplete data were excluded, and no imputation was performed. Terminal stage and palliative care status were determined based on physician documentation and available clinical records, including documented palliative-care follow-up and relevant clinical notes regarding the patient’s overall clinical status, when available. A patient flow diagram (Fig. 1) is provided to show the number of screened patients during the study period, the number excluded according to each eligibility criterion, the number excluded due to incomplete data, and the final analyzed sample. The final analysis was performed using data from 169 patients who met the inclusion criteria. No formal a priori sample size calculation was performed; all eligible patients during the study period were included. The study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist for cross-sectional studies. Fig. 1Flow diagram of patient selection during the study period Flow diagram of patient selection during the study period Data collection Data for all included patients were collected prospectively while the patients were hospitalized in the ICU. Before the study, a standardized data collection form was developed to record sociodemographic characteristics, clinical diagnoses, feeding type, medications used, vital signs, laboratory results, and ICU-related parameters. Sociodemographic and selected clinical data were obtained from patient files, bedside observation forms, and the ICU information management system. In addition, the study instruments were applied by the researchers during the patients’ ICU stay. When necessary, relevant clinical and laboratory information was cross-checked using the institutional ICU clinical information system and patient medical records. Data collection instruments The study variables included sociodemographic variables (age, sex, marital status, educational level, and income status), clinical characteristics (primary diagnosis, feeding type, mechanical ventilation support, and medication use), nutritional risk, disease severity, pain, and anxiety. Nutritional risk was assessed using the Nutritional Risk Screening-2002 (NRS-2002), and disease severity was determined using the Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Pain and anxiety levels were assessed using the Critical-Care Pain Observation Tool (CPOT) and the Facial Anxiety Scale (FAS), respectively. Detailed scoring procedures and item-level descriptions of these instruments are provided in the Supplementary Materials. Measurement procedures were standardized across study variables. Nutritional risk, pain, and anxiety were assessed by the researchers at bedside under routine ICU conditions at the time of study inclusion in patients who had completed at least 48 h in the ICU, using the NRS-2002, CPOT, and FAS, respectively. The exact time from ICU admission to nutritional assessment was not recorded as a separate variable; therefore, feeding type, mechanical ventilation support, sedative use, laboratory parameters, and oxygen saturation were interpreted as concurrent clinical characteristics rather than antecedent predictors of nutritional risk. Vital signs and laboratory parameters were obtained from patient files and the ICU information management system, using values recorded on the same day as the study assessment or the closest available measurements. Disease severity was evaluated using APACHE II scores extracted from patient files and the ICU information management system; these scores reflected the worst clinical and laboratory values recorded within the first 24 h after ICU admission. At the time of study assessment, ICU length of stay and duration of mechanical ventilation were recorded as cumulative durations elapsed since ICU admission. Nutritional Risk Screening-2002 (NRS-2002) The NRS-2002, developed by Kondrup et al. in 2002, is a screening tool recommended by the ESPEN to identify nutritional risk and evaluate the need for nutritional support in hospitalized patients. It is widely used in clinical practice, particularly in ICUs. The tool evaluates nutritional status and disease severity, with an additional point added for patients aged over 70 years. A total score of ≥ 3 indicates nutritional risk and the need for appropriate nutritional support, whereas scores < 3 warrant regular reassessment [13]. Acute Physiology and Chronic Health Evaluation II (APACHE II) APACHE II is a classification system used in ICUs to assess disease severity and predict prognosis. It incorporates 12 physiological variables along with patient age and chronic health status. Evaluation is performed using