Hospital ward managers, staff personality data and safety

What question does the research ask?

Before interpreting a study’s coefficient, a ward manager should establish three things: which trait was measured, what counted as safety behavior and who was in the study group. In workplace safety research, both individual factors and working conditions may matter. The association described in the sources concerning workplace safety locus of control therefore does not answer whether a particular nurse will perform well on their next shift. A correlation in a particular sample is not a test that predicts an individual’s error. Unless the research question is distinguished from the practical question, it is easy to give the result more weight than it deserves. Start the conversation with the scope of the data, not an assessment of the employee.

Measuring behavior

The word “behavior” can mean different things. An employee’s statement that they follow the rules is not the same as an action observed during a shift; a supervisor’s assessment has different limitations from a self-report. Check, too, whether participants worked in healthcare or another industry. In a hospital, raising a concern with someone who can act on it and carrying out a procedure correctly are two distinct actions. A review of 294 publications helps clarify the concept of speaking up in healthcare, but does not establish which intervention effectively develops it. Only after establishing what was measured and how can the findings be meaningfully compared with the team’s daily work.

The trap of a fixed label

A questionnaire result can easily become a permanent label, even though behavior may change with a person’s role, available resources and the team’s response. If an incident occurs during a difficult shift, it should not automatically be explained by a high or low score on one trait. Calling an employee “not very conscientious” says nothing about whether they were given a clear procedure, had time to follow it or could ask for support. Before interpreting an individual’s result, it is also necessary to check whether the test is suitable for that purpose at all. This is an important boundary: describing differences within a group does not provide a sound basis for attributing a fixed pattern of behavior to an individual.

Uncertainty in the results

Even a clear statistical association has a margin of uncertainty. When reading a result, check the confidence interval and possible measurement error. When reviewing publications, check whether similar analyses draw on the same sample. A cross-sectional study cannot, on its own, show that a trait caused a particular behavior. Three studies involving a total of 792 people examined workplace safety locus of control, including in nursing samples. They did not, however, validate staff profiling for future clinical errors. A group-level finding can help frame questions for further analysis, but it does not predict that a named person will make an error in patient care.

The team’s role

Safety also depends on what happens between people. A meta-analysis covering 136 samples, more than 22,000 people and nearly 5,000 groups links psychological safety with behavior and learning. Because many of the included studies were correlational and the analyses covered different levels, this association should not be presented as straightforward proof of causation. For a ward manager, it raises a practical question: can every team member flag a risk, ask questions and correct an action, regardless of their personality profile? It is also worth checking whether procedures are workable for everyone on a given shift. Relationships and working conditions may help shape behaviors that we attribute solely to the individual.

Developing skills

Rather than using a questionnaire to draw up a list of supposedly high-risk employees, choose a behavior that can be practiced and observed. This might be briefly raising a concern during a handover or checking that an instruction has been understood correctly. Feedback should address how the task was carried out: what the person said or did, whether they received a response and what could be improved next time. This approach makes it possible to discuss a skill without assigning an employee a fixed trait based on a single situation. Progress can then be assessed in everyday work, taking account of the task’s difficulty and conditions on the shift, rather than treating a test result as an assessment of the whole team.

Limits on using the data

Before measuring traits, explain the purpose to employees, how consent works and who will see the results. A voluntary developmental assessment should not quietly become a staff evaluation; check local rules on data processing if needed. In a development plan, an employee’s own result can be a starting point for discussing what they want to practice. Safety behaviors, by contrast, should be assessed through observations of work and reports, with the circumstances taken into account. Without validation for the specific use, a trait measure should not, on its own, determine scheduling, promotion or permission to work with patients. Everyone must understand the boundary between support for development and a staffing decision.

A questionnaire result can start a conversation about development, but it cannot replace observing work, assessing conditions on a shift or checking the quality of the measure.

Empatyzer in discussions about safety behavior data

A nurse notices that she puts off asking about an unclear instruction during handover, even though she later has to return to it. In the “About me” view, she can explore her own preferences without treating them as an explanation for everything she does. She then practices a short sentence with Em to ask for clarification and prepares to use it at the next handover. Similarly, she can practice raising a specific concern with someone able to respond. If the manager uses a group view, they may notice different ways of working together within the team and consider how to make it easier to ask questions. This is not a basis for identifying who will make an error. A conversation with Em can instead help an employee reflect on their communication style and prepare to discuss observable behavior: what to say, when to ask for advice and how to review what happened afterward.