Table 6.
Statistical evaluation of the resistance to antibiotics, showing significant differences among treatments, days, and interactions between treatments and days.
| Antibiotic | Effect of day | Effect of treatment | Effect of interactions |
|---|---|---|---|
| Amoxicillin—clavulanic acid*** | 38>17 (P = 0.036) | AB>NC, CC, FA, SA (P < 0.0001) | n.a. |
| Ampicillin** | 38>17 (P < 0.0001) | AB>NC, CC, FA, SA (P < 0.0001) | n.a. |
| Cefoxitin*** | P = 0.36 | AB>NC, CC, FA, SA; SA<CC(P < 0.0001) | n.a. |
| Ceftriaxone*** | P = 0.69 | AB>NC, CC, FA, SA (P < 0.0001) | n.a. |
| Chloramphenicol** | P = 0.65 | P = 0.49 | n.a. |
| Gentamicin* | 38>17 (P < 0.0001) | P = 0.70 | P = 0.17 |
| Sulfisoxazole* | 38>17 (P < 0.0001) | P = 0.11 | P = 0.75 |
| Tetracycline* | 38>17 (P < 0.0001) | P = 0.04**** | P = 0.30 |
| Trimethoprim-sulfamethoxazole* | 38>17 (P = 0.0002) | P = 0.16 | P = 0.48 |
NC, negative control without E. coli challenge; CC, E. coli challenge control; AB, ampicillin; FA, feed additive based on organic acids; SA, multistrain synbiotic; n.a., not available.
*GLIMMIX procedure and multiple comparisons of the resistance results adjusted according to the Tukey–Kramer test at a significance level of 5%.
**GLIMMIX procedure without interaction terms and in-depth analysis of treatment effects by day according to the Tukey–Kramer test.
***Interpretation by contingency analysis (dependencies in contingency tables of treatments were tested by Pearson's Chi-squared test, and tests on subgroups were based on the Bonferroni correction to comply with the type I error rate).
****In-depth analysis showed significant differences between AB-SA on day 17 only.