TABLE 5

Summary of Adjusted Multivariate Associations Between Antibiotic Exposure and IBD Development According to Antibiotic Class and Exposure Measurea

Exposure MeasureBroad PenicillinsFluoroquinolonesTetracyclines
HR (95% CI)PHR (95% CI)PHR (95% CI)P
Any exposureb1.72 (1.32–2.24)P < .0013.70 (2.25–6.08)P < .0011.05 (0.65–1.69)P = .85
Per course1.06 (1.02–1.11)P = .0061.07 (0.99–1.16)P = .0640.97 (0.82–1.16)P = .78
No. of coursesb
 0 1 (Reference)1 (Reference)1 (Reference)
 1–2 1.66 (1.26–2.19)P < .0013.71 (2.22–6.19)P < .0011.02 (0.58–1.78)P = .95
 >2 2.42 (1.13–5.18)P = .0233.57 (0.50–25.53)P = .201.13 (0.47–2.75)P = .78
Per week1.02 (0.99–1.04)P = .0611.01 (0.98–1.05)P = .511.00 (0.98–1.02)P = .91
No. of weeksb
 0 1 (Reference)1 (Reference)1 (Reference)
 1–2 1.62 (1.22–2.15)P = .0013.81 (2.24–6.48)P < .0010.75 (0.24–2.33)P = .61
 >22.52 (1.37–4.62)P = .0033.07 (0.76–12.36)P = .111.14 (0.68–1.93)P = .62
Exposure MeasureMetronidazole
HR (95% CI)P
Any exposurec337.78 (37.42–3048.96)P < .001
Per course1.15 (1.07–1.22)P < .001
No. of coursesc
 0 1 (Reference)
 1–2 50.84 (12.71–203.39)P < .001
 >2 35.82 (0.02–60450.42)P = .35
Per week1.04 (1.02–1.07)P < .001
No. of weeksc
 0 1 (Reference)
 1–2 57.25 (13.79–237.67)P < .001
 >2 14.51 (0.14–1530.87)P = .26
  • —, not applicable.

  • a Excludes exposures during the median latency period (3.9 months) before cohort censorship for all subjects. The model included family history of IBD, gender, chronic granulomatous disease, primary sclerosing cholangitis, and socioeconomic deprivation for all listed associations.

  • b Represents the increased hazard of developing IBD with exposure before 1 year of age; hazard decreased with increasing age.

  • c Hazard did not vary according to age.