Untreated ADHD: The Real Costs and Choosing Treatment
Two questions sit behind this topic and they deserve separating. What actually happens to people whose ADHD is never treated, and how does anyone decide which treatment to use.
The first has better evidence behind it than most people realise, drawn from national registries following hundreds of thousands of people over decades. The second is more constrained than it looks, and being honest about the constraints is more useful than pretending the choice is wide open.
📊 The finding that most surprises people. Large registry studies comparing the same individuals during periods on and off medication — which removes the usual objection that treated and untreated people differ in other ways — found substantially fewer road traffic collisions during medicated periods, in the order of a third to two-fifths fewer. Similar within-person designs found lower rates of accidental injury and of criminal offending. Comparing a person with themselves is a much stronger method than comparing groups, and these are among the more robust findings in the field.
⚖️ Before the list, a caveat
Everything that follows describes what happens across populations. It is not a prediction about any individual, and this distinction gets lost in a lot of writing on the subject.
Plenty of people with untreated ADHD do well — often those whose circumstances happen to suit them, who found work that rewards their strengths, or who built compensating systems early. The associations below are real, they are worth acting on, and they are not a forecast.
The reason to state this plainly is that frightening parents into treatment produces bad decisions. The case for treatment is that it works and that the risks of not treating are meaningful, not that disaster is certain.
📉 What untreated ADHD is associated with
| Domain | What the evidence shows |
|---|---|
| Road traffic collisions | Higher crash risk, and substantially lower during medicated periods in within-person analyses |
| Accidental injury | More fractures, burns, poisonings and emergency attendances in children; reduced with treatment |
| Education | Lower attainment than ability predicts, more repeated years, higher rates of leaving early |
| Employment | More frequent job changes, more dismissals, lower earnings across a career |
| Substance misuse | Higher rates when untreated — and see the section below, because the direction is the opposite of what most people assume |
| Offending | Registry data show lower rates of offending during medicated periods within the same individuals |
| Relationships | Higher rates of relationship breakdown; the day-to-day burden of forgotten commitments and unfinished tasks accumulates |
| Money | Impulsive spending, missed payments, difficulty with long-horizon planning |
| Mood | Markedly higher rates of depression and anxiety, much of it arising from years of failing at things others manage easily |
| Mortality | Modestly increased in cohort studies, driven largely by accidents rather than by illness |
Read together, a pattern emerges that is not really about attention. Most of these outcomes are consequences of impaired impulse control and poor sense of future consequence, operating in situations where a moment of poor judgement has an expensive result.
🚗 Driving
This deserves separating out because it is the most concrete and most actionable finding here.
Driving demands sustained attention over long uneventful periods, resistance to distraction, and inhibition of impulsive decisions — a combination that maps almost exactly onto the difficulties in question. Young drivers with ADHD have measurably higher collision rates, and the reduction during medicated periods is among the largest treatment effects reported in this field.
Practical implications for an adolescent or adult who drives: take medication on days involving driving rather than only on school or work days, be aware of when a short-acting dose wears off in relation to an evening drive, and treat this as a specific matter to discuss rather than something that will sort itself out.
🍺 The substance-use question, inverted
The commonest reason families decline treatment is a fear that giving a stimulant to a child leads to drug problems later. It is a reasonable-sounding worry and the evidence points the other way.
Untreated ADHD is associated with higher rates of substance misuse, which makes sense given impulsivity, the tendency to self-medicate difficulty concentrating, and the accumulated demoralisation of years of underachievement.
Treatment does not increase that risk. Meta-analyses of stimulant-treated children followed into adulthood find no increase, and some analyses suggest a reduction.
The genuine concerns are different ones, and they are manageable: diversion of medication to other people, particularly at university; misuse of a controlled medicine within a household where someone has an existing problem; and the fact that some people misuse stimulants recreationally, which is a reason for secure storage and supervision rather than a reason to withhold treatment. Where that risk is genuinely present in a household, a non-stimulant is a reasonable choice specifically because it is not a controlled substance.
😔 The cost nobody counts
The registry outcomes are measurable. The most damaging consequence is not.
A child with untreated ADHD hears an enormous amount about what is wrong with them — corrections, reprimands, comparisons with siblings, disappointed remarks — across years, from adults who mean well and are responding to genuinely difficult behaviour. Very little of it changes anything, because the difficulty is not one of knowledge or motivation.
