The Pleasure Leaves First. The Wanting Stays.

What you'll learn

Whether sugar is genuinely addictive — and why the honest answer is more useful than the confident one

What separates addiction from a strong habit, and why gambling settles the argument

The recovery statistic almost nobody knows

Which withdrawals are medically dangerous — this section could matter more than the rest combined

Why two words in a clinical vignette changed how clinicians judged the same patient

How SmartBar applies the deficiency framework

SmartBar's educational framework begins with two organizing terms, and applies them wherever the body has inputs and burdens — starting with what we can verify in food: ingredient composition, level of processing, and nutritional data. Deficiency means something the body needs and isn't getting. Toxicity means a burden the body must process and clear — not a poison.

This is the article where our framework does the least work, and we'd rather say so than stretch it. Addiction is not a nutrient problem. SmartBar's expertise reaches the food end of it and stops. What follows is education, not treatment, and it is not medical advice.

If you are currently dependent on alcohol or benzodiazepines, read the section on withdrawal before you act on anything else here. Stopping either abruptly, without medical supervision, can kill you.

Start where we actually have standing, because a lot of readers arrive here from the same place.

Is sugar addictive?

The confident answer is everywhere. Sugar lights up the brain like cocaine. Sugar is eight times more addictive than heroin. You have seen the headlines.

The honest answer is more interesting.

The animal evidence is real. Rats given sugar show bingeing, withdrawal-like signs, craving and cross-sensitisation — the behavioural signature of a substance of abuse.[1] But there's a condition attached that almost never survives the headline: those behaviours appear only when the animals are deprived for 12 to 16 hours and then given access. Rats with unlimited sugar didn't show them.[2]

A Cambridge review examining both the animal and human literature concluded there is little evidence to support sugar addiction in humans, and that the addiction-like behaviours in animals arise from intermittent access to sweet, highly palatable food rather than from the neurochemical effects of sugar itself.[2]

In those animal models, deprivation and repeated cycles of restriction and access generate much of the pattern. That is one reason to be cautious about reading cravings produced during restrictive dieting as proof that sugar itself is addictive.

None of that means cravings aren't real. A contested label means researchers disagree about whether "addiction" is the right explanation — not about whether people experience powerful cravings.

We'll take this apart properly in a dedicated article — sugar under all its names, from ordinary carbohydrate craving through to genuinely compulsive eating. This page is about the wider thing that question belongs to.

What addiction actually is

Addiction isn't liking something a lot. Across diagnostic systems, the recurring features include: impaired control over use, increasing priority given to it over other activities, and continuation or escalation despite negative consequences.[3]

Notice what's missing: no particular quantity automatically establishes addiction. The diagnosis turns on impaired control, priority, persistence despite harm and the resulting impairment — not simply how much someone consumes, or which substance is involved.

Gambling settles this, and it's the most clarifying fact in the field.

Gambling disorder is the only non-substance disorder placed in the DSM-5-TR addictions chapter, moved there in 2013 from "impulse control disorders"; ICD-11 also recognises it.[3] Nothing is ingested. No substance crosses the blood-brain barrier. And yet it produces impaired control, escalation, and persistence despite mounting harm — alongside neurobiological changes in the same reward and cognitive-control circuits implicated in substance use disorders.[4]

So addiction can occur with no substance at all. Which reframes the sugar question entirely: asking whether a molecule is addictive may be the wrong question. The pattern is the thing.

The dopamine part — and the bit that gets left out

You've heard that addiction is about chasing dopamine. That's roughly right, and the way it's usually told gets one thing backwards — and the correction explains something people actually experience.

Dopamine is not the pleasure chemical. It's closer to the wanting chemical.

"Liking" is the pleasure a reward produces. "Wanting" is the pull toward it — the urge, the reaching, the itch. They feel joined from the inside, but rely on partly separable systems: dopamine is central to wanting, while pleasure itself is not simply a dopamine signal.[16]

Here's why that matters. In the incentive-sensitization model of addiction, those two can come apart: wanting becomes amplified even when liking does not.[16]

That matches a common experience: I don't even enjoy it anymore. I don't know why I keep doing it. In this model, cues acquire disproportionate motivational pull even after much of the pleasure has faded — which is not denial, and not an excuse.