the worst clinical and laboratory values recorded during the first 24 h after ICU admission. Higher scores indicate an increased mortality risk and a poorer prognosis. APACHE II is a reliable prognostic tool widely used in both clinical practice and research [14, 15]. Critical-Care Pain Observation Tool (CPOT) The CPOT was developed by Gélinas et al. in 2006 in Canada to assess pain in ICU patients [16]. It has since been validated in multiple languages; the Turkish version was adapted in 2016 by Gündoğan et al. [17]. These studies demonstrated that CPOT is a valid and reliable instrument for pain assessment in critically ill patients. The tool includes four behavioral domains, with a total score ranging from 0 to 8, and scores above 2 generally indicate the presence of pain, with acceptable internal consistency demonstrated in the current sample (Cronbach’s alpha = 0.757). Facial Anxiety Scale (FAS) The FAS was developed by McKinley et al. in 2003 to assess anxiety levels in ICU patients [18]. The Turkish validity and reliability study was conducted by Iyigun et al. in 2016 [19]. The scale consists of five facial expressions representing different levels of anxiety. In this study, FAS scores of 3 or higher were considered to indicate at least moderate anxiety. Statistical analysis Statistical analyses were performed using IBM SPSS Statistics for Mac, version 28.0 (IBM Corp., Armonk, NY, USA). The normality of continuous variables was assessed using the Shapiro–Wilk test. Variables that did not follow a normal distribution were presented as median (interquartile range, IQR), and group comparisons were conducted using the Mann–Whitney U test. Categorical variables were expressed as numbers (percentages), and differences between groups were analyzed using the chi-square test. To examine adjusted associations with high nutritional risk, modified Poisson regression with robust variance was used, and adjusted prevalence ratios (aPRs) with 95% confidence intervals were reported. Variables with p < 0.05 in univariable analyses and those considered clinically relevant were considered for inclusion in the adjusted model. Multicollinearity was assessed using tolerance and variance inflation factor (VIF) values. No relevant multicollinearity was detected, with tolerance values ranging from 0.848 to 0.943 and VIF values ranging from 1.061 to 1.180. Model fit was assessed using deviance and Pearson chi-square goodness-of-fit statistics. In the final model, the deviance/df and Pearson chi-square/df values were 0.486 and 0.316, respectively. The final model included six predictors for 116 events, corresponding to approximately 19.3 events per predictor. Feeding type was not retained in the final adjusted model because it may indirectly overlap with the reduced oral intake component of the NRS-2002 scoring structure. Age, ICU length of stay, mechanical ventilation (MV) support, and APACHE II score were not retained in the final adjusted model because age is a component of the NRS-2002 score, ICU length of stay, although recorded at the time of study assessment, reflects a time-dependent clinical course variable rather than an antecedent factor, and MV support showed conceptual overlap with oxygen saturation, and APACHE II score may partly reflect the disease-severity domain considered within the NRS-2002 framework. p < 0.05 was considered statistically significant. Ethical considerations Ethical approval was obtained from the Mardin Artuklu University Non-Interventional Clinical Research Ethics Committee (Decision No: 2023/12-19; Date: 05 December 2023). Written institutional permission was obtained from the hospital where the study was conducted. Additionally, informed consent was obtained from the participants or their representatives. The study was conducted in accordance with the principles of the Declaration of Helsinki. Results The median age of the ICU patients was 74 years (IQR = 21), and patients in the high nutritional risk group were significantly older than those in the low nutritional risk group (p = 0.018). No statistically significant differences were observed between groups with respect to gender, marital status, educational level, income level, or primary diagnosis (p > 0.05). Regarding feeding type, the proportion of patients receiving enteral or parenteral nutrition was significantly higher in the high nutritional risk group (p < 0.001). Mechanical ventilation support was also significantly more common among patients with high nutritional risk (p < 0.001). While no significant difference was found between groups in terms of analgesic medication