The SSRI-Violence Link: Myth or Menace?

Selective serotonin reuptake inhibitors (SSRIs) have been controversial nearly since the introduction of fluoxetine (Prozac) to the US market in 1988. They're well known to cause side effects such as nausea, insomnia, and sexual dysfunction, and the evidence is mixed for their benefit in people with mild to moderate depression.

Recent findings reported by Medscape Medical News suggest another dent in the SSRI reputation, drawing an association between the drug class and an increased risk for violent crime. Some past research has linked SSRIs with violence and aggressive behavior as well. On the basis of the collective literature, however, we don't believe that there is sufficient evidence to support the link. Here we present how we came to our conclusion.

Serotonin and Aggression: Conflicting Findings

Unfortunately, the literature on the adverse neurobehavioral side effects of SSRIs, as well as serotonin-norepinephrine reuptake inhibitors (SNRIs), is complicated by vague and overlapping terminology, specifically the terms activation, irritability, anger attacks, and aggression. Generally, activation refers to increased psychomotor activity or hyperarousal. Irritability entails having a low threshold for experiencing frustration or anger. Anger attacks denote bouts of anger that start suddenly, are disproportionate to the situation, are not part of the patient's usual behavior, and are associated with autonomic activation. Finally, aggression may be defined as forceful physical, verbal, or symbolic action which may be appropriate and self-protective or inappropriate.

Serotonin (5-HT) appears to mediate the inhibition of such behaviors and may regulate emotional expression and social functioning. In a comprehensive review, Walsh and Dinan noted: "Since the introduction of the SSRIs to the market in the late 1980s, studies on both non-human primates and on smaller animals have bolstered the theory of serotonergic dysfunction in aggression." Several lines of preclinical evidence suggest that the 5-HT1A and 5-HT1B receptors are involved in aggressive behaviors, and that agonists at these receptors may reduce aggression. A recent study found that treatment with the SSRI citalopram mitigated "unmanageable violence" in an adult male chimp.

Neurotransmission of 5-HT appears to play a major role in modulating impulse control and aggressive behavior in humans as well. Indeed, recent evidence suggests that enhancing serotonergic neurotransmission reduces aggressive behavior in humans, perhaps by altering brain regions and circuits that mediate aggression. Some lines of clinical evidence, however, seem to implicate serotonergic agents in increasing or precipitating aggression or violence.

Using the US Food and Drug Administration Adverse Event Reporting System (FAERS) data, Moore and colleagues found that antidepressants with serotonergic effects were "strongly and consistently" implicated in reports of acts of violence toward others. Moreover, specific concerns regarding younger patients began to arise as early as the 1990s. Constantino and colleagues carried out a prospective study of aggressive behavior in 19 psychiatrically hospitalized adolescents who received open clinical trials of the SSRIs fluoxetine, paroxetine, or sertraline. Verbal aggression, physical aggression toward objects, and physical aggression toward oneself occurred significantly more frequently while on SSRIs than off. Of note, no increase was observed in physical aggression toward others.

A number of studies show an association between use of SSRIs and various forms of noncriminal aggression or violence. We hasten to add that 'association' does not demonstrate causation.

We should point out that in our experience, the FAERS remains poorly understood. First, the mere reporting of an (unverified) "event" to the FDA does not constitute a scientific finding. The FDA itself explicitly states that reports do not always contain enough detail to properly evaluate an event, which may be due to underlying disease, other substances, etc. Indeed, the FAERS website states, "Importantly, the FAERS data by themselves are not an indicator of the safety profile of the drug." When an association becomes clear via sound research or overwhelming anecdotal reports, the FDA can and will remove a product from the market. To date, no antidepressant has been removed.

We also wish to emphasize that so-called "SSRIs" are not a homogenous group of agents, nor are they all exclusively "selective" serotonin reuptake inhibitors. For example, animal studies show that sertraline has modest dopaminergic effects, whereas paroxetine has modest noradrenergic effects and clinically significant anticholinergic side effects. These pharmacodynamic issues are relevant to any claim that lumps all "SSRIs" together and attributes adverse (or favorable) behavioral effects to the entire class.

SSRIs, Aggression, and Violence

A number of studies show an association between use of SSRIs and various forms of noncriminal aggression or violence. We hasten to add that "association" does not necessarily imply causation. Nevertheless, these reports have raised concerns regarding the safety of SSRIs, particularly in younger populations.

Many variables may explain these widely varying conclusions, including...differences in underlying psychiatric diagnosis and psychopathology among subjects.

