5 principles for taking action in the face of uncertainty
- Faced with the unknown, organisations are forced to make decisions without reference to prior knowledge.
- Management of such is based on pragmatic rationality based on five principles, including taking practical action and rapidly assessing its effectiveness.
- During the COVID-19 crisis, for example, the method used consisted of testing hypotheses in the field, updating them as the first results became apparent.
- By agreeing to take part in large-scale collective surveys, stakeholders must also develop an attitude of humility and caution in the face of simplistic assertions.
- These principles form a specific management style within organisations known as High-Pragmatic-Organisation (HPO).
While there are known principles for managing situations of relative uncertainty, it is less clear how to deal with situations of great uncertainty, i.e. where the unknown reigns supreme. In these rare cases, there is no longer any reference to existing knowledge (these are known as ‘unknown unknown’ situations, as opposed to ‘known unknown’ situations where there is an existing reference). In such cases, management requires rules to those already used in the past.
The first wave of the COVID-19 crisis was an exemplary case of this type of situation1. During this period, it was impossible to refer to a past event. Dubbed the ‘flu bug’ for a short time, the high mortality rates observed quickly contradicted this assertion. What’s more, hospital stays in intensive care were much longer than those usually observed for other infectious respiratory viruses, and loss of smell was a previously unknown symptom. There were innumerable ‘surprises’, making any reasoning by reference to existing knowledge tricky.
In response, organisations grappled their way through the process and made, often rushed, decisions. Some considered that these actions were based on intuition and flair. However, on closer inspection, they reveal a certain form of management. This is shown by a study based on interviews with more than 120 players in the French hospital system, published recently in the European Management Review2.
The management in question is based on a rationality that is neither Cartesian nor close to the concept of High-Reliability-Organisation (HRO), which is often used in uncertain situations. Rather, it is a pragmatic rationality, which consists of conducting a collective investigation to test hypotheses in the field. Five principles emerge, reflecting the ability of French hospital players to demonstrate realism in the face of the unknown.
#1 Undertake practical actions as part of a survey
This first principle encourages us to design actions without waiting for perfect information. The actions envisaged at this stage represent hypotheses. They are derived from the initial results of the survey, which selects them on the basis of the following methods : (i) learning from unusual events that can be observed in the field (anomalies such as the abnormally long duration of stays in intensive care for patients with COVID-19, which is surprising given our knowledge of the effects of infectious viruses of the respiratory system); (ii) consolidating the reliability of the information gathered by triangulating different sources or by assessing the profile of the issuer of the alert (what some people call ‘epistemic vigilance’); (iii) and when the hypotheses are contradictory, to organise debates between the various stakeholders in order to reach a collective decision (as in the case of the reception of the Chinese curves on the incidence of the pandemic, which led to several medical specialists and epidemiologists being brought together to compare their points of view). In this activity, any modelling effort is also useful, but it is not sufficient alone, because it can neither provide reliable context (the relationship between national projections and a local situation), nor provide a sufficiently consistent prediction, due to a variety of new criteria that disrupt the ‘model’.
#2 Test the hypothesis in the field, ensuring rapid feedback of findings
This principle requires the hypothesis to be tested in the field, which is the only way to judge its relevance. Such recourse to the field takes place despite the surrounding ignorance, and conditions for action that are rarely optimal. The objectives are guided by a search for evidence and learning, while the implementation consists of circumscribing the test in various small stages, each of which is followed by a rapid return. This incremental approach optimises the assessment of the relevance of the action, wasting as little time as possible and avoiding obvious errors. One example at the start of the COVID-19 crisis was the priority actions taken in the health hospital sector (limiting patient visits to hospital, setting up quarantine measures, staff protection procedures, increasing the number of intensive care beds, to name but the main ones).These measures quickly proved their worth. At the same time, they revealed the weakness of actions taken in EPHAD establishments, where many elderly and vulnerable patients were exposed to the risk of the virus. This observation led to the rapid extension and reinforcement of actions in this sector, as recommended in the following principle.
#3 Revisit initial hypotheses through collective deliberation
This principle ensures that organisations translate the results of field tests appropriately, adapting their actions if necessary. Depending on the feedback and the collective deliberation that follows, the players maintain the initial hypothesis or modify or even reformulate it. The multidisciplinary nature of the collective deliberations and the open-mindedness of the participants are essential criteria in this revision. The more the deliberation involves a variety of expert viewpoints, the more likely it is that the update will be relevant. Similarly, the more the attitudes expressed accept the conclusions of the deliberation, the more likely it is that the chosen hypothesis will be adopted.
