**Background:** Malnutrition is a common and serious complication in cancer patients, associated with reduced chemotherapy response, increased hospital-acquired infections, longer hospital stays, higher costs, and increased mortality. Despite this, data on malnutrition prevalence and its determinants in provincial hospital settings in Vietnam remain limited. This study aimed to determine the prevalence, severity, and factors associated with malnutrition among cancer inpatients at Long An General Hospital in Southern Vietnam.
**Methods:** A descriptive cross-sectional study was conducted from May to September 2020. All 118 cancer patients undergoing inpatient treatment who met inclusion criteria (confirmed pathological diagnosis, ability to communicate, and consent) were enrolled. Exclusion criteria included mental illness, cognitive disorder, aphasia, or death during treatment. Nutritional status was assessed using the Subjective Global Assessment (SGA) tool, which classifies patients into SGA-A (normal), SGA-B (mild/moderate/suspected malnutrition), and SGA-C (severe malnutrition). Data were collected via face-to-face interviews, physical measurements (weight, height), and medical record review. Variables included demographics (age, sex, residence, education, economic status, BMI), cancer type, comorbidities, gastrointestinal symptoms lasting 2 weeks, reduced motor function, stress, weight changes, dietary changes, and laboratory markers (albumin, lymphocyte count). Univariate analysis used chi-square or Fisher's exact tests. Variables with p<0.2 were entered into a multivariable logistic regression model with stepwise forward selection. Statistical significance was set at p<0.05.
**Key Results:** The mean age was 65.6±12.3 years (range 20–89). The overall prevalence of malnutrition (SGA-B or SGA-C) was 84.7% (100/118; 95% CI: 78.2–91.3%). Of these, 33.0% (39/118) were severely malnourished (SGA-C) and 51.7% (61/118) had mild-to-moderate malnutrition (SGA-B). Pancreatic and gallbladder cancers had the highest malnutrition rate (100%), followed by lung cancer (95.8%) and liver cancer (86.7%). In multivariable logistic regression, four factors remained independently associated with malnutrition: gastrointestinal symptoms lasting 2 weeks (OR: 6.10, 95% CI: 1.12–33.35, p=0.037), reduced motor function (OR: 13.73, 95% CI: 2.56–73.86, p=0.002), serum albumin <35 g/L (OR: 6.42, 95% CI: 1.54–26.82, p=0.011), and lymphocyte count ≤1700 cells/mm³ (OR: 5.36, 95% CI: 1.31–21.97, p=0.020). No significant association was found with age, sex, residence, education, economic status, hypertension, diabetes, stress, or BMI. Notably, 70.3% of patients had albumin <35 g/L, 60.2% had lymphocytes ≤1700 cells/mm³, 55.1% had reduced motor function, and 39.0% had gastrointestinal symptoms lasting 2 weeks.
**Clinical Implications:** This study demonstrates an alarmingly high prevalence of malnutrition (84.7%) among hospitalized cancer patients in a provincial Vietnamese hospital, with over one-third severely malnourished. The strong independent associations with modifiable factors—prolonged gastrointestinal symptoms, reduced motor function, hypoalbuminemia, and lymphopenia—highlight clear targets for screening and intervention. The findings suggest that reliance on BMI alone is insufficient, as no association was found between BMI and malnutrition. Routine nutritional assessment using SGA, early identification of at-risk patients, and multidisciplinary interventions including nutritional counseling, symptom management, and physical rehabilitation are urgently needed. Limitations include the single-center design, small sample size per cancer type, lack of data on cancer stage and treatment duration, and use of SGA rather than the more cancer-specific PG-SGA. Larger, multicenter studies using PG-SGA are warranted to confirm and extend these findings.