Dazzled by Science Part 8 – Behavior Medications – FLUOXETINE

Lana Kaiser MD, DVM


January 25, 2024

I often hear complaints that “the vet only prescribed fluoxetine,” as if prescribing this drug is a “bad decision.” I also hear that “boarded veterinary behaviorists rarely prescribe fluoxetine,” as if because they use it rarely it shouldn’t be used at all. We hope to provide insight into why veterinarians might go to fluoxetine as a first-choice drug, and why BV often need a more nuanced approach.


Before talking about individual behavior medications, we need to understand what goes into the veterinarian’s decision to prescribe a specific medication. We are going to assume that all medical issues that can adversely affect behavior, including pain, have been dealt with. As a disclaimer we are only talking about prescription behavior medications that can be used as an aid in managing behavioral issues in dogs, we will not be discussing nutraceuticals, supplements, pheromones, or cats!


When prescribing a medication for a behavior issue the veterinarian needs to know what the abnormal behavior is and when it occurs. So (1) Is it predictable, like thunderstorm or fireworks phobia? If so a “short-term” (aka “game changer”) medication can be given before the predictable event. (2) Is it “all the time,” like anxiety? If so a long term (aka long game) medication can be used daily. (3) Is it all the time and also predictable? If so a combination of game changer and long game meds can be used. (4) Is it “all the time” and the dogs and humans are stressed to the max? In this situation a game changer med can be given to “provide” rapid relief until the long game med becomes effective.  Choosing medications for a specific dog and their humans is part art and part science.


So, what are game changer meds? These are drugs with a relatively rapid onset and short duration of action. That means that they start to work “quickly” (within an hour +/-) and they only last a “short time.” They are useful when you can predict the cause of the behavior, but they also can be effective when combined with other game changers or long game meds. They can be considered quite versatile. Game changer meds include trazodone, gabapentin, clonidine, and dexmedetomidine. Game changer meds will be covered in a later article.


What are long game medications? In the broadest sense these are drugs developed to treat human depression, anxiety, etc. that have found use for behavior issues in dogs. These drugs take weeks or months to have an effect on behavior, which is interesting since the neurotransmitters are altered almost immediately after ingesting the drug. This makes us really wonder what we really know about the mechanism of action of these medications.


We are going to talk about the long game meds and the science that describes their use for dogs with behavioral issues over the next few articles. We will cover the peer reviewed literature, the studies that describe their safety and efficacy in dogs – how do we know the drugs are safe? how do we know they work? This is what the peer reviewed science tells us. We won’t cover the myriads of rodent and human studies. We also won’t cover abstracts and presentations that have not been published as full articles nor will we cover individual case reports. We also won’t cover multiple behavior medications used together (that will also come in a later installment). What we will cover is prospective clinical trials and retrospective studies of a specific drug used for behavior issues in dogs. [If you need a refresher on how drugs for dogs are approved, what the label means, or what extra-label drug use is you can find that information here https://aggressivedog.com/2023/06/19/dazzled-by-science-part-5-behavior-medication/]


Long game medications used in veterinary medicine fall into three broad categories: SSRIs = selective serotonin reuptake inhibitors; TCAs = tricyclic antidepressants, and SNRIs = serotonin and norepinephrine reuptake inhibitors. You will recall that reuptake inhibitors increase the concentration of neurotransmitter in the synaptic space by blocking the return of the neurotransmitter to the nerve. The most commonly used long-acting behavior medications in dogs are fluoxetine (SSRI, labeled and branded for dogs as = Reconcile ®) and clomipramine (TCA, labeled and branded for dogs as = Clomicalm® & Caniquel®). As you know, drugs that are labeled for use in dogs have been tested for efficacy and safety in dogs. One way to review some of the early research on behavior medications labeled for dogs is to look at the package insert [1]. It will give you the specific label indication (“RECONCILE chewable tablets are indicated for the treatment of canine separation anxiety in conjunction with a behavior modification plan);” side effects observed/reported; and the number of animals in each group (control = no drug vs treatment = medication being tested). Fluoxetine is absorbed orally and metabolized in the liver to norfluoxetine which is an equipotent metabolite, which means that the active metabolite extends the duration of action of the drug.


Fluoxetine is an SSRI that was developed to treat depression in humans and marketed as Prozac in 1988. By inhibiting the reuptake of serotonin, the serotonin concentration in the synapse increases (see Dazzled #7 for more info https://aggressivedog.com/2023/10/31/dazzled-by-science-part-7-behavior-medications-continued/). Of behavior medications in veterinary medicine, fluoxetine has been the most studied, both in placebo-controlled  and retrospective studies. Some of the reasons veterinarians prescribe fluoxetine are that it has been studied in dogs, is considered “reasonably safe,” is efficacious, is approved by the FDA for use in dogs for specific conditions, and can legally be used extra label.


