Objective To construct a scientific and objective clinical management sensitive indicators for cancer nutrition in order to provide references for evaluating and monitoring the quality of care in cancer nutrition therapy. Method The structure-processoutcome model was used as the theoretical framework. Constructing a search strategy and searching for evidence. Literature screening and data extraction were conducted according to the inclusion and exclusion criteria by two researchers independently. Level of evidence and methodological quality were evaluated using relevant tools. The preliminary draft of clinical management sensitive indicators for cancer nutrition was formed based on evidence. 14 experts from 9 grade A tertiary hospitals in Beijing Shanghai Guangzhou Chongqing and Sichuan in China was consulted by Delphi method. To test the reliability and representativeness of expert consultation indicators such as expert positive coefficient authority coefficient variation coefficient and coordination coefficient were used. The analytic hierarchy process was used to determine the weight of indicators at all levels. Result The effective recovery rates of the 2 rounds of expert consultations were 100% and 92. 8% respectively. In the two rounds the expert authority coefficients were 0. 946 and 0. 979 the coefficients of variation were 0. 135 and 0. 084 and the Kendall harmony coefficients were 0. 241 P<0. 01 and 0. 284 P<0. 01 respectively. The final sensitive indicators included 7 structural indicators 20 process indicators and 4 outcome indicators. Conclusion The clinical management sensitive indicators for cancer nutrition is scientific and reliable which was helpful to improving the clinical quality in cancer nutrition.