Behavioural Medication: Tool, Treatment or Crutch?

Behavioural medication in dogs is one of those subjects that can become very polarised very quickly.

Some people see medication as a shortcut. Others see it as an essential part of helping certain dogs cope and learn.

As usual, I think the reality is somewhere in the middle.

Medication isn't a magic pill, and it certainly isn't a replacement for behaviour modification. But for some dogs, it can be another tool that makes behaviour modification possible in the first place.

Give the medication a fair chance

One thing that is easy to overlook is that behavioural medication isn't necessarily an immediate solution.

Depending on the medication and the individual dog, it can take two or three months to see the full effect. Finding the right dose can also take time. We don't necessarily get the perfect dose first time, and adjustments may be needed along the way.

That means we have to give the medication a reasonable opportunity to work before deciding whether it has been useful.

And while that is happening, the dog still needs behaviour modification, management and appropriate environmental changes.

The medication isn't doing the training.

It is potentially helping the dog reach a state where he is actually capable of benefiting from it.

That distinction matters.

So how long should a dog stay on medication?

This is where I need to make something very clear.

There isn't a magic number.

The approach I'm describing here is my own way of working, not a veterinary guideline or a universal rule that every dog should follow.

Personally, if I'm using medication as part of a behaviour modification plan, I want to see whether we're actually making meaningful progress.

Around the six-month point, I would normally start thinking about the eventual exit strategy.

By around nine months, if we're still not seeing the progress we hoped for, I'm starting to prepare for the possibility that we may need to try something else.

That doesn't mean I stop the medication at nine months.

Quite the opposite.

Any reduction or discontinuation needs to be done safely and in consultation with the prescribing vet. We would normally spend the following months gradually reducing the medication and monitoring how the dog responds.

So in reality, by the time the dog has been assessed, the medication has had a proper opportunity to work, the dose has been adjusted if necessary, behaviour modification has been established, and the medication has then been safely reduced, we can easily be looking at a year or more.

And that's an important lesson for clients.

Behaviour change doesn't happen overnight

This is often the point where families begin to realise just how much work behaviour modification can involve.

They start to see progress, but they also see that it isn't always linear.

There will be good days and bad days.

There may be setbacks.

There may be environmental factors that make things harder.

And eventually, quite understandably, someone will often ask:

"Are there quicker ways?"

The honest answer is that sometimes there are.

But this is where we need to have an honest conversation rather than pretending there is only one possible approach.

Some methods and tools can produce much faster behavioural suppression, particularly aversive methods.

I don't personally choose to use those tools, because I have concerns about their welfare implications and the potential effect they can have on the dog's emotional state and relationship with his handler.

But that doesn't mean I should pretend they don't exist.

Clients deserve honest information

This is an important part of being a trainer.

I have my own boundaries about what methods I am prepared to use.

That doesn't give me the right to deliberately withhold information from a client because I don't like one of their options.

If a client asks whether there are faster ways of addressing a behaviour, they deserve an honest answer.

I can explain the alternatives.

I can explain why I don't use them.

I can explain the potential risks and compromises involved.

And then we can make an informed decision about what is appropriate for their dog and their circumstances.

My job isn't to hide information from people in order to make them choose my preferred method.

My job is to give them enough information to understand the choices they're making.

That doesn't mean I have to agree with every choice.

It means I respect the client's ability to make an informed decision.

What does the research actually tell us?

The evidence around behavioural medication is not perfect, and that's something we should be comfortable admitting.

A 2025 analysis of 52 randomised, placebo-controlled trials involving more than 10,000 people found that the median antidepressant trial duration was only eight weeks, with 88.5% lasting 12 weeks or less.

That is interesting when these medications may subsequently be used for years in real-world situations.

But it would be wrong to take that finding and conclude that we know nothing about antidepressants or their long-term use. There is longer-term evidence; the point is that the controlled trial evidence doesn't always match the duration of real-world use.

The canine evidence is also much smaller, but there is controlled research supporting the use of medications such as fluoxetine for certain behavioural problems.

A Veterinary Evidence review found moderate evidence that fluoxetine can help with some anxiety-related and compulsive behaviours in dogs, while emphasising the importance of behavioural and environmental modification alongside medication.

Fluoxetine is also an approved veterinary treatment for canine separation anxiety when used alongside behaviour modification.

So the sensible position isn't:

"Medication is completely safe and we know everything about it."

