Skip to main content
. 2015 Sep 10;5:13925. doi: 10.1038/srep13925

Table 4. Multiple regression analyses of elderly patients with and without acute kidney injury, with in-hospital complications as the dependent variables.

Results Odds ratio 95% Confidence Interval p value
Model 1a – All in-hospital complications
 AKI at presentation 3.51 1.32–9.3 0.01
 SBP at presentation 0.99 0.98–1.0 0.08
Model 1b – All in-hospital complications
 AKI KDIGO stage 1 3.44 1.11–10.7 0.03
 AKI KDIGO stage 2 6.39 1.01–51.2 0.05
 AKI KDIGO stage 3 6.39 1.01–51.3 0.05
Model 2 – New-onset hospital-associated infections
 DM as comorbidity 2.35 1.09–5.07 0.03
 AKI at presentation 1.99 0.92–4.29 0.08
Model 3a – Incident electrolyte imbalances
 AKI at presentation 7.1 3.19–15.8 <0.01
Model 3b – Incident electrolyte imbalances
 AKI KDIGO stage 1 5.37 2.1–13.7 <0.01
 AKI KDIGO stage 2 7.08 1.8–27.9 <0.01
 AKI KDIGO stage 3 39.9 4.4–362.9 <0.01

Models 1 and 3 include variables from demographic data, all comorbidities, Charlson comorbidity index, vital signs, admission diagnoses, and AKI (with vs. without for 1a and 3a) or AKI (divided into tertiles for 1b and 3b).

Model 2 includes variables from demographic data, all comorbidities, Charlson comorbidity index, vital signs, admission diagnoses, and AKI (with vs. without).

Abbreviations: AKI, acute kidney injury; DM, diabetes mellitus; KDIGO, Kidney Disease Initiative Global Outcome; SBP, systolic blood pressure.