Introduction
In recent years, cannabidiol (CBD) is gaining growing consumer attention [
1]. The substance belongs to one of at least 130 naturally occurring phytocannabinoids that have been identified in
Cannabis sativa L. [
2]. In industrial hemp, CBD is the main cannabinoid [
3]. In contrast to tetrahydrocannabinol (THC), CBD is not intoxicating [
4] and – according to a decision of the Court of Justice of the European Union – should therefore not be considered a narcotic [
5]. Notwithstanding, CBD interacts with several molecular targets in the organism, potentially resulting in beneficial but also adverse health effects [
6].
In consequence, products containing high CBD doses clearly exerting a pharmacological activity or being intended for the treatment of diseases are considered to be medicinal products in the European Union (EU). They require authorization according to EU regulations on pharmaceuticals [
7]. The rationale behind this is to protect consumers from potential health risks that may arise from the products themselves or use of non-approved products with unproven effects instead of effective medicines.
Only low-dose CBD products without a pharmacological activity and without the intention to treat diseases may potentially considered to be foods. Such CBD-containing foods, however, are generally classified as novel in the EU [
8]. They require a safety assessment carried out by the European Food Safety Authority (EFSA), followed by authorization by the European Commission before they can be marketed in the EU [
9]. Numerous novel food applications for CBD and CBD extracts have been submitted during the last years and are partly under risk assessment [
10]. EFSA recently identified some potential hazards but was not able to fully assess the actual risks due to several data gaps. Consequently, it was concluded that “
the safety of CBD as a novel food cannot currently be established” [
11]. Therefore, no foods containing CBD have yet been authorized and the marketing of such products as foods, including supplements, is currently considered illegal. Despite these rules, consumers can buy an increasing number of illegally marketed CBD-containing products, also in the form of food supplements, that are available mostly online or in retail stores. In some cases, such products are mislabeled, for example as “aroma oils” or “cosmetics” to circumvent legal rules [
12].
As described, the regulatory context for CBD is complex and it might be put into question whether consumers are aware of both the legal aspects as well as the potential health risks. Therefore, the aim of the current study was to gain insight into the user group of CBD in Germany, to investigate reasons for consumption, and to gain information about the risk perception of the population and its knowledge about CBD-containing products in order to target risk communication and increase awareness for the substance.
Methods
Study design and participants
The study was conducted in two parts: In study part I the prevalence for CBD awareness and usage was determined while study part II examined the usage and perception of CBD for users vs. non-users in detail. The two studies consisted of independent samples. Data collection was conducted by German market research institutes. All respondents expressed their consent to participate in the surveys.
Study part I was carried out in Germany via telephone survey from November 25 to November 26, 2020 by trained market research assistants. The survey was conducted with a sample of
n = 1,011 respondents aged 14 years and older who were randomly selected via a random digit dialing procedure including mobile and landline telephone numbers. For mobile phone numbers, the person answering the call was selected for the interview. For landline telephone numbers, the Kish selection method [
13] was used to choose the respondent within a household. To achieve population representativeness, data were statistically weighted using an iterative process and a 2-step-procedure: [
14] In a first step, weighting took into account the number of mobile and landline telephone numbers respondents could be reached by (design weighting). In a second step, data were weighted based on the German microcensus regarding gender, education, age, employment, size of household and city, West-/East-Germany and German federal state (adjustment weighting). Within the questionnaire, participants were asked whether they had already heard of products containing CBD and, if so, whether they had consumed or used such products (Suppl.
A, Questionnaire Q1).
Study part II was carried out in Germany via an online panel survey from March 11 to March 23, 2021, using a quota sample. Quotas for gender, age, and education were determined based on the results of study part I. Respondents were randomly selected from the panel until the desired number of participants meeting the quota targets were obtained. An online questionnaire was developed to determine CBD-containing product consumption patterns, frequency and reasons for use, the perception of health risks and benefits, and knowledge about CBD-containing products. Respondents received remuneration in accordance with the usual incentive structure of a panel. The sample consisted of n = 2,000 respondents with an age of 16 years and older that indicated that they had already heard of products containing CBD.
Online questionnaire
The online questionnaire of study part II included questions on sociodemographic measures such as gender, age, and education, as well as use, product groups, its purchase, and frequency of use. Additional questions addressed the intention of future use and the perception of health risks and benefits. Furthermore, open questions asked participants about reasons for use and assumed health risks and benefits. Responses to these open-ended questions were coded using code frames that were developed within an inductive process. Multiple codes were used for responses that consisted of more than one aspect. The coding was cross-validated by two researchers.
All questions included “don’t know” and/or “no answer” options. For more details see the questionnaire (Suppl.
A).
Statistical analysis
Data were processed and analyzed using SPSS (Version 26). Descriptive statistics including the calculation of means (
M) and standard deviations (
SD) were used to describe socio-demographics as well as product and usage characteristics. In addition to descriptive statistics, differences were analyzed for sociodemographic groups (gender, age, education) and user status (user vs. non-user). For these analyses, we grouped respondents into three age categories (up to 29 years, 30 to 59 years, 60 years and older) as well as three educational categories (low: without graduation / student / lower secondary school; medium: secondary school without high school diploma; high: high school / university degree). Analyses for Study part I are based on a weighted sample (see “Study design and participants”). Therefore, we applied Rao-Scott adjustment [
15] for complex survey data for all tests of significance within study part I. Effect sizes were calculated using common thresholds for interpretation (Suppl.
