{"id":188752,"date":"2026-09-30T08:38:51","date_gmt":"2026-09-30T05:38:51","guid":{"rendered":"https:\/\/fendaf.ee\/?p=188752"},"modified":"2026-09-26T13:27:33","modified_gmt":"2026-09-26T10:27:33","slug":"medication-safety-in-dogs","status":"publish","type":"post","link":"https:\/\/fendaf.ee\/en\/health\/medication-safety-in-dogs\/","title":{"rendered":"Medication Safety: Why \u201cSafe\u201d Does Not Mean the Same Thing for Everyone"},"content":{"rendered":"

A while ago, when I was updating FendaF\u2019s article about chemical tick treatments<\/a>, one seemingly very simple word kept bothering me: safe.<\/p>\n

We use it all the time when talking about medicines. \u201cThis medicine is safe.\u201d \u201cA lot of dogs use it.\u201d \u201cSide effects are rare.\u201d \u201cThe medicine is approved.\u201d<\/p>\n

But what does the word \u201csafe\u201d actually mean?<\/p>\n

For a pet owner, it can sound quite absolute: if a medicine is safe, then nothing serious should happen to my dog.<\/p>\n

In medicine and pharmacovigilance, however, \u201csafe\u201d usually does not mean \u201crisk-free\u201d.<\/strong> The European Medicines Agency\u2019s veterinary medicines committee assesses, among other things, whether a medicine has a positive benefit-risk balance in the target animal population. The FDA describes an approved veterinary medicine as one whose safety and effectiveness have been demonstrated for the intended animal species under the specified conditions of use. In both cases, benefit, risk, dose, indication and conditions of use are part of the equation. It is not a promise that a bad reaction is impossible.<\/p>\n

And this is where the same word starts to mean different things to different people.<\/strong><\/p>\n

\u201cSafe\u201d does not mean \u201crisk-free\u201d<\/h2>\n

Almost every medicine that has an effect can also have unwanted effects. So the question is not simply whether a medicine carries risk, but how large that risk is, how serious the possible harm is, how much benefit the medicine offers, what alternatives exist and which animal is receiving it.<\/p>\n

A regulator\u2019s task is not to prove that no animal will ever experience an adverse effect. Quality, effectiveness and safety are assessed, together with whether the expected benefit outweighs the risks under the proposed conditions of use. EMA guidance treats benefit-risk assessment as a separate, systematic process.<\/p>\n

That is a very different question from the one a pet owner asks: \u201cAm I willing to accept this risk for my dog?\u201d Both questions are legitimate. They are simply not the same question.<\/p>\n

Safe for whom, and in what sense?<\/h2>\n

Medication safety can be viewed from at least four different perspectives.<\/p>\n

From the regulator\u2019s perspective, the central question is whether the benefit-risk balance is good enough in the target population for the medicine to be authorised.<\/p>\n

The manufacturer must provide data on the medicine\u2019s quality, safety and effectiveness and, after authorisation, continue monitoring safety. In the European Union, marketing authorisation holders for veterinary medicines must collect information about adverse events, assess safety signals and continuously monitor the benefit-risk balance of their products.<\/p>\n

For the veterinarian, population statistics are joined by the individual animal sitting in front of them. What disease does this animal have? How severe is it? How much could the medicine help? Could other illnesses or medicines change the risk? Can the response to treatment be monitored? Can the dose be adjusted?<\/p>\n

Modern professional veterinary guidance also emphasises treatment in context. For example, the RCVS states that a treatment plan should take into account the animal\u2019s health and welfare, the circumstances of the individual patient and the owner\u2019s circumstances as well.<\/p>\n

And then there is the owner. The owner lives with the consequences at home. They see whether the dog eats. Whether the dog sleeps. Whether the dog vomits. Whether the stomach is okay. Whether the dog wants to go for a walk. Whether thirst changes. Whether urination increases. Whether the dog still seems like themselves.<\/p>\n

A medication decision may be made during ten or twenty minutes in the consultation room, but the real life of the treatment unfolds during the next hours, days, months or years at home.<\/p>\n

Population risk is not the same as one individual dog\u2019s risk<\/h2>\n

If an adverse effect is rare, that tells us something important about the population. It does not mean the adverse effect is unimportant.<\/p>\n

