Table 2.
Poisson Mixed Effects Model for Intensive Care Unit–Acquired Bloodstream Infection
| Variable | Incidence Rate Ratio (95% Confidence Interval) | P Value |
|---|---|---|
| Patient-level characteristics | ||
| Age (per 10 years) | 0.95 (.94, .96) | <.001 |
| Sex | … | <.001 |
| Male | 1.00 | … |
| Female | 0.89 (.86, .92) | … |
| Immunosuppression due to treatment | … | <.001 |
| No | 1.00 | … |
| Yes | 1.18 (1.10, 1.27) | … |
| Immunosuppression due to illness | … | .001 |
| No | 1.00 | … |
| Yes | 1.15 (1.06, 1.25) | … |
| Oncology reason for admission | … | <.001 |
| Nononcology | 1.00 | … |
| Tumor or malignancy | 0.81 (.74, .89) | … |
| Reason for admission | … | <.001 |
| Respiratory | 1.00 | … |
| Cardiovascular | 1.22 (1.14, 1.30) | … |
| Cardiovascular surgical | 1.19 (1.09, 1.30) | … |
| Dermatological | 2.39 (2.08, 2.75) | … |
| Dermatological surgical | 1.34 (1.05, 1.71) | … |
| Endocrine, metabolic, thermoregulation | 0.95 (.85, 1.05) | … |
| Endocrine, metabolic, thermoregulation surgical | 0.89 (.57, 1.41) | … |
| Gastrointestinal | 1.63 (1.53, 1.73) | … |
| Gastrointestinal surgical | 1.32 (1.25, 1.41) | … |
| Genitourinary | 1.26 (1.17, 1.37) | … |
| Genitourinary surgical | 0.99 (.82, 1.19) | … |
| Hematological/immunological | 1.66 (1.45, 1.89) | … |
| Hematological/immunological surgical | 1.10 (.49, 2.45) | … |
| Musculoskeletal | 1.31 (1.15, 1.50) | … |
| Musculoskeletal surgical | 1.04 (.88, 1.23) | … |
| Neurological (including eyes) | 1.09 (1.01, 1.17) | … |
| Neurological (including eyes) surgical | 1.05 (.94, 1.17) | … |
| Respiratory surgical | 0.98 (.83, 1.15) | … |
| Intensive Care National Audit & Research Centre Physiology Score (per point) | 1.03 (1.03, 1.03) | <.001 |
| Financial year of admission | … | <.001 |
| 2007–2008 | 1.00 | … |
| 2008–2009 | 0.74 (.69, .79) | … |
| 2009–2010 | 0.53 (.50, .57) | … |
| 2010–2011 | 0.41 (.38, .44) | … |
| 2011–2012 | 0.29 (.27, .32) | … |
| 2012–2013 | 0.25 (.23, .27) | … |
| 2013–2014 | 0.24 (.22, .26) | … |
| 2014–2015 | 0.21 (.20, .23) | … |
| 2015–2016 | 0.21 (.19, .23) | … |
| ICU-level characteristics | ||
| Number of ICU beds (per additional bed) | 1.02 (1.01, 1.02) | <.001 |
| SHA/country | … | <.001 |
| East Midlands | 1.00 | … |
| East of England | 0.48 (.38, .60) | … |
| London | 1.02 (.84, 1.25) | … |
| North East | 0.64 (.49, .82) | … |
| North West | 0.64 (.52, .80) | … |
| South Central SHA | 1.37 (1.07, 1.77) | … |
| South East Coast | 0.51 (.40, .65) | … |
| South West | 0.51 (.40, .64) | … |
| West Midlands | 0.49 (.38, .61) | … |
| Yorkshire and the Humber | 0.83 (.67, 1.04) | … |
| Northern Ireland | 0.55 (.41, .72) | … |
| Wales | 0.46 (.34, .62) | … |
Data from 617 336 patients staying longer than 48 hours on the ICU.
Abbreviations: ICU, intensive care unit; SHA, Strategic Health Authority.