Global ADHD Network — One of the most highly reviewed ADHD CPD training providers in ADHD diagnosis and neurodiversity training. Global ADHD Network — One of the most highly reviewed ADHD CPD training providers in ADHD diagnosis and neurodiversity training. Global ADHD Network — One of the most highly reviewed ADHD CPD training providers in ADHD diagnosis and neurodiversity training.
View Courses →
Global ADHD Network
Join Today →
0
Global ADHD Network
✕
1,500+ Reviews
Rated 4.8 on Trustpilot by verified clinicians →
November 18, 2026

Beyond First-Line: Atomoxetine, Guanfacine and Treatment Failure in Adult ADHD

A UK prescriber's guide to second-line adult ADHD treatment: stimulant failure, NG87 order, atomoxetine, off-label guanfacine and documentation.
Header image

A 34-year-old accountant has had six weeks of lisdexamfetamine and six of methylphenidate, both at what you judged adequate doses. One left her unable to sleep and the other made no difference she could name, and she now asks what comes next.

The next step depends on how precisely you can describe what went wrong. Non-response, intolerance and unsuitability lead to different decisions.

Inadequate response is not the same as intolerance

NICE NG87 frames non-response in adults as a 6-week trial at an adequate dose without enough benefit “in terms of reduced ADHD symptoms and associated impairment”. Intolerance differs: the drug may work, but the person cannot take it. Unsuitability is a third: a contraindication, or a misuse or diversion risk that rules out some preparations.

Before you record a stimulant as failed, rule out the common false failures.

  • Dose. Was the dose actually optimised against symptoms and adverse effects? Our adult titration protocol covers this.
  • Preparation and timing. NG87 reminds prescribers that effect size, duration of effect and adverse effects vary from person to person. A drug that wears off at 3pm has not necessarily failed.
  • Adherence. NG87 notes that ADHD symptoms themselves can make it hard to order, collect and take medication.
  • Manageable adverse effects. For weight loss, NG87 lists options such as taking medication with or after food. Changing medication is one option among several. See our guide to ADHD medication side effects.
  • The diagnosis. If behaviour worsens on treatment, NG87 advises adjusting medication and reviewing the diagnosis.

NICE NG87 sequencing for adults

NG87 sets out the order in which medicines are offered to adults. It recommends lisdexamfetamine or methylphenidate first line, and switching to the other after a 6-week trial at an adequate dose without enough benefit. It suggests considering dexamfetamine (off-label) where symptoms respond to lisdexamfetamine but the person cannot tolerate its longer effect profile.

Atomoxetine comes next. NG87 recommends offering it to adults who cannot tolerate lisdexamfetamine or methylphenidate, or whose symptoms have not responded to separate 6-week trials of both, “having considered alternative preparations and adequate doses”. Guanfacine is not in the adult sequence: NG87 says not to offer it to adults without advice from a tertiary ADHD service. It was last updated in September 2019, so check your local formulary.

Atomoxetine: licence, onset, monitoring and key warnings

The atomoxetine SmPC licenses it for ADHD in children aged 6 and over, adolescents and adults, with treatment initiated by a specialist. In adults, childhood symptoms should be confirmed, and the SmPC says atomoxetine should not be started when verification of childhood symptoms is uncertain. Our article on retrospective recall of childhood symptoms covers how that evidence is gathered.

Atomoxetine is not a psychostimulant, and in an adult abuse-potential study it showed no pattern suggesting stimulant or euphoriant properties. That matters when diversion risk is the reason stimulants are unsuitable.

Onset is slower, and patients need to hear that. The adult efficacy trials in the SmPC ran for ten to sixteen weeks. A pooled analysis by Sobanski and colleagues of 13 Eli Lilly-sponsored studies found that many adults responded relatively quickly, while others responded gradually without a clear plateau at 24 weeks. The analysis was industry-sponsored and limited to completers, but a two-week review is too early to call failure.

Monitoring. The SmPC asks for heart rate and blood pressure before treatment, after each dose adjustment and then at least every 6 months. Most patients have a modest rise, but combined trial data show more pronounced changes in 6 to 10% of adults. Baseline cardiac assessment is covered in our guide to cardiovascular checks before ADHD medication. NG87 also asks prescribers to monitor adults for sexual dysfunction as a potential adverse effect of atomoxetine.

Key warnings from the SmPC:

  • Contraindications include use with or within 2 weeks of an MAOI, narrow-angle glaucoma, severe cardiovascular or cerebrovascular disorders, and phaeochromocytoma.
  • Suicide-related behaviour has been reported. Adult double-blind trials showed no difference from placebo, but monitoring is still required.
  • Liver injury is very rare. Stop atomoxetine in patients with jaundice or laboratory evidence of liver injury, and do not restart it.
  • CYP2D6. With inhibitors such as fluoxetine or paroxetine, exposure may increase 6- to 8-fold. Slower titration and a lower final dose may be needed.
  • Serotonin syndrome has been reported with serotonergic drugs, including SSRIs, SNRIs and tricyclics.

