Measuring safety culture requires a combination of validated surveys, behavioral observations, and operational data that together reveal how people in your organization actually think and act around safety. No single tool captures the full picture. The most effective approach layers a standardized survey with qualitative methods like interviews or walk-throughs, then tracks leading indicators over time to see whether culture is shifting.
What Safety Culture Actually Measures
Safety culture is not one thing. It’s a collection of shared beliefs, attitudes, and behaviors that determine how seriously an organization treats risk. When you measure it, you’re trying to quantify things like whether people feel comfortable reporting mistakes, whether leadership visibly prioritizes safety, and whether teams communicate well enough to catch problems before they cause harm.
This makes measurement inherently tricky. You can’t take a single reading the way you’d check a temperature. Instead, you need to assess multiple dimensions and look at how they interact. The good news is that well-tested frameworks exist for exactly this purpose.
Validated Survey Frameworks
Surveys are the most common starting point, and two instruments dominate the field.
AHRQ Hospital Survey on Patient Safety Culture 2.0
The most widely used tool in healthcare is the AHRQ Hospital Survey on Patient Safety Culture, now in its second version. It contains 40 items organized into 10 composite measures, each targeting a distinct dimension of culture:
- Teamwork: whether staff help each other, work effectively together, and treat one another with respect
- Communication Openness: whether people speak up when they see something unsafe and feel comfortable asking questions
- Communication About Error: whether staff are informed when errors occur and when changes are made in response
- Response to Error: whether people are treated fairly after mistakes and whether the focus stays on learning rather than blame
- Reporting Patient Safety Events: whether near misses and caught errors actually get reported
- Organizational Learning: whether processes are regularly reviewed and improvements are evaluated
- Staffing and Work Pace: whether there are enough people to handle the workload without rushing
- Handoffs and Information Exchange: whether critical information transfers reliably across shifts and units
- Hospital Management Support: whether leadership treats safety as a genuine priority and provides resources
- Supervisor Support for Patient Safety: whether frontline managers act on safety concerns and discourage shortcuts
Each composite produces a “percent positive” score that you can compare against national benchmarks. AHRQ maintains a database with data from hundreds of hospitals. In 2024, the overall composite average was 71% positive. Teamwork scored highest at 81%, while Staffing and Work Pace lagged at 55%. These numbers give you a realistic target: if your teamwork score is below 74% (the 10th percentile), you have a clear problem area. If it’s above 85%, you’re performing well relative to peers.
Safety Attitudes Questionnaire
The Safety Attitudes Questionnaire (SAQ) takes a slightly different angle, measuring six domains: teamwork climate, safety climate, perceptions of management, job satisfaction, working conditions, and stress recognition. It’s particularly useful for capturing how frontline providers feel about their immediate work environment and is widely used in surgical and intensive care settings.
Psychological Safety Surveys
If your concern is whether people feel safe enough to speak up, a psychological safety instrument may be more targeted. One validated version uses 12 items across four stages: inclusion safety (do people feel accepted and respected), learner safety (can they ask questions and learn from mistakes), contributor safety (are their contributions valued), and challenger safety (can they disagree or take risks without punishment). Items like “I can take reasonable risks without being punished” and “I feel safe disagreeing with the way my team does things” get directly at the behavioral signals that matter most for safety culture.
Why Surveys Alone Aren’t Enough
Healthcare has historically relied almost exclusively on surveys to assess safety culture. Other high-risk industries, like aviation and nuclear energy, routinely combine surveys with direct observation, document analysis, and interviews. A systematic review published in BMJ Open found that surveys alone “do not expose rich insights into dimensions of culture” and that qualitative methods are better suited to uncovering the deeper, contextual layers of how safety actually plays out in daily work.
More than 60% of studies that used qualitative methods did so alongside surveys in a mixed-methods design. The qualitative piece typically involved interviews, focus groups, or both. In one approach, researchers used open-ended comments from a survey to develop follow-up interview questions, letting the survey results guide deeper exploration. In another, nurses led photo-documented walk-throughs of their units, then discussed what they found in focus groups. Five studies analyzed organizational documents, such as safety checklists from management rounds, and triangulated those findings with survey data.
The practical takeaway: run your survey first, then use the results to focus qualitative follow-up on the dimensions that score lowest or show the most variation across units. This tells you not just what people reported on paper, but why.
Leading Indicators to Track Over Time
Surveys give you a snapshot. Leading indicators give you a trend line. Unlike lagging indicators (injury rates, workers’ compensation claims, lost-time incidents), leading indicators measure the behaviors and systems that prevent incidents from happening in the first place.