use (p = 0.832), sedative medication use was significantly higher in the high nutritional risk group (p = 0.035) (Table 1). Table 1Socio-demographic and clinical characteristics of the patients according to nutritional risk status (n = 169)CharacteristicTotal(n = 169)Low nutritional riskNRS < 3(n = 53)High nutritional riskNRS ≥ 3(n = 116)Test statistic p Age, median (IQR)74 (21)66.0 (20.0)75.0 (19.75)2375.5 0.018 Gender, n (%)1.3480.246 Male94 (55.6)26 (49.1)68 (58.6) Female75 (44.4)27 (50.9)48 (41.4)Marital status, n (%)0.4410.507 Single27 (16.0)7 (13.2)20 (17.2) Married142 (84.0)46 (86.8)96 (82.8)Education level, n (%)4.0670.131 Literate118 (69.8)42 (79.2)76 (65.5) Primary school30 (17.8)5 (9.4)25 (21.6) High school21 (12.4)6 (11.3)15 (12.9)Income level, n (%)0.8490.654 Low29 (17.2)7 (13.2)22 (19.0) Moderate134 (79.3)44 (83.0)90 (77.6) High6 (3.6)2 (3.8)4 (3.4)Primary diagnosis, n (%)4.0550.669 Neurological emergencies28 (16.6)9 (17.0)19 (16.4) Trauma and post-operative conditions26 (15.4)5 (9.4)21 (18.1) Respiratory failure33 (19.5)9 (17.0)24 (20.7) Cardiovascular diseases29 (17.2)11 (20.8)18 (15.5) Renal failure18 (10.7)8 (15.1)10 (8.6) General condition impairment16 (9.5)5 (9.4)11 (9.5) Other19 (11.2)6 (11.3)13 (11.2)Feeding type, n (%)50.900 < 0.001 Oral54 (32.0)37 (69.8)17 (14.7) Enteral or parenteral115 (68.0)16 (30.2)99 (85.3)MV support, n (%)27.829 < 0.001 No100 (59.2)47 (88.7)53 (45.7) Yes69 (40.8)6 (11.3)63 (54.3)Analgesic medication use, n (%)0.0450.832 No40 (23.7)12 (22.6)28 (24.1) Yes129 (76.3)41 (77.4)88 (75.9)Sedative medication use, n (%)4.453 0.035 No115 (68.0)42 (79.2)73 (62.9) Yes54 (32.0)11 (20.8)43 (37.1)Values are presented as frequency (n) and percentage (%) or median and interquartile range (IQR). Among patients classified as receiving enteral or parenteral nutrition, 89 received enteral nutrition and 26 received total parenteral nutrition. MV Mechanical ventilation, NRS Nutritional Risk Screening. Group comparisons were made using the Chi-square test for categorical variables and the Mann–Whitney U test for continuous variables. p < 0.05 in bold Table 1Socio-demographic and clinical characteristics of the patients according to nutritional risk status (n = 169)CharacteristicTotal(n = 169)Low nutritional riskNRS < 3(n = 53)High nutritional riskNRS ≥ 3(n = 116)Test statistic p Age, median (IQR)74 (21)66.0 (20.0)75.0 (19.75)2375.5 0.018 Gender, n (%)1.3480.246 Male94 (55.6)26 (49.1)68 (58.6) Female75 (44.4)27 (50.9)48 (41.4)Marital status, n (%)0.4410.507 Single27 (16.0)7 (13.2)20 (17.2) Married142 (84.0)46 (86.8)96 (82.8)Education level, n (%)4.0670.131 Literate118 (69.8)42 (79.2)76 (65.5) Primary school30 (17.8)5 (9.4)25 (21.6) High school21 (12.4)6 (11.3)15 (12.9)Income level, n (%)0.8490.654 Low29 (17.2)7 (13.2)22 (19.0) Moderate134 (79.3)44 (83.0)90 (77.6) High6 (3.6)2 (3.8)4 (3.4)Primary diagnosis, n (%)4.0550.669 Neurological emergencies28 (16.6)9 (17.0)19 (16.4) Trauma and post-operative conditions26 (15.4)5 (9.4)21 (18.1) Respiratory failure33 (19.5)9 (17.0)24 (20.7) Cardiovascular diseases29 (17.2)11 (20.8)18 (15.5) Renal failure18 (10.7)8 (15.1)10 (8.6) General condition impairment16 (9.5)5 (9.4)11 (9.5) Other19 (11.2)6 (11.3)13 (11.2)Feeding type, n (%)50.900 < 0.001 Oral54 (32.0)37 (69.8)17 (14.7) Enteral or parenteral115 (68.0)16 (30.2)99 (85.3)MV support, n (%)27.829 < 0.001 No100 (59.2)47 (88.7)53 (45.7) Yes69 (40.8)6 (11.3)63 (54.3)Analgesic medication use, n (%)0.0450.832 No40 (23.7)12 (22.6)28 (24.1) Yes129 (76.3)41 (77.4)88 (75.9)Sedative medication use, n (%)4.453 0.035 No115 (68.0)42 (79.2)73 (62.9) Yes54 (32.0)11 (20.8)43 (37.1)Values are presented as frequency (n) and percentage (%) or median and interquartile range (IQR). Among patients classified as receiving enteral or parenteral nutrition, 89 received enteral nutrition and 26 received total parenteral nutrition. MV Mechanical ventilation, NRS Nutritional Risk Screening. Group comparisons were made using the Chi-square test for categorical variables and the Mann–Whitney U test for continuous variables. p < 0.05 in bold Socio-demographic and clinical characteristics of the patients according to nutritional risk status (n = 169) Values are presented as frequency (n) and percentage (%) or median and interquartile range (IQR). Among patients classified as receiving enteral or parenteral nutrition, 89 received enteral nutrition and 26 received total parenteral nutrition. MV Mechanical ventilation, NRS Nutritional Risk Screening. Group comparisons were made using the Chi-square test for categorical variables and the Mann–Whitney U test for continuous