Guanfacine in adults: an off-label decision

The Intuniv SmPC licenses guanfacine for ADHD in children and adolescents aged 6 to 17 for whom stimulants are not suitable, not tolerated or ineffective. It is not silent on adults. It states that safety and efficacy in adults with ADHD have not been established and that guanfacine “should not be used in this group”. Elsewhere it notes that guanfacine has not been studied in adults with ADHD.

Adult use is therefore off-label, and NG87 places it behind tertiary advice. If a tertiary service advises it, the SmPC warnings are the starting point:

  • Hypotension, bradycardia and syncope. NG87 advises reducing the dose or switching if a person has sustained orthostatic hypotension or fainting episodes.
  • Sedation and somnolence, mainly early in treatment and typically lasting 2 to 3 weeks. Patients should not drink alcohol.
  • Stopping. Taper by no more than 1 mg every 3 to 7 days. Blood pressure and pulse can rise after discontinuation, and hypertensive encephalopathy has been very rarely reported after abrupt withdrawal.
  • Interactions. A 50% dose reduction is recommended with moderate or strong CYP3A4/5 inhibitors. Use with QT-prolonging drugs is generally not recommended.
A failed trial is a clinical finding. Record the drug, the dose, the duration and exactly why it failed.

Switching and combination decisions

When the problem is non-response to one stimulant, NG87 points to the other stimulant first, not a non-stimulant. When the problem is intolerance, ask whether it is the molecule or the profile. NG87’s dexamfetamine option is for people who respond to lisdexamfetamine but cannot tolerate its longer effect.

When you move to atomoxetine, plan for the slower onset. Neither the atomoxetine nor the Intuniv SmPC gives a cross-titration schedule from a stimulant, so any overlap or gap is a clinical decision. Agree it with the patient and write it down.

Combination treatment needs more caution. NG87’s adult recommendations describe single agents in sequence, and it advises against offering medication outside them without tertiary advice. The Intuniv add-on study in stimulant partial responders enrolled 6 to 17 year olds, not adults. The atomoxetine SmPC advises caution with drugs that may raise blood pressure.

When to seek specialist review

  • Treatment-resistant symptoms. NG87 recommends a second opinion or tertiary referral if symptoms are unresponsive to one or more stimulants and one non-stimulant.
  • Guanfacine, or anything outside the NG87 adult sequence. Tertiary advice first.
  • Cardiovascular findings. Sustained resting tachycardia over 120 beats per minute, arrhythmia, or a clinically significant rise in blood pressure on 2 occasions: NG87 advises reducing the dose and referring to an adult physician. See our annual review and physical health monitoring guide.
  • Acute psychotic or manic episode. NG87 advises stopping ADHD medication. Our guide to the ADHD and bipolar differential covers the diagnostic question.
  • New or worsening seizures. Review and stop any medication that might be contributing.

NG87 says all ADHD medication should be initiated by a professional with training and expertise in ADHD, with shared care only after titration and dose stabilisation. See our guide to shared care agreements for ADHD medication.

Documenting off-label decisions

The MHRA’s advice on off-label prescribing says the prescriber should be satisfied that off-label use would better serve the patient than a licensed alternative, and that there is enough evidence or experience to show safety and efficacy. The prescriber owns monitoring and follow-up, and should record the reasons where common practice is not followed. The MHRA confirms that nurse and pharmacist independent prescribers can prescribe a licensed medicine off-label, subject to their competence, professional codes and employer policies (see our overview of prescribing training for nurses and pharmacists). Doctors should also follow the GMC’s prescribing guidance.

Guanfacine is not the only case. NG87 notes that atomoxetine is off-label for adults with no ADHD symptoms in childhood, and the Elvanse Adult SmPC likewise requires pre-existing childhood symptoms. A good record covers:

  • each previous trial, with drug, dose, duration and reason for stopping
  • why licensed options are unsuitable for this patient
  • who gave tertiary advice, and when
  • what the patient was told, including serious or common adverse reactions
  • the monitoring plan, who prescribes, and the review date

Frequently asked questions

Can guanfacine be prescribed for adults with ADHD in the UK?

Only off-label. Intuniv is licensed for ages 6 to 17, and its SmPC says it should not be used in adults. NG87 advises against offering it to adults without advice from a tertiary ADHD service.

How long should an adult atomoxetine trial last?

The SmPC sets no fixed figure. Its adult efficacy trials ran for ten to sixteen weeks, and some adults respond gradually over 24 weeks. Agree a review point before starting.

When has a stimulant trial actually failed?

NG87 describes a 6-week trial at an adequate dose without enough benefit. Check dose, preparation, timing and adherence before recording it as failed.

Conclusion

Second-line prescribing starts with an accurate account of why first-line treatment failed. Atomoxetine is the licensed non-stimulant in the NG87 adult sequence. Guanfacine in adults is off-label and needs tertiary advice, and every off-label decision needs a record that shows your reasoning.

Global ADHD Network’s ADHD prescribing and management course covers stimulant and non-stimulant pharmacology, including atomoxetine and guanfacine, alongside monitoring and documentation with reference to NG87.

‍

Trusted by 100's of ADHD clinicians