Young People’s Substance Use: From Diagnosis to Developmental Trajectory

I have argued for some time that young people’s substance use is often better understood as a trajectory than as a fixed category. The key question is not simply, “what drug is the young person using?” or “how often are they using it?” Those questions matter, but they are not enough. The more useful question is: what kind of developmental pathway is this substance use sitting inside?

In broad terms, I have suggested that young people’s substance use often falls into three overlapping but clinically distinct trajectories: internalised, externalised, and normative.

In the internalised trajectory, substance use is closely linked to distress. The young person may be using to manage anxiety, low mood, trauma, shame, loneliness, social fear or emotional overwhelm. The substance has a function: it soothes, numbs, quietens, helps them feel less exposed, or gives temporary relief from an inner state that feels hard to tolerate.

In the externalised trajectory, substance use is more closely linked to impulsivity, risk-taking, conduct problems, anger, peer conflict, school exclusion, sensation-seeking, or difficulty with boundaries and consequences. Here, substance use may serve a different function. It may be about stimulation, status, defiance, belonging, thrill, performance, or acting out. The drug is not just relieving distress; it is part of a wider pattern of dysregulated action.

In the normative trajectory, use is not necessarily organised around pathology. It may be linked to curiosity, peer culture, identity formation, social experimentation, weekends, parties, music, festivals, or the ordinary developmental push toward independence. This does not make it risk-free, but it does mean the practitioner should be careful not to over-clinicalise every episode of adolescent substance use.

This three-trajectory model matters because different pathways require different responses. Internalised use may need emotional regulation, safety, relational trust and work on the distress beneath the behaviour. Externalised use may need structure, boundaries, impulse control, alternative status, family, school or system work, and risk containment. Normative use may need credible harm reduction, developmental guidance and proportionate intervention rather than unnecessary pathologising.

What is interesting is that this kind of thinking now sits within a much wider scientific movement.

A major recent Nature study has added weight to the idea that mental health conditions are not always as separate as our diagnostic systems make them appear. Grotzinger et al. (2025) analysed genetic data across 14 psychiatric disorders and found that much of the genetic risk could be organised into five broad genomic factors, rather than neat disorder-by-disorder boxes. These factors explained around 66% of the genetic variance on average and were associated with 238 pleiotropic loci — genetic regions linked to more than one condition.

The five genetic groupings included internalising disorders such as depression, anxiety and PTSD; substance use disorders; neurodevelopmental conditions such as ADHD and autism; compulsive conditions such as OCD, anorexia and Tourette’s; and a schizophrenia/bipolar grouping. The study also found that the schizophrenia/bipolar and internalising factors showed high levels of polygenic overlap and very few disorder-specific loci, suggesting that some conditions traditionally treated as separate may share deeper biological organisation (Grotzinger et al., 2025).

The relevance here is not that genetics can explain a young person’s substance use in any simple or deterministic way. That would be a mistake. The important point is broader: the science is moving away from rigid categories and toward shared underlying systems.

This fits with what many practitioners already see. Young people rarely arrive with neat, single-problem presentations. Substance use may appear alongside anxiety, trauma, school exclusion, ADHD traits, family conflict, peer risk, self-harm, low mood, exploitation, impulsivity or social isolation. Traditional systems often separate these into different service pathways or diagnostic labels. But the lived presentation is usually more integrated than that.

The wider research literature has long recognised the importance of internalising and externalising pathways in young people’s development. Externalising patterns are commonly associated with earlier and higher-risk substance use, particularly where conduct problems, impulsivity and risk-taking are stable over time. Internalising patterns are more complex: anxiety, depression and distress may inhibit substance use in some young people, while in others they may increase the likelihood of self-medication, coping-motivated use or later escalation. Augustyn et al. (2021), for example, found that elevated externalising behaviours were associated with higher levels of substance use through age 18, while internalising showed a more complicated relationship.

That complexity is exactly why the three-trajectory model is useful. It does not reduce young people to one cause. It allows us to ask whether the substance use is mainly functioning as relief, action, or developmental experimentation.

Put simply:

Internalised use asks: what is the young person trying not to feel?

Externalised use asks: what pattern of action, risk or social identity is the use part of?Normative use asks: what developmental, peer or identity process is this use expressing?

These questions help prevent two common errors. The first is treating all youth substance use as evidence of disorder. The second is treating all youth substance use as harmless experimentation. Both positions are too blunt. Some young people are experimenting and need proportionate harm reduction. Some are showing early signs of distress-organised use. Others are moving into high-risk externalising patterns where consequences can accumulate quickly.

