Where Teens Get Misused Prescription Drugs Shapes Their Risk Profile
A nationally representative study of US high school seniors identifies five distinct patterns of prescription drug diversion among adolescent nonmedical users, revealing that the source of misused medications is strongly associated with sex, motives, and the depth of co-occurring substance use involvement.
Why This Matters
Nonmedical use of prescription drugs among adolescents remains a persistent public health concern, with roughly one in nine US high school seniors reporting past-year misuse of opioids, stimulants, or anxiolytics. Prevention efforts have historically treated all nonmedical users as a single group, but the channels through which teenagers obtain these medications vary widely and may signal very different levels of risk. Understanding those source patterns is essential for targeting interventions, from medication disposal campaigns aimed at lower-risk users to more intensive substance use screening for teens who purchase drugs or access them through multiple channels.
Clinical Summary
Nonmedical use of prescription drugs (NMUPD) in adolescents encompasses a range of behaviors, from a teenager taking a leftover antibiotic-era opioid from their own medicine cabinet to actively purchasing stimulants from peers. Ford and colleagues, publishing in Addictive Behaviors (2018), applied design-based latent class analysis to pooled 2009 through 2016 data from the Monitoring the Future (MTF) study to identify subgroups of high school seniors defined by how they obtained misused prescription opioids, stimulants, and anxiolytics. The rationale is straightforward: if source patterns cluster in meaningful ways, they may serve as practical indicators of misuse severity and co-occurring risk.
Among the 1,917 past-year nonmedical users in the analytic sample, the analysis identified five source classes: friend or relative sources (17.5%), friend or purchased sources (14.6%), own leftover prescription (14.3%), multiple sources (44.2%), and other sources (9.4%). Boys were significantly more likely to fall into the friend or purchased class, while girls were overrepresented in the own-leftover-prescription class. Multiple-source users demonstrated the highest rates of binge drinking, cigarette smoking, and marijuana use, while own-leftover users showed the lowest. Critically, all findings are cross-sectional, meaning the study cannot determine whether source type drives broader substance involvement or simply reflects it. The authors acknowledge that longitudinal research is needed before source-based risk stratification can inform clinical or policy recommendations.
Dr. Caplan’s Take
This study does something genuinely useful by moving past the oversimplified question of whether a teenager misuses prescription drugs and asking how they obtain them. The five-class model feels clinically intuitive: a teen who raids their own leftover hydrocodone from a dental procedure is in a different place than one who is buying benzodiazepines from multiple contacts. What the study cannot tell us, and what I need to be honest about when patients or parents ask, is whether the source pattern actually predicts where someone is headed. We are looking at a snapshot, not a trajectory.
In practice, I ask about sources when screening for prescription drug misuse in adolescents. A teen who reports using their own leftover medication gets a conversation about safe disposal and a closer look at whether their prescriber is overprescribing. A teen who is purchasing drugs or accessing them through multiple channels gets a more intensive substance use assessment and, often, a referral. This study supports that clinical instinct, even if it does not yet prove it is the right algorithm.
Clinical Perspective
This study sits at an important juncture in the adolescent NMUPD literature. It confirms what prior work has suggested, that nonmedical users are not a monolith, and extends those findings by using a person-centered analytic approach on nationally representative data. The identification of a large multiple-source class (44.2%) is particularly striking and suggests that nearly half of teen nonmedical users have diversified their access channels, a pattern associated with substantially higher co-occurring substance use. However, the cross-sectional design means clinicians should not assume that restricting a particular source will reduce downstream risk. The evidence supports using source inquiry as a screening signal, not as a causal lever.
From a pharmacological and safety standpoint, the three drug classes examined, opioids, stimulants, and anxiolytics, carry distinct toxicity profiles and interaction risks, particularly when combined with alcohol and marijuana use as seen in the multiple-source class. Clinicians working with adolescents should recognize that a teen who endorses obtaining prescription drugs from several channels may be polydrug-involved and warrants comprehensive screening beyond the index substance. One immediately actionable step: integrate a brief source-of-drug question into existing adolescent substance use screening protocols such as the CRAFFT, as this information appears to add meaningful risk stratification at minimal clinical cost.
