Fig. 3

Trial sequential analysis for the 9 studies providing the calculation of correlativity between DM and IPF. The diversity-adjusted required information size (RIS) was 47,415 patients, with α = 5% (two-side), power of 80%, relative risk reduction of 35% and incidence of 3% in the control arm. The cumulative Z curve got across both the conventional boundary (green line) and the trial sequential monitoring boundary (blue line), indicating that the information size was sufficient and the outcome was reliable