Two Drugs Tie in First Trial for Compulsive Sexual Behavior Disorder

Eighty volunteers in Stockholm who struggle to control compulsive sexual urges took part in the first trial to test two medications against each other. One group received fluoxetine, an antidepressant; the other received naltrexone, a drug normally prescribed for addiction. After eight weeks, both groups reported a similar drop in symptoms. The findings, published in the journal BMJ Mental Health, suggest the two medications work about equally well for compulsive sexual behavior disorder, though they take different routes to get there.

Compulsive sexual behavior disorder is a recognized diagnosis under the World Health Organization's international disease classification. It involves a persistent pattern of failing to control intense sexual impulses, leading to distress and harm in everyday life. Researchers estimate it affects two to three percent of the population. Talk therapy is the usual first treatment, and medication is sometimes added when therapy alone is not enough. Until now, the evidence for drugs was thin: only a pair of small randomized trials had tested medication for the condition, and none had compared the two most commonly prescribed options directly.

How the trial was run

The trial ran at ANOVA, a sexual health clinic at Karolinska University Hospital in Stockholm. It enrolled adults diagnosed with the disorder under international criteria, most of whom had lived with symptoms for roughly fifteen years. Participants were randomly assigned to either fluoxetine or naltrexone for the length of the treatment period. The fluoxetine group started at twenty milligrams a day, with the dose doubled for some participants; the naltrexone group took fifty milligrams a day. When the treatment period ended, everyone stopped the medication for six weeks so the researchers could track what happened without the drugs. The work was supported by the Swedish Research Council, regional grants, and private foundations, the authors reported.

What the results showed

By the end of the eight weeks, symptom scores on a standard clinical scale had fallen by roughly half in both groups, with no meaningful difference between them, according to the researchers. But the path to that point looked different. The naltrexone group started improving within the first couple of weeks, while improvements in the fluoxetine group took several weeks longer to appear. Once the drugs were stopped, symptoms crept back in both groups, and roughly four in ten volunteers in each group wanted to keep taking their medication.

Guidelines already suggest trying an SSRI or naltrexone when therapy alone is not enough, but doctors have been prescribing largely on educated guesswork, since the two options had never been tested side by side. That gap is what the Stockholm team set out to close. The researchers also noted that compulsive sexual behavior disorder appears in the World Health Organization's disease classification but not in the American Psychiatric Association's diagnostic manual, a split that has complicated research funding and public awareness of the condition.

The side effects told a different story. People taking fluoxetine commonly reported sexual side effects: about seven in ten experienced lower libido, nearly half reported delayed ejaculation, and about a quarter felt fatigued. Two volunteers in the fluoxetine group stopped treatment because of adverse events, one with hives and one with elevated liver enzymes. Nausea was more common with naltrexone, though most side effects in both groups were mild to moderate. Reduced sexual desire showed up in both groups.

"Knowledge about pharmacological treatment for compulsive sexual behavior disorder is still limited," said Josephine Savard, a consultant psychiatrist at ANOVA and a researcher at Karolinska Institutet and Umea University, in the study report. "The findings suggest that treatment choices may need to be based on the patient's individual circumstances, preferences and which adverse effects are considered acceptable." The authors cautioned that the trial was open-label, meaning everyone knew which drug they were taking, and that almost all participants were men, so the results cannot be assumed to apply to women. They say longer studies are needed to find out who benefits most from each medication. MedicalXpress and the journal BMJ Mental Health reported the findings in October 2026.

Sources: BMJ Mental Health abstract and MedicalXpress