Introduction
The proliferation of synthetic opioids in illicit drug markets has fueled high overdose rates and triggered HIV and hepatitis C virus outbreaks among people who inject drugs (PWID) across the US1. Recently, overdose deaths involving psychostimulants—particularly methamphetamine—have increased sharply, often in combination with opioids2. Stimulant injection is of particular concern given its association with more frequent injection and syringe sharing, and elevated risks of HIV and hepatitis C virus3. There are limited data on national trends in injecting methamphetamine among PWID. We examined trends in injecting methamphetamine and opioids among PWID entering drug treatment in the US.
Methods
We conducted a serial cross-sectional analysis of the 2012–2023 Treatment Episode Data Set-Admissions (TEDS-A) of drug treatment episodes at facilities receiving state/public funding in the US, collected by the Substance Abuse and Mental Health Services Administration (eMethods). We included first-encounter treatment admissions among persons aged 18 years or older who reported injection drug use in 40 states with data reported each year (eFigure;eTable). Each admission record indicated up to three substances used at admission and their respective routes of use. To characterize temporal trends in substances injected among first-encounter PWID admissions, we examined the annual prevalence of reporting (1) injection of methamphetamine but no opioid injection and (2) injection of both methamphetamine and opioids. To assess changes between 2012 and 2023, we estimated prevalence differences (PD) and 95% confidence intervals (95%CI) using binomial regression. Subgroup analyses were stratified by demographic characteristics. Data were analyzed using R, v4.4.1. This study of public de-identified data was deemed exempt from review by the Johns Hopkins University School of Medicine Institutional Review Board and followed the STROBE reporting guidelines.
Results
There were 831,571 first-encounter PWID treatment admissions between 2012 and 2023, rising from 58,643 admissions in 2012 to 111,048 in 2018 and declining to 33,071 by 2023. The prevalence of injecting methamphetamine but not opioids significantly increased from 14.3% (n=8,367) in 2012 to 31.4% (n=10,389) in 2023 (PD=17.1 [95%CI=16.6–17.7]); the prevalence of injecting both methamphetamine and opioids significantly increased from 4.1% (n=2,397) in 2012 to 12.2% (n=4,029) in 2023 (PD=8.1 [95%CI=7.7–8.5]) (Figure;Table). Significant increases in both outcomes were observed in nearly all age, sex, and race-ethnicity groups and Census regions/divisions examined (Table). Notably, the prevalence of injecting methamphetamine but not opioids increased by over 25 percentage points among those aged 18–24 years and among those in the East North Central and East South Central divisions.
Figure. Trends in the prevalence of injecting methamphetamine and opioids among first-encounter drug treatment admissions of people who inject drugs in the United States, Treatment Episodes Data Set-Admissions, 2012–2023.
Note: Error bars indicate 95% confidence intervals. Percentage of PWID entering treatment adds to 100% across the 4 groups.
Table.
Changes in the prevalence of injecting methamphetamine and opioids among first-encounter drug treatment admissions of people who inject drugs in the United States, Treatment Episodes Data Set-Admissions, 2012–2023.
| Outcome: Injected methamphetamine but not opioids | Outcome: Injected methamphetamine and opioids | |||||
|---|---|---|---|---|---|---|
| 2012 | 2023 | 2023 vs. 2012 | 2012 | 2023 | 2023 vs. 2012 | |