What it does change is what the child concludes about themselves. By adolescence a great many have settled on an explanation: that they are stupid, lazy, or a disappointment. That belief long outlasts the hyperactivity and shapes what they attempt for the rest of their life.
This is why a diagnosis, even arriving late, is so often described as a relief rather than a burden. It replaces a verdict about character with an explanation about function, and those are very different things to live with.
👤 Adults who were never diagnosed
An entire generation grew up when only visibly hyperactive boys were recognised, so large numbers of adults have this and do not know it — particularly women, and particularly anyone whose presentation was inattentive rather than disruptive.
What it looks like in adulthood is rarely hyperactivity. It is chronic disorganisation, chronic lateness, a long history of unfinished projects begun with enthusiasm, an inability to start dull tasks until a deadline creates panic, careers well below apparent ability, financial chaos, and a persistent sense of underperforming while working harder than everyone else.
Recognition frequently comes second-hand — a parent reading about their own child, or an adult reading about a sibling. Diagnosis in adulthood requires evidence that difficulties were present in childhood, which usually means school reports and family recollection rather than current symptoms alone.
Treatment in adults works. It is not too late in the way people assume.
💊 How the choice is actually made
Here it is worth being straightforward about what is and is not on the table.
Stimulants — methylphenidate and the amphetamines — are first-line for school-age children and adults in every major guideline. They have the largest average effect, they work within hours, and the correct dose can be found within days or weeks.
They are also controlled substances in almost every country, with restrictions on prescribing, dispensing and shipping. They are not available here and cannot be, and it is better to say that than to imply the available options are equivalent. Anyone whose treatment should be a stimulant needs a local prescriber and a local pharmacy.
What is available is the main non-stimulant option, and it is a genuine treatment with specific advantages rather than a consolation prize.
| Stimulants | Atomoxetine | |
|---|---|---|
| Class | Psychostimulant | Selective noradrenaline reuptake inhibitor — not a stimulant |
| Average effect | Larger | Somewhat smaller |
| Time to work | Hours | Four to six weeks |
| Coverage | Hours per dose, with a wearing-off period | Continuous, including evenings and weekends |
| Controlled substance | Yes | No |
| Appetite and sleep | Appetite suppression and insomnia common | Less appetite effect; may cause drowsiness or insomnia |
| Rebound | Irritability as the dose wears off is common | None, since levels are steady |
| Tics | Can worsen them | Does not |
| Coexisting anxiety | Sometimes worsened | Often improved |
Reading the right-hand column as a list of situations rather than a list of weaknesses gives the real answer. Atomoxetine is specifically preferred where stimulants have not been tolerated or have not worked, where there are tics, where anxiety coexists, where evening and weekend coverage matters, where the wearing-off of a stimulant in the late afternoon is itself the problem, where a controlled medicine in the household is a concern, and where the family prefers a non-stimulant.
⚠️ The four-to-six-week timeline is the thing that most often causes atomoxetine to be abandoned unnecessarily. Benefit builds gradually and can continue improving for two to three months, which is unrecognisable to anyone expecting the same-afternoon change a stimulant produces. Two weeks is not a trial. Keep a written record or a repeated rating scale, since gradual improvement is genuinely hard to perceive from inside a household. Report any new or worsening low mood, agitation or thoughts of self-harm promptly, particularly in the early weeks, and be aware of the rare possibility of liver problems — unexplained abdominal pain, dark urine, jaundice or persistent nausea should be reported rather than waited out.
❌ Two things that are not ADHD treatments
Both appear in this context and both are worth correcting directly.
Modafinil and armodafinil. These are wakefulness-promoting agents licensed for narcolepsy, sleep apnoea and shift work sleep disorder. Modafinil was trialled in ADHD and did not gain approval for it; armodafinil has no established role at all. They are not equivalent alternatives to ADHD medication, they should not appear in a comparison table alongside licensed treatments, and someone with ADHD taking one is likely to feel more awake without addressing the impulse control and executive function difficulties that constitute the condition.
Aripiprazole. This is an antipsychotic used in schizophrenia, bipolar disorder, as an addition in depression, and for severe irritability in autism. It is not a treatment for ADHD. It is occasionally used alongside ADHD treatment where severe aggression or mood instability coexists, which is a specialist decision about a separate problem — not a way of treating attention or impulsivity. Its side effect profile, including weight gain, metabolic effects and movement disorders, is in a different category from the drugs above and makes casual use inappropriate.
