This study employed a cross-sectional survey method, enrolling 477 hospitalized patients with decompensated cirrhosis from September 2021 to February 2022. Inclusion criteria were: (1) meeting diagnostic criteria for decompensated cirrhosis; (2) age ≥ 18 years; (3) voluntary participation with signed informed consent. Exclusion criteria included: (1) comorbidities such as malignancies or severe cardiovascular/cerebrovascular diseases; (2) cognitive impairment preventing assessment cooperation; (3) recent major surgeries or trauma; (4) incomplete clinical or assessment data. With 16 independent variables, the study required a sample size 5 - 10 times larger than the variables, necessitating at least 160 participants. The protocol was approved by our hospital’s ethics committee, and all participants provided written informed consent.

This study employed an observational cross-sectional design. All data were collected and assessed by two specialized liver disease research nurses during the patients’ hospitalization, who had undergone systematic and standardized training. To ensure data quality and consistency, the training covered standardized interpretation of various scales, proper use of assessment tools (e.g., grip strength testers), standardized testing procedures, and a unified data recording format. Data collection was conducted after the patients’ admission, when their condition stabilized following initial treatment and they could cooperate with the assessments, typically within 48 to 72 hours post-admission.

2.2.1. General Data Collection Form

Researchers used a self-designed general data collection form for information gathering. The questionnaire primarily included: (1) Demographic data: age, gender, marital status; (2) Anthropometric measurements: height and weight were measured using calibrated scales, with body mass index (BMI, kg/m²) calculated; (3) Clinical data: cirrhosis etiology (e.g., viral or alcoholic) was identified through medical records, along with current complications (ascites, hepatic encephalopathy, upper gastrointestinal bleeding) and their severity; (4) Laboratory parameters: fasting venous blood samples were collected the morning after hospital admission, and the hospital laboratory analyzed serum prealbumin, albumin, hemoglobin, and potassium levels using standard methods.

2.2.2. Anxiety and Depression Scale

The Anxiety Self-Rating Scale (SAS) and Depression Self-Rating Scale (SDS) were used to assess patients’ anxiety and depressive symptoms. Both SAS and SDS are self-administered scales comprising 20 items, requiring patients to select responses based on their actual feelings during the past week. Scoring criteria: A 4-point Likert scale was employed, with items primarily evaluating symptom frequency: 1 (none or very few), 2 (a few), 3 (quite a few), and 4 (most or all). Total scores were calculated by summing individual items, with a score > 40 or a standard score exceeding 60 indicating the presence of anxiety or depressive symptoms. Higher scores indicate more severe symptoms. The Chinese version of this scale has been validated in multiple chronic disease populations, including chronic liver disease, demonstrating good reliability and validity, with Cronbach’s α coefficients typically reported above 0.80 and strong construct validity [6].

2.2.3. The NRS2002 Scoring System

The NRS2002 scoring system utilizes the 2002 Nutrition Risk Screening Tool recommended by the European Society for Parenteral and Enteral Nutrition. This tool evaluates three dimensions: (1) Disease severity score (reflecting increased nutritional demands); (2) Nutritional status impairment score (based on BMI, recent weight loss, and reduced dietary intake over the past week); (3) Age score (1 point added for patients aged ≥ 70). The total score ranges from 0 to 7, with a score of 3 or higher indicating nutritional risk and necessitating the development and implementation of a clinical nutrition support plan. NRS2002 has been validated through large-scale multicenter trials internationally, demonstrating high validity in predicting clinical outcomes, along with reliable reliability and clinical applicability [7].

2.2.4. Morse Scoring System

The Morse Fall Risk Assessment Scale is employed to predict fall risks during hospitalization. This scale comprises six key components: fall history, number of comorbidities, use of walking aids, current intravenous therapy, gait characteristics, and cognitive status. Each component has clear scoring criteria, with a total range of 0 - 125 points. According to widely adopted clinical thresholds, a total score of ≥45 points indicates a high fall risk. The Morse scale, validated through extensive clinical practice, demonstrates excellent sensitivity and specificity in fall prediction, making it a reliable tool for nursing assessments [8].

2.2.5. Daily Living Activity Assessment

The Barthel Index Scale is used to evaluate patients’ daily living activity capabilities. This scale comprises 10 basic activities: feeding, bathing, personal hygiene, dressing, bowel control, bladder control, toilet use, transfer from bed to chair, walking on flat surfaces, and stair climbing. Each activity is scored based on the level of assistance required, with a total score ranging from 0 to 100. Higher scores indicate greater independence and better ADL performance. The Barthel Index is one of the most widely used ADL assessment tools in rehabilitation medicine and geriatric medicine. Its inter-rater reliability and criterion validity have been repeatedly validated, demonstrating high reliability and validity [9].

