This article covers clinically sensitive territory. If substance use has become a primary way of managing the stress of building a business, support is available without judgement. Alcoholics Anonymous can be reached on 0800 9177 650. FRANK provides confidential drugs advice on 0300 123 6600.

Substance use as affect regulation, not moral failure

Edward Khantzian’s self-medication hypothesis, developed at Harvard through the 1980s and 1990s, reframed how clinicians understood substance use in high-achieving, high-stress populations. The argument is that people do not primarily use substances for pleasure or because of weak character. They use them because specific substances match specific emotional states they are trying to manage — and the match is pharmacologically coherent.

Someone experiencing anxiety and hypervigilance reaches for alcohol because alcohol is a GABA agonist that produces temporary anxiolysis. Someone experiencing depletion and cognitive fatigue reaches for caffeine because caffeine blocks adenosine receptors that accumulate sleepiness signals. Someone experiencing low mood, low energy, or difficulty concentrating reaches for stimulants because stimulants increase dopamine and norepinephrine in the prefrontal cortex, producing temporary improvements in mood, energy, and focus.

These are not random choices. They are implicit pharmacological decisions about which states are most unbearable and which substances most efficiently address them.

A review of longitudinal data by Bolton and colleagues, covering over 43,000 participants, found that mood and anxiety disorders consistently preceded substance use disorders temporally — supporting the self-medication causal direction. The substance use is downstream of the psychological state, not its cause. In a sample of 2,435 self-employed professionals, elevated psychological distress was directly associated with increased alcohol and cigarette use.

Freeman and colleagues’ 2019 entrepreneur mental health study found that 12% of entrepreneurs reported substance use disorders compared to their comparison group. Given the self-report limitations of that study, the true figure is likely higher.

Why specific substances are specifically appealing to entrepreneurs

The three substances named in the title map onto the three dominant stress states of entrepreneurship with enough specificity to warrant explaining separately.

Alcohol addresses the anxiety and hypervigilance that chronic uncertainty produces. The entrepreneurial stress response keeps the amygdala activated and the HPA axis (body’s stress response system) running — a sustained threat-detection state that alcohol temporarily counters through its sedative and anxiolytic effects. The evening drink after a high-demand day is pharmacologically rational as an anxiety regulation tool. That is precisely what makes it difficult to recognise as self-medication.

Caffeine addresses the depletion state. Sustained cognitive effort produces adenosine accumulation — the brain’s fatigue signal. Caffeine blocks those receptors, restoring the felt sense of alertness and focus. A 2025 meta-analysis of eight studies covering 546 participants found that caffeine intake above 400mg daily significantly increased anxiety risk. Moderate doses may alleviate depressive symptoms; higher doses produce the anxiety they were initially recruited to manage. The dose-response relationship is an inverted U, and most heavy caffeine users are operating somewhere past the peak.

Stimulants — prescription or otherwise — address the executive function demands of building a business: focus, decision-making, sustained attention under pressure. The cognitive gains at low-to-moderate doses are real. The prefrontal cortex’s dopamine and norepinephrine systems genuinely improve in function at optimal doses. The problem is that optimal doses are narrow, the inverted-U relationship means performance degrades past a threshold that is not easily visible to the user, and tolerance develops through the same mechanism that applies to every dopamine-mediated substance.

How functional use becomes something else

The escalation trajectory follows a predictable pattern across all three substance categories. When a substance produces relief or enhancement, the brain’s homeostatic mechanisms downregulate the receptors it acts on — requiring higher doses to produce the same effect. What began as one coffee becomes six. What began as a drink to wind down becomes the amount required to feel normal.

Lembke’s tolerance model, applied to any dopamine-mediated substance, describes this precisely: the baseline shifts downward as tolerance develops, such that the absence of the substance now produces a negative state rather than simply the absence of a positive one. At this point, the substance is no longer primarily enhancing performance or providing relief. It is preventing a withdrawal state that the original baseline did not include.

The recognition test the research supports is practical: if removing the substance produces a recognisable negative state — caffeine headaches, alcohol-related irritability and sleep disruption, stimulant-related inability to focus — tolerance has developed and mild dependency is already present. Withdrawal symptoms are the evidence, not the dose level.

The entrepreneurial context is particularly vulnerable to this trajectory for three reasons: the stressors driving initial use are persistent rather than episodic; the professional culture normalises high use of performance substances; and the functional framing — “it helps me work” — is accurate enough in the short term to prevent recognition of the pattern forming.

If the pattern described here is recognisable in your own relationship with a substance — and particularly if the absence of it now produces a negative state — speaking with a GP or addiction counsellor is the appropriate next step. The self-medication dynamic is workable, but addressing it properly requires professional support rather than willpower applied to a tolerance-based dependency.

A book worth reading alongside this

In the Realm of Hungry Ghosts by Gabor Maté provides the most compassionate and rigorous account of why people become dependent on substances that initially served a genuine regulatory function. Written from his clinical experience working with severely addicted patients, but grounded in the same self-medication hypothesis this article describes, his core argument — that addiction is a response to pain rather than a moral failure — is the normalisation the topic requires. For any entrepreneur who recognises a self-medication pattern in their own behaviour, it is the most honest and least judgemental starting point available.

This article discusses psychological patterns documented in research on substance use, affect regulation, and entrepreneurial wellbeing. It is not designed to identify, diagnose, or assess any substance use disorder or psychological condition, and it is not a substitute for professional support. If substance use is significantly affecting your work, relationships, or health, speaking with your GP or a specialist addiction counsellor is the appropriate next step. UK resources: Alcoholics Anonymous (0800 9177 650). FRANK (0300 123 6600). Samaritans (116 123, free, 24/7).

This article is for educational and informational purposes only. It is not a substitute for professional psychological advice, diagnosis, or treatment. If you are experiencing significant psychological distress, please consult a qualified mental health professional.

Sources: Khantzian, E.J. (1997), Harvard Review of Psychiatry, 4(5). Bolton, J.M. et al. (2009), Journal of Clinical Psychiatry, 70(9), N=43,093. Leignel, S. et al. (2014), International Archives of Occupational and Environmental Health, 87(7), N=2,435. Freeman, M.A. et al. (2019), Small Business Economics, 53(2). Frontiers in Neuroscience (2025), cognitive stimulants meta-analysis, N=546.