Several studies have evaluated the efficacy of fluoxetine for separation anxiety (SA) in client owned dogs. In a multicenter, double blind, placebo-controlled study, Simpson et al. [2] evaluated the effect of daily oral fluoxetine on SA behaviors (destructive/rearranging behavior, inappropriate urination or defecation, hypersalivation). This study took place at multiple veterinary clinics and neither the client nor the veterinarian knew if the dog received fluoxetine or the sham pill. The sham pill was formulated to look and taste like Reconcile® (placebo = sham medication). Humans in both groups (treatment, i.e. Reconcile® = fluoxetine; n=122 and control = tablet identical looking to Reconcile = placebo; n=120) were taught behavior modification techniques for SA. To be included in the study, the SA behaviors had to have been present for at least one month, and the diagnosis confirmed by a veterinary behaviorist. Each owner identified four triggers for SA (for example departure for work, leaving for the store etc.). The frequency of these behaviors was recorded daily for 2 weeks and served as “baseline” SA behaviors. Behaviors were then recorded daily for an additional 8 weeks. Overall and specific SA behaviors were scored by owners (0 = absent; 1 = mild; 2 = moderate; 3 = severe ). The measure of effectiveness was change (improved overall severity score) from baseline, not a comparison between the two groups. The percentage of dogs with improved overall severity score was higher in dogs treated with fluoxetine than controls (Figure 1). Eight weeks after treatment, 72% of dogs treated with fluoxetine showed improvement in overall severity score, compared with 50% of placebo-treated dogs. Adverse events including calm/lethargy/depression were noted more frequently in the fluoxetine group, as was anorexia/decreased appetite. Seizures were identified in three dogs in the treatment group and one in the control. Although one does need to consider the “placebo effect” on the humans (who want their dogs to improve), the lack of robust statistical significance, and the source of funding (the company that makes Reconcile), this study strongly suggests that fluoxetine may be useful, when combined with behavior modification, for dogs with SA. The authors concluded that given the potential gravity of outcome of dogs with SA (euthanasia), the benefits of fluoxetine treatment outweigh the risks. The study also evaluated safety and efficacy of Reconcile, was funded by Lilly Animal Health, and 5 of 14 authors were affiliated with Elanco, a subsidiary of Lilly at the time.



Figure 1 -Norwegian Institute for Nature Research


In a similarly designed study Landsberg et al. [3 ] evaluated the effect of fluoxetine on SA behaviors, without behavior modification. Compared with placebo, fluoxetine treated dogs had a higher incidence of improved overall SA severity score at every weekly interval during the treatment period, “with significant differences occurring at weeks 1 and 4.” They also noted decreased destructive/rearranging behavior and decreased inappropriate urination in the fluoxetine group. Seizures were reported in one dog in each group. They concluded that “once-daily fluoxetine dosage of 1–2 mg/kg has some efficacy in improving overall SA severity scores in dogs, even when used without a formal behavior modificationplan.” The study was also funded by Elanco and 2 of the 6 authors were affiliated with the company.


These two studies suggest that fluoxetine is a generally safe medication for dogs, and can be useful in treating SA, both with and without behavior modification. Seizures were reported in both groups, causation was not determined. Anorexia/decreased appetite is noted in dogs on fluoxetine, which may decrease with decreasing dose or switching to another SSRI. In the Simpson study [2], aggression was reported more frequently in controls (n=9) than fluoxetine (n=5). In the Landsberg study [3], aggression was reported in 4 dogs from each group (control, n=4; fluoxetine, n=4).


The FDA requires, by law, that warnings, precautions, contraindications, and adverse events be listed on the “prescribing information for veterinarian”[4]. You might note in the prescribing information for Reconcile under precautions that it is “not recommended for the treatment of aggression” and also that it has not been “clinically tested for the treatment of other behavioral disorders” and that studies on “breeding, pregnant, or lactating dogs, and in patients less than 6 months have not been conducted.” Precautions are basically a statement of awareness, not a prohibition, where the veterinarian should weigh the risks against the potential benefits. In the case of Reconcile®, this does not mean that the drug cannot be used in cases of aggression, breeding, pregnant, or lactating animals, or those less than 6 months of age, but that the drug should be used “with caution” in those cases, weighing the risks and the potential benefits.