On "burnt-out dopamine receptors." You'll see the claim that addiction fries your dopamine receptors and you spend years rebuilding them. There is something real underneath it: brain imaging does show reduced D2 receptor availability in people with addiction. But its meaning and causal direction are contested — the reduction may develop partly as a consequence of prolonged heavy use rather than simply preceding and causing addiction.[17]

Which is why willpower is such a poor tool against it. You're not simply arguing with something you enjoy. You're working against a system that has learned to assign it disproportionate attention and motivational pull.

The race doesn't start level

Risk is unevenly distributed long before the first drink or pill. Adverse childhood experiences — including abuse, neglect and household dysfunction — show a graded association with later substance-use problems: as exposure accumulates, risk rises.[18]

That does not make addiction inevitable or establish causation in any individual. It means people can arrive at the same apparent choice carrying very different histories.

Why "just stop" misdescribes the problem

Tolerance means needing more to achieve a previous effect, or getting less effect from the same amount. Physical dependence means the body has adapted enough that reducing or stopping can produce withdrawal. Neither by itself establishes addiction — both can occur during appropriate prescribed treatment. Craving is the wanting described above, running on its own regardless of whether you still enjoy the thing.

The willpower frame fails for a simple reason: it predicts the wrong things. If addiction were a willpower deficit, you would expect people with more discipline in other areas of life to be protected. They aren't. You would expect relapse to be rare among the determined. It isn't.

Recurrence is common in substance use disorders, as it is in many long-term conditions. But the rate varies enormously by substance, severity, treatment, follow-up period, and what a given study counts as relapse — a systematic review found no consensus definition, with nearly a third of studies using the term not defining it at all.[5]

What recurrence does not establish is a lack of resolve.

It is information: risk remains, and the support or treatment plan may need to change. That is all it has ever meant.

Is one thing more addictive than another?

There is real science here, and one finding most readers don't expect.

Researchers asked a specific question of Americans aged 15–54: of everyone who ever tried a substance, what proportion went on to become dependent? They called it the capture rate. About one in three tobacco smokers became dependent — the highest in the survey. Roughly one in seven drinkers, and roughly one in seven users of other controlled substances, which that survey grouped together rather than reporting drug by drug.[22]

The substance with the highest capture rate was the legal one sold in petrol stations. Not the one with the frightening reputation.

But "addictive" is not one measurement, which is where rankings mislead. How likely you are to develop a disorder, how hard it is to stop, how fast dependence develops, and how much harm it causes are four different questions that rank in four different orders. Capture rate also reflects availability, dose, delivery and era — not an intrinsic property of a molecule. And that survey is from 1994, using older diagnostic criteria.

So two things, and they're what we'd want you to leave with. Quitting anything genuinely addictive is hard, and nobody's difficulty is invalidated by where their substance sits on a chart. But the consequences are not equal. Nicotine withdrawal is unpleasant and not usually dangerous. Alcohol withdrawal can kill you. Opioid withdrawal is rarely fatal by itself, but lost tolerance can make return to use deadly. The difficulty may be comparable; the danger is not — and that difference decides whether you need a doctor before you start, not just support along the way.

This is education, not medical advice, and none of it can tell you what your own situation requires.

The statistic almost nobody knows

Recovery is far more common than almost anyone believes.

A nationally representative survey found that 9.1% of US adults — roughly 22.3 million people — have resolved a significant alcohol or drug problem.[6] A CDC and NIDA analysis of separate national data found that among US adults who reported ever having had an alcohol or drug problem, roughly three in four described themselves as in recovery or recovered — more than 20.5 million people.[7]

The exact estimate depends on how "problem" and "recovery" are defined, and both surveys rely on self-identification across problems of varying severity. What they overturn is the belief that lasting improvement is rare.

And the routes vary more than the culture admits. In the survey data, about 54% used some form of assistance — mutual-help groups most commonly, then formal treatment. The other 46% resolved it without any of that.[6] Assisted recovery was associated with greater problem severity and complexity — consistent with people facing more severe problems being more likely to seek or require support.

This is not an argument against treatment. It is an argument against despair. Lasting resolution is common, and almost nobody knows it — least of all the people in the middle of it, who are told constantly that they are facing a life sentence.

Recovery may take repeated attempts, although experiences vary widely. Needing another attempt is not evidence that recovery has failed.

What works, and what almost nobody is offered

Effective medications exist for alcohol use disorder and for opioid use disorder.

