"‘Cognitive Aid’? ‘Cognitive Aids’?

Cogito Ergo Sum

Yet, I and we all have limited physical and cognitive abilities. Our environment is constantly and rapidly evolving. The health situations we encounter are often more complex than complicated. Teams change. With them, individual and collective experiences.

In short :

entropy is our enemy,

and adaptation (of professionals) is part of the ‘survival’ (of patients and caregivers).

However, while delegating physical actions to colleagues is usually not a problem (“I only have two hands, could you please prepare...?”), a question we ask ourselves less often, too infrequently, is: “Which part of my cognitive load, my expertise, my know-how, could I delegate to fully concentrate on... what I cannot delegate?”. And in this case, to whom can I delegate these cognitive tasks? Who to ask for ‘cognitive aid’? What can I ‘regulate’ to better ‘manage’?

A perspective that greatly inspires me in our work on developing digital cognitive aids in health, stems from the work of Daniel Kahneman and Amos Tversky: “My system 1, intuitive and experiential, indicates that I am probably facing a certain situation. If my brain were ideal, I would start my intervention by conscientiously unfolding a certain procedure thanks to my system 2, I would submit my reevaluation to my system 1 to see what it evokes, then I would continue with the same procedure, or change course, or open other procedures in addition. And iterations until completion.” However, my brain is lazy, and my rigorous and logical system 2 is little solicited because it is time and energy consuming. What ‘cognitive aid’ could I therefore introduce in my daily practice to force interactions between systems 1 and 2? How to move from a 95:5 Economy Class to a 50:50 First Class?

An interest of this perspective is that we can naturally integrate tools (the ‘famous cognitive aids’) and an indisputable methodology into it: the ‘Bayesian’ reevaluation of the plausibilities of each of our hypotheses by injecting new data. Don’t rush for a painkiller; this simply means: “In this situation, I consider and rank several hypotheses to which I assign ‘plausibilities’, and I reevaluate these plausibilities over time based on new information (clinical & paraclinical) arriving on my cerebral teletypewriter.”

?I will develop these concepts more fully in a future series of posts, as they are at the heart of the initial development of our digital cognitive aids, and their current evolutions.

?Returning to the delegation of cognitive tasks, such an aid (a cognitive aid) would therefore be contextual, not imposing, and should even be co-constructed with end-users to perfectly fit their environment, their local habits, and of course, the expertise they want to delegate. The need for content customization is self-evident, as it is possible to follow the same (inter)national consensus in as many ways as there are teams, individuals, environments, oral expressions, or different accessible materials...

?In practice, if I wanted to develop ‘cognitive aids’ tools and an ad hoc methodology to help caregivers, I would prefer:

1. Starting from field practice to assess what would be (or not) usable and operational for professionals managing mainly routine and sometimes the unexpected;
2. Then construct tools that align with known brain functions, i.e., in accordance with ‘cognitive sciences’, to optimize ease of use in situ;
3. Finally, and only after demonstrating that a tool would work in a given environment, dare to call it a 'cognitive aid' for that environment. This is, to me, the only way to avoid going in circles: “testing a cognitive aid” is an oxymoron for me. It seems to me that one can test a “tool in an environment”, and if it brings benefit, say a posteriori that “it is a cognitive aid”. To quote a well-known phrase:

Science is not an opinion.

So, what about the basics that led to the creation of MAX by MEDAE ???????? , and the results published in recent years?

Why do you spot at first glance that the person driving the car in front of you is responding to a text or adjusting his radio? Because his brain is sequential and cannot do several conscious things at once. Specifically, being attentive to the road while simultaneously typing on a keyboard or thinking of a response is actually very complicated (yes, really!), and you immediately notice it in the erratic trajectory of his car. Similarly, after a complex or urgent care situation, who hasn’t thought: “damn, I knew it. Why didn't I do it?”. We may delude ourselves into thinking we can manage multiple actions simultaneously, but the results are clear: in case of fatigue or stress we ‘commit’ 30 to 50% of errors and omissions of things otherwise known by heart, regardless of our experience. In our work, we have only found in anesthesia-resuscitation results known in other complex fields. However, it immediately appeared that using a specially designed digital sequential tool by and for caregivers (MAX in this case) improved technical AND non-technical performances (Lelaidier et al, Brit. J. Anaesth., 2017, ref.1 below). MAX by MEDAE thus officially became a ‘cognitive aid’ for managing urgent situations in anesthesia-resuscitation. And thereby, Stuart Marshall , a inspiring pioneer in cognitive aids in health with his seminal 2013 article (Ref.2), honored us with an editorial in the same issue of the BJA (2017), validating our concept of ergonomics:

In this issue, Lelaidier and colleagues demonstrated that non-technical (teamwork) skills displayed by anaesthetists in a number of anaesthetic emergencies were improved by the provision of a cognitive aid on a mobile phone. The majority of research on cognitive aids in anaesthesia to date has been performed using paper-based cognitive aids. Although the use of personal digital assistants and computers have been shown in some instances to improve technical performance and knowledge recall during emergencies, they have not always fulfilled their promise. Few studies have looked at the effects of electronic cognitive aids on non-technical skills and this study represents the first to suggest an improvement in this area of performance.

Cocorico! Thank you, Stuart, we love Australia.

Focusing on the non-technical aspects of the job is for us THE priority since humans are at the origin of almost all accidents. However, apart from the fact that in the same context of emergencies in anesthesia-resuscitation, we were able to demonstrate that paper is not very effective (in fact, as ineffective in these conditions as the brain alone: Donze et al, Brit. J. Anaesth., 2019, ref.3, work for which we won a SFAR - Société fran?aise d'anesthésie et de réanimation (French Society of Anesthesia and Resuscitation) award in 2018), it turns out that work seeking to demonstrate an oxymoronic ‘efficacy of a cognitive aid’ paper, possibly read to the team by a ‘reader’, has not been conclusive on ‘Non-Technical Skills’ (NTS: roughly, the ‘Human and Organizational Factors’ = HOF part of the interventions). On the other hand, all the work we have done with digital cognitive aids MAX has improved these NTS. This suggested to us that, while not being perfect, our tools were at least operational. We will return in future posts to the identified reasons for the (in)effectiveness of different tools on NTS.

Benefiting from the experience and field feedback of our work to evolve the ergonomics and functionalities of MAX, we have gone further in the evaluation, with research projects covering various health fields (anesthesia, resuscitation, first aid, combat rescue, pediatrics, pharmacotechnics...), and several other criteria than technical and non-technical performances in situations ranging from routine to extreme emergency. I summarize the results obtained, and for enthusiasts, the links are at the bottom of the article:

·? ??Military first aid: technical and non-technical improvement, Truchot et al, Mil Med., 2020 (ref.4)

·? ??Fighting the forgetting curve: Learning memorization doubled at 3 months, among very experienced military doctors-nurses: Paraschiv et al, Mil Med, 2021 (ref.5).

·? ??Combat rescue (doctor-nurse level): improved technical and non-technical performances, Paraschiv et al, Simul Healthc, 2021 (ref.6).

·? ??Improved neonatal resuscitation compared to the reference paper poster, (reviewed).

·? ??Implementation of cognitive aids in high-stakes pharmaceutical activities: Car-T-cells, Elisabeth STASSAERT , M.Sc. thesis “Quality and Risk Management in Health”, under the direction of Prof. Rémy Collomp , Nice University Hospital, (ref.5).

·?????? Improved technical performances superior to paper & maintenance of skills over time, Isma?l Kabir-Idrissi & Clément PERRIGUEY , State Diploma of Nurse Anesthetist, Thesis and abstract SFAR 2023 (ref.6).

These works and the following, by comparing the medians of performance scores achieved to the expected ones, address the ‘biased’ aspect of our responses, as detailed in the works of Kahneman and Tversky:

We think we are rational, our decisions are not.