Study at a Glance
- Study Type
- Cross-sectional epidemiological study using design-based latent class analysis
- Population
- US high school seniors (modal age 17 to 18 years), nationally representative
- Intervention
- Not applicable (observational design)
- Comparator
- Five latent source classes compared on sociodemographic and substance use variables
- Primary Outcomes
- Source class membership; associations with sex, race, motives, binge drinking, cigarette smoking, and marijuana use
- Sample Size
- N=18,549 total respondents; n=1,917 past-year nonmedical prescription drug users
- Journal
- Addictive Behaviors
- Year
- 2018
- DOI or PMID
- Data from Monitoring the Future (MTF) study, pooled 2009 through 2016 cohorts
- Funding Source
- National Institute on Drug Abuse (NIDA)
What Kind of Evidence Is This
This is an original cross-sectional epidemiological study using pooled nationally representative survey data and latent class analysis. Cross-sectional studies occupy a moderate position in the evidence hierarchy: they can identify associations and characterize population subgroups but cannot establish temporal ordering or causation. The single most important inference constraint is that the study cannot determine whether source type precedes, follows from, or merely co-occurs with differences in broader substance use involvement.
How This Fits With the Broader Literature
Prior work on adolescent NMUPD sources, including earlier MTF analyses by McCabe and colleagues, has typically examined individual sources in isolation, finding that purchasing and peer-mediated diversion are associated with greater substance use involvement. The present study extends that literature by using a person-centered approach that acknowledges teens often obtain drugs from multiple channels simultaneously, a reality that variable-centered analyses obscure. The finding that own-leftover-prescription users represent a lower-risk subgroup aligns with research distinguishing self-treatment motives from recreational motives in adolescent NMUPD. However, no longitudinal study has yet tested whether source-based subgroups follow distinct risk trajectories over time, which limits the translational reach of these findings.
Common Misreadings
The most likely overinterpretation is to conclude that the source of misused prescription drugs causes differences in substance use risk, and therefore that blocking specific sources (such as eliminating leftover prescriptions through medication disposal programs) will reduce overall misuse severity. The data do not support this causal chain. Source class membership and substance use involvement were measured simultaneously, so it is equally plausible that teens with greater substance use involvement seek out more sources, rather than that having more sources escalates their use. The public health recommendations in the original paper, while reasonable, extend modestly beyond what the cross-sectional design alone can justify.
Bottom Line
This study demonstrates that adolescent nonmedical prescription drug users are not a uniform group and that source patterns meaningfully distinguish subpopulations with different risk profiles. It does not establish that source type causes risk differences, and longitudinal confirmation is needed. For clinicians now, asking how a teenager obtains misused medications provides a practical, low-cost signal that can inform the intensity of further assessment and intervention.
References
- Ford JA, Pomykacz C, Veliz P, McCabe SE, Boyd CJ. Sports involvement, injury history, and non-medical use of prescription opioids among college students: an analysis with a national sample. Am J Addict. 2018;27(1):15-22.
- McCabe SE, West BT, Boyd CJ. Sources of nonmedical prescription drug misuse among US high school seniors: differences in motives and substance use. J Am Acad Child Adolesc Psychiatry. 2013;52(2):154-162.
- McCabe SE, West BT, Teter CJ, Boyd CJ. Trends in medical use, diversion, and nonmedical use of prescription medications among college students from 2003 to 2013: connecting the dots. Addict Behav. 2014;39(7):1176-1182.
- Johnston LD, Miech RA, O’Malley PM, Bachman JG, Schulenberg JE, Patrick ME. Monitoring the Future national survey results on drug use, 1975-2016: Overview, key findings on adolescent drug use. Ann Arbor: Institute for Social Research, The University of Michigan; 2017.
- Ford JA, Pomykacz C, Szalewski A, McCabe SE, Boyd CJ. Friends, family, and pills: the source of nonmedical prescription drug misuse among US high school seniors. Addict Behav. 2018;87:220-226.
Have thoughts on this? Share it:
