Can We Improve the Quality of Care at the End of Life in Israel?

This study will evaluate whether end-of-life simulation training of critical care physicians and nurses throughout Israel using standardized patients improves communication and interpersonal skills leading to improvements in ICU care process, teamwork and family satisfaction.

Design- Simulation training will include approximately 10-15 staff members from 4-5 ICUs during the first 12 months of the study (intervention group) who will be compared using a stepped wedge design to staff members from another 4-5 ICUs who will not receive initial simulation training (control group) in the first year but will receive training during the second year. To evaluate whether the simulation training with simulated standardized patients improves ICU professional's communication and interpersonal skills, a baseline evaluation before the training and an evaluation after the training will be required. Training using 9 essential skills for communication and interpersonal relationships that the study Steering Committee chose include: How to elicit patient preferences (including patient vs. family goals of care)? How does one listen? How to be more compassionate, show empathy and respect (including how to address emotions)? How to handle unrealistic patient/family expectations (including a family requesting "that everything be done")? How to handle prognostic uncertainty? How to handle decision making? How to handle conflicts between siblings or within the team? Physicians and nurses working together, and Becoming knowledgeable and understanding the Israeli Terminally Ill Law.

Simulation sessions- The simulation sessions include 6 scenarios. The simulation training is composed of several teaching sessions interspersed between clinical scenarios with simulated standardized patients (actors) performed with the ICU doctors and nurses and several subsequent debriefing sessions. The entire interchange between the trainee(s) and actor is video-taped for educational purposes and used in the subsequent debriefing.

Evaluations-

1. Baseline evaluation will consist of a questionnaire for self-evaluation, ICU process, teamwork and family satisfaction and performance testing using a simulation evaluation of the 9 skills before the teaching and training session. The trainee questionnaire will include demographic information including age, gender, country of birth, profession, education, years practicing, hospital type (private, governmental, health fund), number of ICU beds, religion, religiosity, previous experiences with death and the Israeli Terminally Ill Law. The simulation evaluation will access level of knowledge, skills and attitude. The baseline questionnaire will be administered on the morning of the simulation training or prior to that date if possible before the training begins. The questionnaires will be for the baseline and repeat questionnaires 3-6 and 12 months after the training in their units or by mail. The control group will receive their questionnaires (for the baseline questionnaire and the repeat questionnaire 3-6 and 12 months later) during the first year of the study (before the second year when their simulation training occurs) in their ICUs or by mail. As this is a National Project and we hope to train most ICU doctors and nurses in Israel over the next several years, we will try to obtain baseline data for ICU professionals around the country. The Israeli Society of Critical Care Medicine has agreed to participate in this study. Therefore, we will also be obtaining baseline information from all ICUs - those included and not included in the study. ICU doctors and nurses will also answer the doctor-nurse questionnaire in their ICU or by mail. It will be important to see if there are any differences between units in the study or not. It also will be important to repeat the questionnaires in ICUs not in the study after 1-2 years.
2. Post- training questionnaires- immediately after the simulation training and 3-6 and 12 months later for the self-assessment questionnaire and 3-6 and 12 months later for the ICU process, teamwork and family satisfaction questionnaires. The control group will also have post-training questionnaires, ICU process and teamwork evaluations 3-6 and 12 months after the first assessment during the first 12 months of the study period. It is important to evaluate whether the training intervention affects a change for a short period (3-6 months) or a more prolonged period (12 months). Potential variables for questionnaires include:

ICU process questionnaires- terminal illnesses as defined by the Law, advanced directives or durable power of attorney, designated surrogates, physician or nurse asked about patient preferences, documentation patient and/or surrogate preferences, documented family conferences, receiving CPR, ventilation, vasopressors or dialysis, deaths preceded by DNRs, limitations of life-sustaining treatments or CPR, time from ICU admission to first discussion about limitations, to DNR and to death, time from first discussion about limitations until death; teamwork questionnaires- will include nurse initiated discussions with doctors, nurse initiated discussions with families for patient/family preferences, unrealistic family expectations, family conflicts, inappropriateness of care, decisions shared by nurses and physicians, nurse involvement in EOL decisions and work autonomy;

family satisfaction questionnaires- will be completed by family members of ICU patients 2-4 months after the patient dies or is discharged from the ICU. They include completeness of information, listening, trust, respect, asking patient preferences and family opinions, inclusion in decision making, quality and frequency of communication, adequate time for asking questions, express your fears and emotions, emotional support, compassion shown to family, support family decisions, consideration of family needs, achieving appropriate level of care and perceptions of the quality of palliative care and quality of death and dying.

Sample size justification and statistical analyses- The power analysis shown below is not for limiting sample size to the smallest required to prove or disprove the hypothetical effectiveness of an intervention, rather to estimate our chance of reproducing a significant improvement with the sample size suggested by a limited budget and trying to best learn from the process. Power analyses will be performed using PowerAndPrecision (version 4.1) using one tail and alpha set at 0.05. The following are examples of questions that are likely to be answered with current stepped wedge design where each cluster of units will have 30 to 40 team members:

1. knowledge of teams regarding the obligation by current law to provide support to families of dying patients- Current surveys show that only 50% of teams are aware of this legal obligation and that this clearly improves after the workshop. A power analysis shows that with an n=30 in each group (control and intervention) we will have a 91% chance of showing an increase from 50% to 85% (Chi square or Fisher's exact tests).
2. skill for listening- If at baseline, the capacity to listen is rated by self-assessment at 3.5 on a Likert scale of 1-5, power analysis shows that with n=40 in each group (control and intervention), we will have a 84% chance of showing a statistical significant increase to 4.1 (nearly 20%) by the workshop (t-test).

We will compare the change in basic skills (self-assessment questionnaires and performance testing), ICU processes, teamwork of care and family satisfaction from the baseline period to the post-training periods which will serve as the primary outcome of this study. Professional demographic and organizational variables will be correlated with the actual improved skill performance, ICU processes of care, teamwork and family satisfaction using a multivariate analysis to determine the important variables leading to change. The variables that will be assessed include: age, gender, profession, country trained, years practicing, religion, religiosity, previous experiences with death, large or small ICU (< 10 vs. >10 beds) and hospital type (private, governmental, health fund).