Of people with past-year alcohol use disorder in 2024, only 7.6% received any treatment — and just 2.5% received medication for it.[8]

That is not a story about people who won't get better. It's a treatment gap — the same shape as the finding on our Sleep page, where the guideline-recommended first-line treatment for chronic insomnia is widely under-offered. When effective care exists and very few people receive it, the failure is not adequately explained by patient motivation.

If you take one practical thing from this page: if you or someone close to you has a problem with alcohol or opioids, ask specifically about medication. Don't assume it will be offered.

The treatment people laugh at

Paying someone for meeting a recovery goal sounds simplistic — even offensive. Yet contingency management is among the best-supported behavioural treatments in addiction medicine, with decades of evidence behind it.[19]

Small incentives tied to verified goals — a negative test, showing up to an appointment — consistently improve abstinence, treatment retention and medication adherence. It is considered the current standard of care for stimulant use disorder, for which no approved medication exists.[19] Benefits are strongest while the programme is active, and evidence on how well they persist afterwards is mixed.

Despite the evidence, access remains limited in routine care; the US Veterans Health Administration is one of the best-known large-scale implementers. Much of the resistance isn't about whether it works — it's discomfort with rewarding behaviour people feel should happen anyway. That's a moral objection, not an evidence-based one.

The section that matters most: some withdrawals can kill you

Read this before acting on anything else here.

Abruptly stopping some substances is not merely difficult — it can be lethal. With most things, going cold turkey is just hard. With two, it can kill you.

Alcohol. Among people with sustained heavy use or physical dependence, abrupt cessation can trigger withdrawal.[11] Seizures typically occur 8 to 48 hours after the last drink, and can happen in people with no seizure history.[12] Delirium tremens — severe confusion, hallucinations, fever, autonomic instability — typically begins 48 to 72 hours in and is a medical emergency.[11]

On the mortality figures — and this is worth knowing about how such numbers travel. You will see delirium tremens death rates quoted with great confidence: 15%, 20%, 35%, 37%. The clinical literature does not actually agree on them. Published estimates for untreated or historical mortality range from around 15% to 40%, one peer-reviewed review argued the widely repeated 15–20% figure was never borne out in practice, and contemporary estimates with treatment range from about 1% to 5%.[11][12]

What the literature does agree on is the part that matters to you: delirium tremens is a medical emergency, it can be fatal, and prompt medical treatment markedly reduces the risk of death.[12] That is the sentence to act on. Anyone quoting you a precise percentage — in either direction — is more certain than the evidence.

Benzodiazepines. Withdrawal can produce seizures even in people taking therapeutic doses as prescribed. People who have taken benzodiazepines regularly for longer than about a month may be physically dependent and should not abruptly discontinue them without clinical guidance. Current joint guidance recommends an individually adjusted taper where tapering is indicated.[13]

Medical management reduces the risk of severe complications. Benzodiazepines are first-line medicines for alcohol withdrawal, including prevention and treatment of seizures and delirium in people at risk.[13]

So the message is not "don't quit." It's "don't quit cold turkey on your own." With alcohol or benzodiazepines, the safe first step is a conversation with a doctor — before the last drink, not after the seizure.

Opioid withdrawal, by contrast, is intensely unpleasant but rarely directly life-threatening — though the danger afterwards is severe, because tolerance falls fast and a previously ordinary dose can become fatal.

Staying alive is not the opposite of getting better

There's a persistent idea that keeping people alive and helping them recover are competing goals. The evidence says they're the same project, because dead people don't recover.

Naloxone temporarily reverses opioid-induced respiratory depression by blocking opioid receptors, buying time for emergency care — and it can wear off before the opioids do. It has been used in hospitals for over fifty years. Community take-home programmes put it in the hands of the people most likely to witness an overdose — friends, family, other users — with training on how to use it.[20]

Syringe services programmes provide sterile equipment and connect people with testing, vaccination, naloxone and treatment. Decades of evidence do not support the claim that they increase drug use or crime; comprehensive programmes likely reduce HIV transmission and may reduce hepatitis C transmission.[21]

Harm reduction is not the opposite of recovery. It keeps recovery possible.

Addictions without a substance

If addiction is a pattern rather than a chemical, the obvious question is how far it extends. The honest answer is that the diagnostic manuals disagree, and the disagreement is worth seeing.

