When we're now teaching people how to look for pressures, conflicts, and adaptations in their system, they need to start at the blunt end, distally and trace forwards in time and through layers of the organization towards the incident. The role of monitoring in proactive safety and performance management. Christine Jeffries @ LFI Conf 2023. youtube 
YOUTUBE 39dGxBjAmyo
The role of monitoring in proactive safety and performance management. Christine Jeffries @ LFI Conf 2023. youtube
The system is always changing. A common approach is to monitor how things are changing, graphing signals and trends over time. Jeffries proposes to monitor adaptive behavior instead by looking at patterns of pressures, conflicts, and adaptations.
Traditional trending of data doesn't appreciate these kinds of complexity in the system. It encourages overly linear decision-making. With pressures in your system driving you forward you leverage your resources to find and fix so you can get to the next task. This is how people in all kinds of domains get stuck in whack-a-mole actions—solving one problem only for a similar problem to happen in another area.
Traditional monitoring about things that happen and the way they come together just isn't enough. Knowing about your system in this way is necessary, but insufficient.
A sequence of shapes represent things that happen—the stuff we typically monitor. An empty spot on the right represents the unknown thing about to happen next. A green box highlights the space between the last known event and the thing about to happen. Jeffries identifies this place as where people adapt to change and the pressures and conflicts in the system.
Rasmussen warned early on that if investigators search for answers going back from the incident they would stop at the first plausible point and that would be a person. Do more. Try harder. Do more right.
So Woods and Decker later explicitly stated that to avoid this trap you must start distally and trace the sharp end actions forward to the incident—get in the tunnel with them to see how their behavior was rational.
However it was always thought that analysts moving forward through the investigation would naturally see the blunt end pressures and constraints that that made sharp end behavior rational. However this didn't happen.
Historical progression of accident investigations. Rasmussen warned about blaming human error. Woods and Dekker advised starting with sharp end actors to follow their journey "through the tunnel." New recommendations call for deliberately looking for the pressures from the blunt end towards the incident.
When we're now teaching people how to look for pressures, conflicts, and adaptations in their system, they need to start at the blunt end, distally and trace forwards in time and through layers of the organization towards the incident.
Systemic Contributors and Adaptations Diagramming (SCAD). Four quadrant diagram. Left to right axis is distal to proximal. Top to bottom axis is blunt to sharp. A region identifies distal pressures from the blunt end. Another region identifies conflicts, mostly distal, both blunt and sharp. In the lower right is a region identifying proximal and sharp adaptations.
Jeffries begins examples from healthcare which identify pressures conflicts and adaptations. Several specific adaptations by nurses provided a signal of pressures and conflicts. The lever available to leaders is to provide more resources or to time-shift workloads or introduce human factors changes to enable the current level of work to proceed more safely. 15m40s
Another example featured a social network analysis to identify expert practitioners in particularly high demand. The adaptation was how people were seeking advice and from whom. The lever available to leaders is to monitor which phones are ringing off the hook as a way of anticipating the overload. Also, anticipate risk of burnout among heavily burdened expert clinicians by identifying ways to create recovery periods for these experts. 16m41s
David Provan's safety maturity model survey has over 500,000 points in it. The data shows this pattern that the more compliance-focused senior leadership is in the organization, the less deference to expertise happens in the business. That is, decision-making in the organization becomes more driven by hierarchy than by expertise. We know skilled practitioners find ways to create successful work despite bureaucratic constraints. However, leadership preoccupied with compliance may indicate undervaluing of that crucial expertise. 17m40s
Military wanted to know what was holding back innovation among their people. We looked for adaptations to competing pressures and goals that would support innovative behavior, like room for failure, autonomy, and collaboration. We found those adaptations were eroded by cyclical time pressures, procedural adherence, and how overtly prioritized innovation actually was. 18m28s
Katie Walker's work has identified patterns of cascades in pressures, conflicts, and adaptations that reliably described the erosion of adaptive capacity that eventually manifested in major accidents. 20m41s
Indicators include: increasingly unable to anticipate changes in their reserve of adaptive options; indicators that the reserve adaptive options itself has changed—this may show up as selective attention to specific signals which limit information flow. Monitor these indicators as alarms you don't want to dismiss despite the intense pressures you will encounter to dismiss it.
.
Blank template for SCAD pressures, conflicts, and adaptations.