How to choose a CBD dose - drops, mg, effects - guide 2026

The most common question about CBD oil is: how many drops to take. This is a question without a good answer because a drop is not a unit of dosage. The same drop from a five percent bottle and a thirty percent bottle differs in content by six times, while manufacturers provide milligrams on the label. There is a second unknown: how much of the taken dose actually enters the bloodstream. A systematic review of the pharmacokinetics of cannabidiol in humans showed that this value was measured only for one route of administration (Millar, Frontiers in Pharmacology, 2018). This guide shows how to convert the content of a bottle into milligrams, what doses were actually administered in clinical trials, and what daily amount is currently considered safe by EFSA.

KEY INFORMATION
• EFSA established a provisional safe dose in 2026 of 0.0275 mg per kilogram of body weight per day, which is about 2 mg for a person weighing 70 kg (EFSA, 2026).
• The safety of cannabidiol cannot be established in individuals under 25 years of age, in pregnant and breastfeeding women, and in those taking medications.
• A drop is a volume, not a dose. Milligrams are calculated from the concentration and volume of the bottle.
• In humans, bioavailability was measured only after smoking and was 31%; for drops and capsules, no one has determined it.

Dlaczego nie ma jednej dawki CBD dla wszystkich?

Because a dose that worked in one study may not work in another, and this is not due to differences between people, but due to the substance itself. In a randomized study involving 57 healthy men, single oral doses of 150 mg, 300 mg, and 600 mg were compared with placebo. Anxiety before public speaking was reduced only by the middle dose (Linares, Revista Brasileira de Psiquiatria, 2019).

This is a result that cannot be translated into a simple rule of 'more means stronger.' The authors called it a U-shaped curve and emphasized that optimal doses must be determined separately for each application before research results are applied in practice.

The second reason is pharmacokinetic. Cannabidiol is metabolized in the liver, and its concentration in the blood depends on the form of the preparation, the presence of food, and other medications taken. A review of the pharmacokinetics and pharmacodynamics of cannabinoids describes this variability as the main reason why doctors lack data for simple dose conversions between patients (Lucas, British Journal of Clinical Pharmacology, 2018).

The third reason is the most mundane. Most people do not know how many milligrams they are taking because they count drops. As long as the unit remains a drop, no self-observation of one's body is repeatable, as it changes along with the bottle, pipette, and oil temperature. Therefore, this guide starts with arithmetic, not with a table of recommendations.

How many milligrams of CBD are in one drop of oil?

It depends solely on the concentration and volume of the bottle, and it is calculated in two steps. The percentage concentration indicates the mass of cannabidiol in 100 ml of the preparation, so a 5% oil contains 5 g in 100 ml, which is 500 mg in a 10 ml bottle. Dividing by the volume gives you milligrams per milliliter. Dividing by the number of drops in a milliliter gives you milligrams per drop.

The table below is an arithmetic conversion, not a dosage recommendation. It assumes 30 drops per milliliter, which is a conventional value for typical pipettes. The actual number of drops depends on the viscosity of the oil, temperature, and the angle of holding the bottle, so treat the result as an approximation, not a measured portion.

It shows why the instruction 'ten drops in the morning' means nothing without specifying the concentration. Ten drops of 5% oil is about 17 mg, while ten drops of 30% oil is about 100 mg. The difference is sixfold and falls within the entire range that clinical trials distinguish. If you want to compare two products, compare milligrams per milliliter, not the number of drops on the label.

How to convert any bottle into milligrams?

You need two numbers from the label: the total content of cannabidiol and the volume of the preparation. The content divided by the volume gives milligrams per milliliter, and that number divided by 30 gives an approximate content per drop. A 30 ml bottle with 1500 mg contains 50 mg per milliliter and about 1.67 mg per drop, which is exactly the same as 5% oil in a smaller package.

This example is worth remembering because it shows that the volume of the bottle and concentration are two independent pieces of information. The manufacturer can sell the same preparation in a 10 ml and a 30 ml package, describing the first as five percent and the second only providing the total mass. Without conversion, it looks like two different products.

