HealthDNA™ technical whitepaper
LeaderDNA Just Culture & Psychological Safety Assessment™
Will staff on this unit speak up, report errors, and be treated fairly when they do?
- Reporting level
- Aggregate only, minimum 5 respondents
- Audience
- Clinical, allied health, and laboratory staff; reported at unit level only
- Structure
- 7 dimensions × 5 items = 35 items
- Overall index
- Just Culture Index
- Response scale
- 5-point Likert; the fifth item in every dimension is reverse-scored
- Score range
- Dimension 5-25 · total 35-175
1. Construct and design
LeaderDNA Just Culture & Psychological Safety Assessment™ is built on the shared HealthDNA™ framework: 7 dimensions, five items each, with one reverse-scored anchor per dimension to reduce straight-line responding. Every item is written in behavioral, observable language for clinical and laboratory settings, so respondents rate conditions and behaviors rather than abstractions.
Audience: quality, safety, and nursing leaders reading a unit level just culture report. Tie every finding to reporting behavior, harm prevention, and survey exposure. Never identify individuals.
2. Dimensions
Speaking Up Safety
PSAFWhether staff will raise a concern, including to a physician or leader, without fear.
When strong
Staff stop the line when something looks wrong, regardless of who is in the room.
When low
Publicly back the next staff member who stops the line, and say why.
Operational risk
Silence in the moment is how preventable harm reaches the patient.
Coaching move
Model asking for dissent in every huddle and thank the person who gives it.
Error Reporting
ERRWhether errors and near misses are entered rather than absorbed quietly.
When strong
Near misses are reported at a higher rate than actual events, the sign of a healthy system.
When low
Simplify the reporting pathway and time it: if it takes more than three minutes, fix it.
Operational risk
Underreporting hides the risk that eventually becomes a sentinel event.
Coaching move
Treat rising report volume as good news publicly for one full quarter.
Fair Response to Error
RESPWhether the response distinguishes human error, at risk behavior, and reckless behavior.
When strong
Staff can predict how an error will be handled, and the prediction is fair.
When low
Adopt a written just culture algorithm and apply it to every event review.
Operational risk
Punishing human error trains staff to hide it.
Coaching move
Review the last ten events against the algorithm and correct any misapplied discipline.
Leader Response & Follow-Through
LEADWhether leaders act on what is raised and tell staff what happened.
When strong
Staff can name a change that came from something they reported.
When low
Publish a monthly 'you said, we did' summary on the unit.
Operational risk
Nothing kills reporting faster than silence after a report.
Coaching move
Close the loop on every report within 14 days, even if the answer is no.
Learning Systems
LEARNWhether reviews produce system fixes rather than reminders and re-education.
When strong
Event reviews change process, not just policy memos.
When low
Require one system level change from every serious event review.
Operational risk
Re-education as the only fix guarantees repetition.
Coaching move
Ban 'staff re-educated' as a sole corrective action for one quarter.
Respect & Incivility
RESPECTFreedom from disrespect, intimidation, and disruptive behavior across disciplines.
When strong
Disruptive behavior is addressed regardless of the person's role or seniority.
When low
Name the specific behaviors that are out of bounds and enforce them at every level.
Operational risk
Incivility suppresses speaking up more powerfully than any policy encourages it.
Coaching move
Address the highest status offender first, or nothing else you do will be believed.
Equity of Treatment
EQUIWhether response to error and behavior is consistent across roles, shifts, and identities.
When strong
Staff believe the same standard applies to nights as days and to physicians as nurses.
When low
Audit event responses by role, shift, and tenure and share what you find.
Operational risk
Inconsistent enforcement is read as favoritism and ends candor.
Coaching move
Publish the standard and then apply it visibly to the hardest case.
3. Scoring and banding
Each item scores 1-5, with reverse-scored items inverted before summing. A dimension raw score runs 5-25 and is banded as At Risk (5-10), Developing (11-15), Solid (16-20) or Strong (21-25). The total across all 7 dimensions runs 35-175 and is converted to a percentage of scale to place the respondent or group into an overall Just Culture Index tier.
| Tier | Threshold | Interpretation |
|---|---|---|
| Strong Just Culture | 75% of scale and above | Staff speak up and trust the response. |
| Developing | 55% of scale and above | Reporting happens, but follow-through is uneven. |
| Fragile | 35% of scale and above | Staff are filtering what they report. |
| Unsafe to Speak | 0% of scale and above | Silence is the norm. Treat this as a patient safety exposure. |
4. Reporting and the 90-day plan
Every report pairs the Just Culture Index with a dimension breakdown, the three highest and three lowest dimensions and a phased action plan. The plan is deliberately short so a leader can act inside one quarter.
Days 1-30
Close the feedback loop
- Report unit results back to staff.
- Publish what happened to the last ten reported events.
Days 31-60
Change the response
- Adopt one written just culture decision framework for error response.
- Train charge staff and leaders in it.
Days 61-90
Prove it
- Track reporting volume as a positive metric.
- Re-pulse and compare the fear of reporting items.
5. Where it is used
- n/aTesting whether a low event-report count reflects safety or reporting reluctance.
- n/aBaseline before a just-culture policy rollout, so the rollout can be measured rather than announced.
- n/aPairing with quality or laboratory data to explain why a process metric is not improving.
6. Interpretation guidance
- n/aLook at the gap between willingness to report and belief that reporting leads to change. A wide gap predicts reporting decay.
- n/aCompare units. Psychological safety is local and system averages hide the units that need attention most.
- n/aImprovement here usually shows up as more reported events before it shows up as fewer harm events.
7. Privacy model
- No protected health information is collected at any point.
- Results are reported at group level only. A report unlocks only once at least 5 people in that group have responded and individual responses are never shown back to the employer.
- Demographic capture is optional and deliberately coarse, so no combination of fields re-identifies a respondent.
- Free-text comments are never attributed to a named respondent.
8. Stated limitations
- n/aPerception measure. It captures what staff believe will happen when they speak up, not an audit of what did happen.
- n/aUnit-level only; never use it to identify or evaluate an individual respondent.
- n/aNot a regulatory compliance instrument and not a substitute for event review or root cause analysis.
Normative benchmarks and reliability statistics are reported from the live respondent pool as it grows. If you need current sample sizes, internal consistency figures or validation documentation for an RFP, request the technical packet and we will send what exists today rather than an estimate.
9. Related case studies
Run this instrument
Questions about this instrument?
Tell us what you need to review, items, scoring, anonymity thresholds or sample output and we will walk your team through it.
No PHI is ever collected. We only use this to schedule a walkthrough.