Skip to main content
. Author manuscript; available in PMC: 2014 Jul 1.
Published in final edited form as: Inflamm Bowel Dis. 2013 Jul;19(8):1671–1680. doi: 10.1097/MIB.0b013e318281f506

Table 5.

Multiple Logistic Regression Results for Stricturing/Penetrating Behavior.

Factors P-values
Age at diagnosis (>16yrs) 0.6526
Gender 0.1788
NOD2 risk allele carriage 0.7334
Cohort 0.0205
Smoking 0.8690
Elevated GM-CSF Ab 0.0001
Ileal Location 0.0481
Perianal location 0.2250
Disease Duration (>3yrs) 0.0146

Cohort* Ileal Location 0.0722

Factors By Cohort OR (95%CI)

Ileal Location Yes vs. No Brisbane 2.48(1.01–6.10)
NIDDK 7.09(3.47–14.49)
Disease Duration >3yrs vs. ≤3 2.46 (1.20–5.07)
Elevated GM-CSF Ab Yes vs. No 2.37 (1.53–3.67)

A multiple logistic regression model was fitted with Stricturing/Penetrating behavior as the response variable. Main effects for the variables age at diagnosis, gender, NOD2 allele, Cohort, Smoking, elevated GM-CSF Ab (≥ 5mcg/mL), Perianal location and disease duration were included in the model automatically, while interaction terms between these main effects were included only if significant after backward elimination. Table 5 displays the p-values for all fitted terms in the final model, as well as the estimated odds ratios for association of Stricturing/Penetrating CD behavior with Ileal CD location, disease duration longer than 3 years, and Elevated GM-CSF-Ab. The Ileal location odds ratio estimates were broken down by cohort as the significant association was defined by interaction of Ileal location with cohort. OR estimates are reported for the individual factors disease duration and elevated GM-CSF Ab as significant interactions for these factors were not identified.