Table 1.
Characteristics of the Intensive Care Unit (ICU) Population, According to Study Period and Group.*
| Variable | 12-Mo Baseline Period (N = 48,390) | 18-Mo Intervention Period (N = 74,256) | ||||
|---|---|---|---|---|---|---|
| Group 1 | Group 2 | Group 3 | Group 1 | Group 2 | Group 3 | |
| Admission with ICU stay (no.) | 15,816 | 15,218 | 17,356 | 23,480 | 24,752 | 26,024 |
| Attributable ICU patient-days (no.) | 63,135 | 57,418 | 69,668 | 88,222 | 92,978 | 101,603 |
| ICU type (no.)† | ||||||
| Medical | 3 | 5 | 5 | 3 | 5 | 5 |
| Surgical | 1 | 2 | 6 | 1 | 2 | 6 |
| Mixed medical and surgical | 19 | 14 | 18 | 19 | 15 | 17 |
| Hospital stay (days) | ||||||
| Median | 7 | 7 | 8 | 7 | 7 | 7 |
| Interquartile range | 5–12 | 5–12 | 5–12 | 5–12 | 5–12 | 5–12 |
| ICU stay (days) | ||||||
| Median | 3 | 3 | 3 | 3 | 3 | 3 |
| Interquartile range | 2–5 | 2–5 | 2–5 | 1–5 | 2–5 | 2–5 |
| Age (yr) | ||||||
| Median | 65 | 66 | 65 | 65 | 66 | 65 |
| Interquartile range | 52–77 | 53–77 | 51–77 | 52–77 | 53–77 | 52–77 |
| Female sex (%)‡ | 47.2 | 47.2 | 47.9 | 47.6 | 47.2 | 47.5 |
| Nonwhite race (%)§ | 25.9 | 22.1 | 30.8 | 25.9 | 23.5 | 31.7 |
| Coexisting condition (%) | ||||||
| Diabetes | 31.3 | 33.0 | 30.7 | 31.8 | 32.7 | 31.5 |
| Renal failure | 20.0 | 20.4 | 19.0 | 20.3 | 22.2 | 19.7 |
| Cancer | 10.4 | 10.8 | 14.1 | 9.9 | 10.8 | 13.0 |
| Liver failure | 3.4 | 4.4 | 3.9 | 4.0 | 4.1 | 4.2 |
| History of MRSA infection (%)¶ | 10.2 | 11.5 | 10.6 | 9.7 | 11.1 | 3.9 |
| Surgery during hospitalization (%) | 40.5 | 38.6 | 47.5 | 38.7 | 37.7 | 46.2 |
Group 1 implemented methicillin-resistant Staphylococcus aureus (MRSA) screening and isolation; group 2, targeted decolonization (i.e., screening, isolation, and decolonization of MRSA carriers with chlorhexidine and mupirocin); and group 3, universal decolonization (i.e., no screening and all patients underwent decolonization). At baseline, there were no significant between-group differences. For additional details, see the Supplementary Appendix.
Differences in the number of ICUs in the groups between the baseline and intervention periods reflect the fact that one ICU in group 2 opened during the trial and one in group 3 closed.
Data were missing for eight patients.
Race was determined from electronic administrative data at each hospital.
A history of MRSA infection was identified with the use of all available screening and clinical cultures, with the history defined as MRSA carriage documented by the Hospital Corporation of America during the period from 1 year before admission to day 2 of the ICU stay. Data from group 3 during the intervention period are not comparable to data from the other groups because universal decolonization, without screening, was performed for all patients in this group. As the intervention progressed, patients who were readmitted to the ICU were less likely to be identified as MRSA-positive.