Antidepressants for under 18s: what NICE actually says

Antidepressant prescribing has been back in the national press this week, and the figures underneath the coverage are worth reading properly. In England in 2025/26, an estimated 63,600 patients aged 17 and under received at least one antidepressant item dispensed in the community. That is a rise of 1.6 per cent on the previous year and the first increase in that age group since 2021/22 (NHS Business Services Authority, 2026). If your child has been offered medication, or if you are about to raise low mood with a GP now term has started, the useful question is not whether that national number is too high or too low. It is whether what is being proposed for your child matches what the guidance for this age group actually says, because that guidance is narrower and more specific than most of the argument around it.

What the figures do and do not cover

The NHS Business Services Authority counts a child as anyone aged 17 or under on 30 September of the relevant financial year, and counts only prescriptions dispensed in the community. Anything issued in secondary care, in prisons, or by a private prescriber sits outside the total, so 63,600 is a floor rather than a ceiling. The age band detail is where it gets more interesting. Among 10 to 14 year olds, 6,320 girls and 4,630 boys were prescribed an antidepressant. Among 15 to 19 year olds the numbers jump to 124,434 female and 51,397 male patients, with 71 per cent of that band female (NHS Business Services Authority, 2026). Read that second figure carefully, because it includes 18 and 19 year olds and is not an under 18 statistic. One more thing gets lost when these numbers are summarised: antidepressants are not prescribed only for depression. Obsessive compulsive disorder is the clearest example, and there are others. A prescribing count is not a diagnosis count.

Medication is not where the guidance starts

NICE guideline NG134, Depression in children and young people, is the reference point for this age group and was published on 25 June 2019. For mild depression it is unambiguous: antidepressant medication should not be used for the initial treatment of children and young people with mild depression. For moderate to severe depression the sequence starts with psychological therapy. Fluoxetine is offered to a young person aged 12 to 18 following multidisciplinary review, if their depression has not responded to a specific psychological therapy after four to six sessions. For a child aged 5 to 11 the same step is to be considered cautiously, because the evidence for fluoxetine in that age group is not established. NG134 also allows combined therapy, fluoxetine alongside psychological therapy, as an initial treatment for 12 to 18 year olds, noting that this was an off label use in June 2019 for young people who had not previously tried therapy on its own (National Institute for Health and Care Excellence, 2019).

The rule that matters most to parents

Recommendation 1.6.14 is the one to know. Do not offer antidepressant medication to a child or young person with moderate to severe depression except in combination with a concurrent psychological therapy. Medication and therapy are meant to run together, not as alternatives you pick between. The guideline does allow for families who decline therapy, and says medication may still be given in that situation, but it then puts the monitoring obligation squarely on the prescribing doctor, precisely because nobody will be seeing the young person weekly in a therapy room. It also sets out what monitoring should look like at the start: specific arrangements for reviewing adverse drug reactions, mental state and general progress, with weekly contact with the child and their parents or carers for the first four weeks given as the worked example (National Institute for Health and Care Excellence, 2019). If a prescription has been issued with no therapy attached and no review date, that is the gap worth raising.

Why it is specifically fluoxetine

The reason one drug keeps appearing is not habit. A network meta-analysis in The Lancet pooled 34 double blind randomised controlled trials covering 5,260 children and adolescents and 14 different antidepressants. Only fluoxetine was statistically significantly more effective than placebo, with a standardised mean difference of -0.51 and a 95 per cent credible interval running from -0.99 to -0.03. Patients given imipramine, venlafaxine and duloxetine discontinued because of adverse events more often than those given placebo. The authors rated the quality of evidence as very low across most comparisons, and their conclusion was measured: antidepressants as a class do not seem to offer a clear advantage in the acute treatment of major depressive disorder in this age group, and fluoxetine is probably the best option to consider when a pharmacological treatment is indicated (Cipriani et al., 2016). NG134 reaches the same place, describing fluoxetine as the only antidepressant for which clinical trial evidence shows the benefits outweigh the risks, and noting that as of June 2019 it was the only antidepressant with a UK marketing authorisation for this indication in 8 to 18 year olds. Sertraline and citalopram are named as second line options if fluoxetine fails or is not tolerated, both off label in under 18s, and NG134 attaches a list of conditions to that step, including advice from a senior child and adolescent psychiatrist, usually a consultant, and a signed consent form.

