Table 3.
Results of Resampling and Permutation Analyses of Outbreaks of Species-Groups and Resistance
| Location | No.of Outbreaks |
P Value for Test Statistics |
||||
|---|---|---|---|---|---|---|
| Resampling |
Permutation |
|||||
| Start Dates | Overlap Days | Start Dates | Overlap Days | |||
| Species-Groups | ||||||
| ICU-1 | All | 15 | .379 | .178 | … | … |
| ICU-2 | All | 9 | .885 | .257 | … | … |
| Combined | All | 21 (18 species) | .666 | .116 | … | … |
| Antibiotic resistance | ||||||
| ICU-1 | Amikacin | 3 | .432 | .513 | .638 | .380 |
| Amoxicillin | 1 | … | … | … | … | |
| Ceftazidime | 2 | .324 | 1.000 | .350 | .112 | |
| Cefuroxime | 2 | .664 | 1.000 | .764 | .237 | |
| Ciprofloxacin | 3 | .090 | .075 | .163 | .074 | |
| Co-amoxiclav | 3 | .966 | 1.000 | .959 | .373 | |
| Gentamicin | 4 | .119 | .434 | .253 | .368 | |
| Imipenem | 3 | .288 | .154 | .698 | .397 | |
| Piperacillin/tazobactam | 3 | .023* | .098 | .028* | .014* | |
| Tobramycin | 0 | … | … | … | … | |
| ICU-2 | Amikacin | 2 | .467 | 1.000 | .553 | .342 |
| Amoxicillin | 0 | … | … | … | … | |
| Ceftazidime | 2 | .122 | .183 | .208 | .212 | |
| Cefuroxime | 1 | … | … | … | … | |
| Ciprofloxacin | 1 | … | … | … | … | |
| Co-amoxiclav | 2 | .101 | .076 | .149 | .153 | |
| Gentamicin | 3 | <.005* | .069 | .014* | .313 | |
| Imipenem | 1 | … | … | … | … | |
| Piperacillin/tazobactam | 2 | .122 | .255 | .210 | .209 | |
| Tobramycin | 1 | … | … | … | … | |
| Combined | Amikacin | 4 | .248 | .620 | .718 | .714 |
| Amoxicillin | 1 | … | … | … | … | |
| Ceftazidime | 4 (3 species) | .072 | .043* | .296 | .085 | |
| Cefuroxime | 2 | .673 | 1.000 | .794 | .274 | |
| Ciprofloxacin | 4 | .029* | .078 | .085 | .028* | |
| Co-amoxiclav | 4 | .874 | .482 | .953 | .451 | |
| Gentamicin | 6 (5 species) | .032* | .033* | .148 | .033* | |
| Imipenem | 3 | .284 | .154 | .581 | .305 | |
| Piperacillin/tazobactam | 5 (4 species) | .001* | .012* | .005* | .001* | |
| Tobramycin | 1 | … | … | … | … | |
P values reflect the proportion of resampling or permutation iterations with fewer days between outbreak start dates, and with greater percentage of days with overlapping outbreaks than observed from the data. Outbreaks of resistance were observed in multiple species-groups in 8 of 10 antibiotics and were detected in several species-groups simultaneously in 7 of 10 antibiotics with resistance outbreaks.
A total of 84 resampling and permutation tests were carried out. P < .05 was noted for 18% of these tests (15/84). If the null hypothesis were in all cases true, only 4 (5%) of these tests would have been expected to show effects significant at the P = .05 level.
Abbreviation: ICU, intensive care unit.
*P < .05.