2.2.6. Liver Frailty Index (LFI)

This study employs a liver-related frailty index to evaluate patients’ physical frailty. The index comprises three tests: grip strength, timed standing test, and balance ability, which respectively indicate nutritional status, lower limb strength, and motor balance function. (1) Grip Strength Test: A calibrated electronic grip strength meter measures the maximum grip strength of the patient’s dominant hand. The patient sits with elbows flexed at 90˚ and forearms in a neutral position. Three consecutive measurements are taken at 30-second intervals, with the maximum value recorded for calculating the Liver Frailty Index (LFI). (2) Timed Standing Test: This evaluates lower limb muscle strength and endurance. The patient sits on a standard 43 cm-high armless chair with arms crossed in front of the chest. The time required to complete five consecutive “stand-sit” movements as quickly as possible upon hearing the “start” command is recorded. This test reflects lower limb explosive power and functional strength. (3) Balance Ability Test: This assesses static balance. Patients are required to perform three progressively more challenging standing postures in sequence, recording the maximum duration they can maintain each posture without support. The three postures are: a) Standing with feet together; b) Half tandem stance (one heel touching the midfoot of the other); c) Tandem stance (heel-to-toe alignment). Records are kept for the patient’s performance. The ability to maintain each posture for seconds (0-10 seconds), with a total score range of 0-30 seconds. Substitute the three test results into the structured formula for frailty performance: LFI = (−0.330 × grip strength (kg)) + (−2.529 × log_5 (chair stands time (s))) + (−0.040 × log_5 (balance time (s))) + 6. The final calculation determines the severity of the patient’s frailty. Patients with LFI ≥ 4.5, 3.2 - 4.5 and <3.2 are categorized as frail, pre-frail, and normal. This index is a well-validated tool for assessing frailty in cirrhosis patients [10].

2.2.7. Statistical Analysis Methods and Procedures

Statistical Analysis Statistical data were analyzed using SPSS 26.0 software. For quantitative data, normality testing was performed first. Data conforming to a normal distribution were described as mean ± standard deviation, with an independent samples t-test used for inter-group comparisons. For non-normally distributed data, the median and quartiles were reported, and the Mann-Whitney U test was employed for inter-group comparisons. Categorical data were presented as frequency and percentage, with the chi-square test or Fisher’s exact probability method used for inter-group comparisons. Spearman’s rank correlation analysis was conducted to examine associations between frailty and continuous variables such as nutritional risk scores and depression scores. To identify independent factors affecting frailty, a binary logistic regression model was constructed using variables with P < 0.1 from univariate analysis and clinically significant variables as independent variables, with frailty status as the dependent variable. The forward stepwise method (likelihood ratio test) was applied for variable selection, with alpha levels set at 0.05 for inclusion and 0.10 for exclusion. The odds ratios and 95% confidence intervals for selected independent variables were calculated. All statistical analyses were performed with two-tailed testing, and differences were considered statistically significant when P < 0.05. The binary logistic regression model was selected as the dependent variable (frailty status) was dichotomous (frail vs. non-frail), which is suitable for identifying factors associated with the presence or absence of an outcome.

This study ultimately enrolled 477 patients with decompensated cirrhosis who met the inclusion criteria. The cohort comprised 387 males (81.13%) and 90 females (18.87%), with an age range of 22 to 83 years and a median age of 51.00 years. According to the standardized Liver Frailty Index (LFI) assessment, among the 477 patients, 74 (15.51%) were categorized as normal (LFI < 3.2), 301 (63.10%) as pre-frail (3.2 ≤ LFI < 4.5), and 102 (21.38%) as frail (LFI ≥ 4.5). The overall prevalence of frailty was 21.38%. For subsequent regression analysis, patients were dichotomized into a frail group (n = 102) and a non-frail group (n = 375, combining normal and pre-frail categories).

The study involved 477 participants. The frailty group and non-frailty group showed statistically significant differences in age, body mass index (BMI), prealbumin, albumin, hemoglobin, serum potassium, Morse score, NRS2002 score, ADL score, depression score, anxiety score, gender, ascites, and bleeding (P < 0.05). However, marital status and hepatic encephalopathy showed no statistically significant differences (P > 0.05), as shown in Table 1.

Table 1. Single factor analysis of frailty in patients with decompensated cirrhosis.

The Spearman rank correlation analysis showed that the LFI value, a continuous variable representing the degree of frailty, was significantly positively correlated with the self-rated depression score and significantly negatively correlated with the hemoglobin level (P < 0.001). Additionally, there was a significant positive correlation between the depression score and the NRS2002 nutritional risk score (P < 0.001).

A logistic regression analysis was conducted with frailty status as the dependent variable and age, gender, body mass index (BMI), NRS2002 score, anxiety score, depression score, prealbumin, albumin, hemoglobin, serum potassium, Morse score, ADL score, bleeding, and ascites as independent variables. The interpretation of the odds ratios (ORs) provides clinical insights. For the depression score, the OR of 1.03 (95% CI: 1.01 - 1.07) indicates that for each one-point increase in the depression score, the odds of being frail increase by approximately 3%, holding other factors constant. For hemoglobin, the OR of 0.98 (95% CI: 0.97 - 0.99) suggests that each one-unit (g/L) increase in hemoglobin level is associated with a 2% decrease in the odds of frailty. Similarly, each additional year of age increases the odds of frailty by about 4% (OR = 1.04), each one-point increase in Morse score increases the odds by about 2% (OR = 1.02), and each one-point increase in ADL score decreases the odds by about 3% (OR = 0.97). Depression score, hemoglobin level, age, Morse fall risk score, and ADL score were identified as independent factors affecting frailty status in decompensated cirrhosis patients. See Table 2.

