Feasibility of prediction protocols for EP in PUL
In order to evaluate the impact of the complexity of each protocol on clinical work efficiency, we propose to introduce the concept of “the rate of lost”, which has never been put forward before. The rate of lost refers to the ratio of the number of patients who have not received all of the diagnostic approaches included in protocols before follow-up, management or final diagnosis to the number of patients with complete data. Table 2 shows the number of examination items and visits required by each protocol. When it comes to the relationship between the sum of the number of visits and the number of inspection items and the rate of lost, table 2 shows that the average rate of lost in each category. The average rate of lost is 11.19% (95%CI 4.67-17.72) for the sum of 3, 18.63% (95%CI 9.67-17.71) for the sum of 5 and 9.29% (95%CI cannot be counted) for the sum of 6.
We also introduce average production utility, an economic concept, to quantify the cost of improving prediction accuracy. It was used to evaluate the effectiveness of each protocol under the expenditure of unit cost. Considering the different economic development level, medical insurance policies and cultural concepts of different countries and regions, the sum of the number of visits and the number of examinations required by the protocols were used to define the total cost, and AUC to define the total benefit. Table 3 shows the average production utility of the protocols. Among them, M1 is 0.181, progesterone cut-off is 0.242, M4 is 0.174, hCG ratio is 0.165, hCG cut-off is 0.195, and M6 is 0.157.