Nor is it:

"Medication is dangerous because we don't have perfect long-term evidence."

The sensible position is:

What does the available evidence tell us? What are its limitations? Is medication appropriate for this individual dog? Is it helping? Are we monitoring the dog properly? And are we continually reassessing the plan?

Why I'd rather work with what we can actually measure

This is where I'll draw a comparison, because I think it matters.

If you've read my post on e-collars, you'll know my position there: I don't use them in pet dog training, partly because we simply don't have meaningful research telling us what a dog actually experiences at a low-intensity stim. A dog showing no outward reaction doesn't mean no pain. It might just mean nothing about pain at all, one way or the other. That absence of a clear reaction gets treated as evidence of "barely felt," when it isn't evidence of anything.

Medication is a different situation. It isn't perfect, and I've been upfront above about where the evidence has real limitations, particularly around trial duration versus real-world use. But it's still a body of controlled, published, peer-reviewed research that we can actually look at, question and weigh up. We know roughly what fluoxetine does, how it behaves in the body, what the trial data shows, and where the gaps are.

There's also a more basic problem with "low level" as a concept. A 2013 study examined the electrical characteristics of 13 different e-collar models on the UK market and found large, inconsistent differences between them in voltage, pulse count and stimulus energy, even at settings that were supposedly comparable. The strength of the stimulus also depended on the dog's neck impedance, which changes depending on whether the dog is wet or dry. So "low level" on one collar isn't necessarily "low level" on another, and it isn't even reliably "low level" on the same collar from one moment to the next. That's a difficult starting point for anyone confidently telling you what a low setting does or doesn't do to a dog.

That's the difference for me. I'm not choosing medication over an e-collar because one is "kind" and the other is "cruel." I'm choosing the option where we have genuine research to lean on, over the option where the honest answer to "does this hurt the dog?" is that nobody has properly measured it.

Given a straight choice between a tool with real, examinable research behind it and a tool built on an assumption nobody can actually verify, I know which one I'd rather be defending to a client.

Another tool, not a magic pill

That's ultimately how I view behavioural medication.

It's another tool.

Sometimes that tool can make a huge difference to a dog's quality of life and ability to learn.

Sometimes a particular medication or dose won't work.

Sometimes something else needs to be tried.

And sometimes, after the dog has made enough progress, we can start looking at whether he can manage without it.

There isn't one answer that applies to every dog.

There shouldn't be.

Every dog is an individual, and behavioural problems have different causes, different histories and different levels of complexity.

Medication should never replace good behaviour modification.

But equally, refusing to consider medication simply because it isn't a perfect solution doesn't make much sense either.

The goal isn't to keep a dog medicated forever.

The goal is to give that dog the best possible opportunity to improve, while continually asking whether what we're doing is actually working.

And sometimes, the best route forward is slower than we'd like.

That doesn't make it the wrong route.

It just means we're working with the dog in front of us rather than trying to force every case into the same timetable.

---

Sources referenced in this article:

• Ward, W., Haslam, A., Prasad, V. (2025). Antidepressant Trial Duration Versus Duration of Real-World Use: A Systematic Analysis. The American Journal of Medicine.

• Echeverri, N.S., Govendir, M. (2022). Does the selective serotonin reuptake inhibitor (SSRI) fluoxetine modify canine anxiety related behaviour? Veterinary Evidence.

• Landsberg, G., Melese, P., Sherman, B.L., Neilson, J.C., Zimmerman, A., Clarke, T.P. (2008). Effectiveness of fluoxetine chewable tablets in the treatment of canine separation anxiety. Journal of Veterinary Behavior.

• Irimajiri, M., Luescher, A.U., Douglass, G., Robertson-Plouch, C., Zimmermann, A., Hozak, R. (2009). Randomized, controlled clinical trial of the efficacy of fluoxetine for treatment of compulsive disorders in dogs. Journal of the American Veterinary Medical Association.

• FDA / Reconcile (fluoxetine hydrochloride) prescribing information — approved for canine separation anxiety in conjunction with a behavior modification plan.

• Lines, J.A., van Driel, K., Cooper, J.J. (2013). The characteristics of electronic training collars for dogs. Veterinary Record.

• Rafter-Ritchie, M., Radosta, L., Levine, E., Horst, C., Yin, S., Hammerle, M., Overall, K. (2015). AAHA Canine and Feline Behavior Management Guidelines. Journal of the American Animal Hospital Association.