Burnout and network centrality as proxies for assessing the human cost of resilient performance

This study explored the human cost of Resilient Performance (RP), in the form of burnout, among healthcare professionals.

That is, calls to enhance the potential for resilience (resilient potentials) may have an underappreciated toll on people, as they adapt to system deficiencies which further become masked and normalised.

Providing background, it’s noted:

·????????RP is defined as the “ability of the healthcare system to adjust its functioning prior to, during, or following changes and disturbances, so that it can sustain required performance under both expected and unexpected conditions” (p1)

·????????RP is said to arise from both internal self-organisation of people and via the deliberate work system design – definitions aligned with resilience engineering and systems thinking

·????????Anticipate, monitor, respond and learn potentials within RP play out at several scales – organisational, team and individual. At individual and team levels, potentials can play out during social interactions like asking for info, attending rounds and shift handovers

·????????It’s said that these interactions often occur without adequate organisation support. Despite recognising the lack of support, it’s said that “the resilience engineering movement has paid scant attention to the price that actors pay for supporting system resilience” (p1)

·????????Further, the resilience literature “often conveys that the individual is called upon to adapt in the face of risk to secure system performance” and moreover, “RP is normally reactive, implying excessive effort of [workers] due to flawed work system design, lack of basic resources, and ineffective management” (pp1-2)

·????????Others maintain similar concerns around the human cost of resilient performance, whereby workers’ put in extra effort under adverse conditions and this goes unnoticed due to successful outcomes and thereby workers’ “stressful working conditions can become “normal work” (p2)

This paper draws on proxy measures for RP and its human cost (burnout) via a social network analysis. They see an individual’s network centrality in advice-seeking as a proxy of their contribution to system RP.

The network centrality (discussed below) and two non-network attributes were used to calculate a resilience score. Professional burnout, a proxy of the human cost of RP, was assessed via the Maslach burnout inventory.

Key concepts for the network analysis are:

1) in-degree centrality: the number of ties directed to a given node. This indicates how many connections one node (person) has with others; e.g. well-connected or popular individuals or people who hold particular information or can quickly connect with the wider network.

2) Betweenness centrality: the number of node pairs connected by a given node. This indicates how particular people bridge connections between other parties.

3) Closeness centrality: a measure of how close a node is to all others in the network. Lower values indicate an individual is more reachable by others.

Results

Key findings included:

·????????A weak correlation was found between emotional exhaustion and the resilience score

·????????The score was also weakly correlated with working overtime

·????????Overall they note that RP, as measured in this study, was relevant to burnout

·????????Proxies of RP and its human cost seem to be more suitable metrics for assessing the human cost than with workplace accidents

·????????Concepts like resilience engineering “emphasizes RP to cope not only with acute problems (e.g., disasters) but with every day and chronic problems (e.g., insufficient staff) (Hollnagel, 2014). Health-related metrics such as the burnout score are consistent with this focus as they reflect undesired conditions that develop gradually over time, being associated with chronic problems result of acute crises” (p8)

Differences between RP as described by social interactions versus personal resilience were found. RP here was weakly correlated with overtime and emotional exhaustion, whereas personal resilience has been found to either be negatively or positively correlated with burnout.

They argue these differences may lay in how resilience is measured. RP is said to be “inherently interactive and somehow less “selfish” than personal resilience, in the sense of enhancing collaborative work” and therefore, this greater generosity of RP comes at a higher personal cost.

They found that burnout can be lessened by the practitioner’s experience; supporting previous work. They suggest a greater use of scenario-based training within an explicitly designed resilience engineering framework may assist. In any case, it’s argued that “accelerating the acquisition of expertise might be useful” (p8).

They emphasise the resilience engineering framework for training for its focus on “realistic working conditions, stresses teamwork, and produces insight into the improvement of the work system design rather than only improving individual performance” (p8).

Some practitioners of the ICU, namely physician and nurse ICU chiefs displayed high resilience and low burnout. The authors suggest this is indicative of the law of fluency: “well-adapted cognitive work occurs with a facility that belies the difficulty of the demands resolved and the dilemmas balanced” (p8).

Not everybody had this fluency. Some had high burnout and low resilience, pointing to uneven distribution of the human costs of RP in the workforce; or as the authors put it “some people carrying a heavier burden than others” (p8).

Based on this, it’s argued that perhaps some people, intentional or not, “lower their RP human cost by borrowing or diverting too many resources from their co-workers. This would be a source of system brittleness, as the reciprocity that characterizes a resilient system would be in jeopardy” (pp8-9).

Link in comments.

Authors: Terra, S. X., Saurin, T. A., Fogliatto, F. S., & de Magalh?es, A. M. M. (2023). Applied Ergonomics, 108, 103955.