Sharma and colleagues published a large meta-analysis reporting a doubling in both suicidality and aggression in children and adolescents taking duloxetine, fluoxetine, paroxetine, sertraline, or venlafaxine. In contrast, an extensive review by Walsh and Dinan of all published papers linking serotonin, SSRIs, and aggression found "no convincing evidence to link the use of fluoxetine or other SSRIs with violent or suicidal behavior...On the contrary, there is a considerable body of evidence from the last decade to suggest that fluoxetine may be associated with improvement in anger and aggression both towards oneself and others."

Similarly, a randomized, placebo-controlled study using a validated irritability rating scale found that adult patients with major depressive disorder who were randomly assigned to 8 weeks of treatment with sertraline showed a greater reduction in irritability than did patients randomly assigned to placebo. Bouvy and Liem reported a significant decrease in lethal violence over 15 years of increased SSRI exposure in the community; while Finkelhor and Johnson looked at psychiatric medication prescription rates and crime trends over time, finding that the latter declined in the United States while 9% of youth and 20% of adults were taking medications.

Claims that psychiatric medications are causally connected to more serious or criminal violence have steadily increased with the rise of social media. Even though "involuntary intoxication" is sometimes raised in these cases as a defense in crimes in which the defendant was taking a psychiatric medication, there is no convincing evidence that these medications play a role in causing criminal or violent behavior. Nevertheless, such narratives are promoted via the internet and social media as a result of confirmation bias, lack of prevalence awareness, and misunderstanding of how FAERS works.

Why Such Disparate Conclusions?

Many variables may explain these widely varying conclusions, including differing study methods; the particular drugs and drug dosages studied; use of concomitant medications; and perhaps most important, differences in underlying psychiatric diagnosis and psychopathology among participants. Very few studies of SSRIs and aggression control for the latter variable. For example, aggression is a frequent manifestation of antisocial personality disorder, which is also characterized by high levels of impulsivity, psychopathic traits, and a high prevalence of comorbid substance use disorders. Additionally, evidence suggests that antidepressant treatment may be poorly tolerated in youth with a familial risk of developing mania, many of whom may develop antidepressant-related irritability, aggression, impulsivity, or hyperactivity.

There is also the issue of "discontinuation" or "withdrawal" syndromes related to SSRIs and SNRIs, which can cause irritability, impulsiveness, aggression, and anger. This constellation of signs and symptoms represents an indirect effect of treatment, usually in situations where insufficient time has been allowed for tapering and discontinuation of the SSRI (usually paroxetine) or SNRI (eg, venlafaxine). A putative association between the timing, onset, and course of SSRI withdrawal and a violent act should be a relatively straightforward determination.

In weighing the existing evidence, we do not feel that a direct connection between antidepressants and violence has been reliably established. Beyond just the findings we've presented, consider the fact that about 13% of the total US population has been exposed to antidepressants. Rates were even higher in women (16.5%) and those over age 60 (19.1%). Given that antidepressant prescribing in the United States has increased substantially in recent years, we would expect to see significantly increasing rates of violent crime as a result, yet violence remains a low base rate event, even demonstrating significant declines over the decades. Most important is that there is no compelling evidence that violence has increased as a result of antidepressant use.

Even the authors of the recent study linking SSRIs with violent crime concede, "[W]e do not know how far the association between SSRI medication and violent crime reflects causation." Furthermore, in speaking with Medscape Medical News, the lead author cautioned, "Our findings should not be used as grounds for individuals to go off their [SSRI] medication or for clinicians to withhold medication from those who might benefit from it."

Accordingly, we find no basis for withholding or discontinuing SSRI treatment in patients who are clearly benefitting from it, owing to concerns about other-directed violence.

That said, we urge caution in prescribing SSRIs for young males with a known history of violent offenses, as well as for patients with a personal or family history of bipolar disorder. Finally, for patients with mild to moderate, nonmelancholic episodes of major depressive disorder, cognitive-behavioral or interpersonal psychotherapy ought to be considered as a first-line treatment.

Ronald W. Pies, MD, is professor emeritus of psychiatry and a lecturer on bioethics and humanities at SUNY Upstate Medical University, and clinical professor of psychiatry at Tufts University School of Medicine. His chief interests include the philosophy of psychiatry, psychiatric ethics, and mood disorders. He is the author of Handbook of Essential Psychopharmacology (2nd ed).

James L. Knoll IV, MD, is professor of psychiatry and director of forensic psychiatry at SUNY Upstate Medical University, and clinical director of the Central New York Psychiatric Center in Marcy, New York. He is teaching faculty for the American Academy of Psychiatry and the Law (AAPL) Forensic Psychiatry Review Course. He has served as an expert witness for the courts in cases alleging involuntary intoxication with psychotropic medications.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.