If the reaction happens to your dog, it is not especially comforting for that dog to know that it happened very rarely to other dogs.<\/p>\n

In addition, one individual animal\u2019s risk may not be exactly the same as the average risk in the study population. Age, concurrent disease, kidney or liver function, other medicines, dose, duration of treatment and many other factors can change it.<\/p>\n

That is why population statistics cannot simply be turned into the statement: \u201cIf it is safe for most dogs, it must also be safe for mine.\u201d Population data provide an essential foundation for decision-making. The individual patient shows how that knowledge applies in real life.<\/p>\n

A rare adverse effect can still matter<\/h2>\n

This brings us to another issue that, in my view, is discussed too little: probability and severity are two separate axes.<\/p>\n

An adverse effect may be common but minor, such as a short-lived gastrointestinal upset. Another may be very rare but potentially life-threatening. You cannot judge both by one number alone.<\/p>\n

If a medicine is necessary to treat a severe or life-threatening disease and the alternatives are poor, even a relatively unpleasant or serious risk may be acceptable. If, however, a medicine is given preventively to a healthy animal, or in a situation where several reasonable alternatives exist, the threshold for accepting the same risk may be completely different. The same benefit-risk thinking is very clear with anti-inflammatory pain medicines as well.<\/p>\n

This does not mean one medicine is automatically \u201cgood\u201d and another \u201cbad\u201d. It means the meaning of risk cannot be assessed separately from the benefit of the medicine.<\/p>\n

An adverse effect is not automatically a \u201cbig\u201d or \u201csmall\u201d problem<\/h2>\n

The same adverse effect can mean completely different things in different situations.<\/p>\n

If a medicine helps a dog breathe, controls severe heart disease or saves the dog\u2019s life, an owner may knowingly accept an adverse effect that they would never accept from a preventive medicine or a medicine that only improves a minor inconvenience.<\/p>\n

In one situation the thought may be: \u201cThis is unpleasant, but without this medicine the situation would be much worse.\u201d In another: \u201cThis risk is too high for me because I have other options.\u201d<\/p>\n

So choosing a medicine is not just a matter of reading a list of side effects. It is always a relationship between at least three things: what the medicine gives, what it may cost, and what happens without it.<\/p>\n

The RCVS highlights the BRAN framework as one practical decision-support tool for veterinarians: Benefits, Risks, Alternatives and what would happen if Nothing were done.<\/p>\n

But adverse event reports do not prove the medicine caused it, do they?<\/h2>\n

No, they do not. And that distinction is very important.<\/p>\n

In veterinary pharmacovigilance, an adverse event report can describe an unwanted event that occurred after a medicine was used, even when a causal relationship with the medicine has not yet been proven.<\/p>\n

A dog may have several diseases at the same time. The dog may be receiving several medicines. A symptom may be caused by the underlying disease, another medicine or an entirely separate factor.<\/p>\n

The FDA also stresses that an individual adverse event report usually cannot prove that the medicine caused the event. Nor can the number of reports alone be used to calculate the true incidence of an adverse effect, because it is often not known exactly how many animals used the medicine.<\/p>\n

But sometimes the next leap is too large in the opposite direction: \u201cIf the report does not prove causation, the report means nothing.\u201d That is not correct either.<\/p>\n

One of the main purposes of adverse event reporting is signal detection. If a particular event repeatedly appears after use of a medicine, it may be the first clue that the phenomenon deserves closer investigation. Regulators use such data to detect new safety risks and, where necessary, medicine information may be changed or other regulatory measures taken.<\/p>\n

So two things can be true at the same time: an adverse event report is not automatic proof of causation, but neither is it a worthless anecdote.<\/p>\n

And then there is underreporting<\/h2>\n

Adverse event databases have another major problem: not everything gets reported.<\/p>\n

The FDA states directly that reporting to its veterinary adverse event database remains incomplete. A European survey of veterinarians likewise found substantial underreporting.<\/p>\n

Even more interesting is a study published in 2024 that examined electronic health records from first-opinion veterinary practices in the United Kingdom. Researchers identified 827 suspected adverse drug reactions, and in about 90% of those cases there was no note in the clinical record indicating that the reaction had been formally reported.<\/p>\n

