Assessing Recovery from Delirium: An International Survey of Healthcare Professionals Involved in Delirium Care.

Abstract

A crucial part of delirium care is determining if the delirium episode has resolved. Yet, there is no clear evidence or consensus on which assessments clinicians should use to assess for delirium recovery. To evaluate current opinions from delirium specialists on assessment of delirium recovery. Online questionnaire-based survey distributed internationally to healthcare professionals involved in delirium care. The survey covered methods for assessing recovery, the importance of different symptom domains for capturing recovery, and local guidance or pathways that recommend monitoring for delirium recovery. Responses from 199 clinicians were collected. Respondents were from the UK (51%), US (13%), Australia (9%), Canada (7%), Ireland (7%) and 16 other countries. Most respondents were doctors (52%) and nurses (27%). Clinicians worked mostly in geriatrics (52%), ICUs (21%) and acute assessment units (17%). Ninety-four percent of respondents indicated that they conduct repeat delirium assessments (i.e., on ≥2 occasions) to monitor delirium recovery. The symptom domains considered most important for capturing recovery were: arousal (92%), inattention (84%), motor disturbance (84%), and hallucinations and delusions (83%). The most used tool for assessing recovery was the 4 'A's Test (4AT, 51%), followed by the Confusion Assessment Method (CAM, 26%), the CAM for the ICU (CAM-ICU, 17%) and the Single Question in Delirium (SQiD, 11%). Twenty-eight percent used clinical features only. Less than half (45%) of clinicians reported having local guidance that recommends monitoring for delirium recovery. The survey results suggest a lack of standardisation regarding tools and methods used for repeat delirium assessment, despite consensus surrounding the key domains for capturing delirium recovery. These findings emphasise the need for further research to establish best practice for assessing delirium recovery.