| n/N (%) | n/N (%) | PD (95% CI) | n/N (%) | n/N (%) | PD (95% CI) | |
| Overall | 8367/58643 (14.3) | 10389/33071 (31.4) | 17.1 (16.6, 17.7) | 2397/58643 (4.1) | 4029/33071 (12.2) | 8.1 (7.7, 8.5) |
| Age group (years) | ||||||
| 18–24 | 1692/16965 (10.0) | 820/1867 (43.9) | 33.9 (31.7, 36.2) | 766/16965 (4.5) | 289/1867 (15.5) | 11.0 (9.3, 12.6) |
| 25–29 | 1886/14707 (12.8) | 1629/4755 (34.3) | 21.4 (20.0, 22.9) | 623/14707 (4.2) | 723/4755 (15.2) | 11.0 (9.9, 12.0) |
| 30–39 | 2832/15657 (18.1) | 4376/14603 (30.0) | 11.9 (10.9, 12.8) | 689/15657 (4.4) | 1961/14603 (13.4) | 9.0 (8.4, 9.7) |
| 40–49 | 1471/6997 (21.0) | 2580/8019 (32.2) | 11.2 (9.8, 12.5) | 239/6997 (3.4) | 815/8019 (10.2) | 6.7 (6.0, 7.5) |
| 50–64 | 480/4156 (11.5) | 943/3416 (27.6) | 16.1 (14.3, 17.8) | 78/4156 (1.9) | 230/3416 (6.7) | 4.9 (3.9, 5.8) |
| 65+ | 6/161 (3.7) | 41/411 (10.0) | 6.2 (2.1, 10.4) | 2/161 (1.2) | 11/411 (2.7) | 1.4 (-0.9, 3.7) |
| Sexa | ||||||
| Male | 4445/35284 (12.6) | 6373/20903 (30.5) | 17.9 (17.2, 18.6) | 1344/35284 (3.8) | 2639/20903 (12.6) | 8.8 (8.3, 9.3) |
| Female | 3919/23346 (16.8) | 4004/12138 (33.0) | 16.2 (15.2, 17.2) | 1053/23346 (4.5) | 1383/12138 (11.4) | 6.9 (6.3, 7.5) |
| Race-Ethnicitya | ||||||
| Non-Hispanic White | 6875/46136 (14.9) | 7655/24302 (31.5) | 16.6 (15.9, 17.3) | 1960/46136 (4.2) | 3145/24302 (12.9) | 8.7 (8.2, 9.2) |
| Non-Hispanic Black | 145/2639 (5.5) | 479/1657 (28.9) | 23.4 (21.1, 25.8) | 26/2639 (1.0) | 89/1657 (5.4) | 4.4 (3.2, 5.5) |
| Non-Hispanic Otherb | 495/2061 (24.0) | 820/1948 (42.1) | 18.1 (15.2, 20.9) | 131/2061 (6.4) | 288/1948 (14.8) | 8.4 (6.5, 10.3) |
| Hispanic (any race) | 790/6967 (11.3) | 1261/4506 (28.0) | 16.6 (15.1, 18.2) | 263/6967 (3.8) | 417/4506 (9.3) | 5.5 (4.5, 6.4) |
| Census region and divisionc | ||||||
| Northeast | 75/14171 (0.5) | 508/7190 (7.1) | 6.5 (5.9, 7.1) | 32/14171 (0.2) | 429/7190 (6.0) | 5.7 (5.2, 6.3) |
| New England | 11/5108 (0.2) | 87/2640 (3.3) | 3.1 (2.4, 3.8) | 13/5108 (0.3) | 141/2640 (5.3) | 5.1 (4.2, 6.0) |
| Middle Atlantic | 64/9063 (0.7) | 421/4550 (9.3) | 8.5 (7.7, 9.4) | 19/9063 (0.2) | 288/4550 (6.3) | 6.1 (5.4, 6.8) |
| Midwest | 2171/10375 (20.9) | 2447/6022 (40.6) | 19.7 (18.2, 21.2) | 395/10375 (3.8) | 953/6022 (15.8) | 12.0 (11.0, 13.0) |
| East North Central | 152/6341 (2.4) | 1193/3716 (32.1) | 29.7 (28.2, 31.3) | 53/6341 (0.8) | 618/3716 (16.6) | 15.8 (14.6, 17.0) |
| West North Central | 2019/4034 (50.0) | 1254/2306 (54.4) | 4.3 (1.8, 6.9) | 342/4034 (8.5) | 335/2306 (14.5) | 6.0 (4.4, 7.7) |
| South | 2640/19814 (13.3) | 4895/13460 (36.4) | 23.0 (22.1, 24.0) | 697/19814 (3.5) | 1645/13460 (12.2) | 8.7 (8.1, 9.3) |
| West South Central | 1537/4291 (35.8) | 2311/3819 (60.5) | 24.7 (22.6, 26.8) | 228/4291 (5.3) | 406/3819 (10.6) | 5.3 (4.1, 6.5) |
| East South Central | 402/3562 (11.3) | 1342/3310 (40.5) | 29.3 (27.3, 31.2) | 194/3562 (5.4) | 775/3310 (23.4) | 18.0 (16.3, 19.6) |
| South Atlantic | 701/11961 (5.9) | 1242/6331 (19.6) | 13.8 (12.7, 14.8) | 275/11961 (2.3) | 464/6331 (7.3) | 5.0 (4.3, 5.7) |
| West | 3481/14283 (24.4) | 2539/6399 (39.7) | 15.3 (22.1, 24.0) | 1273/14283 (8.9) | 1002/6399 (15.7) | 6.7 (8.1, 9.3) |
| Mountain | 1519/5081 (29.9) | 1145/2484 (46.1) | 16.2 (13.9, 18.5) | 482/5081 (9.5) | 460/2484 (18.5) | 9.0 (7.3, 10.8) |
| Pacific | 1962/9202 (21.3) | 1394/3915 (35.6) | 14.3 (12.6, 16.0) | 791/9202 (8.6) | 542/3915 (13.8) | 5.2 (4.0, 6.5) |
Abbreviations: N, sample size; n, number of people with outcome; PD, prevalence difference
Missing data on sex and race-ethnicity not shown due to small cell sizes (sex: <0.1% in 2012, 0.1% in 2023; race-ethnicity: 1.4% in 2012, 2.0% in 2023).
