A medication conversation can quickly fill with unfamiliar names: methylphenidate, lisdexamfetamine, stimulants, non-stimulants, immediate release and modified release. You may want to understand what each option offers before discussing a prescription, or make sense of a medicine you already take.
This guide brings those questions together. It explains the main medicine groups, how preparations differ, commonly reported unwanted effects, and what large research reviews can tell us. It is written for adults in the UK; children's prescribing guidance and licensing can be different. Brand examples explain particular products and are not recommendations or a check of current pharmacy stock.
Key points
- Start with the active ingredient, then check the particular preparation and its instructions.
- Consider benefits and unwanted effects together when discussing treatment.
- Read research figures alongside the population, outcome and length of follow-up.
- Use the questions below to prepare for an individual prescribing conversation.
Which medicine options are available?
The broad classification is stimulant or non-stimulant. Within those groups, individual medicines and preparations differ. The table introduces five names you may encounter; it does not mean all five are routine or licensed adult choices.
Scroll sideways to read the full table on a small screen.
| Medicine | Classification | Preparation or adult-use distinction |
|---|---|---|
| Methylphenidate | Stimulant | Standard tablets and slow-release tablets or capsules.1 |
| Lisdexamfetamine | Amphetamine-family stimulant; a prodrug | Elvanse capsules have an adult ADHD indication.2 |
| Dexamfetamine | Amphetamine-family stimulant | Amfexa tablets do not have an adult ADHD licence.3 |
| Atomoxetine | Non-stimulant; selective noradrenaline reuptake inhibitor | The linked capsule product includes adult ADHD.4 |
| Guanfacine | Non-stimulant; alpha-2A adrenergic receptor agonist | Intuniv prolonged-release tablets are licensed for ages 6–17.5 |
The mechanism names in the non-stimulant rows describe how the medicines act. They are useful for identifying the class, but do not by themselves tell you how helpful or tolerable a medicine will be for you.
How do these fit into UK adult guidance?
NICE recommends medication when significant impairment persists after environmental changes have been tried and reviewed. Methylphenidate or lisdexamfetamine are first choices. Atomoxetine follows intolerance or inadequate response to adequate trials of both. Dexamfetamine can be considered when lisdexamfetamine helps but its longer effect is not tolerated. Guanfacine for adults is off-label and requires advice from a tertiary ADHD service.6
Does the particular product have an adult licence?
Adult licensing depends on the preparation. Concerta XL includes adult ADHD, whereas standard Medikinet tablets do not have an adult licence.78
Off-label prescribing means using a licensed medicine outside its authorised terms. It can be appropriate, but the prescriber must take responsibility for the evidence, individual circumstances, information given and follow-up. It is different from using an unlicensed product.9
Questions to ask: Is this particular preparation licensed for adult ADHD? If the proposed use is off-label, why is it appropriate for me? What alternatives were considered, and what monitoring will I need?
Immediate release, modified release and long acting
These terms help describe delivery and duration. They do not describe how many months or years someone will remain on treatment.
Immediate release (IR) is a standard-release preparation. Modified release (MR) changes how the ingredient is released over time. For methylphenidate, MR products contain different proportions of immediate- and modified-release components.10
You may also see slow release, sustained release or prolonged release. NHS information describes modified- and sustained-release methylphenidate as slow-release forms; Concerta XL is labelled a prolonged-release tablet. These labels concern the formulation, rather than simply indicating a higher dose.17
How long might a dose work?
NHS Specialist Pharmacy Service lists approximately four hours for immediate-release methylphenidate, and eight or twelve hours for the MR products in its comparison. These are product-dependent reference durations, not guaranteed coverage for an individual. Food requirements and release patterns also vary, so a change between products needs professional advice.10
Why is lisdexamfetamine different?
Lisdexamfetamine is a prodrug: the body converts it to active dexamfetamine, mainly in red blood cells. Its longer effect should not be confused with a modified-release capsule mechanism.2
A useful way to prepare: describe the pattern of your day. When do difficulties matter most? Are you trying to understand changes during work, study, caring responsibilities or the evening? Ask how the proposed medicine fits that pattern and how to report a gap in benefit. Do not rearrange doses to match this article's duration examples.