However, there is another component of human error just as important and yet much less evaluated than biases. This is the intra and interindividual variability of judgments in the face of an identical situation, in other words, the ‘noise’. In other terms, two professionals with the same training and working in the same environment can provide very different judgments on the same case. We're not talking about a 10% variation in your insurance policy or in the decision of two judges, no, no, no! The differences can commonly be up to 50%. This is very well explained in the book 'Noise' by Olivier Sibony and the same Daniel Kahneman . We all know that our decisions in healthcare are subject to inter and intraindividual variations, but we just did not know how to quantify them. It's done, and it's not flattering for us healthcare workers! :-(

The variability in judgments is undoubtedly less intuitive to measure, and therefore less studied. In a clinical study, the main result is generally expressed in a statistical test on N values, validating or not a hypothesis. On the other hand, in the same study, one only obtains a standard deviation value or a coefficient of variation per studied group. It's tough to compare two methodologies because the result will indicate a trend ('it seems that the standard deviation of the treated group is lower than that of the control group'), but it cannot be statistically significant from a single study. Nevertheless, our various studies (published and ongoing) have allowed us to accumulate and analyze more than 380 videos of care situations over the course of 6 different studies, and we have obtained very significant results on reducing the dispersion of practices. But shh, the article is still being written.

At this level, we can therefore affirm that it is now possible for every healthcare professional to integrate field tools that have been demonstrated to systematically reduce all aspects of human error (bias and noise), to improve teamwork and learning. As a bonus: a reduction in perceived stress.

All this raises many very practical questions, which we will detail in a series of posts to follow. Among those from the latest post by Thierry Morineau , and in order not to suggest that the world of healthcare is now perfect thanks to digital cognitive aids, let's quote a few obstacles:

• Change in routines, habits: oh yes! Those who know me will not be surprised that I have lost my last hairs along with my illusions, and that while the topics of FHO, cognitive aids, and human error in health are finally clearly expressed in public, many researchers in these fields have been preaching for many years with mixed success. 'Rome wasn't built in a day,' as they say.

There is certainly no miracle. As far as we are concerned, the absence of a team or institutional project regarding quality, and the lack of a medium to long-term vision is deal-breaker for considering any semblance of success. Quality is not a gadget but a mindset. Without quality training, there can be no proper team use of cognitive aids... which, in fact, will not be such (cognitive aids) under these suboptimal conditions. And without change management, it's likely that any tool will remain unused, any app...just another app in the pocket.

• Cognitive aids and training: if no CRM or HOF training can ignore the need to integrate cognitive aid, no tool (even the most tested CA) can be used without training on this tool AND on its use within a team. It is clear that without specific training, any tool becomes just another stress on the field. No miracles in sight on this side either.
• ‘Digital resistors’: there's no escaping it. But with a little hindsight, it seems possible to develop some 'countermeasures': initial information on the normality of error, co-creation of cognitive aids with teams (hence the force of customizable tools), especially by soliciting the opinion of the 'refractories', and of course, adapted training and change management (see above).
• Attentional processes: health studies that call for a 'reader' have hardly proven their effectiveness on the non-technical skills. On the other hand, the fear of confusion between leader and reader within teams still too infrequently trained in NTS has led us to always place the CA MAX in the hands of the leader in our studies. We are objectively convinced that this explains in large part we consistently observed an improvement in non-technical skills scores (CRM): with a suitable tool in hand, the leader is forced to take two steps back, cannot perform technical gestures, and thus maintains a global vision of the situation. In short, he takes his place and does his job as a leader. Nevertheless, the field prevails, and in some cases of use like military duos, the leader is required to perform technical gestures. This did not pose a problem in our studies because the 'raison d'être, and the functional objective of the cognitive aids was for us to allow the user to devote infinitely less time to his CAs than to the team and the environment. Therefore, no, the leader is not hampered in his global vision of the situation by using MAX as long as he knows his tools and the contents of his personal library before throwing himself into a complex situation. An essential condition and unsurprisingly deal-breaker.
• 'Why would I open a CA when I have to save a life?'; I would also add the often heard: 'why change our practices for one emergency per month?': unavoidable? Not so much. First of all, we advocate for the integration of cognitive aids in the facilitation of routine tasks, to provide daily assistance, to train teams continuously in CA tools and teamwork, and a fortiori to deal with the unexpected. For example (thanks to the pioneers who show the Way: anne lamarkbi , and the CENTRE HOSPITALIER DE SENS ), the opening of the delivery rooms on MAX by MEDAE ???????? vs paper: 'fun and intuitive', personalized to the local environment, possibilities to take photos in case of equipment failure, to call the manager instantly, to edit detailed histories by biomedical staff or for establishment certifications... Secondly, a change in practices requires a change in habits and reflexes. It is clear that if one is faced with a Cardiac Arrest, the first instinct is not to take out their phone from their pocket, or open the MAX platform on the computer in the room. However, it is the responsibility of the trainers to emphasize that once the first 2-3 reflex actions adapted to an emergency situation have been verbalized (right in the thought freeze moment!), the next reflex action for the whole team is to ensure that the leader has a cognitive aid in hand quickly. We will come back another time on the "read and do" (‘I scrupulously follow the CA without any effort of imagination... at least at the beginning’) and "do-verify" (‘I do as usual and check the CA from time to time to validate that I did well’) uses that we have also evaluated (ref.3).