A more accurate method than counting drops is a graduated pipette. Measuring 0.25 ml or 0.5 ml gives a repeatable result, independent of viscosity and temperature. With 10% oil, half a milliliter is 50 mg regardless of how many drops happen to come out of the pipette. For comparing your own observations from day to day, this is more significant than the precision of the conversion itself.

It is also worth checking whether the declared content is supported by the certificate of analysis of the batch. The label is a declaration of the manufacturer, while the certificate is a measurement result. If you base your dosage selection on the arithmetic from the label, its reliability is the limit of the accuracy of everything else. You can find more about the selection of concentration in the text Which CBD concentration to choose.

How do we know that the bottle contains what the label claims?

It is not known from the label itself, as the label is a declaration of the manufacturer, not a measurement result. All the arithmetic from the previous section is based on two numbers from the packaging, so its accuracy cannot be greater than the accuracy of those numbers. A verifiable confirmation is the certificate of analysis issued for a specific production batch.

The certificate is issued by an external laboratory and provides the measured content of individual cannabinoids, and usually also the result of testing for contaminants. The batch number on the document must match the number embossed on the bottle. A certificate issued for a different batch says nothing about the preparation you have in hand, even though it looks just as credible.

The purity of the product is not a trivial formal matter. The provisional safe dose set by EFSA applies only to dietary supplements in which the purity of cannabidiol is at least 98% and which do not contain nanoparticles. Outside of this scope, the panel does not comment on any value, so relating your own portion to this number requires knowledge of the product's purity.

The safety review of cannabidiol highlights the other side of the same issue. The authors point out that the impact of cannabidiol on liver enzymes and drug transporters is still insufficiently studied, and most clinical data comes from studies on epilepsy and psychotic disorders (Iffland, Cannabis and Cannabinoid Research, 2017). A product with an unknown composition introduces an additional unknown into this equation, which cannot be compensated for by precision in measurement.

How much THC can a legal CBD oil contain in Poland?

The threshold is 0.3% and is calculated as the sum of delta-9-THC and tetrahydrocannabinolic acid, i.e., THCA, based on dry mass, rounded to one decimal place. The basis is Article 4 point 5 of the Act of July 29, 2005 on counteracting drug addiction (Journal of Laws 2023 item 1939), as amended by the Act of March 24, 2022 (Journal of Laws 2022 item 763).

The method of calculation has practical significance because it changes the result of the laboratory test. Plant material mainly contains THCA, which only converts to delta-9-THC under the influence of heat. Therefore, measuring only delta-9-THC would yield a seemingly lower result, while the regulation clearly states the sum of both compounds.

The national threshold corresponds to the EU threshold, but it does not derive from it. As of January 1, 2023, cannabis varieties eligible for support under the Common Agricultural Policy may contain up to 0.3% THC based on Article 4 paragraph 4 of Regulation (EU) 2021/2115. Previously, the EU threshold was 0.2% and stemmed from Regulation 1307/2013, which was repealed on January 1, 2023. These are two separate regulations with the same numerical value.

For someone counting milligrams, the conclusion is that a full-spectrum product contains THC in an amount limited by regulation, but not zero. If you work in a profession that conducts tests for the presence of psychoactive substances, this is significant regardless of the chosen portion of cannabidiol. A separate regulation, expressed in milligrams per kilogram, applies to food from hemp seeds and does not pertain to oils.

Is full spectrum counted differently than isolate?

In terms of arithmetic, no, but in terms of reference to official values, yes. Milligrams of cannabidiol are calculated from the label the same way in every preparation. The difference only appears when you want to compare your own portion with the provisional safe dose set by EFSA, because that value has a specified composition condition.