Fears of “making the dog worse” or making the dog “aggressive” are often expressed by humans with dogs when discussing the use of behavior medications. Granted every time you “mess with neurotransmitters” there is the potential for a change in behavior, however, correlation in time does not indicate causation. In the Simpson study “aggression” was reported in 5 of 117 dogs treated with fluoxetine (4.3%) and 9 of 112 dogs in the control group (8%). In the Landsberg study “aggression” was reported in 4 of 99 dogs treated with fluoxetine and 4 of 99 dogs in the placebo control group. These data suggest several things: (1) there is no evidence that fluoxetine increases aggression; however, fluoxetine may increase aggressive behavior in some individual dogs; (2) there is no evidence to suggest that fluoxetine should not be used in dogs exhibiting aggressive behaviors; (3) for most dogs with behavior issues, the benefit of treatment largely outweigh the risks; and (4) veterinarians assess the dog in front of them, understand their clients financial and personal limitations, and use their background in pharmacology, physiology, and pathology to assess which medications might provide the most benefit with the least risk.


In a retrospective descriptive study, Chutter et al. reviewed medical records of dogs with a variety of behavioral issues [5]. Of the 288 dogs seen in the Cornell Behavior Clinic from 6/14/12 to 12/31/16, 134 were prescribed fluoxetine. Of the 134 clients contacted, 93 responded (66%) and of these 88 were included in the study. Behavioral issues were divided into three broad categories: anxiety, aggression, and other. Owners were contacted by phone or email and asked if their dogs’ behavior improved (positive), did not change (neutral), or worsened (negative) while on fluoxetine. The primary diagnoses of the 88 dogs were fear aggression (44%) and anxiety (22%). Overall, 52 (59%) dogs improved (positive response); 28 (32%); showed no change; and 8 (9%) worsened (negative response). Improvement was noted in all sub categories except territorial aggression (n=4) and self-mutilation (n=1). It should be noted that some of these dogs were on a variety of other medications in addition to fluoxetine. Dogs prescribed trazodone in addition to fluoxetine had the highest percentage of positive responses (67%). Dogs prescribed clonidine in addition to fluoxetine showed 63% positive responses. Dogs treated with fluoxetine alone had 59% positive responses. These results suggest that combination therapy may be beneficial when treating behavior issues in dogs. The authors conclude “Fluoxetine appeared to be effective for all diagnoses and had few adverse effects.” The authors report no conflict of interest. With retrospective studies one has to question the accuracy of the medical records, as well as the recall of the humans, however, this study suggests fluoxetine alone, and in combination with trazodone or clonidine, may be useful for a wide variety of canine behavioral issues.


The effect of fluoxetine on compulsive disorders was evaluated by Irimajiri et al. in a randomized controlled multi-site clinical trial [6]. Participants responded to flyers and advertisements. The diagnosis of compulsive disorder was confirmed by three veterinary behaviorists who individually evaluated each dog. Six categories of compulsive behavior were identified: (1) Bull Terriers with spinning; (2) German Shepherd Dogs with tail-chasing; (3) Doberman Pinschers with flank-sucking; (4) Miniature Schnauzers with hind-end checking; (5) dogs of any breed with another oral compulsive disorder (e.g. acral lick dermatitis, licking of any body part, objects, and air); and (6) dogs of any breed with another locomotory compulsive disorder (e.g. chasing shadows or lights, circling, spinning, pacing, tail chasing, biting at the air, and fixation). To obtain baseline data, owners completed a questionnaire daily for 14 days. On day 14 owners rated behavior as absent, mild, moderate, severe, or very severe. Dogs were randomly assigned to receive fluoxetine or placebo control tablets which were provided by Elanco. The study lasted 42 days and owners completed a daily diary including number, frequency, and duration of compulsive behaviors. Response to fluoxetine and placebo was assessed at 14, 28, and 42 days and compared to each individual dog’s baseline. At 42 days there was a significant decrease in the severity of the compulsive behaviors in 70% of dogs treated with fluoxetine (19/27 ) compared to 21% for controls (6/28 ). However, mean number and duration of obsessive behaviors did not differ between groups. The authors conclude that fluoxetine appears to be safe, well tolerated, and may improve the severity of compulsive behaviors. It is interesting that owners reported improvement, yet the daily diaries showed no difference from baseline. Although failure or errors in recording could be explain the difference, it also highlights the point that from the owners’ perspective 70% of the dogs improved. It should be noted that four of six authors were affiliated with the company that manufactures the medication.