Gambling disorder is recognised in both DSM-5 and ICD-11.[3]
Gaming disorder is recognised in ICD-11, issued by the WHO in 2022. In DSM-5 it appears only as a condition requiring further research — alongside, notably, caffeine use disorder.[3]
Compulsive sexual behaviour, compulsive buying, and problematic social media use are clinically relevant, prevalent, and studied — but formally recognised as addictive disorders in neither manual, with empirical support varying considerably.[4]

Two respected diagnostic systems, looking at the same evidence, reaching different conclusions. That's not a scandal. It's what an unfinished science looks like — and anyone telling you confidently that scrolling is "exactly like heroin" is ahead of both of them.

And heavy use is not automatically addiction, including in children. Gaming disorder requires impaired control, gaming taking priority over other life interests, and continuation despite harm — generally with significant functional impairment over at least twelve months.[3] Most heavy gaming does not meet that bar. The wider dispute over screens, social media and youth mental health is genuinely unresolved, and it belongs in our forthcoming Mental Health article rather than here.

Two words

The most striking finding in this whole field is not about brains. It's about language.

Clinicians were given identical case descriptions differing in two words. Where the person was called a "substance abuser" rather than "a person with a substance use disorder," clinicians were more likely to attribute personal responsibility for the condition, and more likely to endorse punitive rather than therapeutic responses.[14]

Same case. Same facts. Different label. Different judgement.

Provider stigma is linked to delayed help-seeking and to poorer treatment engagement.[10] And stigma gets internalised: people absorb it, and self-stigma becomes its own barrier to seeking care.[9]

Which is why the National Institute on Drug Abuse now publishes explicit guidance on language for clinicians.[15] Not merely as etiquette, but because language can alter clinical perception and recommended care.

What actually helps, starting with free

Mutual-help groups are the most-used route, and they cost nothing. In the national data they were the most common form of assistance among people who resolved a problem — more common than formal treatment.[6]

Ask about medication, particularly for alcohol and opioids. It is under-offered, and asking directly is often what surfaces it.

Reduce the cue load. Craving is triggered by context — places, people, times of day, the ritual around the substance. And cue-triggered craving can persist long after withdrawal has finished, which is exactly what the wanting-versus-liking research predicts. Changing your environment isn't weakness; it's removing the trigger rather than out-arguing it.

Tell one person. Recovery doesn't have to begin publicly. It can begin with one person who knows what's happening.

And if this is urgent, contact a doctor, a local addiction service, or an emergency number now rather than finishing this article. Nothing here is a substitute for that.

Key Takeaways

Addiction is a pattern, not a measure of character. Impaired control, increasing priority, and continuation despite harm matter more than which substance is involved — gambling qualifies with no substance at all.

Wanting and liking can separate. A person may feel intensely driven toward something they no longer enjoy. That isn't an excuse; it's the mechanism.

Recovery is common. Among US adults reporting a past alcohol or drug problem, roughly three in four described themselves as recovered or in recovery — though definitions and severity vary.

Recurrence is not a character verdict. Rates vary by substance and by definition, but renewed use means risk or treatment needs reassessing.

Effective treatment is underused. Medication for alcohol and opioid use disorders, and contingency management for stimulant use disorder, remain far less available than the evidence warrants.

Keeping someone alive supports recovery. Naloxone and syringe services reduce preventable harm and create routes into care.

Alcohol and benzodiazepine withdrawal can be life-threatening. Neither should be stopped abruptly without medical guidance where physical dependence may be present.

How to think about addiction in thirty seconds

Look for the pattern, not the substance. Impaired control, rising priority, continuing despite harm. That's the definition — and it applies with or without a chemical.
Treat relapse as information. It means the approach needs adjusting, not that the person failed. Chronic conditions behave this way.
Ask about medication explicitly, especially for alcohol or opioids. Assume it won't be offered.
Never withdraw from alcohol or benzodiazepines alone if you may be physically dependent. Talk to a doctor first — this is the one item on this list that can kill.
Keep recovery in view. National surveys show millions resolve serious alcohol and drug problems, by many different routes — with and without formal treatment.

What SmartBar does with this

Very little, and it would be dishonest to claim otherwise.

SmartBar scores packaged food. It cannot treat addiction, detect it, or advise on it. There is no scan that solves this.

Where the platform genuinely touches this subject is at the narrow end: highly processed foods are engineered to be easy to overeat, and our front-of-package and Nutrition pages show what's in them and what the label doesn't say. That's a real contribution to one small corner of a very large problem, and it is the whole of what we can offer.