variables. p < 0.05 in bold Comparisons of clinical and laboratory parameters presented in Table 2 revealed that ICU length of stay (p < 0.001) was significantly longer in patients with high nutritional risk. Among mechanically ventilated patients, duration of mechanical ventilation (p = 0.001) was also significantly longer in the high nutritional risk group. In addition, oxygen saturation levels were significantly higher in the high nutritional risk group (p = 0.008). Statistically significant differences were also observed between groups in potassium (p = 0.017), serum albumin (p = 0.007), hemoglobin (p < 0.001), and alanine aminotransferase (ALT) levels (p = 0.013). No significant differences were detected for the remaining parameters (p > 0.05). Table 2Clinical and laboratory parameters of the patients according to nutritional risk status (n = 169)VariableTotal(n = 169)Low nutritional riskNRS < 3(n = 53)High nutritional riskNRS ≥ 3(n = 116)Test statistic p ICU length of stay (days)13.0 (11.5)10.0 (5.0)15.0 (24.0)1602.5 < 0.001 MV duration (days)(n = 69)20.0 (61.0)8.0 (2.0)20.0 (62.0)37.5 0.001 SBP (mmHg)125.0 (34.0)117.0 (38.5)126.0 (29.0)2846.50.441DBP (mmHg)66.0 (18.0)71.0 (13.5)63.5 (18.0)2761.00.288Oxygen saturation (%)98.0 (4.0)98.0 (4.0)99.0 (4.0)2306.5 0.008 Temperature (°C)36.5 (0.3)36.5 (0.2)36.5 (0.3)2738.50.251Laboratory tests Blood glucose (mg/dL)138.0 (73.0)122.0 (100.0)150.0 (51.0)2639.50.141 Potassium (mmol/L)3.83 (0.8)4.3 (1.3)3.8 (0.8)2371.0 0.017 Sodium (mmol/L)139.0 (10.0)139.0 (9.0)141.0 (12.0)2899.50.554 Serum albumin (g/dL)2.9 (1.0)3.1 (1.05)2.8 (1.0)2275.5 0.007 Prealbumin (mg/dL)27.0 (8.5)29.0 (9.0)27.0 (7.75)2561.00.081 Hemoglobin (g/dL)9.5 (2.7)9.6 (3.2)8.9 (2.95)2011.5 < 0.001 BUN (mg/dL)34.0 (42.0)31.0 (31.0)35.0 (43.0)2910.50.579 Urea (mg/dL)70.62 (80.48)70.62 (73.54)70.62 (84.76)2739.50.257 Creatinine (mg/dL)1.11 (1.01)1.07 (0.87)1.2 (1.47)2927.00.618 AST (U/L)25.0 (24.5)25.0 (29.5)25.0 (23.0)2784.00.325 ALT (U/L)25.0 (23.0)19.0 (23.0)31.0 (24.5)2338.0 0.013 Values are presented as median and interquartile range (IQR). ALT Alanine Aminotransferase, AST Aspartate Aminotransferase, BUN Blood Urea Nitrogen, DBP Diastolic Blood Pressure, ICU Intensive Care Unit, MV Mechanical Ventilation, NRS Nutritional Risk Screening, SBP Systolic Blood Pressure. Group comparisons were performed using the Mann–Whitney U test. p < 0.05 in bold Table 2Clinical and laboratory parameters of the patients according to nutritional risk status (n = 169)VariableTotal(n = 169)Low nutritional riskNRS < 3(n = 53)High nutritional riskNRS ≥ 3(n = 116)Test statistic p ICU length of stay (days)13.0 (11.5)10.0 (5.0)15.0 (24.0)1602.5 < 0.001 MV duration (days)(n = 69)20.0 (61.0)8.0 (2.0)20.0 (62.0)37.5 0.001 SBP (mmHg)125.0 (34.0)117.0 (38.5)126.0 (29.0)2846.50.441DBP (mmHg)66.0 (18.0)71.0 (13.5)63.5 (18.0)2761.00.288Oxygen saturation (%)98.0 (4.0)98.0 (4.0)99.0 (4.0)2306.5 0.008 Temperature (°C)36.5 (0.3)36.5 (0.2)36.5 (0.3)2738.50.251Laboratory tests Blood glucose (mg/dL)138.0 (73.0)122.0 (100.0)150.0 (51.0)2639.50.141 Potassium (mmol/L)3.83 (0.8)4.3 (1.3)3.8 (0.8)2371.0 0.017 Sodium (mmol/L)139.0 (10.0)139.0 (9.0)141.0 (12.0)2899.50.554 Serum albumin (g/dL)2.9 (1.0)3.1 (1.05)2.8 (1.0)2275.5 0.007 Prealbumin (mg/dL)27.0 (8.5)29.0 (9.0)27.0 (7.75)2561.00.081 Hemoglobin (g/dL)9.5 (2.7)9.6 (3.2)8.9 (2.95)2011.5 < 0.001 BUN (mg/dL)34.0 (42.0)31.0 (31.0)35.0 (43.0)2910.50.579 Urea (mg/dL)70.62 (80.48)70.62 (73.54)70.62 (84.76)2739.50.257 Creatinine (mg/dL)1.11 (1.01)1.07 (0.87)1.2 (1.47)2927.00.618 AST (U/L)25.0 (24.5)25.0 (29.5)25.0 (23.0)2784.00.325 ALT (U/L)25.0 (23.0)19.0 (23.0)31.0 (24.5)2338.0 0.013 Values are presented as median and interquartile range (IQR). ALT Alanine Aminotransferase, AST Aspartate Aminotransferase, BUN Blood Urea Nitrogen, DBP Diastolic Blood Pressure, ICU Intensive Care Unit, MV Mechanical Ventilation, NRS Nutritional Risk Screening, SBP Systolic Blood Pressure. Group comparisons were performed using the Mann–Whitney U test. p < 0.05 in bold Clinical and laboratory parameters of the patients according to nutritional risk status (n = 169) MV duration (days) (n = 69) Values are presented as median and interquartile range (IQR). ALT Alanine Aminotransferase, AST Aspartate Aminotransferase, BUN Blood Urea Nitrogen, DBP Diastolic Blood Pressure, ICU Intensive Care Unit, MV Mechanical Ventilation, NRS Nutritional Risk Screening, SBP Systolic Blood Pressure. Group comparisons were performed using the Mann–Whitney U test. p < 0.05 in bold APACHE II scores were significantly lower in the high nutritional risk group compared with the low