ROAR-00110

Key Takeaways

• Unit-level perception data surfaces retention risk that facility-wide engagement scores and incident reports miss entirely
• Charge nurse communication coaching is the highest-leverage intervention a CNO controls directly, with the shortest distance between action and perception shift
• A focused 90-day measurement cycle on one high-turnover unit proves the model faster than a system-wide rollout

Your incident reports show nothing alarming. Your engagement survey scores look acceptable. Yet the resignations keep coming from your behavioral health units, and exit interviews keep circling back to safety.

The disconnect is a measurement problem. A nursing safety program built on incident counts and annual engagement composites can’t surface what’s actually driving departures: how safe your nurses believe they are, and whether they trust the organization to respond when something happens. This guide walks through how to measure safety perception at the unit level, coach the charge nurses who shape it daily, and coordinate the response systems that prove commitment. For the research behind why perception predicts retention, see the complete guide to staff safety in psychiatric hospitals.

What You Need Before You Start

Building a unit-level perception baseline takes about 90 days to establish and get a first quarterly comparison. You need your current engagement survey data (with safety-specific items identified), exit interview summaries, incident reports, and turnover data broken out by unit.

Your team: your CHRO or HR lead for survey infrastructure (the CHRO measurement framework covers the corporate side of this), your CSO for incident and response time data, and your directors of nursing for unit-level context.

If your exit interviews don’t currently include safety-specific questions, add two or three before moving forward: “Did safety concerns influence your decision to leave?” and “How would you rate our response to safety incidents?” The first round of responses tends to surface units no one flagged as high-risk.

Measuring Your Nursing Safety Program at the Unit Level

The critical shift here is moving from facility-wide scores to unit-level data. Your organization-wide average may look acceptable while specific units are in crisis. CNOs who have pulled safety-specific items from engagement surveys and scored them by unit often discover that their highest-turnover units share one trait: not the most incidents, but the lowest confidence that leadership will act on what gets reported.

Start with what you already have. Pull safety-related questions from your existing engagement survey and score them separately by unit. If your engagement instrument doesn’t include safety-specific items, add three to five targeted questions to your next pulse survey focused on organizational response, not just incident frequency.

Then add the question that connects perception to retention: “I would consider leaving this organization due to safety concerns.” That single item turns perception measurement into a workforce planning tool with documented outcomes. Cross-reference the results. Which units show the largest gap between low perception scores and high intent to leave? That’s where your retention risk concentrates.

If survey infrastructure doesn’t exist yet: Focus on your single highest-turnover unit first. Three to five safety-specific questions on a pulse survey takes under 10 minutes per nurse. One unit measured well proves the model faster than a facility-wide rollout.

Coaching Charge Nurses to Move Perception

Perception doesn’t shift because of policy memos or annual training refreshers. It shifts when staff experience visible, rapid organizational response to their safety concerns. And the person who shapes that daily experience on each unit is the charge nurse.

Leadership quality accounts for about 34% of the variation in whether nurses stay or leave [1]. One-third of your retention outcome depends on something you directly control: how your charge nurses communicate commitment to safety on every shift.

This means explicit coaching, not general encouragement. Your charge nurses need specific language for three moments:

• Shift handoff: A sentence acknowledging current safety status and any open concerns from the prior shift. Not a policy reading. A direct statement: “We had an escalation on this unit yesterday, the response took under two minutes, and here’s what we’re doing differently today.”
• After an incident: Visible follow-up that the reporting nurse can see. When a nurse reports an incident and nothing visibly happens, the lesson they learn is that reporting is pointless. Directors of nursing describe a pattern where a single failed response undoes months of goodwill.
• During routine rounding: Asking one safety-specific question per round. Not “do you feel safe?” (too broad). Something like “is there anything about safety response on this unit you’d change?”

In behavioral health settings where this kind of visible communication was paired with documented safety systems, staff reporting they felt “very prepared” to respond to incidents went from 38% to 76% [2]. Peer CNOs using unit-level perception data describe charge nurse coaching as the intervention with the shortest distance between action and measurable perception shift.

Coordinating Response Systems With Your CSO

The other half of the perception equation is what happens when staff actually call for help. Charge nurses describe a telling detail: what registers with staff isn’t the difference between 30 seconds and three minutes on a stopwatch. It’s whether the person who called for help can still see the situation escalating when backup walks through the door. That visual, help arriving while the moment is still live, is what staff remember when asked whether the organization takes safety seriously.

Work with your CSO to verify actual response times on your target unit. Is there timestamped data, or are you relying on estimates? The nursing safety brief on unit-level perception data provides the specific talking points to bring into that conversation.

Your 90-Day Unit-Level Action Plan

Start with your single highest-turnover behavioral health unit. Each percentage point of nursing turnover costs roughly $289,000 annually [3], so even one unit’s improvement builds the financial case for scaling.

• Pull safety-specific engagement items and score them by unit this week. Can you identify your three lowest-scoring units without requesting new data?
• Add two intent-to-stay questions to your next pulse survey cycle, distinguishing between “planning to leave the organization” and “planning to leave this unit”
• Script three sentences of safety commitment language for charge nurses to use at shift handoff, and test the language with a charge nurse before rolling it out
• Verify actual response times on your target unit with your CSO using timestamped data
• Schedule a 90-day re-measure on your target unit with a comparison point, not just a single snapshot

See how one behavioral health provider documented these results across their facilities.

Your charge nurse on that high-acuity unit doesn’t need another policy update. She needs to see that when her team calls for help, help arrives fast, and that the organization measures whether she feels protected, not just whether an incident was filed. A nursing safety program that tracks perception at the unit level gives you the lead time to intervene before the next resignation letter lands on your desk. Start with one unit. Measure it well. The retention data will make the case for every unit after.

References

1. PMC. Leadership Quality and Nurse Retention. https://pmc.ncbi.nlm.nih.gov/articles/PMC10806563/
2. ROAR for Good. Internal data, 2024. Internal data
3. NSI Nursing Solutions. 2025 NSI National Health Care Retention & RN Staffing Report. https://www.nsinursingsolutions.com/documents/library/nsi_national_health_care_retention_report.pdf