This does not mean that \u201conly 10% of all adverse effects worldwide make it into official statistics\u201d. That figure cannot be used universally. But it does show very clearly why the number of official reports cannot be treated as a complete count of everything that has happened in real life.<\/p>\n

And this works in both directions. We should not use the report count as though every report proves medicine-caused harm. But neither should we say: \u201cThere are only this many reports, therefore there cannot be any more cases.\u201d<\/p>\n

There is another reason for underreporting that is discussed less often: the timing may not make the connection obvious.<\/p>\n

An unwanted effect does not necessarily appear on the first, second or third day of treatment. With some medicines, problems can develop during long-term use and only become apparent weeks or months later. Some longer-term effects may take even longer to notice.<\/p>\n

And the longer the interval between starting a medicine and the appearance of a health problem, the harder it may be to notice a possible connection. If a dog starts vomiting an hour after a dose, the medicine\u2019s role comes to mind quickly. If a change develops gradually over months, it may be attributed to age, the underlying disease or something else entirely.<\/p>\n

That does not mean a health problem that appears later was automatically caused by the medicine. Causality assessment has to consider the disease itself, other medicines and many other factors.<\/p>\n

But the opposite conclusion would also be wrong: a long time interval does not exclude a possible role for the medicine.<\/p>\n

This is why monitoring becomes especially important during long-term treatment. What was the animal\u2019s condition before treatment? What has changed during treatment? Have blood results, kidney or liver function, digestion, behaviour or overall wellbeing changed? Could the change be related to the disease, the medicine, the dose, the combination of medicines or something else?<\/p>\n

And this may be one mechanism behind underreporting: if a possible connection with the medicine is never recognised, the event naturally never reaches an adverse event report either.<\/p>\n

Informed choice must come before the medicine<\/h2>\n

For me, one of the most important principles in this entire topic is this:<\/p>\n

Informed choice means that the owner knows about relevant and potentially serious adverse effects before the medicine is used, not only after their animal experiences one.<\/p>\n

This does not mean a veterinarian has to read every line of every package leaflet aloud. It means the owner receives the information that matters for making a decision.<\/p>\n

What is the goal of the treatment? How large is the likely benefit? What are the more common risks? Is there a rare but serious risk worth knowing about? What alternatives exist? What should be monitored during treatment? Which symptom should prompt a quick call to the veterinarian? And what may happen if the medicine is not used at all?<\/p>\n

For example, the UK RCVS guidance on informed consent says that clients should have an opportunity to consider reasonable treatment options and should be told the purpose of the treatment, the likely outcomes and both common and serious risks.<\/p>\n

That is not \u201canti-medication\u201d. It is genuinely informed decision-making.<\/p>\n

The owner is not simply the person who puts the tablet in the dog\u2019s mouth<\/h2>\n

Faddy\u2019s last few months have made this especially clear to me.<\/p>\n

A veterinarian sees the dog during a consultation and has medical knowledge, diagnostic tools and experience. But I live with this dog 24 hours a day.<\/p>\n

I notice tiny changes that may never be visible in the clinic. Does he get up happily? How long a walk can he manage? Is his breathing calm while he sleeps? Is his abdomen starting to enlarge again? Has his appetite changed? How much does he drink and urinate? Does he sleep through the night or do we have to go outside several times?<\/p>\n

Those observations alone do not prove why a change happened. But they are part of the clinical picture.<\/p>\n

This is exactly why collaboration between the pet owner and the veterinarian<\/a> matters so much: one side sees the medical picture, the other sees the animal\u2019s real life. Shared decision-making models in veterinary medicine likewise start from combining clinical knowledge with the owner\u2019s knowledge, values and possibilities.<\/p>\n

Faddy and furosemide: four completely different questions<\/h2>\n

Furosemide, a diuretic whose job is to help the body remove excess fluid, has become part of our life because of Faddy\u2019s heart condition<\/a>.<\/p>\n

And his example shows very clearly why the word \u201csafe\u201d alone is not enough.<\/p>\n

Recently I spoke with the owner of another large dog who also receives furosemide because of a heart problem. That dog receives considerably more tablets than Faddy.<\/p>\n

Can we conclude from that that Faddy could safely receive the same amount? No.<\/p>\n