nutritional risk group (p < 0.001). There were no statistically significant differences between groups in terms of median CPOT and FAS scores, nor in the distribution of CPOT and FAS scores categorized as < 3 or ≥ 3 (p > 0.05) (Table 3). Table 3Comparison of APACHE II score, CPOT, and FAS levels between low and high nutritional risk groups (n = 169)VariableTotal(n = 169)Low nutritional riskNRS < 3(n = 53)High nutritional riskNRS ≥ 3(n = 116)Test statistic p APACHE II score, median (IQR)13 (8)16 (8)11 (6)U = 2038.5 < 0.001 CPOT score, median (IQR)1 (4)1 (4)1.5 (4)U = 2948.50.664CPOT score, n (%) < 3 points109 (64.5)33 (62.3%)76 (65.5%)χ² =0.1680.682 ≥ 3 points60 (35.5)20 (37.7%)40 (34.5%)FAS score, median (IQR)3 (2)3 (2)3 (1.75)U = 2906.00.555FAS score, n (%) < 3 points67 (39.6)23 (43.4%)44 (37.9%)χ² = 0.4540.500 ≥ 3 points102 (60.4)30 (56.6%)72 (62.1%)Values are presented as frequency (n) and percentage (%) or median and interquartile range (IQR). APACHE II Acute Physiology and Chronic Health Evaluation II, CPOT Critical Care Pain Observation Tool, FAS Facial Anxiety Scale, NRS Nutritional Risk Screening. Mann–Whitney U test was used for continuous variables, and Chi-square test for categorical variables. p < 0.05 in bold Table 3Comparison of APACHE II score, CPOT, and FAS levels between low and high nutritional risk groups (n = 169)VariableTotal(n = 169)Low nutritional riskNRS < 3(n = 53)High nutritional riskNRS ≥ 3(n = 116)Test statistic p APACHE II score, median (IQR)13 (8)16 (8)11 (6)U = 2038.5 < 0.001 CPOT score, median (IQR)1 (4)1 (4)1.5 (4)U = 2948.50.664CPOT score, n (%) < 3 points109 (64.5)33 (62.3%)76 (65.5%)χ² =0.1680.682 ≥ 3 points60 (35.5)20 (37.7%)40 (34.5%)FAS score, median (IQR)3 (2)3 (2)3 (1.75)U = 2906.00.555FAS score, n (%) < 3 points67 (39.6)23 (43.4%)44 (37.9%)χ² = 0.4540.500 ≥ 3 points102 (60.4)30 (56.6%)72 (62.1%)Values are presented as frequency (n) and percentage (%) or median and interquartile range (IQR). APACHE II Acute Physiology and Chronic Health Evaluation II, CPOT Critical Care Pain Observation Tool, FAS Facial Anxiety Scale, NRS Nutritional Risk Screening. Mann–Whitney U test was used for continuous variables, and Chi-square test for categorical variables. p < 0.05 in bold Comparison of APACHE II score, CPOT, and FAS levels between low and high nutritional risk groups (n = 169) Values are presented as frequency (n) and percentage (%) or median and interquartile range (IQR). APACHE II Acute Physiology and Chronic Health Evaluation II, CPOT Critical Care Pain Observation Tool, FAS Facial Anxiety Scale, NRS Nutritional Risk Screening. Mann–Whitney U test was used for continuous variables, and Chi-square test for categorical variables. p < 0.05 in bold In the adjusted analysis, absence of sedative medication use (aPR: 0.800, 95% CI: 0.646–0.990) and higher hemoglobin level (aPR: 0.898, 95% CI: 0.852–0.948) were associated with a lower prevalence of high nutritional risk. Oxygen saturation, potassium, serum albumin, and ALT were not significantly associated with high nutritional risk (Table 4). Table 4Factors associated with high nutritional risk (n = 169)VariableaPR95% CI p Sedative medication use (no vs. yes)0.8000.646–0.990 0.040 Oxygen saturation (%)1.0380.995–1.0830.081Potassium (mmol/L)0.9330.780–1.1160.447Serum albumin (g/dL)0.9220.847–1.0030.059Hemoglobin (g/dL)0.8980.852–0.948 < 0.001 ALT (U/L)1.0010.999–1.0030.344Adjusted prevalence ratios (aPRs) were estimated using modified Poisson regression with log link and robust variance. ALT Alanine Aminotransferase, CI Confidence Interval. p < 0.05 in bold Table 4Factors associated with high nutritional risk (n = 169)VariableaPR95% CI p Sedative medication use (no vs. yes)0.8000.646–0.990 0.040 Oxygen saturation (%)1.0380.995–1.0830.081Potassium (mmol/L)0.9330.780–1.1160.447Serum albumin (g/dL)0.9220.847–1.0030.059Hemoglobin (g/dL)0.8980.852–0.948 < 0.001 ALT (U/L)1.0010.999–1.0030.344Adjusted prevalence ratios (aPRs) were estimated using modified Poisson regression with log link and robust variance. ALT Alanine Aminotransferase, CI Confidence Interval. p < 0.05 in bold Factors associated with high nutritional risk (n = 169) Adjusted prevalence ratios (aPRs) were estimated using modified Poisson regression with log link and robust variance. ALT Alanine Aminotransferase, CI Confidence Interval. p < 0.05 in bold Discussion Malnutrition in intensive care units (ICUs) is a significant clinical condition that has been associated with poorer treatment processes and recovery outcomes in critically ill patients. This study was conducted to evaluate NRS-2002-defined nutritional