Table 4. Multivariable Logistic Regression Output for CLABSI.
| Variable | Adjusted Risk Ratio (95% CI) | |
|---|---|---|
| Model 1a | Model 2b | |
| NHSN CLD with concurrence, %c | ||
| 0 | 1 [Reference] | 1 [Reference] |
| ≤66 | 0.70 (0.48-1.08) | 0.58 (0.36-0.92) |
| >66 | 1.62 (1.10-2.33) | 0.73 (0.45-1.16) |
| NHSN CLDs, per 1-unit increased | 1.08 (1.07-1.10) | NA |
| Total CLDs, per 1-unit increasee | NA | 1.06 (1.05-1.07) |
| Age, per 1-y increase | 1.00 (0.99-1.01) | 1.00 (0.99-1.01) |
| Male sex | 1.36 (0.99-1.89) | 1.35 (0.98-1.86) |
| Propensity scoref | 1.23 (0.23-6.24) | 1.25 (0.23-6.35) |
Abbreviations: CLABSI, central line–associated bloodstream infection; CLD, central line–day; CVC, central venous catheter; NA, not applicable; NHSN, National Healthcare Safety Network.
Adjusted for age, sex, propensity score, NHSN CLD, and proportion of NHSN CLD with concurrence (categorized as 0%, ≤66%, or >66%).
Adjusted for age, sex, propensity score, total CLD, and proportion of NHSN CLD with concurrence (categorized as 0%, ≤66%, or >66%).
Concurrence was defined as any 2 CVCs present for 2 or more of the same days. Percentage of concurrence was calculated as ([total CLD / NHSN CLD] / NHSN CLD) × 100 (median [interquartile range] for patients with concurrent CVCs, 66.7% [40.0%-93.8%]).
Calculated as 1 day for each day any CVC is present (ie, a patient can contribute a maximum of 1 NHSN CLD per calendar day, regardless of the number of CVCs present).
Calculated as each day for each CVC (eg, a patient with 2 CVCs for for 2 days would be considered to have 4 total CLDs).
Calculated as the output of a logistic regression model in which the outcome is any concurrence of CLABSI; the risk factors include Charlson Comorbidity Index score and receipt of hemodialysis, chemotherapy, or total parenteral nutrition.