Key Takeaways
- 3 million people die each year from harmful use of alcohol globally, equivalent to 5.3% of all deaths
- 70.2% of adults aged 18+ in the U.S. reported having used alcohol at some point in their lives (2019), showing the scale of exposure relevant to alcohol use disorder
- 9.1% of U.S. adults aged 18+ had an alcohol use disorder in 2019 (about 23.3 million people), based on NSDUH estimates
- USD 39.6 billion U.S. spending on substance use disorder treatment in 2020 (includes mental health and substance use disorder services categories in national health expenditure accounting)
- USD 2.7 billion U.S. federal funding for substance use disorder and related activities in FY2023 (HHS and related agencies appropriations and grants reported for SUD-related programs)
- USD 6.7 billion U.S. revenue opportunity for medications for opioid use disorder by 2030 (forecast from market research on pharmacotherapies), reflecting spending growth potential
- 68% of people who met criteria for substance use disorder in the U.S. in 2019 did not receive any specialty treatment in the past year (barrier and treatment gap metric)
- Less than 30% of adults with substance use disorder in the U.S. received specialty treatment in 2020 (treatment penetration figure from SAMHSA NSDUH analyses)
- Naltrexone extended-release reduces relapse rates in alcohol use disorder; in a key randomized trial, relapse occurred in 25% with naltrexone vs 45% with placebo (56-day window used in trial reporting)
- 53% of Americans report that they have ever experienced addiction-related problems in their household or with a close family member/friend (survey-based metric in major public opinion polls), reflecting social awareness
- In 2021, 74% of opioid treatment programs reported offering take-home naloxone kits or related overdose education (survey-based operational metric cited by SAMHSA)
- In 2022, 55% of substance use disorder treatment facilities reported using electronic health records (EHR) (facility technology adoption metric from SAMHSA or national health IT surveys)
- The U.S. economic cost of substance abuse was estimated at USD 442 billion in 2017 (SAMHSA/NSDUH-based cost accounting synthesis used widely in later policy summaries)
- USD 1.5 billion in workplace costs attributed to substance abuse in 2019 (reported in insurer and workforce analyses focusing on absenteeism and healthcare)
- In the U.S., indirect costs (lost productivity) account for 73% of the total costs of opioid misuse according to a national economic estimate report
Millions die and many go untreated, yet proven medicines and harm reduction can save lives.
Related reading
01 · Category
Prevalence & Burden6 stats
Prevalence & Burden Interpretation
02 · Category
Market Size6 stats
Market Size Interpretation
03 · Category
Treatment & Outcomes6 stats
Treatment & Outcomes Interpretation
04 · Category
Industry Trends4 stats
Industry Trends Interpretation
05 · Category
Cost Analysis8 stats
Cost Analysis Interpretation
More related reading
06 · Category
Epidemiology5 stats
Epidemiology Interpretation
07 · Category
Treatment Access3 stats
Treatment Access Interpretation
08 · Category
Outcomes & Effectiveness7 stats
Outcomes & Effectiveness Interpretation
09 · Category
Market & Spend6 stats
Market & Spend Interpretation
10 · Category
Policy & Risk4 stats
Policy & Risk Interpretation
Addiction burden: exposure vs treatment gap
Large-scale exposure to addictive substances is paired with a substantial share of people who do not receive specialty treatment.
Cite This Report
This report is designed to be cited. We maintain stable URLs and versioned verification dates. Copy the format appropriate for your publication below.
Aisha Okonkwo. (2026, February 13). Addiction Statistics. Gitnux. https://gitnux.org/addiction-statistics
Aisha Okonkwo. "Addiction Statistics." Gitnux, 13 Feb 2026, https://gitnux.org/addiction-statistics.
Aisha Okonkwo. 2026. "Addiction Statistics." Gitnux. https://gitnux.org/addiction-statistics.
Sources & references
55 datasets cited across this report · attribution is report-level
+29 additional datasets cited (not shown individually)

