Table 1 Number of significant pairs (binomial test P<1.21 × 10−9) in the temporal analysis given different combinations of patient stratification.

From: Temporal disease trajectories condensed from population-wide registry data covering 6.2 million patients

 

Age and gender

Not age and gender

Type of encounter

62,821

72,832

Not type of encounter

72,937

81,058

  1. Stratifying for either type of hospital encounter or age and gender reduces the total number of significant pairs by 8,121 and 8,226, respectively, compared with no stratification, and by 18,237 when stratifying for both. This indicates that stratification for type of hospital encounter is as important as stratification by age and gender, and that the two contribute independently.
Back to article page