Table 3

 Primary and secondary outcomes (positive differences indicate higher values at the intervention hospital)

Pre-intervention n % (95% CI)Post-intervention number % (95% CI)Adjusted absolute change* (95% CI, p value)Unadjusted absolute change, 95% CI, p value)Adjusted difference in absolute change from baseline† %Unadjusted difference in absolute change from baseline %
Hospital:Study, n = 181Control, n = 60Study, n = 209Control, n = 53
*Absolute change  =  intervention site post implementation−control site post implementation.
†Difference in absolute change from baseline  =  (intervention post implementation–intervention pre implementation) – (control post implementation–control pre implementation).
‡Statistically significant pre intervention difference between intervention and control groups (p = 0.03).
§Excludes deaths and patients with a length of stay >30 days.
Number of patients receiving appropriate antibiotics within 4 h of admission60 33% (26% to 41%)19 32% (20% to 45%)118 56% (49% to 63%)19 36% (23% to 50%)17% (1% to 32%), 0.03520% (5% to 35%), 0.0091719
Number of patients receiving appropriate antibiotics within 8 h of admission90 50% (42% to 57%)27 45% (32% to 58%)157 75% (69% to 81%)38 72% (58% to 83%)−1% (−14% to 14%), 1.03% (−10% to 17%), 0.6−1−2
Number of patients receiving appropriate initial antibiotics136 75% (68% to 81%)37 62%‡ (48% to 74%)182 87% (82% to 91%)41 77% (64% to 88%)10% (−1% to 22%), 0.0810% (−1% to 21%), 0.086−3
Time to first dose of antibiotic given in hospital—number within 4 h of admission91 50% (43% to 58%)30 50% (37% to 63%)150 72% (65% to 78%)27 51% (37% to 65%)15% (2% to 29%), 0.02821% (7% to 35%), 0.0041320
Length of stay§:
Mean (95% CI)7 (6 to 8)7 (5 to 9)7 (6 to 8)8 (6 to 9)p = 0.5−1 mean day1 mean day
Median (95% CI)5 (4 to 6)5 (3 to 7)6 (5 to 7)6 (5 to 7)Cox regression:
30-day mortality34 19% (14% to 26%)13 22% (12% to 35%)38 19% (14% to 25%)5 10% (3% to 21%)5% (17% to −7%), 0.49% (20% to −2%), 0.1−4−12
Data not available5151