Other bulletins in this series include:

Breast Surgery

Wednesday, 22 February 2017

A Direct Observation Checklist to Measure Respect and Dignity in the ICU

A Direct Observation Checklist to Measure Respect and Dignity in the ICU
Carrese, J A et al
Critical Care Medicine: February 2017 - Volume 45 - Issue 2 - p 263–270

Objective: Treating patients and family members with respect and dignity is a core objective of health care, yet it is unclear how best to measure this in the ICU setting. Accordingly, we sought to create a direct observation checklist to assess the “respect and dignity status” of an ICU. Design: A draft checklist based on previous work was iteratively revised to enhance accuracy and feasibility. Setting: Seven ICUs within the Johns Hopkins Health System. Subjects: A total of 351 patient-clinician encounters with 184 different patients. Interventions: Four study team members pilot tested the checklist between January and August 2015. 
Measurements and Main Results: Standard psychometric analyses were performed. The direct observation checklist exhibits strong content and face validity as well as high reliability and internal consistency. All items load on one factor that supports the unidimensionality of the total index. Furthermore, concurrent validity of the direct observation checklist is demonstrated by statistically significant differences in mean scores between ICUs, between types of clinicians, and between patients’ clinical status and mood. 
Conclusions: We rigorously developed, pilot tested, and analyzed a direct observation checklist designed to assess the extent to which patients and families in the ICU setting are treated with respect and dignity. Future research should validate this checklist in other settings and compare its results with other measures. Data gathered about individual items on the direct observation checklist could be used to target areas for training and education; doing so should help facilitate more respectful treatment of patients and their families.

No comments: