Drug therapy for anorexia nervosa

Drugs that have been tried

• SSRIs (for example fluoxetine): based on the serotonin explanation and because anxiety and depression often occur with anorexia.
• Antipsychotics (for example olanzapine): may reduce anxious, obsessional thinking about food and help weight gain.

NICE (summarised by ELFT, 2024): medication is not recommended as the sole treatment for eating disorders but can be used as an adjunct. Clinicians must consider that malnutrition can reduce drug effectiveness or increase side effects, and offer ECG monitoring for drugs that could affect the heart (GPnotebook).

Evidence

• Walsh et al. (2006): an RCT of 93 weight-restored patients given fluoxetine (49) or placebo (44) for up to a year. Similar proportions stayed well and in the study: 26.5% on fluoxetine and 31.5% on placebo; there was no significant difference in time to relapse. The authors concluded antidepressants are unlikely to offer substantial benefit for most patients.
• Attia et al. (2019): 152 adult outpatients randomly given olanzapine or placebo for 16 weeks at five North American sites. BMI rose by 0.259 kg/m² per month on olanzapine versus 0.095 on placebo (significant), but obsessional thinking did not differ: a 'modest' effect on weight only.

Evaluation

• Weak evidence: SSRIs did not prevent relapse; olanzapine gives modest weight gain without psychological benefit.
• Side effects and physical risk are greater in malnourished patients (heart rhythm, electrolytes).
• Easy to give, and may help co-occurring depression or anxiety.
• Symptoms not causes: drugs do not change beliefs about weight and shape, so psychological therapy remains the main treatment.
• Ethics: people with anorexia may resist weight-gain treatment, raising questions of consent and patient choice.

Worked example

Attia et al. (2019) found BMI increased by 0.259 kg/m² per month on olanzapine and 0.095 on placebo. Over the 16-week (about 4-month) trial, what was the difference in expected BMI gain?

1. Difference per month = 0.259 - 0.095 = 0.164 kg/m².
2. Over 4 months: 0.164 x 4 = 0.656 kg/m².
3. This is a small, though statistically significant, extra gain, which is why the authors called the effect modest.

Answer: About 0.66 kg/m² extra over 4 months: a modest effect.

Check yourself

1. In Walsh et al. (2006), 26.5% on fluoxetine and 31.5% on placebo maintained their weight and stayed in the study. What is the difference in percentage points? Give your answer to 1 decimal place.
   Show answer

   5.0 percentage points. 31.5 - 26.5 = 5.0; placebo was slightly higher.

2. Walsh et al. had 93 participants, 49 on fluoxetine. How many were on placebo?
   Show answer

   44 participants. 93 - 49 = 44.

3. Using Attia et al.'s figures (0.259 vs 0.095 kg/m² per month), how many times faster was BMI gain on olanzapine? Give your answer to 1 decimal place.
   Show answer

   2.7. 0.259 / 0.095 = 2.7.

4. What does NICE guidance say about medication for eating disorders?
   • It should not be the sole treatment, but can be used alongside psychological treatment
   • It is the first and only treatment
   • It is banned
   • It should be given to everyone at diagnosis
   Show answer

   It should not be the sole treatment, but can be used alongside psychological treatment. Adjunct, not sole treatment.

5. Why must drug therapy be used with extra care in people with anorexia?
   • Drugs work faster when malnourished
   • People with anorexia cannot swallow tablets
   • Malnutrition can reduce effectiveness and increase side effects, including heart risks
   • There are no drugs that affect serotonin
   Show answer

   Malnutrition can reduce effectiveness and increase side effects, including heart risks. Physical state changes how drugs act.

6. Explain why drug therapy alone is unlikely to be enough to treat anorexia nervosa.
   Show answer

   Drugs do not change the beliefs about weight and shape that maintain anorexia, and trials such as Walsh et al. found fluoxetine did not prevent relapse, so psychological therapy is needed.. Drugs do not tackle beliefs; trial evidence shows little benefit.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification (issue January 2026)
• Pearson IAL Psychology component guide: WPS04 (Issue 2, November 2024)
• East London NHS Foundation Trust research summary of the NICE eating disorders guideline (Vasey et al., 2024)
• GPnotebook: medication in the treatment of anorexia nervosa
• Walsh et al. (2006), Fluoxetine after weight restoration in anorexia nervosa: a randomized controlled trial, JAMA 295
• Attia et al. (2019), Olanzapine versus placebo in adult outpatients with anorexia nervosa, American Journal of Psychiatry

Last updated 8 November 2026. Spotted a mistake? Tell us.