Discussion and Conclusion
During the COVID-19 lockdown, people self-reported drinking slightly more alcohol compared to before the lockdown, which was negatively associated with age and positively associated with the number of children, unemployment, and homework. The top motive for drinking more was conviviality. The number of cigarettes smoked per day only marginally increased during the lockdown, again with younger respondents having increased odds of smoking more during the lockdown. Similarly, living alone, having a lower educational background, and being technically unemployed increased the odds of smoking more cigarettes during the lockdown as well. The top motive for smoking more cigarettes was boredom.
Cannabis use did not change in times of social isolation. This may indicate a true similarity in the point estimates before and during the lockdown, but it may also reflect a zero net-effect of increased need and reduced availability of cannabis in times of social lockdown (borders are closed, dealers have a harder time to go around unnoticed, and prices increase due to increased demand). Similar to smoking, the top for an increased use of cannabis was boredom. The similarities between ranked motives of cannabis and cigarette smoking are most probably an artefact because most cannabis users are also cigarette smokers. This explains why the ranking of motives is almost the same for smoking cigarettes and cannabis use, and is the result of the fact that no separate questions about motives were asked for the 3 separate substances. It might be an interesting future research question to investigate what drives certain individuals to consume more, less, or equal during a pandemic, and individual assessment of different motives for use, per substance, would here be valuable. Also, focussing on more positive factors might reveal now underreported factors for better resilience and consequently more adequate coping mechanisms than substance use.
This study sample is not optimally representative of the overall Belgian population with 50% men, less urban people, an average lower education, an estimate of 12% active in healthcare, and different jobs (possibly with higher technical unemployment rates). In this survey, we mainly note highly educated females (70%), with 36.8% being healthcare workers. We found no effect of gender and living in a city on changes in substance use and no effect of educational background for changes in alcohol use. However, our data show that non-healthcare workers were 40% more likely to consume more alcohol compared to healthcare workers during the lockdown. The respondents from this survey included more healthcare workers as would be expected from a true representative sample, which may infer that, in a true representative population, the differences in alcohol use may be larger than presented here. This needs to be confirmed in larger surveys.
This study has limitations, including the very brief time during which this web-based survey was made accessible (20 days), which yielded a relatively small sample size (N = 3,632), while being representative for a uniform period of full lockdown measures. In addition, we did not follow-up on the self-reported substance use after this 3-week period, and use may stabilize or increase further with increased duration of lockdown. Another limitation was that we only assessed alcohol, cigarettes, and cannabis and the mean use of alcohol per day, rather than per week, such that individuals consuming only a few drinks per week may have over- or underreported their drinking. This was counterbalanced by the possibility to add “drinking more, or drinking less than usual, or about the same” in the next question regarding their change in alcohol use.
As a whole, these data show that specific groups of individuals, especially younger individuals with lower educational background or those technically unemployed, have a higher risk of increasing their alcohol and/or cigarette use during times of social isolation. It is therefore of importance that care providers and especially general practitioners discuss the use of alcohol, tobacco, and cannabis with their patients, as this often has a beneficial effect on behaviour.