Delirium Screening of the Elderly in the Emergency Department

Delirium Screening of the Elderly in the Emergency Department

This study will refine and pilot test an innovative, emergency department (ED)-based intervention for delirium screening, initial management, and communication with inpatient providers, and examine the impact of this intervention on the rate of documentation of delirium in the electronic health record by ED and inpatient providers. To achieve this, the study will develop and pilot test an intervention, Delirium Screening (SCREEN-ED) in the ED, involving 300 older patients (150 newly enrolled intervention participants, 150 historical controls (comparison group) currently being collected), that has 4 key components: systematic screening for delirium using the Confusion Assessment Method (CAM), informing providers of the screening result, a checklist protocol for initial delirium management based on clinical guidelines, and documentation in the Electronic Health Record (EHR) and communication with inpatient providers of delirium screening results.

The study has two primary aims. The first is to refine and test the feasibility and acceptability of the SCREEN-ED intervention. The second is to examine the impact of SCREEN-ED on rate of delirium documentation and secondary outcomes (length of hospital stay, repeat ED visits and hospitalizations and mortality over 6-months) in the 150 newly enrolled intervention group compared with 150 already collected historical controls.

The investigators will develop and pilot test an intervention, Delirium Screening (SCREEN-ED) in the ED, involving 300 older patients (150 newly enrolled intervention participants, 150 historical controls (comparison group) currently being collected), that has 4 key components: systematic screening for delirium using the CAM, informing providers of the screening result, a checklist protocol for initial delirium management based on clinical guidelines, and documentation in the Electronic Health Record (EHR) and communication with inpatient providers of delirium screening results. The short-term goal is to test the feasibility and acceptability of the intervention, and its immediate impact on documentation of delirium in the EHR. The long-term goal is to utilize the information gleaned from this developmental study to inform a large-scale multi-center randomized clinical trial to test the impact of the SCREEN-ED strategy on clinical outcomes including length of stay, repeat ED visits, rehospitalization, and mortality.

The study will enroll 150 patients to the intervention arm and will screen them for delirium using the CAM. The screening results (positive or negative) will be provided verbally and in the EHR to ED providers, and, in screen positive cases, ED physicians will be provided with a checklist protocol for delirium management that will be embedded in the ED EHR. Additionally, in admitted patients, the screening result will be entered into the inpatient EHR. The study team will review ED and inpatient medical records and compare physician documentation of delirium in SCREEN-ED compared with a historical comparison group currently being collected.

The study will follow all patients (intervention and comparison) through medical record review for 6 months to collect clinical outcomes, which will be examined as secondary outcomes. The study will also examine the acceptability of the intervention with ED and inpatient providers through focus groups prior to the roll-out of SCREEN-ED to refine the intervention components and again in year 2 of the study, after the intervention period is complete, to identify barriers and facilitators to intervention implementation.

Using the same system used to enroll the historical comparison group, patients will be identified using the real-time ED EHR (PulseCheck™), which is populated on arrival with detailed clinical information and links with the hospital EHR. After initial eligibility screen, patients will be approached by trained study staff and introduced to the study. Informed consent will be obtained and study staff will use a standardized Capacity for Informed Consent Instrument that combines capacity assessment questions with interviewer observation and that is used in a prior ED delirium study. If the patient fails the Capacity assessment, a proxy will be asked to provide consent (with patient assent).

The PIs have worked to seamlessly integrate their staff into the workflow of the ED. They expect enrollment to last ~40 weeks, requiring 4 patients be enrolled per week to meet a sample size of 150 in the intervention group (recruitment rate of <10%, assuming 200 patients per week of which 25% are ineligible and 50% refuse).

The pilot SCREEN-ED intervention will test whether screening for delirium in the ED combined with a checklist protocol for initial management is feasible and acceptable and increases EHR documentation of delirium. Consenting participants enrolled in SCREEN-ED will be administered a brief (<10 minutes), standardized questionnaire by a trained study staff member from which the CAM will be scored. Immediately following the interview, the study staff member will score the CAM and the screening result (positive or negative) and, in positive cases, the severity stage of delirium (mild, moderate, severe) will be entered into the EHR and verbally communicated to providers. For patients who screen positive, the checklist protocol for delirium management will automatically appear in the EHR and the ED physicians will be prompted to review and complete all applicable checklist items. In cases where the patient is admitted, the diagnosis and severity will be entered into the EHR and the admitting physicians will be directly informed through existing structured verbal and written communication between the ED and admitting providers.

After the intervention period, the study team will conduct additional focus groups with 10 providers (MDs and nurses), including representatives from inpatient services , who will provide qualitative data on the feasibility of implementation of the intervention and contribute to systematic evaluation using the REAIM (reach, effectiveness, adoption, implementation, maintenance) framework. The investigators will use a widely used focus-group methodology designed to elicit a range of ideas, attitudes, experiences and reactions from providers on the SCREEN-ED strategy. Semi-structured interview scripts will be developed to generate discussion about providers' experiences with SCREEN-ED and changes in the management of older patients as a result of its use. The checklist items will provide the overall structure to the focus group discussions. Information from the focus groups, providing direct accounts of provider experience, will compliment adherence data collected from medical record review. Inpatient providers will describe how checklist protocols (e.g., diagnosis, treatment & communication) impacted the care they provided (e.g., Did they do anything differently knowing the patient was diagnosed with delirium? Did results from initial work-up contribute to their approach to the patient?). It is anticipated that with three focus groups saturation will be reached (i.e., no new ideas are being contributed). However, if this is not the case additional providers will be enrolled and conduct additional groups until saturation is achieved.

Participants enrolled in an ongoing study of delirium that is validating a tool for family members to identify delirium against the CAM will serve as the historical comparison group. This study, which began enrollment in September 2013, has enrolled 120 participants to date by enrolling approximately 2 days a week. The study will enroll intervention patients as soon as possible after completion of comparison group enrollment to minimize temporal separation and enhance comparability. Participants in the comparison group are screened for delirium using the CAM following the same procedures described above for the intervention group. Because the CAM is being used as the gold-standard against which to compare the new screening instrument and was not being administered for screening purposes, CAM results are not communicated to providers, are not incorporated into the EHR, and providers are not given a delirium management checklist. Medical record abstraction elements in the intervention and comparison groups will be identical.

Inclusion Criteria:

• 65 years presenting to the ED between 2pm and 11pm (highest ED census) 7 days a week

Exclusion Criteria:

• Patients who are being evaluated for severe head injury and those who present with delirium tremens due to the difficulty of distinguishing delirium from head injury and because delirium tremens has a distinct etiology, course and outcomes.
• Patients who are aphasic, comatose, terminally ill, deaf, or are non-English speaking.