Impact of an online training module and decision aid on the involvement by intensivists of older patients in shared decision making for goals of care discussions in an intensive care unit

Rationale: Although there exist various decision aids about end-of-life treatment options, these are not widely used in hospitals across Canada. Lack of adaptation to local context and culture are among the reasons these aids are not used widely. Previously, we worked with patients and health professionals to create an open-source and free decision aid that can be adapted to the local context of different settings (www.wikidecision.org). We now need to train clinicians to use our decision aid.

Hypothesis: We hypothesize that training professionals to use context-adapted decision aids can be an effective strategy to increase adoption of decision aids that lead to better decision making

Objectives: (1) Develop and deliver a training program for a context-adapted decision aid, (2) evaluate its effect on professionals’ adoption of the aid, (3) evaluate the impact of our training on the quality of patient’s decisions and (4) identify underlying mechanisms to explore how our wiki-based decision aid can be used elsewhere for the benefit of all Canadians.

Research plan: (1) Observe discussions between elderly patients and their health professional about the type of care they would like to receive in the intensive care unit, (2) develop and implement the training program, (3) evaluate the impact of our intervention by observing new discussions between elderly patients and their health professional and (4) We will perform qualitative interviews with professionals and patients about how our intervention was implemented and its perceived benefits and challenges.

Purpose: Determine the impact of a context-adapted decision aid and online training module about shared decision making (SDM) on intensivists’ involvement of elderly patients in goals of care discussions.

Methods: Between May 2017 and January 2018, we conducted a three-phase before-after study in a closed ICU in the province of Québec, Canada. We video- and/or audiotaped patient-intensivist encounters discussing goals of care during three phases: without access to the decision aid (Phase 1), with access to the decision aid without online training (Phase 2), with access to the decision aid and online training (Phase 3). We included patients aged 65 and older, mentally competent, and for which a goals-of-care discussion was planned. We excluded patients who were intubated, facing urgent decisions or who did not speak French. Two graduate students administered the third observer OPTION scale to determine the level of patient involvement in decision making in each video/audiotaped encounters. We used descriptive statistics and a Kruskal-Wallis test to compare OPTION scale scores between phases.

Results: Among 46 eligible patients, we recruited 21 patients (7 per phase) and five intensivists. Six patients declined to participate, 3 discussions were missed and 16 planned discussions never occurred. We videotaped 20 clinical encounters and one was audiotaped. Patients were mostly male (n=15 (71%)); median age (interquartile range (IQR)) was 77 years (68-82); without complete high school education (12 (57%). Intensivists were mostly male (4 (80%); all fellowship-trained; and median age (IQR) was 35 (33-43). All intensivists completed the online training after phase 2. None of the intensivists used our DA in phase 2 or 3. We did not find any differences in OPTION scale scores with overall median (IQR) scores low in each phase (25 (21-29) (Phase 1) vs. 21 (15-25) (Phase 2) vs. 19 (17-29) (Phase 3) (p=0.27)).

Conclusion: Our online training module did not result in greater involvement of patients in SDM during goals of care discussions. Further investigation is needed to understand why clinicians did not use our context-adapted DA and how SDM can be used in the ICU context to improve elderly patients involvement in goals-of-care decisions.

Key Findings

• Our online training did not change the way doctors discuss goals of care with older patients.
• The doctors in our study did not use our decision aid to support discussions with patients. They explained that they did not have the time and did not find our aid to be appropriate for the patients included in our study.
• Patients and doctors have concerns about goals-of-care discussions. The way doctors discuss goals of care needs to be improved.

Evidence gap addressed by study/Why this study was needed

Discussions about goals of care with older patients are not well structured in Intensive Care Units. Training doctors to discuss goals of care is needed. We also need to understand how discussions about goals of care are perceived by patients and doctors.

Suggestions on how these findings could impact older adults living with frailty and/or their family caregivers and how this might be measured

• Our findings could improve how doctors discuss goals of care with older patients in the Intensive Care Unit.
• Patients and doctors considered video-recording of these discussions acceptable.
• Video-recording could be used to help doctors improve their goals of care communication skills.

Brief comment on type of study in lay terms/plain language

• Our study wanted to find out if teaching doctors how to talk about goals of care with older patients would improve their discussions. We also wanted to find out if a decision aid would help doctors and patients during these discussions.
• We video-recorded patient and doctor encounters in the intensive care unit. We video-recorded encounters before and after doctors were trained. We also met the doctors at the end of the study to find out their general thoughts about our training, the decision aid, and goals of care discussions.
• Our study included patients aged 65 and older and their doctors in the Intensive Care Unit at the Hôtel-Dieu de Lévis in Québec.
• We evaluated the impact of training doctors on the quality of their discussions with patients in the Intensive Care Unit. We also listed the most frequent things that patients did not understand during goals of care in the Intensive Care Unit.

Key Findings

• Online training about shared decision making and the use of a decision aid did not modify the way physicians discussed goals of care with older patients in the Intensive Care Unit.
• Physicians did not use the decision aid developed to help them discuss goals of care with their patients. Many barriers to using the decision aid were identified including: lack of an integrated organizational implementation strategy, lack of interprofessional involvement, lack of adequate discussions before being admitted to the intensive care unit, insufficient time to change habits, and concerns that the decision aid was not adapted to their patients’ needs.
• Patients and physicians have concerns about issues discussed during goals of care conversations. The approach physicians use to discuss goals of care with their patients in the Intensive Care Unit still needs to be improved.

Evidence gap addressed by study/Why this study was needed

Goals of care discussions with older patients are not well structured in the Intensive Care Unit. Training about how to discuss goals of care using shared decision making is needed to help improve these discussions in the Intensive Care Unit. We also needed to understand how goals of care discussions are perceived by patients and physicians.

Suggestions on how administrators or policy makers could use the findings

Our findings can be used to inform the development of improved protocols and policies that address the barriers to clear and thoughtful goals of care discussions in the Intensive Care Unit.  Our study also proposes that there is a need for strategies and policies that facilitate and stimulate goals of care discussions before patients arrive in the Intensive Care Unit and on a more frequent basis.

Brief comment on type of study in lay terms/plain language

• We aimed to understand how physicians discuss goals of care using shared decision making and whether a decision aid with older patients improved the quality of these discussions.
• We video-recorded patient and physician encounters in the Intensive Care Unit. We recorded the encounters before and after physicians were trained on how to use shared decision making and a decision aid. We also interviewed physicians at the end of the study to understand their perceptions about the training, decision aid, and goals of care discussions in the ICU.
• Our study involved physicians and their Intensive Care Unit patients aged 65 and older at the Hôtel-Dieu de Lévis (Lévis, Québec).
• We evaluated the quality of goals of care discussions and the level of patient engagement in decision making.  We identified frequent misunderstandings that patients have during goals of care discussions and areas where physicians can improve their communication skills.

Plante P, Angulo Mendoza GA, Archambault P. (En français) Analyse, développement et évaluation d’une formation médicale en ligne. Mediation & mediatisations. No. 2 (2019): https://revue-mediations.teluq.ca/index.php/Distances/article/view/95

CFN Webinar (May 8, 2019): Shared decision making for goals-of-care discussions in the ICU: Evaluation of an online training module and decision aid