Addiction Isn’t Defined by Withdrawal Symptoms

It’s not only people addicted to alcohol or drugs who experience significant withdrawal symptoms if they stop using. Antidepressant (AD) withdrawal symptoms are also common, can be severe and prolonged, and may last months or years. For example, with antidepressant discontinuation, feelings resembling electric shocks, often called “brain zaps,” may occur in the head or body, along with nausea, vomiting, dizziness, and other symptoms. In some cases, ending antidepressant use may feel as difficult as stopping heroin. But antidepressants are not addictive.

A recent 2025 study by Mark Horowitz, M.D., Ph.D, a clinical research fellow in psychiatry with the National Health Service (NHS) in London, and his colleagues surveyed 310 patients who had been taking antidepressants. They found 79% experienced withdrawal symptoms upon ending antidepressants, with 45% reporting moderate-to-severe symptoms and about 20% experiencing symptoms lasting three months or longer. Longer duration of use was associated with greater likelihood of severe protracted symptoms and being less likely to be able to stop their ADs. Horowitz has stated his career was influenced by his own experience with severe withdrawal symptoms when trying to stop antidepressants.

People aren’t addicted to ADs; instead, their bodies become used to taking them and react if not receiving them or even taking a lower dosage. Antidepressant withdrawal is common—but distinct from addiction. The Horowitz study underscores the urgent need to distinguish between physiological dependence and addictive use.

For much of the 20th century, medical research and teaching equated physical withdrawal symptoms with addiction, and many people still believe it’s true. A drug was assumed “addictive” if patients experienced a recognizable abstinence syndrome when ending use. This framework was shaped by dramatic and sometimes dangerous withdrawal syndromes seen with ending use of alcohol, barbiturates, and opioids among the addicted. However, this interpretation implied physiological dependence was the disorder, and further, presumed drugs without vivid withdrawal symptoms were non-addictive. Contemporary