That's it for this rather detailed response to the different posts that have kindly mentioned us. We thank their authors immensely. And we will also be pleased to exchange with you, and try to provide, if not answers, at least some feedback from our experience to your possible questions.

So,

Cogito ergo sum

or

Cogitamus ergo sumus?

You may now start your papers !

Réferences?:

1.????? Lelaidier et al, BJA, 2017?: https://academic.oup.com/bja/article/119/5/1015/4259258

2.???? Dedicated?Editorial by S. Marshall: https://journals.lww.com/anesthesia-analgesia/fulltext/2013/11000/the_use_of_cognitive_aids_during_emergencies_in.19.aspx

3.???? Donze et al, BJA, 2019?: https://www.bjanaesthesia.org/article/S0007-0912(19)30335-6/fulltext

4.???? M. Truchot et al, Mil Med, 2020 : https://pubmed.ncbi.nlm.nih.gov/32091610/

5.???? Paraschiv et al, Mil Med, 2023?: https://pubmed.ncbi.nlm.nih.gov/33928372/

6.???? Paraschiv et al, Simul Healthc, 2022?: https://pubmed.ncbi.nlm.nih.gov/34934026/

7.???? E. Stassaert et al, 2023, https://www.gerpac.eu/securisation-du-circuit-car-t-cells-chimeric-antigenic-receptor-t-bienvenue-aux-aides-cognitives

8.???? C. Perriguey, I.Kabir-idrissi, J.Rouge, Abstract SFAR 2023, https://www.sfar-lecongres.com/wp-content/uploads/2023/09/Com-orales_IADE_C.pdf

anne lamarkbi ; Benoist Alexandre ; Stéphane KIRCHE ; Jerome Coutet ; Bruno Debien ; Elisabeth STASSAERT Rémy Collomp ; Jean-Jacques LEHOT ; Marc Lilot ; Baptiste Balanca ; Thomas Rimmelé ; Andrei PARASCHIV ; karim Tazarourte ; Véronique DELMAS ; Luc Aigle ; Gilbert MOUNIER ; Guillaume Der Sahakian ; Clément Buléon ; Fran?ois lecomte?; ?? Fouad MARHAR ?? ; Stuart Marshall ; Lucas Denoyel ; Sophie Conte ; Peggy Leplat Bonnevialle ;

EMERGENSIM ; CESITECH CESITECH santé ; CENTRE HOSPITALIER DE SENS ; Hospices Civils de Lyon - HCL ; Université Claude Bernard Lyon 1 ; Centre Hospitalier Universitaire de Nice (CHU de Nice) ; CENTRE HOSPITALIER WILLIAM MOREY ; SoFraSimS ; SFPC - Société Fran?aise de Pharmacie Clinique ; Société Fran?aise de Médecine d'Urgence ; SFPO - SOCIETE FRANCAISE DE PHARMACIE ONCOLOGIQUE ; Association ABASS ; EHESP - école des hautes études en santé publique ; Facteurs Humains en Santé.

#aidescognitivesdigitales?; #aidescognitives?; #qualité ; #cognitiveaids ; #healthcare ; #quality ;

?