Self-Medication of ADHD Symptoms: Does Caffeine Have a Role?

Abstract

Objective: Stimulants are the most effective treatment for Attention Deficit/ Hyperactivity disorder (ADHD). In addition, studies have shown that nicotine dependence in patients with ADHD is probably best explained by self-medication. The question is whether this is also true for caffeine use and caffeine dependence. The aim of our study was, therefore, to examine the relationship of ADHD symptoms, caffeine consumption, caffeine use disorder (CUD) and well-being. We hypothesized that those who have more ADHD symptoms and regularly consume caffeine have higher psychological well-being than those who have more ADHD symptoms, but do not consume caffeine.

Methods: A general population sample (N = 2,259, 70.5% male, mean age 34.0) filled out the 10-item Caffeine Use Disorder Questionnaire (CUDQ), the Adult ADHD Self-report Scale (ASRS) and the WHO-5 Well-Being Index (WHO-5) and were asked about their caffeine consumption habits in an online survey.

Results: There were no associations between ADHD and coffee, tea, energy drink or cola consumption or daily caffeine consumption. However, the results of the path analysis showed that the level of ADHD symptoms was positively associated with the level of CUD (β = 0.350) and negatively with the WHO-5 (β = -0.259).

Conclusions: Caffeine consumption was not associated with ADHD symptom severity and thus not likely to represent self-medication. On the contrary, caffeine use disorder severity is associated with more ADHD symptoms and both caffeine use disorder and ADHD are associated with lower well-being.

Keywords: ADHD; caffeine; caffeine use disorder; self-medication; well-being.

Figure 1

Caffeine use disorder symptom z-scores in the four ADHD groups. ASRS, Attention-deficit/hyperactivity disorder symptoms; CUD, Caffeine Use Disorder symptoms.

Figure 2

Path analysis for the association of ADHD symptoms, caffeine consumption, caffeine use disorder symptoms (CUD) and well-being (N = 2,196). Unstandardized regression coefficients, their standard errors (in brackets) and standardized coefficients (in parentheses) are presented in the figure. Only the significant (p < 0.05) direct paths are presented ^***p < 0.001.

Figure 3

Path analysis for the association of ADHD symptoms, coffee, tea, energy drink and cola consumption, caffeine use disorder symptoms (CUD) and well-being. Unstandardized regression coefficients, their standard errors (in brackets) and standardized coefficients (in parentheses) are presented in the figure. Only the significant (p < 0.05) direct paths and error covariances are presented ^*p < 0.05; ^**p < 0.01; ^***p < 0.001.