Delirium (acute confusion) is common and costly in persons with dementia, resulting in longer hospital stays, more complications, and greater functional decline. This research tests the use of the electronic health record, education, and regular feedback to nurses to improve detection and management of delirium. Ultimately, findings will direct ways to improve acute care of this vulnerable population.

Delirium is an acute, reversible change in mental status that occurs in over 40% of persons with dementia. Delirium superimposed on dementia (DSD) leads to increased mortality, increased costs, nursing home placement, early re-hospitalization, and functional decline. Delirium in persons with dementia appears to substantially worsen outcomes in persons with dementia- who are already burdened with functional decline. The purpose of this study is to improve nursing detection and management of delirium in persons with dementia and decrease the duration and patient complications of delirium. To achieve these objectives, a multi-component intervention strategy called, Early Nurse Detection of Delirium Superimposed on Dementia (END DSD) will be tested. END-DSD employs a paired cluster randomized trial (C-RCT) of three hospitals with a total of six inpatient units and 360 hospitalized persons with dementia to achieve these objectives. END DSD intervention consists of 1) Nursing education regarding DSD, 2) Computerized decision support through standardized delirium assessment and management screens via the electronic health record (EHR), 3) an identified unit champion on each intervention unit who will be utilized to persuade other nurses to implement the innovation, and 4) Weekly feedback to the nursing staff to further facilitate assessment and management of delirium. We are testing the following specific aims: A.1: To determine whether the intervention "END DSD" improves nurse detection and management of DSD. A.2: To determine the effect of "END DSD" intervention on patient clinical outcomes, including duration of delirium and rate of psychoactive medications.

The study focuses on a costly and prevalent problem, and utilizes a novel approach that via the EHR will clearly be replicable across settings of care. END-DSD has the potential to significantly improve quality of life and decrease costs of care by: improving the detection of DSD; increasing the use of non-pharmacological management of DSD; shortening the duration of delirium and hospital length of stay; and decreasing the use of inappropriate medications, thus mitigating the complications of DSD. This project builds on over a decade of funded research and clinical practice by the investigators, and brings their unique and collaborative efforts together in an innovative manner to impact the under addressed problem of DSD in hospitalized older adults.

Inclusion Criteria:

• Persons with dementia will be included if they: a) are on one of the selected medical-surgical units and are age 65 years or older; b) have been hospitalized less than 48 hours; and c) meet the criteria for dementia. The study will include minorities and women.

Exclusion Criteria:

• Persons with dementia will be excluded if they have any significant neurological or neurosurgical disease associated with cognitive impairment other than dementia (due to confounding with dementia or DSD), such as:

  • Lewy Body Dementia
  • Huntington's disease
  • Sormal pressure hydrocephalus
  • Seizure disorder
  • Subdural hematoma
  • Head trauma
  • Known structural brain abnormalities

• Nonverbal and unable to communicate due to severe dementia (MMSE=0)

• Aphasia

• Intubation

• Terminal illness (since interviews are required for the study).

• This study will not exclude persons with pre-existing delirium.

• In addition, subjects will not be excluded on the basis of race or gender.