We built this page anyway, because the alternative was leaving a subject that affects our readers unaddressed simply because it doesn't route through a barcode.

And the reason it's free is the reason it exists. People in the middle of this are frequently sold things — programmes, retreats, subscriptions, certainty. Information shouldn't be one of them.

This article is educational and is not medical advice. If you are dependent on alcohol or benzodiazepines, do not stop abruptly without medical supervision — withdrawal from either can be life-threatening. If you are in crisis, experiencing thoughts of harming yourself, or witnessing a suspected overdose, contact emergency services immediately. For non-urgent help, a GP or primary care doctor is a reasonable first contact, and national helplines for substance use, gambling and mental health exist in most countries and are typically free and confidential. Nothing on this page substitutes for assessment by a qualified professional who knows your history.

Keep exploring: Lifestyle · Nutrition · Sleep · Obesity · Front-of-package claims

Medical Disclaimer: Read our full Medical Disclaimer

References

Avena, N.M., Rada, P., Hoebel, B.G. (2008). "Evidence for sugar addiction: Behavioral and neurochemical effects of intermittent, excessive sugar intake." Neuroscience & Biobehavioral Reviews 32(1):20–39. PMCID: PMC2235907. (Animal model; bingeing, withdrawal, craving and cross-sensitisation demonstrated behaviourally with sugar bingeing as reinforcer. The authors' own framing is conditional: sugar "may be addictive for some individuals when consumed in a binge-like manner" — the effects arise under intermittent access, not ordinary consumption.)
Westwater, M.L., Fletcher, P.C., Ziauddeen, H. (2016). "Sugar addiction: the state of the science." European Journal of Nutrition 55(Suppl 2):55–69. DOI: 10.1007/s00394-016-1229-6. PMCID: PMC5174153. (Cambridge review of animal and human literature. Found little evidence to support sugar addiction in humans; addiction-like behaviours in animals occur only under intermittent access and likely arise from sweetness and palatability rather than sugar's neurochemical effects. Argues against premature incorporation of sugar addiction into scientific literature and public policy. Disclosure: the authors declared no conflicts of interest and reported no financial sponsorship; the article appeared in a supplement sponsored by Rippe Health.)
Diagnostic classification. American Psychiatric Association, DSM-5 and DSM-5-TR: gambling disorder is the only non-substance disorder placed in the DSM-5-TR addictions chapter, reclassified in 2013 from impulse control disorders into "Substance-Related and Addictive Disorders"; internet gaming disorder is listed as a condition requiring further research. World Health Organization, ICD-11 (issued 2022): gambling disorder and gaming disorder both classified under "Disorders due to Substance Use and Addictive Behaviours." Diagnostic features across both systems comprise impaired control, increasing priority given to the behaviour, and continuation or escalation despite negative consequences. Primary sources: American Psychiatric Association (2022), Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR), Washington, DC: American Psychiatric Association Publishing, DOI: 10.1176/appi.books.9780890425787 — "Substance-Related and Addictive Disorders" chapter, which contains gambling disorder as the sole non-substance disorder, with internet gaming disorder listed in Section III, Conditions for Further Study. World Health Organization (2019/2022), International Classification of Diseases, Eleventh Revision (ICD-11), https://icd.who.int/ — 6C50 gambling disorder and 6C51 gaming disorder, both under "Disorders due to substance use or addictive behaviours," each defined by impaired control, increasing precedence over other life interests, and continuation or escalation despite negative consequences, normally evident over at least 12 months.
Brand, M., Antons, S., Bőthe, B., Demetrovics, Z., Fineberg, N.A., Jimenez-Murcia, S., King, D.L., Mestre-Bach, G., Moretta, T., Müller, A., Wegmann, E., Potenza, M.N. (2025). "Current Advances in Behavioral Addictions: From Fundamental Research to Clinical Practice." American Journal of Psychiatry 182(2):155–163. DOI: 10.1176/appi.ajp.20240092. PMID: 39659159. (Narrative review of five potential disorders due to addictive behaviours — gambling, gaming, compulsive sexual behaviour, compulsive buying-shopping, problematic social media use. All described as clinically relevant and prevalent, with empirical support varying across conditions. See also Clark, L. (2014), "Disordered gambling: the evolving concept of behavioral addiction," Annals of the New York Academy of Sciences, DOI: 10.1111/nyas.12558, on how a behaviour becomes addictive without exogenous drug stimulation.)