OSHA recommends tracking a specific set of leading indicators as part of any safety management system:
- Worker participation: level of involvement in safety activities and committees
- Safety suggestions: number of employee-submitted improvement ideas
- Hazard and near-miss reporting: volume of hazards, near misses, and first-aid cases reported
- Response time: how quickly reports are acknowledged and acted on
- Management walkthroughs: number and frequency of leadership safety rounds
- Inspection findings: number and severity of hazards identified during routine inspections
- Training completion: percentage of workers who have completed required safety training
- Corrective action closure: how quickly identified hazards are resolved after discovery
- Preventive maintenance: whether scheduled maintenance activities are completed on time
A rising near-miss reporting rate, counterintuitively, often signals an improving safety culture. It means people trust the system enough to report problems before they escalate. In high-reliability organizations, this is described as “preoccupation with failure,” meaning the organization actively looks for warning signs rather than waiting for something to go wrong. If your near-miss reports are climbing while your actual incidents remain flat or decline, that’s one of the strongest indicators that your culture is maturing.
High Reliability Organization Principles as a Framework
High-reliability organizations (HROs), the kinds of systems that operate under constant risk but maintain extraordinarily low failure rates, organize their safety thinking around five principles. These can serve as an assessment framework in their own right:
- Preoccupation with failure: Does your organization treat the absence of errors as a reason for vigilance rather than complacency?
- Reluctance to simplify: When something goes wrong, do people dig into root causes or settle for surface explanations?
- Sensitivity to operations: Do people recognize how their individual work fits into the larger system and how conditions around them affect safety?
- Commitment to resilience: Does the organization identify threats quickly and respond before problems cause harm?
- Deference to expertise: In a crisis, does authority shift to the person closest to the problem, regardless of rank?
Researchers have developed maturity models based on these principles, tested at Veterans Health Administration hospitals, that score organizations on a progression from reactive to proactive safety management. The key measurement insight from this work is that safety culture is a collective concept. You’re not measuring what individuals think in isolation. You need people who interact with each other daily to characterize how these practices and processes actually function in their shared experience.
How to Run the Assessment
A typical survey administration takes about 10 weeks from planning to final data analysis. The process starts with choosing your instrument, identifying your sample, and building internal support. AHRQ uses a 50% response rate as the threshold for follow-up action. If you’re below that, results may not represent the views of your broader workforce, and you should consider reminder cards, in-person prompts, or extending the survey window. The higher your response rate, the more confidently you can generalize findings to your entire organization.
For the qualitative component, plan focus groups or interviews after you have preliminary survey results in hand. Target units or teams where scores diverge from the organizational average, either high or low. Ask open-ended questions that explore the “why” behind the numbers. If Communication Openness scored poorly in a department, your interviews should probe what happens when someone raises a concern, what the response looks like, and what discouraged people from speaking up.
OSHA recommends evaluating your program at least annually, with scope and frequency adjusted based on changes in regulations, the complexity of your operations, and the maturity of your safety program. The first assessment establishes your baseline. Subsequent rounds let you measure whether interventions are working.
Turning Results Into Action
Measurement without follow-through erodes the very culture you’re trying to build. If people take the time to fill out a survey and nothing changes, they’ll be less likely to participate next time, and less likely to believe leadership takes safety seriously.
AHRQ’s action planning framework provides a clear structure. Start by forming a multidisciplinary team that includes leadership, frontline staff, data analysts, and quality improvement specialists. Use your survey results to identify one or two priority areas, not ten. Set SMART goals: specific enough that everyone understands them, measurable enough to track, and time-bound enough to create urgency.
Before rolling out a large-scale initiative, pilot test it. A Plan-Do-Study-Act cycle lets you try an intervention on a small scale, see what works, adjust, and then expand. Identify process and outcome measures before you begin so data collection fits into daily workflows rather than becoming a separate burden. Build a realistic timeline with milestones, and communicate the plan broadly. People need to see the connection between what they reported on the survey and what the organization is doing about it.
Track both your leading indicators and your composite survey scores over time. The 2024 AHRQ benchmarks show that 53% of respondents reported zero safety events in the past year, which could reflect either a genuinely safe environment or a culture where reporting feels pointless or risky. That ambiguity is exactly why you need multiple data sources. A low event reporting rate paired with high scores on Communication Openness and Response to Error tells a different story than the same reporting rate paired with low scores on those dimensions.