Effectiveness and cost-effectiveness of Crew Resource Management training to improve patient safety at intensive care units.

Objective:
To assess the (cost)effectiveness of CRM training to improve
patient safety at intensive care units (ICU).

Design:
In a paired cluster-randomized trial with one pre-test and one
post-test measurement 3 ICUs with CRM training will be compared with 3
ICUs without training.

Intervention:
The intervention group will take part in CRM training, in
which all icu team members are educated about the limitations of human
performance, nontechnical skills, patient safety culture and leadership
in order to improve detection and management of errors. Furthermore,
participants are trained to assess personal and peer behaviour. The
training consists of an e-learning module (1 hour), a 2-day CRM training
with two half-day comeback sessions are given by two trainers
(behavioural scientist + clinician) to all ICU workers. In addition, a
change-team will be formed to support and coordinate implementation of
CRM.

Outcomes:
The primary outcome is the number of adverse events per
patient day assessed by means direct structured observations and
ICU-specific adverse outcomes assessed by routine administrative data.
Secondary outcomes include patient safety culture, percentage of errors
managed effectively and attitudes towards CRM, assessed by means of
questionnaires and direct structured observations after 0 and 10 months.
Sample size and data-analysis: Sample size calculation is based on
incidence rates transformed to z values of the incidences of adverse
events per patient day. As the paired design compensates for the
inefficiencies of the clustering, no correction for clustering was
applied. If we assume that the incidence of adverse events at ICU
departments is 10%, the control group will reduce adverse events to 9%
and the intervention group to 5%, alpha is 0.05, beta is 0.80, than we
need to observe 1100 patient days per group per measurement moment to
detect a difference. Differences in change in incidence of adverse
events during follow-up between both groups will be assessed using
incidence rates transformed to z values of the incidence percentages.
Changes in patient safety culture, attitude and teamwork behaviour will
be described and tested using t-test and Chi-square tests.

Economic evaluation:
A cost-effectiveness analysis will be performed to
assess the incremental costs per prevented adverse event. In addition a
cost-benefit analysis will be performed to compare incremental costs of
training with incremental costs of ICU stay.

Time schedule:
The study will take 3 years, and measurements will be
performed at 0 and 12 months.

By improving non-clinical skills, such as communication, task coordination, collaboration and leadership of ICU staff, the ICU teams will improve their ability to detect and trap errors and threats and to diminish their harmfull consequences to patients.

Before and 12 months after the training all outcomes will be assessed.

1. 3 intervention icu's and 3 matched control icu's;

2. All staff members and all participants.