Odore et al. evaluated the effect of fluoxetine on owner-directed aggression in eight dogs at the Veterinary Teaching Hospital of the University of Turin [7]. Aggression was present for at least 2 months but not more than 4 months. Diagnosis was confirmed by a behavioral expert. Frequency, intensity, and distance of aggressive behavior were assessed using a 5-point Likert scale (0 = none and 4 = very high). Dogs with a sum score of > 8 were considered for treatment. During treatment a score ≤3 was considered to be positive response to fluoxetine. Cognitive behavioral therapy included ethological management of owner-dog relationship; management of the dogs’ self-control and frustration; attention concentration; and problem-solving activity. Dogs were re-evaluated at 1, 2, 4, and 6 months after fluoxetine treatment. Re-evaluation included physical and behavioral examination and sum scores were updated. Plasma fluoxetine and the active metabolite norfluoxetine and serum serotonin levels were also measured. There was a significant decrease in owner reported aggression scores at all time points (Figure 2). Dogs were considered to have “fully responded” (score ≤ 3) after 2 months of fluoxetine treatment. They remained “fully respondent” throughout the 6 weeks of the study. The authors conclude “our data suggest that fluoxetine at the dose of 1.5 mg/kg/day associated with behavior treatment is effective and safe for long-term control of canine aggression directed toward owners.” They also acknowledge the study limitations including the small ample size. The study suggests that  fluoxetine is safe and efficacious for 6 months. The authors also identified positive behavioral results for dogs with owner directed aggression treated with fluoxetine.



Figure 2 -Norwegian Institute for Nature Research


Bleuer-Elsner et al. examined the effect of fluoxetine on hypersensitivity-hyperactivity syndrome in dogs [8]. Dogs suffering from this disorder show the following signs: “hypermotricity, lack of satiety, and shorter sleep duration with normal cycles.” To identify dogs, 23 of 152 veterinary behaviorists reviewed their records and completed a questionnaire for each individual dog with this diagnosis. The veterinarian assessed the severity of each sign prior to fluoxetine use, and follow-up 4 to 8 weeks after treatment began (Likert scale: 0 = no improvement and 4 = resolution of clinical signs). They also evaluated adverse effects. Two of 89 dogs showed complete resolution; 1 dog showed no improvement; and 86 dogs showed various levels of improvement. The participating veterinarians indicated that 4 dogs worsened: 1 dog developed additional dominance aggression; 1 dog had increased barking; and 2 dogs developed fear. The purpose of detailed methods in research is so other scientists can reproduce the study. While this study confirms positive behavioral effects with fluoxetine and previously described side effects, the methods are confusing, lack clarity, and provide insufficient detail. This may be related to errors in translation, editorial oversight, or as the authors note “The retrospective protocol used for this study shows weaknesses in the data collection.”

In a very early study (1996), Dodman et al. evaluated the effect of fluoxetine on owner directed dominance aggression in 9 dogs [9]. Dominance aggression was confirmed by a detailed behavioral interview and assessed using the canine overt aggression chart. Although the study is described as double-blind cross over, and owners were “blinded” to the medication (fluoxetine or placebo in gel caps), all dogs received placebo for week 1 and fluoxetine for weeks 2 through 5. Owners recorded aggressive behavioral responses (growling, lip curling, snapping, biting) daily. Fluoxetine significantly decreased dominance aggression after 3 weeks of treatment. Eight of 9 owners continued fluoxetine after the study concluded; two of 8 noted an increase in aggression after decreasing or discontinuing the drug, and reinstituted the previous dose. Although the number of dogs is small, this study strongly suggests that fluoxetine decreases owner directed aggression, and that the use of fluoxetine in dogs with aggressive behavior may provide benefits that outweigh the risks.


