How long does a conversation like this really take?

Less than feared, when it is structured. What lengthens the conversation is not the severity of the news but the absence of an order: you double back, repeat yourself, answer questions already asked. A conversation with a prepared context, the information given in one sentence, the silence respected and a plan at the end usually occupies the space of a long consultation, and it clearly reduces the phone calls of the following days.

What do I do if the owner reacts with anger or accusations?

Anger in this situation is nearly always a response to loss of control, not a technical judgement. Letting it finish without interruption, naming it out loud and returning to the facts works better than any immediate defence. If the accusations concern one specific decision, it is better to arrange a dedicated second meeting with the record in front of you than to argue it while the animal is still under care.

Should I give a numerical prognosis when asked directly?

If you have a solid reference you can give a range, stating that it is a population average and not an individual prediction. If you do not have one, saying so is more useful than an invented figure that will be remembered to the exact digit. Either way it helps to add the criterion that will move the estimate, for example the response to the first course of treatment, so the next conversation has something concrete to hang on.

What changes when the owner arrives with data from home?

The starting point changes, often for the better. A run of weights, pain scores or resting respiratory rates offers a trajectory that spoken history cannot, and lets you anchor the conversation to a trend rather than an impression. The method still needs checking before the content, and it should be said explicitly that those data measure the interval between checks while the consultation interprets the trend.

Veterinary practice10 min readUpdated on August 2, 2026

Breaking difficult news: how to structure the conversation

Preparation, sequence, phrases that shut a conversation down and what must be written afterwards. The consultation treated as a clinical procedure.

Audience: Veterinarians
Species: All species
Scope: Valid everywhere, European Union, Italy

A difficult diagnosis is a clinical act performed by speaking. The technical part takes a few minutes; everything else in the conversation decides whether the person sitting opposite will understand what is happening, remember anything tomorrow and manage a decision they can live with. Anyone working in practice knows the usual outcome of the unprepared version: too much talking, the prognosis delivered before the other person is ready to hear it, silence filled in, a slide into a list of options nobody can rank. The next day the phone call arrives and starts again from zero, or no call arrives at all and the animal reappears three weeks later in worse condition. Structure does not make the conversation cold: it removes improvisation from the points where improvisation causes harm.

Why the conversation deserves a procedure

Nobody improvises an anaesthetic induction, yet almost everybody improvises the consultation that delivers a poor prognosis. The difference is not importance but the visibility of the error: when a conversation goes badly no parameter turns abnormal, only an owner who never calls back. Treating the conversation as a procedure means deciding three things in advance: where it happens, who is present and what the minimum objective is before the person leaves the room.

Minimum objective: the person knows what the animal has, in a sentence they could repeat to somebody else.
Second objective: they know what happens over the next forty eight hours and whom to call if something changes.
Third objective: they know which decisions are theirs and by when, separating the urgent ones from the deferrable ones.
An objective that does not belong to this conversation: extracting an immediate final decision about a long treatment or about euthanasia.

That last point is the most often ignored. Pressure to settle everything today comes from the diary, not from the clinical picture, and it produces fragile decisions that get reversed on the phone the same evening. When the situation allows an hour or a day, saying so explicitly improves the quality of the choice and cuts the number of follow up calls.

The minutes before the sentence

Preparation costs little and decides half the outcome. It comes down to checking five things before opening the door, nearly all organisational rather than clinical.

Check what you actually know
Separate confirmed from suspected and prepare the word you will use for it. Reporting as certain something still awaiting confirmation forces a retraction, and the retraction costs more than the initial caution.
Choose the room
A closed room, no reception desk behind you, nobody else waiting. If the practice has no such room, a consulting room free for ten minutes still beats the corridor.
Ask who needs to be there
Many decisions are taken by two people or by a family. A short question about who else decides avoids repeating the whole conversation by phone to somebody who saw nothing.
Put away the phone and the keyboard
Typing while delivering serious news reads as inattention. The note gets written straight afterwards, and the few minutes saved do not justify the effect.
Prepare the opening sentence
One sentence, built beforehand, that introduces without pre empting. It exists to avoid the two worst openings, namely the list of tests and the automatic reassurance.

The shape of the conversation

Human medicine has a six step protocol for delivering bad news, born in oncology and taught for decades. It transfers well to veterinary practice with one substantial difference: the patient is not the person you are talking to, which adds the theme of decision making responsibility over another living being.

Six steps of the conversation, with the objective and the matching error
Step	Objective	Typical error
Set the context	Closed room, stated time, the right people present	Delivering it standing in the waiting room because the day is full
Find out what they already know	One open question about how they see the situation	Starting from the report without knowing where the other person starts
Ask how much they want to know	Check whether they want the whole picture or the immediate steps	Assuming everybody wants everything, now, in detail
Give the information	One clear sentence, then silence	Three minutes of pathophysiology with no pause
Receive the reaction	Acknowledge the emotion before returning to content	Answering tears with more clinical data
Close with a plan	What happens now, whom to call, when you speak again	Closing with a list of options and no priority

The step most often skipped is the fourth, and not from haste but from discomfort. After the sentence carrying the diagnosis comes a silence of several seconds, which feels endless to the speaker and is exactly what the listener needs in order to register what they heard. Filling that silence with technical detail produces a measurable result: the person remembers nothing said afterwards.

