Objective To create an appropriate demand-based feeding mode for full-term newborns hospitalized in unaccompanied ward, to preliminarily put the feeding mode into practice, and to provide evidence for carrying out demand-based feeding in neonatal units. Methods We searched domestic and foreign databases to obtain evidence from clinical practice guidelines, best practices, evidence summaries, systematic reviews and expert consensus.Then a demand-based feeding mode for full-term newborns hospitalized in unaccompanied ward was drafted and subjected to Delphi expert consultation.Nineteen full term hospitalized newborns were included and randomly divided into an experimental group of 10 cases and a control group of 9 cases.The experimental group was subjected to demand-based feeding mode, while the control group was fed on schedule with fixed volume of milk. Results The demand-based feeding mode for full term newborns included such 3 dimensions as cues to start feeding, cues of infant satiety, and safe demand-based feeding, totaling 25 items.A flow chart of demand-based feeding was drawn.The experimental group had shorter period of average daily heart rate ＞160 beats/min, and higher weight gain,than the control group (both P＜0.05）. Conclusion The demand-based feeding mode for full term newborns is beneficial for maintaining stable physiological indicators and promoting weight gain in newborns.