Gambling addiction: definition, DSM-5 criteria and how it is diagnosed

Gambling addiction is a recognized mental health condition in which a person loses control over their gambling and keeps gambling despite the harm it causes. Its clinical name is gambling disorder, and it is diagnosed when at least four of nine specific criteria appear within the same 12 months. Both diagnostic manuals used worldwide place it alongside substance addictions: the DSM-5, since 2013, and the WHO's ICD-11, under code 6C50.

Nothing in that definition refers to how often you gamble or how much you lose. The condition is defined by impaired control and by consequences, which is why a person who bets twice a month can meet the criteria while someone who bets every day does not.

Gambling addiction, gambling disorder, compulsive gambling: which term means what

Five terms circulate for roughly the same territory, and only two of them are diagnostic. Which one a source uses tells you how narrow its claim is.

Population surveys add bands such as "low risk" and "moderate risk" gambling. Those describe scores on a screening questionnaire, not diagnoses.

What changed in 2013

DSM-5 moved gambling out of the impulse-control disorders section and into the expanded chapter on substance-related and addictive disorders, making it the first non-substance behavioral addiction the manual recognized (Rash, Weinstock and Van Patten, 2016). The evidence behind the move was that the reward circuitry and behavior patterns resemble substance addiction rather than impulse-control problems.

Two things changed in the criteria themselves at the same time: the diagnostic threshold dropped from five to four, and the criterion about committing illegal acts to finance gambling was deleted, leaving nine criteria instead of ten (Stinchfield et al., 2016). Anything you read that lists ten criteria or a cut-off of five is describing DSM-IV.

The nine DSM-5 criteria and the four-of-nine threshold

DSM-5 diagnoses gambling disorder when four or more of the following occur in a 12-month period and cause clinically significant impairment or distress. The person:

1. Needs to gamble with increasing amounts of money to get the excitement they want.
2. Is restless or irritable when trying to cut down or stop.
3. Has made repeated unsuccessful efforts to control, cut back, or stop gambling.
4. Is often preoccupied with gambling: reliving past sessions, planning the next one, working out how to get money to gamble with.
5. Often gambles when feeling distressed, such as helpless, guilty, anxious or depressed.
6. After losing money, often returns another day to get even, known as chasing losses.
7. Lies to conceal how much they are gambling.
8. Has jeopardized or lost a significant relationship, job, or educational or career opportunity because of gambling.
9. Relies on others for money to relieve desperate financial situations caused by gambling.

One further condition applies: the gambling must not be better explained by a manic episode. That exclusion exists because mania can produce spending sprees and risk-taking that resolve when the mood episode is treated.

The number of criteria met sets the severity, and the manual attaches two further descriptions to the diagnosis.

Two features of that list catch people out. Three of the nine criteria (2, 3 and 5) say nothing about money; they describe mood and failed attempts to stop. And the threshold is lower than the public image of the condition: someone who chases losses, hides the extent of it, has twice failed to cut back, and gambles when low is already at four.

ICD-11 reaches the same condition through three requirements rather than a checklist: impaired control over gambling; gambling given increasing priority over other interests and daily activities; and continuation or escalation despite negative consequences, normally evident over about 12 months (WHO, 2024). It also splits the diagnosis by setting, with 6C50.1 covering predominantly online gambling.

How gambling addiction differs from gambling a lot

Frequency and stake size are not diagnostic. A person can gamble weekly for years, keep to a set amount, stop when it is gone, and meet none of the nine criteria. Another person can gamble twice a month, borrow to do it, lie about it, and meet five.

The dividing line is control and cost. Three specific things are worth checking against your own behavior, because they separate heavy gambling from disordered gambling more reliably than any figure in a bank statement:

• What happens when you try to stop. Deciding not to gamble this week and then not gambling is different from deciding and gambling anyway. Repeated failed attempts are criterion 3.
• Where the money comes from. Gambling with money set aside for it differs from gambling with rent money and borrowing to replace it. Criterion 9 covers relying on others to dig you out.
• What you tell people. Concealment is criterion 7, and one of the strongest everyday markers, because it means you already know the amount would not survive being said out loud.

A few things gambling addiction is not, since each mistaken belief keeps people from getting help. It is not primarily a financial problem: paying off the debts does not resolve it, which the National Council on Problem Gambling puts first in its own FAQ. It is not classified as an impulse-control disorder any more, and has not been since 2013. It is not the same as gaming disorder, a separate ICD-11 diagnosis about video games. And it is not confined to casinos: the diagnostic criteria are indifferent to the product, whether that is slots, sports betting, lottery, or trading-style apps.

How common gambling addiction is

Those three numbers are not directly comparable, and the organizations publishing them say so. The Gambling Commission warns that its push-to-web survey typically returns higher scores than traditional face-to-face surveys, and that the results should not be lined up against earlier British surveys to claim a trend. The NCPG states plainly that NGAGE is not a prevalence study. The global meta-analysis pooled whatever instrument each country happened to use, which is why its confidence interval is wide.

Read together, they put roughly one to three adults in a hundred at the problem end of the scale in countries with legal gambling markets, with several times that number experiencing harm below any diagnostic threshold. More detail is in our gambling addiction statistics page.

Who is most at risk

Risk is not evenly spread. NGAGE 3.0 identified the strongest correlates in its US sample as taking part in several gambling activities rather than one, gambling weekly or more often, believing gambling is a way to make money, sports betting, gambling online, and being male or under 35. The global meta-analysis found problematic gambling rates were highest among people who gamble on online casino games and slots, which are the fastest products in continuous play.

WHO adds the life circumstances that raise vulnerability: separation, retirement, injury, bereavement, and social stressors including poverty and discrimination. None of these causes gambling disorder on its own. The mechanisms are covered in what causes gambling addiction.

How gambling addiction is screened and diagnosed

Screening and diagnosis are different acts. A screen is a short questionnaire that sorts people into "worth a proper assessment" and "probably not". A diagnosis is made by a clinician who takes a history, applies the nine criteria, and rules out other explanations such as a manic episode. No online test diagnoses anything, including ours. Two screens do most of the practical work:

• The Problem Gambling Severity Index (PGSI) — nine items scored 0 to 3, total range 0 to 27. The Gambling Commission reads 0 as no reported symptoms, 1–2 as low risk, 3–7 as moderate risk, and 8 or more as problem gambling. It is the instrument behind most national prevalence figures, so a PGSI score is comparable with published statistics in a way that a generic quiz is not.
• The Lie/Bet screen — two questions, about needing to bet more and more money, and about lying to people who matter regarding how much you gamble. Validation studies found very high sensitivity and lower specificity, meaning it rarely misses someone who meets full criteria but does flag people who do not (Johnson et al., 1998). A "yes" is a reason to look further, not a verdict.

You can work through the PGSI on our gambling addiction test, which scores it and explains what each band means. If you are reading someone else's behavior rather than your own, the observable markers are in signs of gambling addiction.

What to do with the answer

Gambling disorder is treatable. The evidence-based options are described in gambling addiction treatment, and the practical first steps people take, such as blocking access and handing over card control, are in how to stop gambling.

If any part of this page described your own situation, a free national helpline is the shortest path to someone who deals with this daily. Numbers and chat services by country are in our helplines directory.