Safety Learning Series
The Second Person the Event Harmed
An adverse event harms two people. Support both.
When a patient is harmed by a medical error, everything in a hospital moves toward the patient. Stabilize, treat, escalate, call the owner, document. The response is fast and it is correct, and nobody needs to be told to do it.
We almost never think about the other person the event just injured.
She finishes the shift. She is professional about it, because that is what you do. She drives home, and somewhere on that drive it starts: the replay. Then it happens again at 2 a.m., and again the next night. She questions whether she belongs in this work at all. She second-guesses the following day's cases, hesitates on a decision she has made a thousand times, and tells no one, because admitting how much it shook her feels like admitting she is not cut out for the job.
A name for it
In 2000, Albert Wu wrote a short piece in the BMJ that gave this experience a name: the second victim.
His argument was simple and, at the time, almost unspoken. The patient is harmed first, which is never in question, and the ordering matters. But the professional involved is frequently wounded too, carrying guilt, self-doubt, and shame long after the case is closed and the file is shut.
The term has been contested ever since, and the objection is a fair one worth stating rather than skipping past. Some patient advocates argue that "victim" belongs to the person who was actually harmed, and that extending it to the clinician risks blurring an important line, particularly in conversations with families. Several health systems have moved toward language like affected staff or the clinician involved for exactly that reason.
That debate is legitimate and unresolved. What is not up for debate is the underlying finding: the harm to the clinician is real, it is common, and it is mostly invisible.
In veterinary medicine, it is strikingly common
The best data we have in our profession comes from a 2018 survey of Veterinary Information Network members. Six hundred and six veterinarians completed it, and 73.8% reported involvement in at least one near miss or adverse event. Broken out: 64.2% reported a near miss, 29.5% an adverse event.
Nearly three-quarters.
The aftermath is where it becomes a safety issue rather than only a human one. The same study found meaningful reductions in wellbeing following an adverse event, with more than a third of affected veterinarians reporting problems sleeping and persistent feelings of guilt.
Sit with the arithmetic for a second. In a practice of twelve veterinarians, that survey suggests roughly nine have been through this. Some of them are on shift right now. Almost none of them have talked to anyone about it, because there was no obvious moment to raise it and no obvious person to raise it with.
More recent work has begun looking specifically at veterinary anesthesia, where the second-victim experience has now been characterized across two published parts: the experience itself, and the support structures that might address it. The picture is consistent. This is not a fringe phenomenon affecting the unusually fragile. It is the normal response of a conscientious person to having been part of something that hurt a patient.
Why this belongs in a safety conversation
It would be easy to file this under wellbeing, an HR concern, adjacent to burnout, important but separate from the work of preventing harm.
That filing is a mistake, and here is the mechanism.
A clinician who is hurting and unsupported does not raise the next signal.
They go quiet. They avoid the case type. They stop volunteering observations, because every observation is now a reminder. And a meaningful number of them leave the profession, or the specialty, or the hospital, carrying with them exactly the experience that would have made them the most useful person in the room during your next review.
Look at what that does to everything else in this series. The reporting culture depends on people being willing to speak. The near-miss capture depends on people volunteering things nobody would otherwise know. The closed loop depends on the reporter staying engaged long enough to see it close. Every one of those depends on a person who has not been quietly damaged and left alone with it.
The same silence that buries the report also isolates the person. They are not two problems. They are one problem with two symptoms, and hospitals routinely assign them to two different departments who never speak.
A culture that learns from error and a culture that cares for the people involved are not two projects. They are the same one.
What support actually looks like
The instinct, when a colleague is visibly struggling, is either to say nothing, for fear of making it worse, or to reach immediately for professional referral. Both are understandable and both usually miss.
The published work on peer support suggests the useful intervention is earlier, smaller, and much less clinical than people expect. Programs like RISE (Resilience In Stressful Events), originally developed in human healthcare and now being studied in veterinary settings, are built on a straightforward premise: trained peers, available quickly, offering a confidential conversation with no agenda. Not therapy. Not debriefing. Not an investigation with a kind face.
The active ingredients appear to be:
Speed. The window that matters is hours to days, not weeks. By the time a formal process reaches someone, the isolation has already set.
A peer, not a supervisor. Someone who has stood in the same room. The person who will also be writing your performance review is structurally the wrong person to hear that you are questioning your competence.
No agenda. The conversation is not gathering facts for the case review. If those two functions blur, the support function dies immediately and permanently.
Normalization. Simply hearing that this response is what happens to good clinicians, that it is common, and that it passes, does a substantial amount of the work.
Explicit permission to decline. Offered, never imposed. Mandatory support is a contradiction, and people who are pushed into it disengage from the next offer too.
You do not need a formal program to start. You need someone to notice, and to say something, within a day.
What to do this week
Notice out loud. "That was a hard one. How are you doing with it?" Asked privately, and asked again three days later, because the first answer is almost always fine.
Separate support from review. Whoever is running the case analysis should not be the person offering the shoulder. Name two different people, and say so before anything happens.
Ask one question at your next difficult case discussion. "Who's checking on the people involved?" If nobody has, you have found a gap that costs nothing to close.
Decide in advance. The worst time to design your response is during the event. Write down now: who reaches out, how fast, and what happens if the person says no.
Care for the patient first, always. Then remember there is a second person in the room who may need care too.
A question for your team
When someone here is shaken by a hard outcome, does anyone check on them, or are they left to get over it alone?
If you're the second person
If you or a colleague are struggling after a difficult clinical outcome, support exists. You do not have to wait until you are in crisis to reach for it.
Not One More Vet, Peer Support
Confidential peer support and mental-health resources for veterinary professionals, including opportunities to connect with others who understand the pressures of veterinary medicine.
Veterinary Hope Foundation, Community Groups
Small, virtual community groups for veterinarians and veterinary team members, facilitated by trained leaders and designed around shared roles, career stages, and experiences. Groups provide a structured space for connection, reflection, and support.
AVMA Wellbeing Resources
Wellbeing and mental-health resources for veterinary professionals, including links to support programs and state-specific resources.
You do not have to be in crisis to use these resources. Support after a difficult case, unexpected outcome, error, or near miss can be valuable long before distress reaches that point.
More in the safety-learning series: browse all posts.