This will be the first large cohort study to define the epidemiology of and identify modifiable risk factors for long-term CI and functional deficits of ICU survivors. The investigators will measure the independent contribution of risk factors such as delirium and exposure to sedative and analgesic medications to the incidence of long-term CI, controlling for established risk factors (e.g., age, pre-existing CI, and apoE genotype). Defining the contributions of these risk factors will make it possible to develop preventive and/or treatment strategies to reduce the incidence, severity and/or duration of long-term CI and improve functional recovery of patients with acute critical illness.

Advances in critical care have led to improved survival among those admitted to intensive care units (ICUs). However, survival is lower among those who develop ICU delirium, and the quality of life among survivors may be affected by post-ICU long-term cognitive impairment (CI) that lasts months to years. Long-term CI has been studied predominantly following cardiopulmonary bypass. In the much larger group of medical and general surgical ICU patients, the extent of this problem and its relationship to health-related quality of life is poorly characterized. Evidence from 6 pilot cohorts (including the investigators') totaling ~300 patients suggests that an astonishing 30% to 80% of ICU survivors experience long-term CI functionally equivalent to mild/moderate dementia although it may not be progressive (and thus will be referred to as long-term CI). Interestingly, this cognitive impairment arises independent of severity of illness, and older patients appear particularly prone. The investigators' work and the work of others have shown that delirium is a major independent risk factor for impaired cognitive function at hospital discharge and increased mortality at 6 months. While it is not clear whether delirium itself is injurious to the brain or is simply a marker of brain injury, it is clear that the onset of delirium in the ICU should not be considered innocuous; rather, it may be a determinant of long-term CI and health-related quality of life. Having spent the last 8 years studying delirium and drug exposure during acute phases of critical illness and long-term CI after hospitalization, the investigators are thoroughly prepared to continue the next phases of investigation in VA (Department of Veterans Affairs) patients, many of whom are older and disproportionately at risk for adverse outcomes following ICU care.

Inclusion Criteria:

• Patients will be included if they are adult patients in a medical and/or surgical ICU receiving treatment for any of the following: respiratory failure or cardiogenic or septic shock.

Exclusion Criteria:

Patients who meet the inclusion criteria will be excluded if they meet any of the following criteria:

• Cumulative ICU time > 5 days in the past 30 days, not including the current ICU stay, as this might create a state of flux regarding patients' cognitive baseline.
• Severe cognitive or neurodegenerative diseases that prevent a patient from living independently at baseline, including mental illness requiring institutionalization, acquired or congenital mental retardation, known brain lesions, traumatic brain injury, cerebrovascular accidents with resultant moderate to severe cognitive deficits or ADL dependency, Parkinson's disease, Huntington's disease, severe Alzheimer's disease or dementia of any etiology.
• ICU admission post cardiopulmonary resuscitation with suspected anoxic injury
• An active substance abuse or psychotic disorder, or a recent (within the past 6 months) serious suicidal gesture necessitating hospitalization. This exclusion that will enrich follow-up rates by avoiding patients with whom it is particularly challenging to maintain long-term contact.
• Blind, deaf, or unable to speak English, as these conditions would preclude our ability to perform the follow-up evaluation interviews.
• Overly moribund and not expected to survive for an additional 24 hours and/or withdrawing life support to focus on comfort measures only.
• Prisoners.
• Patients who live further than 200 miles from the study site and who do not regularly visit the study site area.
• The onset of the current episode of respiratory failure, cardiogenic shock, or septic shock was >72 hours ago.
• Patients who have had cardiac bypass surgery within the past 3 months (including the current hospitalization).
• Patients who are homeless and have no secondary contact person available. This exclusion will enrich follow-up rates by avoiding patients with whom it is particularly challenging to maintain long-term contact.

Medical and Surgical ICUs