What is a Safewards Champion?

If you have been asked to “put some champions through” before a Safewards rollout, it is worth being precise about what the role is. The word gets used loosely — sometimes for anyone who attended the training, sometimes for the one person carrying the whole model on a unit. Neither is quite right, and the difference matters for how many people you train and what you ask of them afterwards.

Is a champion for the ward, or for one intervention?

For one intervention. This is the single most useful thing to know about the role, and the thing most often got wrong.

Safewards is ten interventions — from Clear Mutual Expectations and Soft Words through to Calm Down Methods and Discharge Messages. The implementation model attaches a named person to each one. Fletcher and colleagues, writing up the Victorian Safewards trial across 18 inpatient units, describe each intervention as supported by a ward team member “often known as an ‘intervention champion’”. The Victorian Department of Health’s implementation plan for emergency departments carries the same structure: its preparation checklist repeats the task “Champions for this intervention have been appointed” under every one of the ten interventions, and asks services to confirm that all staff know who the lead is for each.

So a unit running Safewards properly does not have a champion. It has a set of them, each holding one piece, and between them they cover the model.

What does a Safewards Champion actually do?

Four things, none of which is running training sessions.

They model the intervention on shift. The champion is the person other staff watch doing it — not in a workshop, but at three in the afternoon on a hard day. The Victorian evaluation describes the Talk Down champion’s work as role modelling and mentoring colleagues individually, which is a fair description of the role generally.

They coach, one person at a time. Most of the skill transfer in Safewards happens in corridor conversations and debriefs, not in formal education. The champion is the person who says “try it this way next time” to a colleague who has just had a difficult interaction.

They keep the intervention visible. Several of the ten interventions have a physical presence on a ward — the Know Each Other folder, the Mutual Expectations board, the Discharge Messages tree. These decay quietly. Someone has to notice that the board has not been updated since March.

They are the point of contact for their intervention. When the organisation’s Safewards Lead needs to know how Soft Words is travelling on your unit, the champion is who they ask.

What is the difference between a Champion and a Safewards Lead?

Scope. A champion holds one intervention on one unit. A Safewards Lead holds the model across a health service.

In the Victorian trial, each participating health service designated a single Safewards Lead for the duration of training and implementation. Where a service had several wards in the trial, the Lead worked through each ward’s intervention champions to pull together how implementation was going. The Lead is a coordination and reporting role, usually held by an educator, a nurse manager or a quality lead. The champion role is a floor role, held by someone who is on the unit with patients.

Services sometimes try to collapse the two into one position. It rarely holds: the Lead ends up doing neither job properly, and the interventions lose their named owners.

How many Safewards Champions does a ward need?

There is no published number, and anyone who gives you one with confidence is guessing. What the structure implies is a floor, not a target: if each of the ten interventions needs a champion, a unit needs enough people to cover ten portfolios without any one person holding four of them.

In practice the services we work with land somewhere between eight and sixteen champions per unit, for three reasons. Rosters mean a single champion per intervention is effectively unavailable most of the week. Some interventions pair naturally and can be held together. And attrition is real — champions move units, go on secondment and leave.

It is also worth training across roles rather than restricting the cohort to nursing. At Geraldton Health Campus, WA Country Health Service trained more than 35 champions across five roles before its new acute psychiatric unit opened. At Central Adelaide Local Health Network, 85 champions were trained across two hospitals, including emergency department nurses, consultants, educators and contracted security staff. Mixed cohorts work because the interventions depend on a shared approach — a de-escalation that works only when the nurse is in the room is not much use.

Who makes a good Safewards Champion?

The implementation research is thin here, but the Victorian evaluation records one ward deciding its champions needed leadership, persistence and a belief that the intervention would work. That matches what we see.

Seniority is not the criterion. Some of the most effective champions are relatively junior staff who are respected on the floor and are around consistently. What does not work is appointing by availability — giving the portfolio to whoever was rostered on the day the decision got made — or appointing someone who is unconvinced and will quietly not do it.

One practical test: would other staff on this unit take a suggestion about their practice from this person without bristling? If not, the portfolio will sit unused.

How are Safewards Champions trained?

Through the model, then through practice. Understanding the ten interventions is a few hours of work. Being able to use one under pressure, and coach someone else in it, is not.

The Xantha Safewards Champion Program runs in two parts: the online Safewards In Practice masterclass, which takes one to two hours and covers the model, the evidence and the ten interventions; then a one-day Interventions Simulation Workshop with two facilitators and professional actors playing simulated patients. The workshop is where champions practise the interventions their service has chosen, in scenarios drawn from its own incident reports, until the response holds. Participants receive a certificate of completion and around 11 hours of CPD.

Two implementation sessions bookend the workshop — one to choose which interventions to start with and who will hold them, one to agree what happens in the first three months.

Why do champion programs fade, and what keeps them going?

Because the role has no protection in the roster and no owner above it.

The Victorian evaluation is direct about what implementation needs to succeed: engagement of key staff including managers, training treated as a priority for all ward staff, and adequate planning. Where that was missing — where the unit management team was not actively involved — implementation was weaker.

Three things make the difference in our experience:

• A named handover process. The Victorian ED implementation plan asks services to put one in place, specifically for when champions go on leave or secondment. A portfolio with no succession plan lasts as long as its first holder.
• Manager involvement that is visible. Not endorsement in a meeting. The nurse unit manager asking a champion how their intervention is going, in front of other staff.
• A point at which someone checks. Which brings us to the last question.

How do you know whether it is working?

Measure fidelity, not attendance.

The UK Safewards trial team developed a Fidelity Checklist — a brief standardised audit scored out of ten, one point per intervention in use. The Victorian evaluation used it in walk-throughs of each ward lasting 30 to 60 minutes, conducted several times across the trial and sustainability periods. It is a blunt instrument and that is its virtue: it asks whether the intervention is actually in use on the unit today, not whether staff were trained in it.

Pair that with the outcome data you already collect — seclusion and restraint episodes, Code Greys or Blacks, occupational violence incidents, staff injury claims. The Safewards randomised controlled trial found a 15% reduction in conflict events and a 26.4% reduction in containment events against a control intervention, so there is a real effect to look for; but it shows up over months, and only where fidelity is reasonable.

If fidelity is low, more training is usually the wrong answer. The question is which champion portfolios have quietly gone vacant.