Moral Injury in Safety-Critical Work

The incident analysis report runs to forty pages. The operator who was on the valve reads one line of it: failure to follow procedure. He knows the procedure had not matched the plant for two years. So does his supervisor. So did the three people who signed the report.

Nobody was hurt that day. The injury came afterwards.

Heavy industry has a name for the first harm, the incident. It rarely has a name for the second. Healthcare and the military do. They call it moral injury, and it belongs in any serious conversation about psychosocial risk in safety-critical work.

What moral injury means

Two definitions anchor the field.

The psychiatrist Jonathan Shay, working with Vietnam veterans, described moral injury as a betrayal of what is right, by someone who holds legitimate authority, in a high-stakes situation (Shay, 2014, Psychoanalytic Psychology). Shay's version puts the leader at the centre. The injury comes from those in authority breaking the moral contract.

Brett Litz and colleagues widened it. Their 2009 paper in Clinical Psychology Review describes potentially morally injurious events as perpetrating, failing to prevent, bearing witness to, or learning about acts that transgress deeply held moral beliefs and expectations. Litz's version includes what a person did, or failed to stop, themselves.

Both describe a wound to conscience and trust. Neither describes ordinary stress. Moral injury is also not a clinical diagnosis, and this article offers none. It is a way of naming a specific kind of harm that people in high-consequence work recognise the moment they hear it.

How it differs from burnout and psychosocial hazards

These three terms get used interchangeably. They should not be.

A psychosocial hazard is a condition of work that can cause harm: high demands, low control, poor support, poor organisational justice, exposure to traumatic events. Hazards describe the work.

Burnout describes exhaustion and detachment that build when demands outrun resources over time. It describes a depleted person. We cover it in Burnout in High-Risk Industries.

Moral injury describes a rupture in what a person believes is right. They did something, failed to stop something, or watched something that violated their moral expectations, often with authority or organisational pressure behind it.

The same system conditions can produce all three. A crew under production pressure, short-handed and blamed after events, can be burnt out and morally injured at once. The difference matters because the fixes differ. Rest helps burnout. Resilience training does nothing for a person whose organisation asked them to do something they knew was wrong.

Moral injury is not a named hazard in ISO 45003 or in Safe Work Australia's hazard list. The hazards that produce it are. Poor organisational justice, poor support, traumatic events, and lack of role clarity all appear on those lists. Moral injury is what those hazards can become when leadership responds badly.

Why safety-critical work is fertile ground

Shay's three conditions fit heavy industry closely.

High stakes. People in maritime, mining and energy work beside energy that can kill. Their moral expectations about safety are concrete: you do not send someone into a space you know is unsafe, and you do not sign off a control you know has failed.

Legitimate authority. Command structures are steep. A master, a shift superintendent or a drilling supervisor holds real authority over whether work continues. Workers on rotation or contract often depend on that authority for their next job.

Repeated exposure. Production pressure, degraded controls and after-the-fact blame are not rare events. They recur, and each recurrence adds weight.

The research gap is real. A 2024 scoping review in the journal Healthcare looked for studies of "second victims", workers harmed by their involvement in adverse events, outside healthcare. It found 18 studies across eight sectors. For oil and gas production, rail and air traffic control, it found support programmes but no specific studies. The authors suggested one likely reason: those industries may describe the experience through other terms, including moral injury.

Three patterns in industrial work

The scenarios below are composites, written to illustrate patterns. They do not describe specific people or companies.

1. Blame after the event

A lifting operation goes wrong. The rigger closest to the load is stood down within hours. The incident analysis focuses on his missed step. It says little about the crane that had been flagged for maintenance twice, or the schedule that compressed a two-day lift into one shift.

The rigger carries two things away. One is the event itself. The other is the discovery that the organisation he trusted chose a story that protected itself. That second discovery is the moral injury. His colleagues learn from it too. Next time something looks wrong, they will think harder before they say so.

2. The second victim on the analysis team

A supervisor sits on the team reviewing a serious near miss. Early in the process, it becomes clear that senior management wants the finding to land on individual behaviour. The supervisor knows the system story. He signs the report anyway, because the alternative looks like career suicide.

He is now a second victim twice over: involved in the event, and complicit in an account he knows is incomplete. Litz's definition covers this directly. Failing to prevent a moral violation, and taking part in it, can wound as deeply as suffering it.

3. Coerced continuance

A vessel is alongside with a charter deadline. The chief engineer reports that a critical alarm is intermittent. The shore office asks whether the ship can sail and fix it on the voyage. Nobody orders it. Nobody has to. The master, the chief and the watchkeepers all understand what refusing would cost.

They sail. Nothing happens. The crew still spent two weeks running a ship they believed was unsafe, because people with authority made it clear what was expected. This is Shay's definition almost word for word.

How this connects to HOP

Human and Organisational Performance rests on a few principles that Todd Conklin, Rob Fisher and Sidney Dekker have done most to shape. Error is normal. Blame fixes nothing. Context drives behaviour. How leaders respond to failure matters.

That last principle is where moral injury lives. An organisation can adopt learning teams, improve procedures and fix its tools, and still injure its workforce if its response to failure breaks trust. If a post-event process punishes honesty, or management overrides a stop-work call with a deadline, workers learn that the HOP language is decoration.