Non-Hispanic Other includes admissions reporting ethnicity as Not Hispanic or Latino and race as Alaska Native, American Indian or Alaska Native, Asian or Pacific Islander, Asian, Two or more races, or Native Hawaiian or Other Pacific Islander.
States included in the analysis by division include: New England (ME, NH, VT, MA, RI, CT); Middle Atlantic (NY, NJ, PA); East North Central (OH, IN, IL, MI); West North Central (MN, IA, MO, SD, KS); West South Central (AR, LA, OK, TX); East South Central (KY, TN, AL, MS); South Atlantic (DC, VA, NC, GA, FL); Mountain (MT, WY, CO, AZ, UT, NV); Pacific (CA, AK, HI).
Discussion
During 2012–2023, the prevalence of injecting methamphetamine only or both methamphetamine and opioids increased significantly among PWID entering treatment, including in areas where methamphetamine injection was previously rare (e.g., the Northeast). These findings build on prior work showing increases in opioid and methamphetamine co-use (regardless of route) among PWID before the COVID-19 pandemic4. Several social and structural factors, including increased availability and lower costs of methamphetamine relative to other substances, may be contributing to rising trends, particularly among marginalized populations5.
From 2018–2023, as the proportion of treatment-seeking PWID reporting methamphetamine injection increased, the absolute number of PWID admissions overall in TEDS-A decreased. This decline may reflect COVID-19-related challenges in data reporting (e.g., limited staffing and resources), access to treatment (e.g., program closures), and engagement in treatment. It may also be related to a concurrent shift from injection to non-injection substance use routes6.
Study limitations include potential temporal changes in data reporting as well as the exclusion of PWID not in treatment or treated outside publicly funded facilities required to report data to TEDS-A, which varies by state. TEDS-A also does not capture whether individuals injected substances simultaneously. Despite these limitations, the study sample serves as a sentinel population for monitoring drug use trends.
The increases in methamphetamine injection with and without opioid injection among PWID seeking treatment underscore the need for developing stimulant use disorder treatment strategies, improving the utilization of medications for opioid use disorder, and expanding harm reduction initiatives (e.g., syringe service programs) to mitigate the risk of injection-related harms.
Supplementary Material
Funding/Support:
This was supported in part from the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK131926 [AART]).
Role of the Funder/Sponsor:
The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Footnotes
Conflict of Interest Disclosures: The authors declare no potential conflicts of interest.
REFERENCES
- 1.Hodder SL, Feinberg J, Strathdee SA, et al. The opioid crisis and HIV in the USA: deadly synergies. The Lancet. 2021;397(10279):1139–1150 [DOI] [PubMed] [Google Scholar]
- 2.Garnett MF, Miniño AM. Drug Overdose Deaths in the United States, 2003–2023. Hyattsville, MD: National Center for Health Statistics. 2024. doi: 10.15620/cdc/170565 [DOI] [Google Scholar]
- 3.Tavitian-Exley I, Vickerman P, Bastos FI, Boily M. Influence of different drugs on HIV risk in people who inject: systematic review and meta-analysis. Addiction. 2015;110(4):572–584 [DOI] [PubMed] [Google Scholar]
- 4.Reid MC, Guthrie BL, Hajat A, Glick SN. National trends in co-use of opioids and methamphetamine among people who inject drugs, 2012–2018. Drug Alcohol Depend. 2025;271:112630. doi: 10.1016/j.drugalcdep.2025.112630 [DOI] [PubMed] [Google Scholar]
- 5.Assaf RD, Morris MD, Straus ER, Martinez P, Philbin MM, Kushel M. Illicit Substance Use and Treatment Access Among Adults Experiencing Homelessness. JAMA. 2025;333(14):1222–1231. doi: 10.1001/jama.2024.27922 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tanz LJ, Gladden RM, Dinwiddie AT, et al. Routes of Drug Use Among Drug Overdose Deaths - United States, 2020–2022. MMWR Morb Mortal Wkly Rep. 2024;73(6):124–130. doi: 10.15585/mmwr.mm7306a2 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.