Quantifying the response relationship of drought statistical data and climatee factors is of great significance for ensuring food security. Aiming the agricultural drought-affected rate in 13 grain producing provinces from 1980 to 2019, the responses of the actual disaster conditions to the climatic factors of different months were constructed by using ensemble empirical mode decomposition (EEMD), and then the possible relationship between important climatic factors and large-scale atmospheric-ocean indices (LAOI) were explored. The main results were as follows:(1) The climate drought intensity reflected by medium-short term fluctuations and the drought vulnerability reflected by trend item were extracted from the annual disaster series by using EEMD. (2) The peak difference of IMF-H indicated that the drought intensity in the three northeastern provinces were significantly higher than those of other provinces. (3) The linear relationship between IMF-H and climatic factors showed that summer climate anomalies had the highest contribution to disaster damage, especially the average daily sunshine hours and precipitation in July were selected as the key drought-influencing factors (KDII) in 10 provinces. (4) The time-series diagnosis results of KDII in various provinces showed that the risk of summer drought in the northern part of Yangtze River had increased since 2000. (5) At the time-lag of 0 to 12 months, the LAOI characterizing ENSO and atmospheric low-frequency oscillation in the northern hemisphere, such as Nino SST, Southern Oscillation and North Pacific teleconnection, had the significant impacts on the KDII in the study area. This study could provide scientific basis for regional drought monitoring and early warning.