The specification of a margin in a non-inferiority trial is often subject to challenge and disagreement. The main reason can be attributed to the subjectivity associated with the current procedures for the non-inferiority margin determination. We introduce the concept of an inferiority index which measures the inferiority level of the study treatment to the control. We establish its link to non-inferiority margin. Through this relationship, the inferiority index can be used as a standard relative to which and it can also provide guidance in the determination of a non-inferiority margin in conjunction with clinical input and historical data. Specifically, this relationship is investigated under survival, normal and binomial distribution. A general theorem is given that establishes the asymptotic normality of a test statistic for the non-inferiority hypothesis defined by the margin so derived. Some examples from non-inferiority trials are used to illustrate the usefulness of inferiority index.