In a study from Mills’ group, Karagiannis et al. investigated “correlates of improvement when using fluoxetine in conjunction with a behaviour modification plan” in dogs with separation anxiety (SA) [10]. The goal was to determine if dogs’ behavioural improvement during treatment included changes in their underlying affective state (optimistic vs. pessimistic outlook). Basically, the dogs’ mood or affective state biases their decision-making about ambiguous things – specifically, dogs in negative states make more negative (‘pessimistic’) judgements about ambiguous things than dogs in positive states [11]. The question they asked was “does fluoxetine and/or behavior modification result in measurable changes in the cognitive bias test to suggest a change in affective state?” Seven dogs with SA on fluoxetine and 8 control dogs (no evidence of SA or other anxiety, not on fluoxetine) were studied. Owners of dogs with SA completed 3 behavioral questionnaires at baseline and every 2 weeks for 8 weeks. After 2-weeks of cognitive bias training, the test was run at 2 and 6 weeks. Cognitive bias training involves two bowls – one with food and one without. For each individual dog the bowl with food (R+) and the bowl without food (R-) were always placed in the same location (left or right). Training was considered complete when the dogs could reliably discriminate the bowl with food (R+) vs the empty bowl (R-) and when the average absolute speed (m/s) towards the R- bowl was less than the average speed to the R+ bowl. For testing 3 “ambiguous bowls” (without food; NR+, MID, NR-) were individually placed and tested (Figure 3). The adjusted speed was calculated (with 0 = speed toward food (R+) and 100 = speed toward empty bowl (R-). When compared to the control group, dogs in the fluoxetine group had a significantly slower baseline score to the NR- bowl. At 6 weeks there was no difference between groups in response to the NR- bowl. This suggests that before treatment dogs with SA may have a more “pessimistic” outlook than dogs without SA. They used a variety of statistics and comparisons to account for the small sample size. As with previous studies, clinical signs of SA improved with fluoxetine. In dogs on fluoxetine, the median scores at 6 weeks toward towards both the NR+ and MID positions were significantly faster than at baseline. There was no significant difference between groups in response to NR-. The authors conclude that dogs with SA on fluoxetine not only showed signs of clinical improvement, but also increased running speed to the MID and NR+ bowls, “which is consistent with the “optimistic” affective bias predicted for an antidepressant.” The authors acknowledge that fluoxetine or behavior modification or both could be responsible for the “improved affect.” The small sample size and the complicated “statistical plan” detract from the study. Despite these limitations it appears that fluoxetine use in dogs with SA may improve affective state and show increased optimism and decreased pessimism.


Figure 3 -Norwegian Institute for Nature Research



Review of the current literature on fluoxetine for behavioral disorders in dogs suggests that fluoxetine is generally safe and efficacious. While any dog can have an adverse effect to any medication, the peer reviewed literature does not support that fluoxetine increases aggression or seizure activity. Fluoxetine may be useful in canine cases of separation anxiety, compulsive disorders, aggressive behavior, other behavioral disorders, and may improve affect. In addition, fluoxetine has been tested and approved by the FDA for use in dogs. Thus, fluoxetine is often the first-choice long game behavior medication for dogs with a variety of behavioral issues.


Thanks to Rene Smith CBDC, Kerrie Hoar M.S., CDBC, CPDT-KA, and Michael Shikashio CDBC for careful review of this article. This article covers the peer reviewed case controlled and retrospective studies as of 1/15/2024. I alone am responsible for any errors.



[1] https://www.reconcile.com/wp-content/uploads/2020/06/prescribing-information.pdf

[2] Simpson BS, et al. Effects of reconcile (fluoxetine) chewable tablets plus behavior management for canine separation anxiety. Vet Ther 8:18-21, 2007.

[3] Landsberg G, et al. Effectiveness of fluoxetine chewable tablets in the treatment of canine separation anxiety. J Vet Behav 3:12-19, 2008

[4] Code of Federal Regulations Title 21  §201.105https://www.ecfr.gov/current/title-21/chapter-I/subchapter-C/part-201/subpart-D/section-201.105

[5] Chutter M, et al. Efficacy of fluoxetine for canine behavioral disorders. J Vet Behav 33: 54-58, 2019.

[6] Irimajiri M et al. Randomized, controlled clinical trial of the efficacy of fluoxetine for treatment of compulsive disorders in dogs. JAVMA 235:705–709, 2009.

[7] Odore R et al. Behavioral therapy and fluoxetine treatment in aggressive dogs: A case study. Animals 10: 832, 2020. doi:10.3390/ani10050832

[8] Bleuer-Elsner S et al. Effect of fluoxetine at a dosage of 2-4 mg/kg daily in dogs exhibiting hypersensitivity-hyperactivity syndrome, a retrospective study. J Vet Behav 44:25–31, 2021.

[9] Dodman N et al. Use of Fluoxetine to treat dominance aggression in dogs. JAVMA 209: 1585-1587, 1996.

[10] Karagiannis C et al. Dogs with separation-related problems show a “less pessimistic” cognitive bias during treatment with fluoxetine (Reconcile™) and a behaviour modification plan. BMC Vet Res 11: 80, 2015.

[11] Mendl M et al. Dogs showing separation-related behaviour exhibit a ‘pessimistic’ cognitive bias. Curr Biol 20:839-840, 2010 https://doi.org/10.1016/j.cub.2010.08.030


General references

Christensen E. https://behaviorvets.mylearnworlds.com/course/the-long-game-medications-fluoxetine-sertraline-clomi,ramine-and-paroxetine

Crowell-Davis SL, Murray TF, de Souza Dantas LM. Veterinary Psychopharmacology. Wiley & Sons, Inc. 2019