This third principle concludes an overall approach. Comprising of three stages : (i) definition of a hypothesis ; (ii) field testing ; (iii) updating of the hypothesis on the basis of the results, it is characteristic of an abduction method. We start with a hypothesis, check its relevance by gleaning observable facts, and then deduce whether it should be maintained or replaced by another. This approach is even more likely to be effective if the investigation is carried out collectively in order to capture as many clues as possible. It is also largely dependent on the attitudes of the members involved, as the following principle makes clear.
#4 Develop an attitude of fallibilism and anti-dualism
According to this principle, players are encouraged to express their doubts and to show humility (fallibilism), because the knowledge they have acquired is extremely fragile, subject to the appearance of new observable facts. Similarly, they are urged to avoid simplification by dichotomies between yes and no (dualist positions), as these generally reduce the ability to select and interpret clues. Without these two attitudes, there is a high risk of making erroneous decisions and giving them too much weight.
An example of the importance of this principle was the debate on hydroxychloroquine as a treatment for the virus. The fact that the hypothesis of such a treatment was put forward was not in itself shocking. It was even justified in the light of what was known about the subject. On the other hand, the failure to question it when the trials carried out had not produced convincing results is evidence of an overly assertive, dualistic position in favour of the ‘yes’ option.
In this quest for appropriate attitudes, the environment outside the players involved in the survey plays an important role, likely to trigger harmful pressures.
#5 Protecting expertise from external pressures
Unknown situations must be managed by those who have the most precise knowledge of the event. In particular, if those on the ground have the expertise that is being built up (which is often the case), they must be given priority in the actions to be taken. One consequence of this is that they need to be protected from external pressures outside the scope of the investigation. The latter can in fact hamper the effort undertaken. The COVID-19 crisis highlighted two such pressures :
- Institutional pressure, which may have called into question local actions in the name of giving priority to decisions from higher hierarchical levels, even though these are irrelevant in terms of their respective expertise.
- Pressure from the media, which may have steered debates towards confrontations devoid of nuance, because of the forms of expression imposed. Once again, the debate on hydroxychloroquine is a good example : the exchanges on TV often turned into caricatured oppositions between the for and against, locking the players into positions of defending a point of view, far removed from the attitudes of fallibility and anti-dualism that are necessary, but also damaging. As a result, people working in the field have found themselves faced with patients who want to undergo treatment at all costs, even though there is no evidence to suggest otherwise.
Unknown situations must be managed by those who have the most precise knowledge of the event.
The preventive measures against these external pressures are to be found at the level of collective deliberations organised by the players holding the expertise. They need to be cautious about making overly simplistic statements about the environment. They must also protect themselves from pressure by uniting collectively. However, this organised protection must not cut off the expertise from information produced elsewhere. The principle of collective investigation means that clues can be gleaned. They must therefore strike a careful balance between protection and selection of surrounding information.
A ”High-Pragmatic-Organisation” to manage the unknown ?
The five principles are inspired by the theoretical approaches of pragmatism, a North American school of thought from the early 20th Century. Taken together, they form a specific form of management at hospital level, known as High-Pragmatic-Organisation (HPO), in reference to this school of thought. An HPO organises collective surveys, engages in abduction processes, relies on players whose attitudes cultivate doubt and humility, and protects itself from external pressure34. By acting in this way, French hospital stakeholders have shown that hospitals can operate according to their own principles to face up to the unknown.
This new conceptual term also alludes to another concept, that of High-Reliability-Organisation (HRO), which is often used to express the management required in situations of uncertainty5. An HRO is an organisation capable of dealing with crisis situations where uncertainty reigns, by applying different principles (a hospital, but also a nuclear power plant or a system for organising air flights, can thus be assimilated to HROs in the event of a crisis).
Without going into detail, the proposed principles differ from those just described on one essential point : the absence of a reference, and therefore of the possibility of expressing reliability. As a reminder, reliability aims to minimise deviations from a normal state that sets performance standards6. It refers to anticipation by triggering preventive actions7. With HROs, nuclear power plant control operators can, for example, shut down reactors if they think that operations have entered these zones. Similarly, aeroplane pilots can refuse to fly if they think the equipment or weather conditions are dangerous.
However, when the players are faced with the unknown, this reasoning is inoperative, because the very purpose of action is to define what these standards are, and by deduction the principles of prevention. For example, when a remote monitoring system was set up for patients with mild forms of COVID-19, the initial feedback showed just how relevant the system was. Previously, managers had admitted these patients without realising that they were occupying beds unnecessarily and increasing the risk of the virus spreading. The preventive nature of the action could only be judged following the empirical test.
Finally, it can be said that although the principles of ORH management can be found, the absence of a reference to normality limits the application of the concept in the event of an unknown situation. In these situations, the reference is not determined in advance but is constructed a posteriori. For this reason, the concept of HPO, and its five principles, seems more appropriate to deal with them.