Table 2. Binary regression analysis of frailty in patients with decompensated cirrhosis.

Nutritional assessment often focuses on albumin and prealbumin, which primarily reflect liver synthesis function and short-term nutritional changes [11][12]. This study revealed that hemoglobin levels demonstrated greater independent predictive significance in multiple regression analysis. These findings suggest that the mechanisms by which nutrition affects frailty may be more complex in the specific pathological context of decompensated cirrhosis. Cirrhotic anemia results from multiple interacting factors, including portal hypertension-induced hypersplenism accelerating erythrocyte destruction, potential or overt gastrointestinal bleeding, deficiencies in hematopoietic substrates such as iron, folic acid, and vitamin B12 due to impaired metabolism or absorption, “chronic disease anemia” under chronic inflammatory conditions, and reduced erythropoietin responsiveness in hepatorenal syndrome [13][14]. Hemoglobin levels, a key indicator of nutritional status, show that lower levels correlate with higher frailty indices, indicating poorer nutritional conditions. Patients with liver cirrhosis often experience complications like portal hypertension, which may cause nausea and vomiting, impairing digestion and absorption and ultimately leading to malnutrition [15]. Long-term malnutrition weakens patients’ physical strength, thereby exacerbating their frailty. Therefore, in clinical practice, nutritional support strategies should go beyond mere calorie and protein supplementation. They should involve systematic screening and intervention for various causes of anemia, improving patients’ nutritional status, and ultimately slowing the progression of frailty.

The results of this study confirm that depression is the strongest psychological predictor of frailty. Patients with decompensated cirrhosis are deeply troubled by disease symptoms, social function regression, economic burden, and fear of liver transplantation or death, and the prevalence of depression is much higher than that of the general population [16][17]. Depression not only affects patients’ mental health but also exacerbates physical frailty through multiple pathways. Specifically, depression-induced symptoms such as appetite loss, sleep disturbances, and poor treatment adherence worsen malnutrition and metabolic disorders, creating a vicious cycle of depression, malnutrition, and physical decline [18][19]. On the other hand, depressive mood may also reduce patients’ treatment adherence, impair therapeutic outcomes, and consequently accelerate the progression of frailty [20][21]. Therefore, recognizing depression as a treatable complication with clear biological consequences is essential. Integrating routine psychological screening into liver disease management pathways and establishing collaborative referral mechanisms with psychiatry departments—through psychological interventions, social support, and safe pharmacotherapy—may be a highly cost-effective strategy to halt the progression of frailty and enhance overall treatment outcomes.

The findings of this study demonstrate that ADL capacity and fall risk represent two distinct yet complementary manifestations of physical frailty. A decline in ADL ability indicates that a patient’s physiological reserves required to maintain basic independence are nearing depletion or have been depleted, serving as a functional outcome of frailty. The elevated fall risk, however, represents a more advanced stage, revealing increased vulnerability in patients when facing dynamic environmental challenges. This serves as an early warning signal for frail individuals who are about to or have already been exposed to severe adverse events. In cirrhosis patients, this functional decline is influenced by multiple factors: portal hypertension-related hypermetabolism, hormonal imbalances, and chronic inflammation collectively contribute to cirrhosis-associated sarcopenia, which forms the biological basis of physical frailty [22][23]. At the same time, mild coordination disorders and postural control abnormalities caused by hepatic encephalopathy, as well as side effects of medications (such as diuretics and sedatives), further increase the risk of falls [24][25]. Clinical interventions must adopt a dual approach: first, through individualized, supervised exercise rehabilitation (combining aerobic, resistance, and balance training) to reverse or delay muscle loss and improve neuromuscular function; second, through standardized fall prevention protocols involving multidisciplinary teams to systematically identify and manage risks, protecting vulnerable patients from secondary injuries.

Aging is the result of the decline of physiological reserve [26]. The results of this study also demonstrate that age remains a robust predictor of frailty. Aging is characterized by a natural decline in anabolic hormone levels, mitochondrial dysfunction, diminished stem cell regenerative capacity, and chronic inflammatory dysregulation in the immune system [27][28]. When the natural physiological decline coincides with cirrhosis—a severe catabolic disorder—these factors create a synergistic effect that accelerates multi-system dysfunction. This means clinicians treating elderly decompensated cirrhosis patients must exercise heightened premature frailty awareness. Even when a patient’s Child-Pugh liver function score resembles that of younger counterparts, their functional reserves and rehabilitation potential may differ substantially. Consequently, treatment goals and supportive strategies should be tailored to individual needs with a conservative approach.