Not everyone who tries drugs becomes addicted, and the percentage varies dramatically depending on the substance. Across all drug types combined, roughly 1 in 10 people aged 12 or older in the United States met the criteria for a drug use disorder in 2024, based on the national survey conducted by SAMHSA. But that broad number hides enormous differences between substances. Heroin hooks about a third of new users within the first year, while prescription opioids lead to addiction in fewer than 3% of patients who take them for pain.
The Overall Numbers
In 2024, 48.4 million Americans aged 12 or older had a substance use disorder of some kind, including both drugs and alcohol. That’s about 16.8% of the population in that age range. Of those, 28.2 million had a drug use disorder specifically, representing 9.8% of people 12 and older. That rate has been climbing: it was 8.7% in 2021.
These numbers count anyone who met clinical criteria for a substance use disorder in the past year, ranging from mild to severe. Severity matters. Among people with a marijuana use disorder, only about 1 in 5 had a severe form. Among people with a stimulant use disorder tied to cocaine or methamphetamine, nearly half (48.3%) had a severe disorder. Opioid use disorders fell in between, with 37.1% classified as severe.
Heroin and Opioids
Heroin carries one of the highest addiction risks of any substance. A Michigan State University study found that roughly 30% of new heroin users become dependent within their first year of use, higher than the previous estimates of 20 to 25%. In some recent years, the rate climbed to nearly 50 to 60%, depending on the population studied. The speed at which heroin creates dependence is part of what makes it so dangerous: there is no long “experimental” phase for many users.
Prescription opioids tell a different story. When taken as prescribed for chronic pain, the addiction rate is likely below 3%. One large review of 67 studies found a mean addiction rate of 3.27% among chronic non-cancer pain patients on opioid therapy, while another review found rates as low as 0.7% and as high as 34.1%, a huge range driven by differences in how studies defined “addiction” and who they studied. A separate analysis found that opioid abuse occurred in about 2.1% of patients prescribed opioids, and in nearly half of those cases, the driving motivation was seeking pain relief rather than getting high.
The gap between heroin and prescription opioids is partly about dose, route of administration (injecting or snorting vs. swallowing a pill), and the speed at which the drug reaches the brain. It’s also about context. People taking prescribed medication under medical supervision are in a fundamentally different situation than people using heroin recreationally.
Cannabis
Cannabis has a lower addiction risk than heroin or stimulants, but it’s far from zero. Among people who have ever used cannabis, the lifetime probability of developing a cannabis use disorder is approximately 27%, based on a large nationally representative study. That means roughly 1 in 4 people who try marijuana will meet the diagnostic criteria for a use disorder at some point in their lives.
Most of those cases are mild. The 2024 national survey confirmed that the majority of people with a marijuana use disorder had a mild form, and only about 1 in 5 had a severe disorder. Still, a 27% lifetime transition rate is higher than many people expect for a substance often perceived as low-risk.
Stimulants
Cocaine and methamphetamine stand out for how severe the resulting disorders tend to be. Among people with a stimulant use disorder in 2024, 48.3% had a severe disorder, the highest severity rate of any drug category in the national survey. Only 28.5% had a mild disorder. This means that when stimulants do cause problems, those problems are more likely to be serious compared to alcohol or cannabis.
Why the Numbers Vary So Much
Addiction risk isn’t just about the drug. It’s shaped by genetics, mental health, age of first use, and environment. Someone who starts using a substance as a teenager faces a higher risk than someone who first tries it at 30. People with untreated depression, anxiety, or trauma are more vulnerable. A family history of addiction increases susceptibility.
The way a drug is used also changes the math. Smoking or injecting a substance delivers it to the brain faster than swallowing it, which produces a more intense high and a stronger pull toward repeated use. Frequency matters too. Someone who uses a substance daily is on a different trajectory than someone who uses it a few times a year.
Definitions also explain some of the wide ranges in the research. Studies that measure “misuse” (any use outside of instructions) will always produce higher numbers than studies measuring “addiction” (compulsive use despite serious consequences). When you see a study claiming 80% of opioid patients develop problems and another claiming 1%, they’re often measuring entirely different things.
What These Numbers Actually Mean
The simplest honest answer is that somewhere between 10 and 30% of people who use drugs will develop a use disorder, depending heavily on which drug and how it’s used. The risk is lowest for prescribed medications taken as directed (under 3% for opioids) and highest for heroin (30% or more in the first year alone). Cannabis falls in the middle, with about 27% of lifetime users eventually meeting the criteria for a disorder, though most cases are mild.
These percentages also mean that the majority of people who try most substances do not become addicted. That fact doesn’t minimize the millions who do develop disorders, but it’s important context. Addiction is not an inevitable outcome of drug use. It’s a probabilistic one, shaped by biology, circumstance, and the specific substance involved.