Abstract
Public hospitals in Thailand recently adopted a new nutrition screening tool to satisfy documentation requirements for reimbursements through the diagnosis-related group system. However, data on the performance of this instrument remains limited. This study was designed to assess the validity and cutoff points of the Society of Parenteral and Enteral Nutrition of Thailand (SPENT) nutrition screening tool against the patient-generated subjective global assessment (PG-SGA) and malnutrition diagnostic criteria proposed by the global leadership initiative on malnutrition (GLIM) in cancer patients receiving outpatient radiation therapy. A cross-sectional study of 350 patients was conducted from August 2018 to September 2020. All patients were screened for malnutrition using the SPENT nutrition screening tool. The instrument's sensitivity, specificity, positive predictive value, negative predictive value, and agreement were calculated using either the PG-SGA or GLIM malnutrition diagnosis as benchmarks. The cutoff that gave the highest sensitivity and specificity of the SPENT nutrition screening tool was selected. The mean age standard deviation of the 350 cancer patients was 59.9 (13.9) years, and 191 (54.6%) were men. Head and neck cancers were the most common type (35.7%). Against PG-SGA and GLIM malnutrition diagnosis, the SPENT nutrition screening tool demonstrated good sensitivity (85.3% and 82.8%), specificity (84.1% and 59.4%), positive predictive value (90.5% and 64.0%), negative predictive value (76.3% and 79.9%), with moderate strength of agreement (Cohen kappa 0.678, P < .001 and 0.414, P < .001). Using only the first 2 out of 4 questions revealed an acceptable sensitivity and specificity. The SPENT nutrition screening tool is an accurate, sensitive, and specific tool for malnutrition screening in cancer patients receiving outpatient radiotherapy.