This is the main argument of The Wellbeing Imperative: wellbeing belongs inside safety because the conditions of work shape both. Moral injury is the clearest case. It is a safety problem and a wellbeing problem with the same root cause, and HOP cannot finish the job of fixing the system while that cause stays in place.

What leaders can change

Moral injury risk falls when conditions and responses change. Telling people to be tougher does not help. Five controls make the biggest difference.

Make stop-work authority real. A stop-work card that costs someone their next rotation is not a control. Leaders need to back every legitimate stop publicly, including the ones that turn out to be unnecessary.

Separate learning from discipline. Run incident analysis to understand what the system failed to prevent. Keep disciplinary processes for genuine recklessness, and keep them apart. A just culture framework that people actually trust does most of the work here.

Support people involved in events. Peer support and structured check-ins after an event reach the operator, the supervisor and the analysis team. Aviation leads other industries here, and the practices transfer.

Govern production pressure. Make decisions to continue with degraded controls visible, recorded and owned at the right level. When a deadline overrides a safety concern, a named senior person should sign for it.

Hold leaders accountable for known degraded controls. If a control has been flagged repeatedly and nothing changes, the failure sits with management. Say so in the analysis.

Counselling and EAP access help someone who is already hurt. They do not replace these controls.

What to measure

Injury rates will not show moral injury. A site can post a clean TRIR while its workforce quietly stops reporting. That is one reason we argue for building capacity instead of counting injuries.

Better signals include speaking-up climate, trust in how incidents are handled, perceived fairness of post-event processes, and the gap between what leaders say about stop-work and what workers believe will happen if they use it. Wellbeing Daily's Six Drivers diagnostic, adapted for industry from Jan-Emmanuel De Neve and George Ward's Oxford research, measures two drivers that carry most of this risk: Culture and Safety, and Leadership and Relationships. Comparing leader and worker scores often shows the problem before an incident does.

The HOP programmes we deliver with Fisher Improvement Technologies cover learning teams and incident analysis, which is where most of the second injury happens or is prevented.

Frequently asked questions

What is moral injury at work?

Moral injury is the lasting harm that follows perpetrating, failing to prevent, witnessing or learning about acts that violate deeply held moral beliefs and expectations. Psychiatrist Jonathan Shay emphasised betrayal of what is right by someone in legitimate authority in a high-stakes situation. It is not ordinary stress and not a clinical diagnosis. In safety-critical work it often follows events where people were blamed, silenced, or pressed to continue work they believed was unsafe.

How is moral injury different from burnout or a psychosocial hazard?

A psychosocial hazard is a condition of work, such as high workload, poor support or unfair processes, that can cause harm. Burnout is exhaustion and detachment that build when demands outrun resources. Moral injury is a rupture in what a person believes is right, often involving organisational pressure or betrayal by authority. The same conditions can produce all three. Resilience training does not address moral injury, because the wound is ethical and organisational.

How does moral injury show up in industrial safety-critical work?

The pattern is under-researched outside healthcare and the military, but operators recognise it. A crew is told to keep running after a control has failed. A worker reports a risk and is punished or frozen out. After an incident, the official account blames the person closest to the event while known system pressures stay untouched. Supervisors and incident analysis team members can become second victims when they have to sign off an account they know is incomplete.

How does moral injury connect to Human and Organisational Performance?

HOP holds that error is normal, that blame fixes nothing, that context drives behaviour, and that how leaders respond to failure matters. Moral injury happens when that response itself violates what people believe is right: blame without learning, production pressure over stop-work, or authority that punishes honesty. A HOP programme that improves tools and procedures but leaves post-incident justice untouched can still injure the workforce.

What can leaders change to reduce moral injury risk?

Leaders reduce moral injury risk by changing conditions and responses. Make stop-work authority real and free of retaliation. Run incident analysis that asks what the system failed to prevent, and keep it separate from discipline. Support everyone involved in an event. Govern production pressure so decisions to continue with degraded controls are visible and owned. Measure speaking-up climate and fairness alongside injury rates. Counselling helps someone already hurt, but it does not replace these controls.

Want to know whether your post-incident culture is building trust or eroding it? Wellbeing Daily's diagnostics measure speaking-up climate, leadership and fairness across your workforce, and our HOP programmes rebuild how your organisation learns from failure. Start a conversation.

References

• Shay, J. (2014). Moral injury. Psychoanalytic Psychology, 31(2), 182–191. https://doi.org/10.1037/a0036090

• Litz, B.T., Stein, N., Delaney, E., Lebowitz, L., Nash, W.P., Silva, C. and Maguen, S. (2009). Moral injury and moral repair in war veterans: A preliminary model and intervention strategy. Clinical Psychology Review, 29(8), 695–706. https://doi.org/10.1016/j.cpr.2009.07.003

• Second victims in industries beyond healthcare: A scoping review (2024). Healthcare, 12(18), 1835. https://doi.org/10.3390/healthcare12181835

• De Neve, J.-E. and Ward, G. (2025). Why Workplace Wellbeing Matters. Harvard Business Review Press.