💵 The same drug at two prices
Worth pointing out because the comparison is easy to miss.
Atomoxetine is available here under two names at the same 40 mg strength. Atomoxet works out at around $0.70 per tablet; Strattera at around $2.44. Both are atomoxetine, both are made by established Indian manufacturers, and there is no published evidence of a difference in effect between them at the same dose.
For a medicine taken daily and indefinitely, that gap compounds substantially. Someone already established on one and doing well has a reasonable case for not changing anything. Someone choosing for the first time should at least know the difference exists.
🎚️ Finding the dose, and knowing whether it works
This part is done badly more often than the choice of drug is.
Start low and increase gradually. Atomoxetine is usually started below the target dose and increased after a week or more, partly for tolerability and partly because a proportion of people metabolise it slowly and reach higher levels than expected on a standard dose.
Take it consistently. Unlike a stimulant, it does not work on the days it is taken and not on others — it needs continuous daily use to maintain the effect, and missed days undermine it.
Measure rather than remember. A standardised rating scale completed by a parent and a teacher before starting and again at six and twelve weeks answers the question far better than anyone’s impression. This is especially true for a drug whose effect appears gradually.
Decide in advance what improvement would look like. Finishing homework without three hours of conflict, fewer notes home, being able to hold a conversation, completing a task once started. Specific targets are assessable; feeling better is not.
Monitor the things worth monitoring — height and weight in children, blood pressure and pulse, sleep, appetite and mood.
Review whether it is still needed. Requirements change with age, circumstances and demands, and periodic reassessment is normal rather than a sign of failure.
🧩 The parts that are not optional
Medication addresses the biological difficulty. It does not teach anybody how to organise a week.
Parent training in behaviour management is first-line in preschool children, ahead of medication, and remains valuable alongside it later.
School and workplace adjustments — seating, extra time, broken-down instructions, written rather than verbal information, deadlines that are visible.
Externalised systems. Visible checklists, alarms, one place for everything. Systems in the environment succeed where systems held in the head do not, and this is the single most useful adult strategy.
Sleep, which is disturbed in a large proportion of people with ADHD and which worsens every symptom when short.
Exercise, which measurably improves attention afterwards.
Treating what else is present. Anxiety, depression, learning difficulties and sleep disorders are all common alongside, and treating the ADHD alone while ignoring them produces a partial result.
🩺 When to act
- A diagnosis made but never treated, particularly in an adolescent approaching driving age
- An adult who recognises the description and has never been assessed
- Falling attainment in a capable child
- Low mood, anxiety or statements about being stupid or worthless
- Alcohol or drug use starting early in an adolescent with untreated symptoms
- Repeated accidents or injuries
- Treatment tried briefly and abandoned, especially a non-stimulant stopped inside a month
- New low mood, agitation or thoughts of self-harm after starting atomoxetine — report promptly
- Unexplained abdominal pain, dark urine, jaundice or persistent nausea on atomoxetine
Treatment is stocked in the ADHD category, including Atomoxet (atomoxetine 40 mg) and Strattera (atomoxetine 40 mg). Stimulant treatment, which is first-line, requires a local prescriber and pharmacy.
❓ Frequently asked questions
What actually happens if ADHD is left untreated?
Across populations, higher rates of road collisions and accidental injury, lower educational attainment than ability predicts, less stable employment, more substance misuse, more relationship breakdown, and markedly higher rates of depression and anxiety. These are associations rather than predictions — plenty of untreated people do well. They are a reason to act, not a forecast about any individual.
Does treatment really reduce accidents?
The evidence here is unusually strong because it compares the same people during periods on and off medication, which removes the objection that treated and untreated people differ in other ways. Those within-person analyses found road collisions reduced by roughly a third to two-fifths during medicated periods, alongside reductions in accidental injury and in offending.
Will medication lead to drug problems later?
The evidence points the opposite way. Untreated ADHD is associated with higher rates of substance misuse, and meta-analyses of treated children followed into adulthood find no increase from treatment, with some suggesting a reduction. The real concerns are different and manageable — diversion of medication to others, and controlled drugs in a household where someone already has a problem. In that situation a non-stimulant is a sound choice.
Why can I not buy stimulants here?