risk in ICU patients and to examine demographic, clinical, and laboratory characteristics associated with NRS-2002-defined nutritional risk. Comparison of the findings with previous studies in the literature supports the importance of nutritional management in the ICU and the need for individualized patient-centered approaches. Socio-demographic and clinical characteristics In this study, advanced age was associated with higher nutritional risk in univariable comparisons. This finding is consistent with previous reports indicating that malnutrition is more prevalent among older individuals [20, 21]. Factors such as reduced appetite, malabsorption, comorbid chronic conditions, and social determinants may be related to nutritional vulnerability in older adults. Furthermore, higher mortality rates observed in elderly ICU patients suggest that this population requires closer nutritional monitoring [1, 22]. This finding should be interpreted with caution, as age is also incorporated into the NRS-2002 scoring structure. Although marital status, educational level, and income status were not significantly associated with nutritional risk in the present study, their inclusion remains important. Social and socioeconomic characteristics may provide a broader understanding of the clinical profile and vulnerability of ICU patients. Although no significant differences were observed in this study, these variables may still be relevant to consider in future research on nutritional risk in critically ill populations. According to the results, high nutritional risk was significantly more common among patients receiving enteral or parenteral nutrition than among those receiving oral nutrition. However, this finding should not be interpreted as evidence that enteral or parenteral nutrition itself affects nutritional risk. Rather, it may reflect impaired oral intake, greater clinical vulnerability, and a higher need for nutritional support in ICU patients. This interpretation is important because reduced oral intake is an explicit component of the NRS-2002 scoring structure, and therefore feeding type may partially overlap with the outcome definition. In this sense, the finding may indicate that the NRS-2002 functioned as expected by identifying patients with impaired oral intake and greater nutritional support needs. The literature emphasizes that enteral or parenteral nutrition constitutes a component of medical nutrition therapy planned according to individual clinical requirements in critically ill patients [6, 23]. Therefore, feeding type should be interpreted within the broader clinical context of nutritional risk screening rather than as a separate determinant of high nutritional risk. Mechanical ventilation support was more common among patients with high nutritional risk in the present study. High nutritional risk may be associated with mechanical ventilation requirement, a relationship that has been frequently emphasized in previous studies [24, 25]. In addition, prolonged duration of mechanical ventilation may be accompanied by infections and complications, which can complicate nutritional management in ICU patients. Sedative medication use was found to be significantly higher among patients with high nutritional risk. This association may reflect clinical conditions related to sedative use, such as reduced gastrointestinal motility, decreased level of consciousness, and impaired oral intake [26]. Moreover, sedative medications are frequently administered in conjunction with mechanical ventilation, which may be associated with poorer nutritional status [27, 28]. Although sedative medication use was more common among patients with high nutritional risk in univariable comparisons, absence of sedative medication use was associated with a lower prevalence of high nutritional risk in the adjusted analysis. This finding suggests that the association between sedative use and nutritional risk may reflect other clinical factors such as route of nutrition and overall clinical severity. Consistent with the literature, sedative use may be related to challenges in nutritional management, including reduced tolerance to enteral feeding and aspiration risk [28]. Therefore, the association between sedative use and high nutritional risk may reflect the broader clinical context rather than a direct causal relationship. These findings support considering medication management alongside nutritional support planning in ICU patients. In addition, patients with high nutritional risk had longer ICU stays in univariable comparisons. This finding aligns with previous evidence indicating