Predictive models of disease burden at diagnosis in persons with adult-onset ulcerative colitis using health administrative data

Table 1 Administrative variables tested in multivariable models

1. Age at Diagnosis
2. Sex
3. Hospitalization for colitis within 30 days of diagnosis
4. Hospitalization for colitis beyond 30 days following diagnosis^a
5. Number of hospitalizations for colitis beyond 30 days^a
6. Total number of hospitalization days for colitis^a
7. Non-hospitalization emergency department (ED) visit for colitis^a
8. Number of non-hospitalization ED visits for colitis^a
9. IBD-related physician encounter after diagnosis^a
10. Number of IBD-related physician encounters after diagnosis^a
11. IBD-related gastroenterologist (GI) encounter after diagnosis^a
12. Number of IBD-related GI encounters after diagnosis^a
13. IBD-related general surgeon (GS) encounter after diagnosis^a
14. Number of IBD-related GS encounters after diagnosis^a
15. Lower endoscopy after initial diagnostic endoscopy^a
16. Number of lower endoscopies after initial diagnostic endoscopy^a
17. Blood transfusion after diagnosis^a
18. Number of blood transfusion occurrences after diagnosis^a
19. Hospitalization associated with colitis-related complication (bowel perforation, peritonitis, peritoneal abscess, megacolon, venous thromboembolism or Clostridium difficile colitis)^a
20. IBD-related death or colectomy^a

^aVariables assessed over 1-, 2-, and 3-year intervals following diagnosis – the definition that demonstrated the greatest bivariate association with an outcome was retained for testing in the multivariable model

ISSN: 1471-230X