CBD Safety Guide
CBD Side Effects: What Does Research Report?
A source-led route through CBD adverse-event, sedation, liver, and medication-interaction evidence without treating every study finding as universal.
The short answer
What should you know first?
CBD is often described as non-intoxicating, but that does not mean every CBD product has no safety questions. Side-effect reports depend on the population, product, dose, route, other medicines, and outcome being measured.
How to read CBD safety findings
Three questions to ask before applying a reported side effect
Key distinction
Reported event versus cause
A reported event in a study is not always proof that CBD alone caused it; the design, comparator, and other treatments matter.
Key distinction
Population matters
Findings from prescription-CBD studies in people with complex conditions may not describe every consumer product or population.
Key distinction
Safety is connected
Side-effect questions overlap with medication interactions, dose, formulation, food effects, and product quality.
What studies reported
Results worth understanding
These findings belong to the named compound or product, dose, route, population, and outcome. They are not one result for every cannabinoid product or person. Open the PubMed links to inspect the original records.
Randomized human trial
Diarrhea and somnolence increased at the higher dose
In tuberous sclerosis complex, diarrhea occurred in 56% and somnolence in 26% at 50 mg/kg per day, compared with 25% and 9% on placebo. PubMed 33346789
Randomized human trial
More adverse events occurred in Dravet syndrome
A 120-person trial reported diarrhea, vomiting, fatigue, fever, somnolence, and abnormal liver tests more often with CBD, with more withdrawals than placebo. PubMed 28538134
Randomized human trial
Liver elevations and discontinuations were recorded
In Lennox-Gastaut syndrome, 14 CBD participants had elevated liver aminotransferases and seven stopped trial medication because of adverse events. PubMed 29768152
Systematic review
Six trials showed a broader safety pattern
A meta-analysis of 1,034 participants found increases in total and serious adverse events, treatment abandonment, and transaminase elevation with CBD versus placebo. PubMed 36417631
Research context
Read the evidence in context
CBD has a documented adverse-event profile
Non-intoxicating does not mean free of side effects. Controlled prescription-CBD trials report diarrhea, somnolence, decreased appetite, fatigue, vomiting, fever, liver-enzyme elevations, and treatment discontinuation in some participants. The pattern is clearest in closely monitored seizure trials using standardized oral formulations and weight-based doses, so the event names are informative while their exact rates remain product-, dose-, and population-specific.
The higher dose sometimes produced more adverse events
In the tuberous sclerosis complex trial, diarrhea occurred in 31% at 25 mg/kg per day, 56% at 50 mg/kg per day, and 25% with placebo. Somnolence occurred in 13%, 26%, and 9%, respectively. Ten people in the higher-dose group, eight in the lower-dose group, and two receiving placebo discontinued because of adverse events. This pattern supports attention to dose without creating a universal consumer dose-response rule.
Seriousness and frequency are not the same question
A side effect can be common but mild, uncommon but important, or lead someone to stop a study. A six-trial systematic review found that CBD increased total adverse events, serious adverse events, treatment withdrawal, and transaminase elevations compared with placebo in the seizure populations studied. Those categories answer different safety questions and should not be collapsed into one label such as 'well tolerated.'
Cause, context, and product identity still matter
Participants in many CBD trials have complex conditions and take other medicines. A reported event may be caused by CBD, another medicine, an interaction, the underlying condition, or a combination. Placebo comparisons, timing, dose patterns, laboratory changes, and dechallenge information can strengthen interpretation, but a list of events alone cannot establish cause. Consumer products add another layer because labeled and measured composition may differ.
Important limits
What can make the answer change?
- 1
Do not call CBD risk-free because it does not cause the same intoxicating high as THC.
- 2
Do not use an adverse-event frequency from one population or product as a universal rate.
- 3
Do not ignore new symptoms, co-medications, or product identity when reading safety information.
Common questions
Questions people ask
What side effects has CBD research reported?
Controlled studies report diarrhea, somnolence, appetite changes, fatigue, vomiting, fever, liver-enzyme elevations, and discontinuation in some participants. PubMed 28538134 PubMed 29768152 PubMed 33346789
Are CBD side effects dose-related?
Some trials reported more events at higher doses, including diarrhea and somnolence. That pattern does not establish one rate or threshold across all products and populations. PubMed 33346789 PubMed 29768152
Is CBD risk-free because it is non-intoxicating?
No. Intoxication and adverse effects are different questions. CBD has a documented safety and interaction record even though it does not produce THC's characteristic intoxication. PubMed 31288397 PubMed 36417631
Do seizure-trial rates apply to store-bought CBD?
Not directly. Those trials used standardized prescription formulations, weight-based doses, defined populations, other medicines, and clinical monitoring. PubMed 28538134 PubMed 33346789