N=1: Where Research on Alcohol and Drugs Becomes Human Again

Worldwide, extensive research is conducted on alcohol and drug use. In most existing alcohol and drug studies, the focus lies on averages.

We look for patterns, correlations, and significances within groups of people. We compare these groups, calculate probability distributions, and draw conclusions that apply to populations. Yet somewhere in those numbers of conventional quantitative research, something gets lost: the individual. The unique way in which one person gives meaning to their substance use, change, or recovery is precisely where the strength of n=1 research lies. But anyone searching scientific databases for alcohol or drug research combined with “n=1” or “single case” will find remarkably few results.

N=1 research (or case study) does not focus on averages but on individual patterns and dynamics.

It studies one person, in detail, over time. That may sound limited, but it offers something that large-scale research by definition cannot: insight into how and why behavior change works, or fails to work, for an individual. Use and recovery are rarely linear processes (1). They fluctuate, respond to context, relationships, mood, hope, and shame. By following one person over time, using for instance diary data, interviews, or physiological measures, we gain insight into the microprocesses that underlie patterns of use. Not in abstract terms, but as the real story.

Still, n=1 research is often dismissed as “not representative.”

In a culture where “evidence-based” has become almost synonymous with “statistically significant,” the single case can appear methodologically suspicious. Yet in alcohol and drug research, paradoxically, the n=1 is often the most relevant unit of analysis. Every patient, every user, every recovery trajectory is unique, in background, motivation, pace, and context. Every therapist will recognize this. Group averages may not be able to capture such complexity at all. What works for one person (e.g., bias modification approaches, medication, or self-help) may have the opposite effect for another. The value of n=1 research lies exactly there: in understanding the exception rather than explaining the rule.

N=1 research is not necessarily a loose case description but a carefully designed study (2). Think of time-series analyses with repeated measurements, visual inspection of trends, qualitative triangulation, or mixed-methods approaches. Such studies can reveal patterns that may later be tested in larger samples (3). In that sense, they are not an alternative to group-based research but its foundation. Moreover, n=1 research creates space for participation and co-creation. The person being studied is not merely a respondent but a co-researcher, someone who reflects, interprets, and gives meaning. This makes the research not only scientifically richer but also clinically and ethically more relevant. In addiction care, where stigma, distrust, and power imbalances can play a role, this approach may help foster reciprocity and respect.

Why, then, is this type of research still so marginal?

Partly for practical reasons: it takes time and yields “little” data in the conventional sense. But also because our research institutions, funding systems, and journals are largely built around the logic of the masses. What cannot be generalized seems unimportant. And precisely because of that, we lose the nuance that policy and practice urgently need. The current shift toward personalized care and lived-experience perspectives calls for a renewed appreciation of the unique (4).

N=1 research reminds us that every data point was once a person, a human being with a history, choices, desires, and failures. In a field that too often talks about people in terms of “target groups”, “addicts,” or “patients,” n=1 research brings back some humanity to science. If we truly want to understand what substance use means, how change unfolds, and what makes recovery possible, then we should not collect more averages but learn to listen more deeply to the story told in the singular.

This blog was written by Daan Hulsmans, Radboud University Nijmegen, for RAD-blog, the blog about smoking, alcohol, drugs, and diet.

References

1. Witkiewitz, K., & Marlatt, G.A. (2007). Modeling the complexity of post-treatment drinking: It’s a rocky road to relapse. Clinical Psychology Review, 27(6), 742-738. doi: http://dx.doi.org/10.1016/j.cpr.2007.01.002
2. Kazdin, A.E. (2019). Single-case experimental designs. Evaluating interventions in research and clinical practice. Behaviour Research and Therapy, 117, 3-17. doi: https://doi.org/10.1016/j.brat.2018.11.015
3. Hekler, E.B., Klasnja, P., Chevance, G., Golaszewski, N.M., Lewis, D., & Sim, I. (2019). Why we need a small data paradgim. BMC Medicine, 17(133). doi: https://doi.org/10.1186/s12916-019-1366-x
4. Flyvbjerg, B. (2006). Five Misunderstandings About Case-Study Research. Qualitative Inquiry, 12(2), 219-245. doi: https://doi.org/10.1177/1077800405284363