The Nature genetics paper does not prove this model. It is not a substance-use intervention study, and it does not tell practitioners how to work with an individual young person. But it does support the broader conceptual direction: presentations that look different on the surface may share deeper organising processes. Diagnosis, behaviour and substance choice are visible outputs. Beneath them may sit common systems of threat, reward, impulse, mood, cognition, development, social belonging and regulation.

This is also consistent with dimensional approaches such as the Hierarchical Taxonomy of Psychopathology, or HiTOP. HiTOP has been developed as a dimensional alternative to traditional diagnostic systems, organising psychopathology across broader spectra and lower-level symptom components rather than relying only on categorical diagnoses (Cicero et al., 2024). It is particularly relevant here because internalising and externalising spectra sit close to the centre of this wider transdiagnostic movement.

For young people’s substance use, the practical implication is clear. We should not begin with the drug alone. We should begin with the pathway.

A young person using cannabis to manage panic and sleep is not on the same trajectory as a young person using ketamine in a peer group organised around thrill, status and risk. A young person drinking occasionally at parties is not the same as a young person drinking alone to cope with shame or trauma. The substances may overlap, but the developmental meaning is different.

This is where trajectory-based formulation becomes more clinically useful than category-based description. It lets us hold several things at once: the substance, the function, the developmental stage, the emotional system, the peer context, the family context and the direction of travel.

The direction of travel is crucial. A trajectory is not a label. A young person can move between pathways. Normative experimentation can become externalised escalation. Internal distress can become risk-taking. Externalised behaviour can conceal depression, shame or trauma. The model is therefore not about fixing young people into boxes. It is about identifying the dominant organising pattern at this point in time, so that intervention can be better matched.

That may be the most important message from both the clinical model and the wider research: complexity does not mean randomness. What looks like a messy combination of symptoms, behaviours and diagnoses may still have a pattern. The task for practitioners is to find that pattern early enough to respond.

In that sense, the three-trajectory model offers a simple but powerful framework:

Is this young person’s substance use primarily organised by internalised distress, externalised risk, or normative developmental experimentation?

Once we can answer that, we are in a much better position to decide what kind of help is needed, how urgent the risk is, and what kind of intervention is likely to make sense to the young person themselves.

References

Augustyn, M.B., Thornberry, T.P. and Krohn, M.D. (2021) ‘The joint development of externalizing and internalizing behaviours in childhood and adolescence and substance use in early adulthood’, Journal of Child & Adolescent Substance Abuse, 30(4), pp. 251–266.

Cicero, D.C., Ruggero, C.J., Balling, C.E., Bottera, A.R., Cheli, S., Elkrief, L., Forbush, K.T., Hopwood, C.J., Jonas, K.G., Jutras-Aswad, D., Kotov, R. and others (2024) ‘State of the science: The Hierarchical Taxonomy of Psychopathology (HiTOP)’, Behavior Therapy, 55(6), pp. 1114–1129.

Grotzinger, A.D. et al. (2025) ‘Mapping the genetic landscape across 14 psychiatric disorders’, Nature. Available online ahead of print.

Kotov, R., Krueger, R.F., Watson, D., Achenbach, T.M., Althoff, R.R., Bagby, R.M., Brown, T.A., Carpenter, W.T., Caspi, A., Clark, L.A. and others (2017) ‘The Hierarchical Taxonomy of Psychopathology (HiTOP): A dimensional alternative to traditional nosologies’, Journal of Abnormal Psychology, 126(4), pp. 454–477.

Lee, P.H., Anttila, V., Won, H., Feng, Y.C.A., Rosenthal, J., Zhu, Z., Tucker-Drob, E.M., Nivard, M.G., Grotzinger, A.D., Posthuma, D. and others (2021) ‘Genomic relationships, novel loci, and pleiotropic mechanisms across eight psychiatric disorders’, Cell, 184(6), pp. 1549–1563.

Krueger, R.F., Hobbs, K.A., Conway, C.C., Dick, D.M., Dretsch, M.N., Eaton, N.R., Forbes, M.K., Forbush, K.T., Keyes, K.M., Latzman, R.D. and others (2021) ‘Validity and utility of Hierarchical Taxonomy of Psychopathology (HiTOP): I. Psychosis superspectrum’, World Psychiatry, 20(2), pp. 151–172.