Table 3.

Statistics of diagnosis trajectory analysis using the set of HIRA.

	Features	Frequency
Trajectorya  = the first and followed nodes linked via edges	No. of trajectories	405
Node  = Diagnosed patients as a disease i	Total type of diagnosis	604
	The 1st diagnosis	170
	Patients in the 1st diagnosis	1,023,171
	Fatal outcomes of diagnosis	2 134
Edge (Directed acyclic edge)  = Sequence between diagnosis i and j (FDR < 0.1)b	Subsequent diagnoses after the 1st diagnosis	80,316
	Mean of steps diagnoses in a trajectory	3.24 ± 1.12

^aThe trajectory consists of the first node, representing a set of patients diagnosed with disease i, succeeding nodes are connected via a directed acyclic edge, representing subsequent diagnoses that occur more frequently than randomly.

^bAll the presented edges were statistically significant (Relative association (RA) for the co-occurrence of diagnosis i and j > 1 and FDR of the binomial test for the co-diagnosis of diagnosis i and j < 0.1; FDR of the binomial test for the sequential directionality of diagnosis dates <0.1). The details of the diagnosis trajectory model are presented on the following website (https://www.youtube.com/watch?v=jJMds31-e2g).

Bold characters to emphasize the numbers of the table.