Moe, F.D., Moltu, C., McKay, J.R., Nesvåg, S., Bjornestad, J. (2022). "Is the relapse concept in studies of substance use disorders a 'one size fits all' concept? A systematic review of relapse operationalisations." Drug and Alcohol Review 41(4):743–758. DOI: 10.1111/dar.13401. PMID: 34792839. (276 studies reviewed; 32% did not define relapse, and no consensus operationalisation was identified. Cited to explain why this article makes no cross-condition relapse-rate comparisons.)
Kelly, J.F., Bergman, B., Hoeppner, B.B., Vilsaint, C., White, W.L. (2017). "Prevalence and pathways of recovery from drug and alcohol problems in the United States population: Implications for practice, research, and policy." Drug and Alcohol Dependence 181:162–169. DOI: 10.1016/j.drugalcdep.2017.09.028. (Nationally representative address-based probability sample; 9.1% of US adults — approximately 22.3 million — reported having resolved a significant alcohol or drug problem. Approximately 54% used some form of assistance: mutual-help groups 45.1%, formal treatment 27.6%; assisted recovery was associated with greater problem severity and complexity. Cross-sectional and retrospective self-report.)
Jones, C.M., Noonan, R.K., Compton, W.M. (2020). "Prevalence and correlates of ever having a substance use problem and substance use recovery status among adults in the United States, 2018." Drug and Alcohol Dependence 214:108169. DOI: 10.1016/j.drugalcdep.2020.108169. PMID: 32682218. PMCID: PMC11060571. (Nationally representative data; 11.1% of US adults reported ever having a problem with alcohol or drugs other than tobacco, and 74.8% of those — more than 20.5 million adults — reported being in recovery.)
National Institute on Alcohol Abuse and Alcoholism. "Alcohol Treatment in the United States." Updated March 2026; data from the 2024 National Survey on Drug Use and Health (NSDUH). (Of people aged 12 or older with past-year alcohol use disorder, 7.6% received alcohol treatment and 2.5% received medication-assisted treatment in the past year.)
Volkow, N.D., Gordon, J.A., Koob, G.F. (2021). "Choosing appropriate language to reduce the stigma around mental illness and substance use disorders." Neuropsychopharmacology 46:2230–2232. DOI: 10.1038/s41386-021-01069-4. PMID: 34276051. PMCID: PMC8580983. (Stigma-related bias among clinicians can contribute to a treatment-averse mindset and failure to implement proven methods; internalised self-stigma creates further barriers to care.)
Bielenberg, J., Swisher, G., Lembke, A., Haug, N.A. (2021). "A systematic review of stigma interventions for providers who treat patients with substance use disorders." Journal of Substance Abuse Treatment 131:108486. DOI: 10.1016/j.jsat.2021.108486. PMID: 34217033. (15 studies; providers frequently endorsed negative attitudes toward patients with substance use disorders. Contact-based interventions involving people in recovery reduced clinician stigma. The "five to six years to seek treatment" figure previously drawn from this review has been removed from the article pending verification of its original source.)
Schuckit, M.A. (2014). "Recognition and management of withdrawal delirium (delirium tremens)." New England Journal of Medicine 371(22):2109–2113. PMID: 25427113. (Primary clinical review of alcohol withdrawal delirium: presentation, timing, risk factors and management.) On the mortality figures specifically: published estimates diverge substantially. Reported historical or untreated mortality ranges from approximately 15% to 40% across sources; a peer-reviewed review (PMID 8244788) contended that the frequently quoted 15–20% rate was not confirmed in practice and that mortality approached zero with benzodiazepine treatment; contemporary treated mortality is variously reported between about 1% and 5%. Because these figures are not consistent across the clinical literature, this article states the agreed clinical position — that delirium tremens is a medical emergency in which prompt treatment markedly reduces mortality — rather than a specific percentage.
StatPearls, "Alcohol Withdrawal Syndrome." NCBI Bookshelf NBK441882. (Alcohol-related seizures typically occur 8 to 48 hours after cessation, with risk of recurrent seizures and progression to alcohol withdrawal delirium. Mortality of alcohol withdrawal delirium historically as high as 20%; with advances in critical care, prompt diagnosis and treatment, now around 1%. Treatment with benzodiazepines reduces risk of progression.)
