Table 3. Evaluation of potential effect-modification by mode of delivery on the association between redemption of antibiotics within the first two years of life and subsequent onset of childhood type 1 diabetes (2 to 14 years, both included).
| Exposed children | Un-exposed childrena | Partly adjusted | Fully adjusted | |
|---|---|---|---|---|
| Number of childhood type 1 diabetes | Modelb | Modelc | ||
| (person-years at risk) | HR (95% CI) | HR (95%) | ||
| Model evaluating the association with any type of antibiotics in | ||||
| Children born by vaginal delivery | ||||
| Exposed vs. un-exposeda | 903 (3,555,860) | 357(1,405,814) | 0.99 [0.88;1.12] | 0.99 [0.87;1.12] |
| Children born by intrapartum cesarean section | ||||
| Exposed vs. un-exposeda | 97 (345,814) | 22 (117,784) | 1.49 [0.94;2.37] | 1.48 [0.92;2.38] |
| Children born by prelabor cesarean section | ||||
| Exposed vs. un-exposeda | 107 (364,335) | 17 (116,462) | 2.00 [1.20;3.33] | 1.91 [1.14;3.20] |
| Model evaluating the association with narrow-spectrum antibiotics in | ||||
| Children born by vaginal delivery | ||||
| Exposed vs. un-exposeda | 657 (2,600,035) | 603 (2,361,639) | 0.98 [0.87;1.09] | 0.99 [0.88;1.10] |
| Children born by intrapartum cesarean section | ||||
| Exposed vs. un-exposeda | 69 (254,976) | 50 (208,622) | 1.11 [0.77;1.60] | 1.19 [0.81;1.72] |
| Children born by prelabor cesarean section | ||||
| Exposed vs. un-exposeda | 77 (268,268) | 47 (212,529) | 1.27 [0.89;1.83] | 1.25 [0.87;1.80] |
| Model evaluating the association with broad-spectrum antibiotics in | ||||
| Children born by vaginal delivery | ||||
| Exposed vs. un-exposeda | 651 (2,494,360) | 609 (2,467,315) | 1.06 [0.95;1.18] | 1.05 [0.94;1.18] |
| Children born by intrapartum cesarean section | ||||
| Exposed vs. un-exposeda | 80 (249,034) | 39 (214,564) | 1.77 [1.21;2.60] | 1.70 [1.15;2.51] |
| Children born by prelabor cesarean section | ||||
| Exposed vs. un-exposeda | 82 (266,914) | 42 (213,883) | 1.57 [1.08;2.28] | 1.63 [1.11;2.39] |
Hazard ratio (HR) for type 1 diabetes diagnosed in children before 15 years of age are reported with 95% confidence interval (95% CI). Estimates are calculated from the interaction-term between mode of delivery and redemption of antibiotics within the first two years of life in three models evaluating the effect of: redemption of any kind of antibiotics, redemption of narrow-spectrum antibiotics and redemption of broad-spectrum antibiotics, respectively, during the first two years of life.
aDefined as: no antibiotics, no narrow-spectrum antibiotics, no broad-spectrum antibiotics, respectively
bPartly adjusted model adjusted for: Year of birth, mode of delivery,redemption of antibiotics within the first two years of life, an interaction-term between mode of delivery and redemption of antibiotics within the first two years of life.
cFully adjusted model adjusted for variables above and: sex, parity, maternal and paternal age and educational level at childbirth, maternal and paternal type 1 diabetes diagnosed before childbirth