First of all, the number of tablets does not even tell us the dose unless we know tablet strength and the dog\u2019s body weight. But even dogs receiving exactly the same milligrams per kilogram may differ in treatment need and tolerance.<\/p>\n

In dogs, furosemide is generally dosed according to effect, not according to the principle that \u201cmore is better\u201d. In heart failure, the dose is adjusted according to how well it controls fluid accumulation and how the individual patient tolerates treatment. Higher doses increase, among other things, the risk of dehydration, reduced circulating blood volume, electrolyte disturbances and worsening kidney function, which is why kidney values and electrolytes may need to be monitored after increasing the dose.<\/p>\n

There is another important nuance here. Increased urination with furosemide is not necessarily a \u201cside effect\u201d in the classic sense. It is part of the medicine\u2019s intended pharmacological action.<\/p>\n

But intended effects also have real-life consequences. If a dose causes extreme thirst, very frequent urination, several night-time trips outside or makes life very difficult for the dog and the whole household, that is still part of what should be considered when assessing whether a treatment plan is sustainable.<\/p>\n

So with Faddy there are at least four different questions: is this dose medically permissible? Is this dose necessary to control his condition? Does his body tolerate it well? And is this treatment plan sustainable in real life while preserving the best possible quality of life?<\/p>\n

Those are not synonyms.<\/p>\n

The fact that a dose can be increased according to a guideline does not automatically mean that a higher dose is the best solution for this particular animal. At the same time, an unpleasant adverse effect or practical problem does not automatically mean that a necessary medicine should be stopped. In severe heart disease, the benefit of medication can be enormous. The question is balance.<\/p>\n

An integrative view does not mean \u201cmedicine or nature\u201d<\/h2>\n

For me, the value of integrative veterinary medicine is precisely that it asks a larger question. Not only: \u201cWhich medicine removes this symptom?\u201d but also: what influences the problem, what is the animal\u2019s overall health, what are their nutrition, body weight, muscle mass, movement, stress, digestion, other medicines and supplements, and what can be done alongside medication.<\/p>\n

That does not mean replacing a necessary medicine with something \u201cnatural\u201d simply because natural sounds more pleasant. A natural substance can also be powerful, interact with medicines or be unsuitable for a particular animal.<\/p>\n

For me, an integrative approach means that we do not automatically choose between two camps. We ask: what does this particular animal actually need?<\/p>\n

Sometimes the answer is medication. Sometimes it is medication together with nutrition, lifestyle, physiotherapy or other support. Sometimes it is reasonable to consider alternatives before starting a medicine. And sometimes a medicine is so important that the risk of not using it is far greater than the risk of the medicine itself.<\/p>\n

What questions can an owner ask before using a medicine?<\/h2>\n

I do not think a pet owner has to become a pharmacologist. But they should feel able to ask questions.<\/p>\n

What is the purpose of this medicine? How much benefit is realistic for my dog? What adverse effects are more common? Which adverse effects are rare but serious? Does my dog have a disease or another medicine that changes the risk? What alternatives are available? What happens if we do not use the medicine now? What should be monitored during treatment? Which symptoms mean I should contact the veterinarian immediately? And if something unexpected happens, should it be reported as a suspected adverse event?<\/p>\n

These are not troublesome questions. They are informed-choice questions.<\/p>\n

\u201cSafe\u201d is simply too small a word for something this big<\/h2>\n

I do not think we should stop using the word \u201csafe\u201d when we talk about medicines. But we should understand what it means.<\/p>\n

\u201cSafe\u201d does not mean: nothing bad can happen.<\/strong><\/p>\n

It often means something much more precise: under the specified conditions of use and in the target population, the medicine has a sufficiently favourable benefit-risk balance.<\/p>\n

For one individual animal, however, a whole set of additional questions must be added. How great is the expected benefit? What is the individual risk? How serious could the consequences be? What alternatives exist? Can the risk be reduced through monitoring or dose adjustment? And after receiving all the information, what level of risk is the owner prepared to live with?<\/p>\n

Because good treatment is not only treatment that works pharmacologically. Good treatment also has to work for that particular animal in real life.<\/p>\n

And perhaps the most important question to ask when you hear the word \u201csafe\u201d is: \u201cSafe for whom, under what conditions, and compared with what?\u201d<\/strong><\/p>\n

Used sources:<\/h3>\n