American Society of Addiction Medicine. The ASAM Clinical Practice Guideline on Alcohol Withdrawal Management. Journal of Addiction Medicine (2020) 14(3S Suppl 1):1–72. DOI: 10.1097/ADM.0000000000000668. PMID: 32511109. (Benzodiazepines first-line for reducing signs and symptoms of withdrawal including incidence of seizure and delirium; preventative pharmacotherapy recommended for patients at risk of severe or complicated alcohol withdrawal; phenobarbital an alternative where benzodiazepines are contraindicated, in appropriate settings. Benzodiazepine tapering: Brunner, E., et al. (2025), Joint Clinical Practice Guideline on Benzodiazepine Tapering: Considerations When Benzodiazepine Risks Outweigh Benefits, Journal of General Internal Medicine 40:2814–2859, DOI: 10.1007/s11606-025-09499-2 — multidisciplinary guideline led by ASAM, warning against abrupt discontinuation in people likely to be physically dependent and emphasising individualised tapering. This article states the warning but deliberately provides no tapering protocol.)
Kelly, J.F., Westerhoff, C.M. (2010). "Does it matter how we refer to individuals with substance-related conditions? A randomized study of two commonly used terms." International Journal of Drug Policy 21(3):202–207. DOI: 10.1016/j.drugpo.2009.10.010. PMID: 20005692. (Clinicians presented with identical vignettes differing only in whether the individual was described as a "substance abuser" or "a person with a substance use disorder"; the former was associated with greater attribution of personal responsibility and greater endorsement of punitive rather than therapeutic measures.)
National Institute on Drug Abuse. Words Matter — Terms to Use and Avoid When Talking About Addiction. NIDAMED guidance for health professionals. (Person-first language guidance issued as a clinical recommendation rather than an etiquette preference, on the basis that language influences provider perception and care.)
Berridge, K.C., Robinson, T.E. (2016). "Liking, wanting, and the incentive-sensitization theory of addiction." American Psychologist 71(8):670–679. DOI: 10.1037/amp0000059. (The brain circuitry mediating "wanting" a reward is dissociable from that mediating how much it is "liked." Incentive salience — "wanting" — is generated by large, robust systems including mesolimbic dopamine; "liking," the actual pleasurable impact of consumption, is mediated by smaller, more fragile systems and is not dopamine-dependent. Neuroimaging studies report dopamine changes correlating more with subjective wanting than with liking. Appears out of numerical sequence relative to first mention — normalise reference order by hand at copyedit.)
Robinson, T.E., Berridge, K.C. (2025). "The Incentive-Sensitization Theory of Addiction 30 Years On." Annual Review of Psychology. DOI: 10.1146/annurev-psych-011624-024031. (Addresses the widely cited D2 receptor finding directly: reduced D2 receptor binding on PET imaging is often presented as evidence of a dopamine deficiency causing addiction, but the downregulation may instead occur later in addiction as a consequence of heavy drug use. Cited here specifically to correct the direction of causation implied by popular "burnt-out receptor" accounts. Appears out of numerical sequence — normalise by hand at copyedit.)
Adverse childhood experiences and substance use disorders. Felitti, V.J., Anda, R.F., Nordenberg, D., et al. (1998), "Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study," American Journal of Preventive Medicine 14(4):245–258, DOI: 10.1016/S0749-3797(98)00017-8 — the original study establishing a graded relationship between childhood adversity and later health outcomes including substance use. Subsequent literature consistently reports dose-response associations between ACE count and substance use disorders. Observational; association does not establish causation in any individual, and many people with high ACE scores never develop a substance use disorder. Modern synthesis: Zhu, J., Racine, N., Devereux, C., Hodgins, D.C., Madigan, S., "Associations between adverse childhood experiences and substance use: A meta-analysis," Child Abuse & Neglect, DOI: 10.1016/j.chiabu.2023.106431, PMID: 37689565 — 102 studies, 901,864 participants, reporting an overall association between ACE exposure and substance-use outcomes.
