Implementing a sentinel event system
Sometimes, drawing a line in the sand is the only option.
Sentinel events, most commonly found in the health and patient care sector, are a subset of bad events that are wholly preventable by an organisation and result in serious harm or death.
Serious harm is typically defined as occurring where, as a direct result of an event:
a life-saving intervention is required (eg resuscitation or surgery)
a person’s life expectancy is shortened
a person suffers an irrecoverable loss (eg physical or financial impairment)
The most important attribute of sentinel events is that they represent a fundamental organisational failure. Even if a person were ignorant, negligent, inebriated, or maliciously motivated in their action, sentinel events signal that the organisation has a non-negotiable responsibility to put sufficiently robust measures in place to prevent those harms from occurring, whether the cause is individual or systemic.
This shift in thinking is critical. Since it is no longer sufficient to find “someone” to blame, the focus of sentinel events shifts to accountability and transparency about what organisational improvements will be made. Indeed, organisations who implement sentinel events are encouraged to implement blame-free reporting; the only non-negotiable stipulation should be that a failure to report a sentinel event will be treated more harshly than any possible consequences of reporting.
In short: Sentinel events draw a line in the sand for the things that absolutely will not be tolerated or accepted as a “cost of business”. Despite the apparent contradiction in seeking out and admitting failure, having a robust reporting and response process for sentinel events acts as a strong foundation for institutional trust.
Sentinel events must meet a strict set of criteria:
They must be specific, easily recognised and clearly defined (eg a surgical procedure being carried out on the wrong patient)
The event must be a real risk (ie have happened before) and not merely be a hypothetical occurrence
It is possible to implement procedural and/or physical barriers intended to make the event “wholly preventable”
Examples of preventative barriers include:
clinical protocols — eg asking for name and date of birth at each handover point
mandatory checklists — eg power down checks, with two people always present
physical protections — eg a switch that makes it physically impossible to connect to two power sources at once
To maximise the effectiveness of sentinel event reporting:
Every occurrence of a sentinel event must be reported to a senior decision maker with accountability for their management and prevention
Every sentinel event should be reviewed and root cause analysis (RCA) carried out, with recommendations implemented and confirmed through subsequent audits
Any trends showing an increase in sentinel event frequency should be investigated, and preventative measures strengthened.
Public reporting of sentinel event occurrences in a transparent and accountable way can be a very effective long-term strategy in building institutional trust. However, even without public accountability, the use of sentinel events is an underutilised but valuable tool to sustain organisational integrity and improve performance over time.