Hospitals collect extraordinary amounts of information. They count infections, medication errors, falls, staffing levels, readmissions and deaths. They convene committees, maintain reporting systems and produce dashboards for senior leaders. Yet the existence of data does not guarantee that an institution understands what is happening inside its walls.
That distinction matters after the BBC News account of the key findings from the Lucy Letby Thirlwall Inquiry. The inquiry found a "complete failure to protect babies on the neonatal unit" where the nurse worked. The particulars belong to that inquiry and the British system. The broader governance question crosses borders: How can hospital leaders determine whether their official picture of safety is incomplete?
For American patients, clinicians and trustees, the answer begins with a less comfortable kind of oversight. Boards should not ask only what the data show. They should also ask what the institution may be failing to capture.
Silence is not evidence of safety
A low number of internal complaints can have several meanings. It may indicate that a unit is functioning well. It may also reflect uncertainty about what deserves reporting, fear of professional consequences, unfamiliarity with the process or a belief that previous concerns went nowhere. A dashboard generally cannot distinguish among those possibilities without further inquiry.
The same problem applies to aggregate measures. A hospital-wide average can look stable while a particular shift, service line or patient group experiences unusual trouble. Small units create an additional challenge because a serious pattern may involve too few cases to stand out statistically. Numbers are essential, but they depend on definitions, reporting practices and human interpretation.
Good oversight therefore includes what might be called negative assurance: deliberate efforts to discover what routine systems do not reveal. Leaders can compare formal reports with anonymous staff surveys, patient complaints, turnover patterns, scheduling disruptions and observations from people who work closest to patients. None of these signals proves misconduct or establishes a cause. Together, however, they can identify questions that deserve an independent review.
Escalation needs a visible destination
Every safety system should make clear where a concern goes after it is raised. That path should not depend entirely on the judgment of one supervisor, especially when the concern involves that supervisor's unit or colleagues. Employees need more than a reporting portal. They need to know who reviews a submission, how urgency is determined, when it moves outside the normal chain of command and whether the person reporting it will receive confirmation that it was examined.
This is not an argument for treating every allegation as established fact. Fair systems protect patients while also applying consistent standards to employees. Preliminary information should be preserved, assessed by people without conflicting responsibilities and separated from conclusions that require fuller investigation. Confidentiality also matters, though it should not become a reason to give reporters no meaningful indication that the process is functioning.
Boards have a distinct role here. They should not manage individual clinical decisions. They should verify that escalation routes exist, that serious concerns can reach an independent authority and that executives cannot quietly redefine alarming information as an ordinary personnel disagreement. Trustees can request periodic reviews of how reports were categorized, how long they remained open and why cases were closed.
Families can ask process questions
Patients and families rarely have enough information to evaluate a hospital's entire safety culture. They can still ask practical questions. Who is responsible for the patient's care at this moment? How is an urgent concern escalated after normal business hours? Where can a family document a problem? Who reviews a complaint if it concerns the immediate care team?
Those questions do not replace clinical judgment, regulatory inspection or formal investigation. They help establish a record and identify the institution's chain of responsibility. Families should keep contemporaneous notes, including dates, names and the substance of responses, while avoiding conclusions unsupported by what they directly observed.
The central lesson is institutional, not personal. A hospital can possess abundant data and still misunderstand its own risk. Safety improves when leaders routinely search for missing information, protect credible escalation and examine whether reassuring reports reflect reality or merely the limits of the reporting system.