AIWW: a new nutrition-screening tool for the oncologic population

Y. -Z., Ge; Z. -M., Fu; Zhang, Q.; Song, M. -M.; Ruan, G. -T.; Zhang, X.; Zhang, X. -W.; X. -R., Li; Zhang, K. -P.; Tang, M.; Liu, X. -Y.; Yang, M.; Liu, T.; Xie, H. -L.; Zhang, H. -Y.; Wang, Z. -W.; C. -L., Hu; Lin, S. -Q.; Zhang, R.; H. -X., Xu; Li, W.; Song, C. -H.; Liu, M.; Chen, J. -Q.; Wang, K. -H.; Bo, L.; Cong, M. -H.; Z. -N., Li; Guo, Z. -Q.; Wang, X. -B.; Wang, B. -Y.; Xu, B.; Qin, X. -H.; X. -P., Xu; Barazzoni, R.; Yao, Q. -H.; Weng, M.; Shen, X.; Shi, H. -P.; Investigation on Nutrition Status and Clinical Outcome of Common Cancers (INSCOC) Group,

doi:10.1007/s11427-022-2292-9

Malnutrition is a common comorbidity among patients with cancer. However, no nutrition-screening tool has been recognized in this population. A quick and easy screening tool for nutrition with high sensitivity and easy-to-use is needed. Based on the previous 25 nutrition-screening tools, the Delphi method was made by the members of the Chinese Society of Nutritional Oncology to choose the most useful item from each category. According to these results, we built a nutrition-screening tool named age, intake, weight, and walking (AIWW). Malnutrition was defined based on the scored patient-generated subjective global assessment (PG-SGA). Concurrent validity was evaluated using the Kendall tau coefficient and kappa consistency between the malnutrition risks of AIWW, nutritional risk screening 2002 (NRS-2002), and malnutrition screening tool (MST). Clinical benefit was calculated by the decision curve analysis (DCA), integrated discrimination improvement (IDI), and continuous net reclassification improvement (cNRI). A total of 11,360 patients (male, n=6,024 (53.0%) were included in the final study cohort, and 6,363 patients had malnutrition based on PG-SGA. Based on AIWW, NRS-2002, and MST, 7,545, 3,469, and 1,840 patients were at risk of malnutrition, respectively. The sensitivities of AIWW, NRS-2002, and MST risks were 0.910, 0.531, and 0.285, and the specificities were 0.768, 0.946, and 0.975. The Kendall tau coefficients of AIWW, NRS-2002, and MST risks were 0.588, 0.501, and 0.326, respectively. The area under the curve of AIWW, NRS-2002, and MST risks were 0.785, 0.739, and 0.630, respectively. The IDI, cNRI, and DCA showed that AIWW is non-inferior to NRS-2002 (IDI: 0.002 (−0.009, 0.013), cNRI: −0.015 (−0.049, 0.020)). AIWW scores can also predict the survival of patients with cancer. The missed diagnosis rates of AIWW, NRS-2002, and MST were 0.09%, 49.0%, and 73.2%, respectively. AIWW showed a better nutrition-screening effect than NRS-2002 and MST for patients with cancer and could be recommended as an alternative nutrition-screening tool for this population.