Contingency management. The ASAM/AAAP Clinical Practice Guideline on the Management of Stimulant Use Disorder, Journal of Addiction Medicine (2024) 18(1S):1–56, DOI: 10.1097/ADM.0000000000001299, PMID: 38669101, PMCID: PMC11105801 — identifies contingency management as the current standard of care for stimulant use disorder. Meta-analytic evidence: Pfund, R.A., Ginley, M.K., Boness, C.L., et al. (2024), "Contingency Management for Drug Use Disorders: Meta-Analysis and Application of Tolin's Criteria," Clinical Psychology: Science and Practice 31(2):136–150, DOI: 10.1037/cps0000121, PMID: 38863566, PMCID: PMC11164545. Foundational trial: Petry, N.M., Peirce, J.M., Stitzer, M.L., et al. (2005), "Effect of prize-based incentives on outcomes in stimulant abusers in outpatient psychosocial treatment programs," Archives of General Psychiatry 62(10):1148–1156, DOI: 10.1001/archpsyc.62.10.1148. (Effects are strongest during active delivery; evidence on persistence after incentives end is mixed.)
Naloxone and overdose reversal. Naloxone is a μ-opioid receptor antagonist that temporarily reverses opioid toxicity and life-threatening respiratory depression, used in hospital settings for over fifty years; community take-home naloxone programmes began in the 1990s. See Moustaqim-Barrette, A., Dhillon, D., Ng, J., et al. (2021), "Take-home naloxone programs for suspected opioid overdose in community settings: a scoping umbrella review," BMC Public Health 21:597, DOI: 10.1186/s12889-021-10497-2, PMID: 33771150, PMCID: PMC8004425, and Fischer, L.S., Asher, A., Stein, R., Becasen, J., Doreson, A., Mermin, J., Meltzer, M.I., Edlin, B.R. (2025), "Effectiveness of naloxone distribution in community settings to reduce opioid overdose deaths among people who use drugs: a systematic review and meta-analysis," BMC Public Health 25:1135, DOI: 10.1186/s12889-025-22210-8, PMID: 40133970, PMCID: PMC11934755. (Most underlying evidence is observational; the umbrella review notes outstanding questions on dosing, route and access.)
Mackey, K.M., Beech, E.H., Williams, B.E., Anderson, J.K., Young, S., Parr, N.J. (2023). Effectiveness of Syringe Services Programs: A Systematic Review. Washington, DC: Department of Veterans Affairs, Evidence Synthesis Program. PMID: 38166137. (Synthesis of 17 systematic reviews and 100 primary studies. SSP use likely lowers HIV transmission and injection-risk behaviour, may lower hepatitis C transmission and facilitate treatment linkage, and does not appear to increase injection frequency, unsafe syringe disposal, or neighbourhood crime. The differing confidence levels for HIV versus HCV are reflected in the body wording.)
Anthony, J.C., Warner, L.A., Kessler, R.C. (1994). "Comparative epidemiology of dependence on tobacco, alcohol, controlled substances, and inhalants: Basic findings from the National Comorbidity Survey." Experimental and Clinical Psychopharmacology 2(3):244–268. DOI: 10.1037/1064-1297.2.3.244. (Americans aged 15–54. Approximately one third of tobacco smokers developed tobacco dependence; approximately 15% of drinkers became alcohol dependent; approximately 15% of users of other controlled substances became dependent. Important limitations: these use DSM-III-R "dependence" criteria rather than current DSM-5 substance use disorder criteria; the data are from 1994 and substance availability, potency and patterns of use have changed since; and the survey grouped controlled substances rather than reporting a separate capture rate for each, so this article does not present per-drug figures. Cross-sectional and retrospective self-report.)

Sourcing note: this article cites peer-reviewed studies, diagnostic manuals, clinical practice guidelines and national surveillance data. Geographic limitation, stated plainly: the recovery, capture-rate and treatment-gap figures come primarily from US national data, while several treatment-access and stigma findings are also US-based, because that is where the largest datasets exist. They are labelled as US throughout. Treatment availability, drug markets and healthcare access differ substantially between countries, and readers elsewhere should treat the proportions as illustrative rather than local. Diagnostic definitions are drawn from both DSM-5-TR and ICD-11, where the international picture is strongest. Where a widely repeated figure differs between clinical sources and commercial ones — as with delirium tremens mortality — the clinical figure is used and the discrepancy is stated in the body. Where a scientific question is unresolved, as with sugar and with most behavioural addictions, the disagreement is presented rather than settled. Commercial interest is disclosed in both directions, including where a finding favours an industry position. This article makes no treatment recommendation and provides no dosing, tapering or protocol information by editorial policy. Cross-link targets (#) resolve once the site structure is live. Reading time is measured, not asserted: body words excluding the reference list and the metadata header, counted at 225 words per minute and rounded to the nearest minute.