Because methylphenidate and the amphetamines are controlled substances almost everywhere, with restrictions on prescribing, dispensing and shipping. They are first-line treatment and they require a local prescriber and a local pharmacy. Saying so is more useful than implying the available options are equivalent. What is available is the main non-stimulant, which is a genuine treatment with its own specific advantages.
When is a non-stimulant the better choice?
Where stimulants have not been tolerated or have not worked, where there are tics, where anxiety coexists and is worsened by stimulants, where evening and weekend coverage matters, where the late-afternoon wearing-off of a stimulant is itself the problem, where a controlled medicine in the household is a concern, and where the family simply prefers a non-stimulant. These are indications, not consolations.
How long before atomoxetine works?
Four to six weeks for meaningful benefit, with continued improvement possible over two to three months. Two weeks is not a trial, and stopping early is the commonest reason it appears to fail. Keep a written record or repeat a rating scale, since gradual improvement is hard to perceive from inside a household. It also needs taking every day — unlike a stimulant it does not work only on the days it is taken.
Is armodafinil an option for ADHD?
No. Modafinil and armodafinil are wakefulness-promoting agents licensed for narcolepsy, sleep apnoea and shift work sleep disorder. Modafinil was trialled in ADHD and did not gain approval; armodafinil has no established role. Someone with ADHD taking one is likely to feel more awake without any effect on the impulse control and executive function difficulties that constitute the condition.
What about aripiprazole?
It is an antipsychotic used in schizophrenia, bipolar disorder, as an addition in depression and for severe irritability in autism. It is not a treatment for ADHD. It is occasionally used alongside ADHD treatment where severe aggression or mood instability coexists, which is a specialist decision about a separate problem. Its side effects, including weight gain, metabolic effects and movement disorders, make casual use inappropriate.
Atomoxet and Strattera are both atomoxetine. Why the price difference?
Both are atomoxetine 40 mg from established Indian manufacturers, and there is no published evidence of a difference in effect at the same dose. Atomoxet works out at roughly seventy cents a tablet against about two dollars forty for Strattera. For a medicine taken daily and indefinitely that compounds substantially. Anyone already doing well on one has a good reason not to change; anyone choosing first should know the gap exists.
How do I know whether it is working?
Measure rather than remember. A standardised rating scale completed by a parent and a teacher before starting and again at six and twelve weeks answers this far better than anyone's impression, especially for a drug whose effect appears gradually. Decide in advance what improvement would look like in specific terms — homework finished without conflict, a conversation held, a task completed once started.
I am an adult who was never diagnosed. Is it too late?
No — treatment works in adults. An entire generation grew up when only visibly hyperactive boys were recognised, so many adults have this unknowingly, particularly women and anyone whose presentation was inattentive. In adulthood it looks like chronic disorganisation and lateness, unfinished projects, inability to start dull tasks without a deadline, and a career below apparent ability. Diagnosis requires evidence the difficulties were present in childhood.
Is medication enough on its own?
No. It addresses the biological difficulty and does not teach anyone how to organise a week. Parent training is first-line in preschool children and valuable alongside medication later. School and workplace adjustments matter, as do externalised systems — visible checklists, alarms, one place for everything, which succeed where systems held in the head do not. Sleep, exercise, and treating anything coexisting all count.
📑 Sources and editorial
- Within-person registry analyses of ADHD medication and motor vehicle crash risk
- Swedish national registry study of ADHD medication and criminality
- Danish cohort study of mortality in ADHD
- Meta-analyses of stimulant treatment in childhood and later substance use outcomes
- NICE guideline on ADHD diagnosis and management
- American Academy of Pediatrics clinical practice guideline on ADHD
- Studies of accidental injury rates in children with ADHD and the effect of treatment
- Literature on adult ADHD recognition, presentation and treatment response
- FDA prescribing information for atomoxetine, including titration, monitoring and hepatic warnings
- Regulatory history of modafinil trials in ADHD
- Related reading: telling ADHD from misbehaviour, what ADHD is and how it is treated
- Related products: ADHD category
- RXshop Editorial Team — reviewed by Thomas Walsh, MD, Psychiatrist and Mental Health Specialist
Medical Disclaimer: The information in this article is for educational and informational purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Always seek guidance from a qualified healthcare provider with any questions you may have regarding a medical condition, and before starting, stopping or changing any medication.