that malnutrition is associated with delayed recovery and increased complication rates [4, 5]. Prolonged ICU hospitalization is also associated with increased healthcare costs and places an additional burden on healthcare systems. However, ICU length of stay was not included in the adjusted analysis because it is a time-accumulating clinical course variable. In the univariable analysis, oxygen saturation was higher in patients with high nutritional risk. However, this association was not maintained in the adjusted analysis. This finding suggests that the relationship between oxygen saturation and nutritional risk may reflect other clinical factors and should therefore be interpreted with caution. In the ICU setting, oxygen saturation primarily reflects arterial oxygenation and may be affected by pulse oximetry-related limitations and measurement conditions [29–31]. Accordingly, this finding should not be interpreted as indicating that higher oxygen saturation itself increases nutritional risk or as a direct marker of nutritional status. Laboratory parameters Lower serum albumin and hemoglobin levels were significantly associated with high nutritional risk in univariable comparisons. Although albumin has long been used as a biomarker of nutritional status, it is not considered a reliable indicator when used alone [32]. Nevertheless, low albumin levels may reflect both malnutrition and disease severity. In ICU patients, reduced hemoglobin levels may have multiple causes, including chronic inflammation, bleeding, hemodilution, chronic kidney disease, and nutritional deficiencies; therefore, hemoglobin should not be interpreted as a specific nutritional biomarker [33]. Significant differences were also observed in potassium and alanine aminotransferase (ALT) levels in univariable comparisons, which may reflect accompanying electrolyte and metabolic disturbances in patients with higher nutritional risk [34]. However, in the adjusted analysis, only hemoglobin remained significantly associated with high nutritional risk, among the laboratory parameters, whereas albumin, potassium, and ALT were not significantly associated with the outcome. This suggests that hemoglobin may represent a clinically relevant correlate of high nutritional risk in this sample; nevertheless, given its multifactorial nature in critically ill patients, it should be interpreted within the broader clinical context rather than as a specific nutritional biomarker. APACHE II score In this study, APACHE II scores were significantly lower in patients with high nutritional risk in univariable comparisons. In the literature, APACHE II scores are generally reported to be positively associated with disease severity, morbidity, and mortality [12, 35]. Indeed, some studies have shown that higher APACHE II scores are associated with an increased risk of malnutrition-related complications and higher mortality rates [11, 36]. APACHE II was based on the worst physiological values recorded within the first 24 h of ICU admission, whereas nutritional risk was assessed after 48 h in the ICU. APACHE II score was not included in the final adjusted model because it may partly reflect the disease-severity domain considered within the NRS-2002 framework. Therefore, the inverse association observed in univariable comparisons should not be interpreted as indicating that greater acute disease severity is protective against nutritional risk. Rather, it may reflect the non-synchronous timing of severity scoring and nutritional assessment, together with possible selection effects related to the study inclusion criteria. Overall, this finding suggests that disease severity scores alone may not always adequately reflect nutritional risk in ICU patients, and that early, structured nutritional risk screening may be useful. FAS and CPOT scores In this study, no significant differences were observed between groups with respect to FAS and CPOT scores. However, previous studies have reported that nutritional status may be associated with anxiety and pain [37, 38]. The discrepancy between our findings and the existing literature may be related to the relatively limited sample size or to the sensitivity of the instruments used to assess psychological variables in critically ill patients. Conclusions Sedative medication use and lower hemoglobin levels were identified as clinical correlates of NRS-2002-defined high nutritional risk in ICU patients. Overall, these findings support the importance of regular nutritional risk screening in ICU patients and suggest that nutritional risk may not always be fully reflected by