Routinely assessed biochemical markers tested on admission as predictors of adverse outcomes in hospitalized elderly patients

• Published in: Hospital practice (1995) 2012-02, Vol.40 (1), p.193-201
• Main Authors: Brown, Siobhan H, Flint, Katie, Storey, Allison, Abdelhafiz, Ahmed H
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Biomarkers - blood; Female; Hospital Mortality; Hospitalization - statistics & numerical data; Humans; Length of Stay - statistics & numerical data; Male; Outcome Assessment (Health Care); Patient Admission - statistics & numerical data; Patient Discharge - statistics & numerical data; Prospective Studies; Treatment Outcome
• ISSN: 2154-8331; 2377-1003
• DOI: 10.3810/hp.2012.02.960

To explore whether routinely assessed biochemical markers tested on admission will predict 3 predefined adverse outcomes for hospitalized elderly patients: discharge to a long-term care facility, in-hospital mortality, and prolonged hospital length of stay (> 14 days). A prospective observational study of elderly patients (aged ≥ 75 years) admitted to an acute-care geriatric ward over a 6-month period. Patients were assessed on admission and baseline characteristics were collected. Activities of daily living were assessed by the Barthel Index and cognitive function by the abbreviated mental test. Results from biochemical markers tested on admission were downloaded from the pathology laboratory database using patient details. Patients were followed-up with until discharge or in-hospital mortality. A total of 392 patients formed the study population. Mean (standard deviation) age was 83.2 (± 5.5) years and 283 (72%) patients were men. Thirty-eight (10%) patients were discharged to a long-term care facility, 134 (34%) had a prolonged hospital length of stay, and 33 (8%) died in the hospital. Results from testing 5 biochemical markers independently predicted in-hospital mortality: hypoalbuminemia (adjusted odds ratio [OR], 2.5; 95% CI, 0.9-6.7; P = 0.04), low total cholesterol level (adjusted OR, 2.9; 95% CI, 1.3-6.3; P = 0.01), hyperglycemia (adjusted OR, 2.9; 95% CI, 1.2-7.4; P = 0.02), high C-reactive protein level (adjusted OR, 4.2; 95% CI, 1.3-13.4; P = 0.01), and renal impairment (adjusted OR, 3.8; 95% CI, 1.7-8.7; P = 0.002). High C-reactive protein level independently predicted prolonged hospital length of stay (OR, 1.7; 95% CI, 1.1-2.9; P = 0.03). Hypoalbuminemia predicted discharge to a long-term care facility independent of confounding factors except for physical dysfunction (OR, 2.4; 95% CI, 1.1-5.1; P = 0.03). Significance was reduced after adjustment for Barthel Index score (OR, 1.9; 95% CI, 0.9-4.1; P = 0.08). Testing of routinely assessed biochemical markers on admission predicted adverse hospital outcomes for elderly patients. Their inclusion in a standardized prediction tool may help to create interventions to improve such outcomes.