How soon should treatment help?
Finding the right methylphenidate type and dose can take weeks. Once established, the NHS says a dose can start working within 30–60 minutes. The time to judge treatment differs from the onset of an individual dose.1
Non-stimulants generally take longer to begin having an effect than stimulants. Ask what review period is appropriate for the medicine you have been offered, and what changes to report sooner.11
Our separate guide to medication titration explains preparing for those reviews. For your notes, use ordinary descriptions of what happened rather than deciding whether you have “passed” or “failed” a trial.
What side effects are usual?
The examples below are reported in the linked product information. They are selected common or very common effects, not a complete safety list. They cannot be used to rank medicines by safety: the underlying populations and reporting methods differ.
Scroll sideways to read the full table on a small screen.
| Medicine | Examples of reported unwanted effects |
|---|---|
| Methylphenidate | Sleep difficulty, headache, reduced appetite, anxiety, nausea and dry mouth.7 |
| Lisdexamfetamine | Reduced appetite, insomnia, headache, dry mouth, nausea, anxiety and weight loss.2 |
| Dexamfetamine | Reduced appetite, insomnia, nervousness, headache, nausea and dry mouth.3 |
| Atomoxetine | Nausea, dry mouth, reduced appetite, sleep disturbance, headache, constipation, and some urinary or sexual problems.4 |
| Guanfacine | Sleepiness, fatigue, headache, abdominal pain, dizziness, low blood pressure and a slow heart rate.5 |
The lisdexamfetamine and atomoxetine examples use adult safety data. The methylphenidate source combines age groups. The dexamfetamine and guanfacine sources do not establish adult-specific frequencies; their licensed ADHD populations are children and adolescents. The table therefore deliberately avoids percentages or claims that one medicine causes fewer problems.
Read the leaflet supplied with your own prescription for its full warnings and instructions. Our monitoring and safety guide explains routine checks and when to seek help, with medicine-specific examples. A side-effect table should not be used to decide whether a new or severe symptom can wait.
Questions to bring to a review: What changed? When did it begin? How often does it happen? What does it interrupt? What advice do I need now? Write down your observations without having to establish the cause yourself.
What do meta-analyses say about effectiveness?
A meta-analysis combines results from studies. A network meta-analysis can also compare treatments indirectly through a shared comparator, such as placebo. An umbrella review examines existing reviews. These methods can reveal patterns across evidence, but their conclusions depend on the included studies and the question being asked.
The following are selected major research syntheses, not an exhaustive review of every trial. They draw on international studies, including research outside the UK. They inform the evidence discussion; UK prescribing guidance remains a separate source.
The 2018 comparison: short-term adult symptom improvement
Cortese and colleagues' network meta-analysis included 8,131 adults in the efficacy analysis closest to 12 weeks. Selected clinician-rated results versus placebo were:12
Scroll sideways to read the full table on a small screen.