disease severity scores alone. Future studies conducted in multicenter settings with larger sample sizes may enable a more comprehensive evaluation of factors associated with nutritional risk. These associations should be interpreted within the limits of a cross-sectional design. Contribution to the literature This study contributes to the existing literature by examining NRS-2002-defined nutritional risk in ICU patients using a range of demographic, clinical, and laboratory variables. The findings suggest that sedative medication use and hemoglobin level may be relevant clinical correlates of nutritional risk in this setting. In this respect, the study supports the value of regular nutritional monitoring and structured nutritional risk screening in ICU patients. Clinical implications The findings of this study suggest that regular nutritional risk screening in ICU patients may be beneficial. Nutritional risk may warrant closer monitoring, particularly in patients using sedative medication and in those with lower hemoglobin levels. Early evaluation of nutritional risk and timely planning of appropriate nutritional support, when indicated, are important considerations in these patients. However, sedative medication use and hemoglobin level should be interpreted as clinical correlates that may support nutritional risk screening, not as independent determinants of nutritional risk. Feeding type should therefore be considered within the broader context of NRS-2002-based nutritional risk screening. Adopting a multidisciplinary approach may facilitate the early identification of nutrition-related risks. Limitations This study has several limitations. First, the single-center design may limit the generalizability of the findings to other intensive care units. Due to the cross-sectional nature of the study, the observed associations cannot be interpreted in terms of causality. Because the study included only patients who remained in the ICU for more than 48 h, survival or left-truncation bias cannot be excluded. Patients with the highest acute severity who died or were discharged before this time point were not represented in the final sample. This selection process may partly explain the inverse association observed between APACHE II score and high nutritional risk. A further limitation is that some evaluated clinical variables, particularly age, feeding-related characteristics, and disease severity, may overlap with the NRS-2002 scoring structure. Therefore, findings involving these variables should be interpreted cautiously and not as independent predictors of nutritional risk. The sample size was relatively limited compared with some studies in the literature, which may have affected the ability to detect significant relationships, particularly between nutritional risk and clinical variables such as anxiety and pain. In addition, no formal a priori sample size calculation was performed; therefore, the adjusted regression model should be interpreted as exploratory. Sedative medication use was evaluated only as a binary variable; therefore, differences related to sedative agent type, dose, duration, and sedation depth could not be examined. Another limitation is that deeply sedated patients were not defined as a separate exclusion category for FAS assessment, and CPOT inter-rater reliability was not assessed in this sample. Therefore, the applicability of FAS in deeply sedated patients and observer agreement for CPOT could not be evaluated in detail. In addition, some factors associated with nutritional risk identified in this study, such as advanced age, have already been reported in the literature and therefore should not be interpreted as entirely novel findings. Moreover, because age is incorporated into the NRS-2002 scoring structure, its association with nutritional risk should be interpreted cautiously. The relatively high prevalence of high nutritional risk may be related to the inclusion of patients who remained in the ICU for more than 48 h, the older age profile of the sample, and the structure of the NRS-2002 scoring system. Furthermore, body mass index (BMI), recent weight loss, recent oral intake, and comorbidity burden were not recorded as separate analyzable baseline variables, although BMI, weight loss, reduced intake, and disease severity were considered within the NRS-2002 scoring process. Nevertheless, the study employed a structured nutritional risk screening approach and provides useful insights into factors associated with nutritional risk in ICU patients. Supplementary Information Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2