Phrases that close a conversation instead of opening it

Some recurring formulas have the opposite effect to the one intended. They fail not because kindness is misplaced but because they shift the weight of the decision onto the wrong person or cut off the chance to ask questions.

Common formulas, real effect and a usable alternative
Formula	Real effect	Alternative
It is your call, he is your animal	Hands a technical choice to somebody without the elements	Here is what I recommend and why, then we decide together
There is nothing more we can do	Deletes the part of care that always remains available	We cannot cure this disease, we can work on pain and on quality of life
Let us see how it goes	Defers with no criteria and no date	We speak again on Thursday, and before then call if one of these three signs appears
If only we had seen it earlier	Adds guilt to an already heavy picture	This disease gives late signs, which is why it almost always arrives at this stage
Is everything clear?	Gets an automatic yes that verifies nothing	Could you tell me how you will explain this at home?

The reverse also holds: short sentences can reopen a stalled conversation. Asking what worries them most right now moves the discussion from facts to priorities, and often reveals that the main fear is not the disease but pain, cost or the practical burden at home. Those are three different themes and each has a different answer.

What stays in writing

The conversation is forgotten, the document remains. After a difficult consultation the person retains a fraction of what was said, and the retained fraction is distorted by emotion. That is why the written part is not defensive paperwork but a continuation of care, and it belongs to the same day.

In the clinical record: what was communicated, in which words, who was present, what was accepted and what declined.
In the document that goes home: the diagnosis in plain language, the plan for the next forty eight hours and the signs that require calling back.
In the consent form, when a procedure follows: alternatives discussed, risks stated, expected outcome and acknowledged limits.
In the follow up plan: the date of the next contact and who owns it, because a follow up with no owner does not happen.
In the home monitoring material: what to measure, how often and with which instrument, whenever the next decision will depend on a trend.

On the tooling side, the Animiyo veterinary workspace includes clinical records, SOAP notes, record templates, a discharge form, signed consent, treatment plans, laboratory reports shared with the owner and a second opinion request between professionals. On the owner side there are the plain language report, the unified clinical timeline, the pain diary and the quality of life scale, all on the web version. There is no audio recording of the consultation, no structured module dedicated to the conversation itself and no video call: that part stays entirely in the room and in the note written afterwards.

When the news concerns an unexpected outcome

The hardest case is not severe disease but a complication, especially where there is a suspicion that things could have gone differently. Here the temptation to soften is greatest and does the most damage, because an ambiguous account is nearly always discovered and turns a clinical problem into a problem of trust.

Tell them first, before they find out from a report or from another colleague.
Describe the facts in chronological order, separating what was observed from what is being hypothesised.
State what was done to limit the consequences and what will be done now.
Express regret for what happened without improvising attributions of blame in the consulting room.
Put the same account in writing, with times and names, on the same day rather than weeks later.

A note on that last point: memory of an intense working day compresses within hours. An account written while it is fresh, even an imperfect one, is worth far more than an elegant reconstruction written ten days later, and this holds both for the clinical quality of the internal review and for any subsequent scrutiny.

Frequently asked questions

How long does a conversation like this really take?: Less than feared, when it is structured. What lengthens the conversation is not the severity of the news but the absence of an order: you double back, repeat yourself, answer questions already asked. A conversation with a prepared context, the information given in one sentence, the silence respected and a plan at the end usually occupies the space of a long consultation, and it clearly reduces the phone calls of the following days.
What do I do if the owner reacts with anger or accusations?: Anger in this situation is nearly always a response to loss of control, not a technical judgement. Letting it finish without interruption, naming it out loud and returning to the facts works better than any immediate defence. If the accusations concern one specific decision, it is better to arrange a dedicated second meeting with the record in front of you than to argue it while the animal is still under care.
Should I give a numerical prognosis when asked directly?: If you have a solid reference you can give a range, stating that it is a population average and not an individual prediction. If you do not have one, saying so is more useful than an invented figure that will be remembered to the exact digit. Either way it helps to add the criterion that will move the estimate, for example the response to the first course of treatment, so the next conversation has something concrete to hang on.
What changes when the owner arrives with data from home?: The starting point changes, often for the better. A run of weights, pain scores or resting respiratory rates offers a trajectory that spoken history cannot, and lets you anchor the conversation to a trend rather than an impression. The method still needs checking before the content, and it should be said explicitly that those data measure the interval between checks while the consultation interprets the trend.

What to do next

Prepare exactly one thing before your next difficult conversation: the sentence in which you will deliver the diagnosis, written out in full, followed by a silence decided in advance. Then always close by asking the person to describe in their own words what they will do over the next forty eight hours, and record in the file what was said and by whom. Three additions at almost no cost that change the outcome of everything that follows.