| Medicine group | Standardised mean difference (95% confidence interval) |
|---|---|
| Amphetamines | −0.79 (−0.99 to −0.58) |
| Methylphenidate | −0.49 (−0.64 to −0.35) |
| Atomoxetine | −0.45 (−0.58 to −0.32) |
A standardised mean difference puts different symptom scales on a common statistical scale. Here, more-negative values favour medication. −0.79 does not mean 79% improvement or a 79% chance of responding. The interval expresses uncertainty around the estimate.12
Amphetamines showed larger average short-term effects, but this is a class result, not a ranking of every brand or formulation. Many comparisons were indirect and confidence varied. These medicines also had more withdrawals due to adverse effects than placebo. Evidence at 26 and 52 weeks was insufficient for the planned comparisons.12
The January 2025 review: outcomes beyond a single rating scale
Ostinelli and colleagues included 113 trials and 14,887 adults, covering medication and other interventions. Around 12 weeks, stimulants and atomoxetine improved core ADHD symptoms compared with placebo on both self-rated and clinician-rated measures. Confidence in those findings ranged from very low to moderate.13
The review did not demonstrate benefit on additional outcomes including quality of life. This does not prove that nobody experiences wider improvements; it shows why symptom scores and broader life outcomes should be discussed separately. Atomoxetine and guanfacine had more all-cause discontinuation than placebo. That measure includes leaving a trial for any reason and differs from stopping specifically because of side effects. Long-term evidence remained limited.13
The November 2025 umbrella review: benefits, tolerability and certainty
Gosling and colleagues reanalysed 221 meta-analyses across children, adolescents and adults. Adult results supported symptom benefits from methylphenidate and atomoxetine with at least moderate certainty; amphetamine findings reached that threshold when restricted to higher-quality trials. The three medicines or medicine groups were less well tolerated than placebo, assessed through stopping because of adverse effects.14
The review found no high-certainty evidence supporting long-term benefit. This describes a limitation of the evidence, rather than proof that benefit ends. Reviews can include overlapping trials, so these three publications should not be counted as three entirely independent sets of patients.14
How should you use that evidence in a consultation?
Use research to frame questions, then describe what matters in your own life. You could ask: Which outcome are we hoping to improve? What would count as a worthwhile change for me? What unwanted effects would make the plan difficult to continue? When will we review the balance?
A simple note can have three headings: what I hoped would change, what I noticed, and what I want to ask. This is an optional way to organise a conversation, not a validated symptom scale or a method for changing treatment yourself.
Medication can sit alongside psychological support. The Royal College of Psychiatrists describes psychoeducation and CBT addressing practical difficulties such as organisation, time management and emotions. Ask what support is available for the concerns that remain important to you.11
Questions people often ask
Is modified release stronger than immediate release?
The label describes delivery over time. Ask about the active ingredient, strength and preparation together; the release label alone does not answer whether a prescription is suitable. See the formulation explanation above.1
Does “long acting” mean I have to take it long term?
These are different timeframes. Ask separately about the expected effect during a day and the plan for reviewing whether to continue treatment. It is reasonable to want both questions answered before starting.
Is a non-stimulant automatically safer?
No. Guanfacine can lower blood pressure and heart rate. Ask which individual risks and checks apply to your prescription.5
Can the research tell me which medicine will work best for me?
The figures describe group averages. Ask your clinician how the evidence applies to your history, priorities and proposed preparation, and how your own response will be reviewed. A larger average effect does not guarantee a better individual result.
What if the packet or formulation changes?
Check with the pharmacist or prescriber. Modified-release methylphenidate products differ and are not all interchangeable. Ask whether instructions or the expected pattern of effect have changed.10
References
Product documents below are current manufacturer summaries of product characteristics (SmPCs) hosted by the electronic medicines compendium. They describe authorised product information and safety; they are distinct from independent comparative research.
-
NHS: About methylphenidate for adults.
Back to text ↑ - Back to text ↑
- Back to text ↑
-
Atomoxetine hard capsules: SmPC.
Back to text ↑ - Back to text ↑
-
NICE NG87: ADHD diagnosis and management.
Back to text ↑ - Back to text ↑
-
Medikinet standard tablets: SmPC.
Back to text ↑ -
MHRA: Off-label or unlicensed use of medicines.
Back to text ↑ -
NHS Specialist Pharmacy Service: Prescribing and switching modified-release methylphenidate.
Back to text ↑ -
Royal College of Psychiatrists: ADHD in adults.
Back to text ↑ -
Cortese et al. (2018): Comparative efficacy and tolerability of ADHD medications—network meta-analysis, The Lancet Psychiatry. DOI: 10.1016/S2215-0366(18)30269-4.
Back to text ↑ -
Ostinelli et al. (2025): Interventions for adult ADHD—component network meta-analysis, The Lancet Psychiatry. DOI: 10.1016/S2215-0366(24)00360-2.
Back to text ↑ -
Gosling et al. (2025): Benefits and harms of ADHD interventions—umbrella review, The BMJ. DOI: 10.1136/bmj-2025-085875.
Back to text ↑
