• in-hospital mortality;
  • risk factors;
  • older people

OBJECTIVES: To identify prognostic factors that are independently predictive of in-hospital mortality in older patients hospitalized in a medical intensive care unit (MICU).

DESIGN: Prospective cohort study.

SETTING: A MICU in an Italian university hospital.

PARTICIPANTS: Patients aged 65 and older consecutively admitted to the MICU directly from the first-aid unit.

MEASUREMENTS: Upon admission, the following variables were examined: demographics, clinical history (diabetes mellitus, active neoplasm, cognitive impairment, immobilization, pressure ulcers, use of nutritional support, home oxygen therapy), physiopathology (Acute Physiology and Chronic Health Evaluation (APACHE) II), and cognition/function (activity of daily living (ADL), instrumental activity of daily living (IADL), Short Portable Mental Status Questionnaire (SPMSQ)). The vital status of the patient at the end of hospitalization was recorded.

RESULTS: Over a period of 10 months, 659 patients were recruited (mean age ± standard deviation = 76.6 ± 7.5; 352 men and 307 women). There were 97 deaths (14.71%). The following factors proved to be significantly associated with in-hospital mortality: old age, low body mass index (BMI) values, low values of albumin, high scores on APACHE II, functional impairment (ADL, IADL), cognitive impairment (SPMSQ), history of cognitive deterioration, history of confinement to bed, and presence of pressure ulcers. Using multivariate analysis, the following variables were independently predictive of in-hospital mortality: lack of independence in ADLs (P < .001), moderate-to-severe cognitive impairment on SPMSQ (P < .001), score on APACHE II (P = .002), and low BMI values (P = .031).

CONCLUSION: The prognosis of older patients hospitalized in medical intensive care units depends not only on the acute physiological impairments, but also on a series of preexisting conditions, such as loss of functional independence, severe and moderate cognitive impairment, and low BMI.