Who should be starting it

NG134 says an antidepressant should only be prescribed following assessment and diagnosis by a child and adolescent psychiatrist. In practice a GP often holds the repeat prescription afterwards under a shared care arrangement, which is ordinary and workable. Specialist initiation followed by GP continuation is a different thing from a first prescription written without a specialist assessment, and it is reasonable to ask which of the two is happening. The guideline is equally specific about the first weeks. A child or young person on an antidepressant should be closely monitored for the appearance of suicidal behaviour, self harm or hostility, particularly at the beginning of treatment. Unless medication needs to start immediately, symptoms that might later be mistaken for side effects should be monitored for seven days before prescribing, so that there is a baseline to compare against. Once treatment starts, any new symptoms of that kind mean urgent contact with the prescribing doctor. The starting dose of fluoxetine is 10mg daily, which can go up to 20mg after a week if clinically necessary, with lower doses considered for children of lower body weight and little evidence supporting anything above 20mg (National Institute for Health and Care Excellence, 2019).

Anxiety is a different question with a different answer

Parents often assume the depression guidance transfers across to anxiety. It does not. For social anxiety disorder, NICE clinical guideline CG159, published on 22 May 2013, is direct: do not routinely offer pharmacological interventions to treat social anxiety disorder in children and young people. There is a medication pathway in that guideline, and it is written for adults. For children and young people the recommendation is cognitive behavioural therapy adapted for their age, with treatments developed for adults considered only for young people typically aged 15 and over who have the capacity to use them (National Institute for Health and Care Excellence, 2013). Obsessive compulsive disorder is the exception that proves the rule. NICE clinical guideline CG31, published on 29 November 2005, accepts that SSRIs are effective in children and young people with OCD, and notes that fluvoxamine and sertraline are the only SSRIs licensed for that use in this age group (National Institute for Health and Care Excellence, 2005). Even there, medication follows cognitive behavioural therapy with exposure and response prevention rather than replacing it. So if an SSRI is being proposed for an anxious child, the question to ask is which condition it is being prescribed for, and what the guidance says for that condition specifically.

If your child is already taking one

Nothing here is a reason to stop a medication that is working, and stopping abruptly is its own risk. NG134 says that when a child or young person responds to fluoxetine, treatment should continue for at least six months after remission, with remission defined as no symptoms and full functioning for at least eight weeks. The same six month rule applies to sertraline and citalopram. Coming off is done gradually and with the prescriber, not unilaterally at the kitchen table. What is worth checking is the shape of the whole plan rather than the tablet on its own. Is a psychological therapy running alongside it, and if it stopped, when and why. When was the last proper review, and who did it. If the answer is that the medication has quietly become the entire treatment, the useful conversation is about adding the therapy back, not about removing the drug.

Questions worth taking to the appointment

Six questions cover most of it. What is the diagnosis, and is it being described as mild, moderate or severe, because that alone determines whether medication belongs in the plan yet. Which psychological therapy is running alongside this, and when does it start. Who is initiating the prescription, and has a child and adolescent psychiatrist assessed my child. Is this drug licensed for this age and this condition, or is it off label, and what does that mean in practice here. What exactly are we watching for in the first four weeks, and who do I contact out of hours if I see it. And how long is this expected to continue, with a review booked rather than assumed. A prescriber who is following NG134 will have ready answers to all six, and asking them changes nothing about the treatment except how well you understand it.

Medication has a real place in treating depression in young people, and for a young person with moderate to severe depression that has not responded to therapy, fluoxetine alongside continued psychological treatment is evidence based care rather than a last resort. The problems come from the edges: a prescription without a therapy, a start without an assessment, or a plan that nobody has reviewed in a year. Blip is a specialist mental health service for children and young people aged 7 to 25. If you want a clear assessment of what is going on for your child, and a treatment plan you can actually follow, get in touch and one of our clinicians will talk it through with you.