B, Table S1).
Discussion
The results of the study provide insights into general awareness and use of CBD-containing products in the German population. 40.2% of the respondents had already heard of such products. This proportion is slightly lower than observed in other studies. In France, about 69% had already heard of CBD [
16], while a recent German study found a proportion of approximately 48% [
17]. Looking at the prevalence of users in the different studies, a relatively consistent picture emerges for European countries with about one in ten who already has used CBD-containing products among the general population (Germany: 11.4% according to our study; France: 10.1%.; UK: 10.9%) [
16,
18]. In contrast, a substantially lower usage rate of 4.3% was observed for Germany in a recently published study by Alayli et al. [
17]. This deviation may be due to a specific focus on recreational use of CBD products in this study whereas our study as well as the other European studies considered a broader product range. In contrast, the proportion of users (just within the past 12 months) was higher in a non-representative study conducted in the US (26.1%) and in Canada (16.2%) [
19]. Classification and accessibility varies strongly between these countries [
20,
21] which – in part – might explain the varying proportions of users between countries. In addition, due to the generally more liberal approach to cannabis in some parts of the US and Canada, it might also be assumed that consumer acceptance to CBD products is higher in North American Countries than in Europe.
In the present study, the age group under 30 years had the largest proportion of users. This finding is consistent with another study that observed a proportion of approximately 34% users among adults aged 18–24 years in the US [
22] pointing out that the use of CBD products is relatively prominent among young adults. These results give valuable insights with regard to relevant target groups for risk communication measures.
More than half of the users interviewed consume CBD orally via oils or tinctures, to a lesser extent via other products like foods or cosmetic and skin care products. Further, almost 13% indicated to use liquids with CBD for e-cigarettes and vaporizers. Interestingly, a study by Alayli et al. [
17] showed that awareness and consumption of recreational CBD products is generally associated with e-cigarette use. At the same time, synthetic cannabinoids in e-liquids have been associated with several health risks including neurological symptoms [
23]. The use of CBD in this form should therefore receive special attention in risk communication.
Beyond, a relatively large part of the users consume CBD products at least once a week (42.0%), indicating a rather regular than occasional usage pattern. Results from the current study and other studies clearly show that, beside curiosity, anticipated health benefits such as pain-relief, stress reduction, and improvement of sleep are important reasons for consumption [
17,
19,
22]. In the present study, main reasons for consumption significantly differed between age groups. While curiosity and stress relief were commonly cited by the youngest age group (16–29 years), pain relief and improved sleep were important motivators for the older respondents, especially those of 60 years and older.
In the present study, health benefits of CBD-containing products were mentioned more often by the participants than potential health risks, especially among users of such products. Similar results were obtained by others [
19,
18]. Indeed, scientific data show that CBD interacts with several molecular targets [
6]. Therefore, it is frequently assumed that CBD can be used as a therapeutic agent [
24]. However, only one medicinal product solely containing CBD as the active ingredient is currently authorized in the US and in the EU, intended for the treatment of certain forms of epilepsy or tuberous sclerosis [
20,
25]. Although there are first indications for therapeutic effects also for other disorders, the clinical efficacy of CBD for those indications has not yet been finally proven [
26,
27].
Health risks were considered as being low or very low by the majority of participants in the present study, especially among users. Similar observations were made in other studies [
16,
19]. However, following the motto “no effects without side effects”, various adverse effects have already been associated with CBD exposure. Results from animal studies and the use as a drug indicate potential hazards like hepatotoxicity, gastrointestinal complaints, neurological symptoms, a negative impact on the endocrine system, reproductive toxicity, as well as potential drug-drug interactions – at least at higher doses [
4,
11]. However, due to data gaps health risks cannot be finally evaluated at the time [
11]. Interestingly, the most frequently mentioned concern among the respondents, namely addiction, has no relevance from a scientific point of view, as long as products are not contaminated with THC. These results indicate that consumers are insufficiently informed about the scientific uncertainties regarding the effects of CBD. Here, targeted risk communication measures could prevent consumers from being misinformed by advertisement and non-scientific sources.
Although CBD-containing foods are currently not legal, a variety of illegally marketed products declared as foods, food supplements, or otherwise intended for oral consumptions can be purchased from several sources – especially online. Beyond potential risks of CBD, the quality of such products may vary. This is of particular note, as almost half of the respondents assume that CBD-containing products have been tested to ensure their safety. However, this is generally not the case for consumer products and foods including food supplements, since these product categories are not tightly regulated by official bodies. In addition to targeted information about potential health risks, clarifying the regulatory circumstances can enable consumers to make an informed decision about whether or not to use CBD-containing products.
Strengths and limitations
One of the study’s strength is its sample size and accounting for various socio-demographic variables. Using a two-stage study design, we were able to assess the usage rate for CBD-containing products based on a population representative sample (study part I), while focusing on more detailed information on usage and perception for specific groups of users vs. non-users within the study part II. However, since the survey for study part II was conducted via an online questionnaire and using quota sampling, generalizability of our findings is restricted to the characteristics of the recruited sample (e.g., people with low affinity for online applications or without internet access are underrepresented). Also, as participants self-reported their behavior, there might be a bias in their answers due to desirability. As this has been a cross-sectional study, the data represent a selective moment. Statements about trends are therefore not possible.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.