The panel noted that the derived number applies only to dietary supplements with a purity of cannabidiol of at least 98%, without nanoparticles, produced by a process deemed safe and excluding genotoxicity. A full-spectrum product, by definition, contains other cannabinoids and terpenes, so it does not fit this description. This does not mean it is more dangerous, only that EFSA's position simply does not cover it and there is no designated reference value for it.

The same caution applies to pharmacokinetic data. The study of the effect of food was conducted on capsules containing cannabidiol with a purity of 99%, while the tolerance study was on an oral solution of purified cannabidiol. Therefore, the results of both studies describe the behavior of a single substance, not a mixture. Transferring them directly to a multi-component preparation is an assumption that neither of these studies tested.

The practical conclusion is not "choose an isolate." It is this: if you reach for a full-spectrum product, know that you are going beyond the scope for which the aforementioned numbers exist, and that your point of reference will then be the certificate of analysis and a conversation with a doctor, not a table from a guide.

Does a higher concentration of oil work stronger?

No. Fifty milligrams of cannabidiol from 5% oil and from 30% oil is the same mass of the same substance. The concentration changes the volume you need to take, not the potency. From 5% oil, it will be about one milliliter, from 30% oil about one-sixth of a milliliter. The difference concerns convenience and precision of measurement.

The practical consequence is the opposite of sales intuition. A low concentration provides greater precision with small portions because one drop is a smaller jump. With 30% oil, one drop is about 10 mg, so the smallest change that can be made with a drop is six times greater than with 5% oil. This matters when gradually adjusting portions.

Higher concentrations, on the other hand, have a volumetric advantage. Those taking large portions would have to consume several milliliters of 5% oil daily, which is impractical and costly in terms of milligrams. In that case, switching to a stronger product is rational, but not because it works differently.

The fat matrix of the product can be more significant than the concentration itself. MCT-based oils deliver cannabidiol along with fat, and the presence of fat alters its absorption in ways described further. When comparing two products with the same concentration, check the carrier as well, not just the percentage on the label.

How much CBD from a drop or a capsule reaches the bloodstream?

The honest answer is: no one has measured this. A systematic review encompassing 24 studies with pharmacokinetic data in humans established absolute bioavailability only after smoking, which was 31%. For other routes of administration, despite the availability of intravenous preparations, no study has attempted such a measurement (Millar, Frontiers in Pharmacology, 2018).

This is important because the numbers repeated in product descriptions, most often "6% orally" and "13-19% sublingually," do not come from measurements in humans. Millar's review explicitly states the lack of such data and cites it as a gap requiring filling. Any bioavailability table that presents these values as established precedes the literature.

What the review established is the half-life and time to maximum concentration. After aerosol to the oral mucosa, the half-life ranged from 1.4 to 10.9 hours, after chronic oral administration from 2 to 5 days, after intravenous administration 24 hours, and after smoking 31 hours. The maximum concentration occurred between zero and four hours after administration.

The area under the curve and maximum concentration increase with the dose, and the maximum concentration is higher after a meal and with lipid-based preparations. This means that the form of administration realistically changes the body's exposure, but the scale of this change is not expressed as a single percentage. The mechanism of loss during swallowing is described in a separate text about the first-pass effect.

Does a meal affect the absorption of CBD?

Yes, and this is one of the best-documented effects in this field. In a phase one study involving healthy volunteers, a single dose of 1500 mg taken after a high-fat meal resulted in a maximum concentration 4.85 times higher, and the area under the curve 4.2 times higher than when taken on an empty stomach. The time to maximum concentration and half-life remained unchanged (Taylor, CNS Drugs, 2018).

A second study checked the same in patients, not volunteers. Eight adults with drug-resistant epilepsy took a single dose of capsules containing cannabidiol with a purity of 99%, once on an empty stomach and once after a meal worth 840-860 kilocalories with a high-fat content. The maximum concentration was on average fourteen times higher after the meal, and the area under the curve four times (Birnbaum, Epilepsia, 2019).

The authors of this work drew a practical conclusion: the fat content in the meal explains part of the variability in exposure observed in the same patient between days. In other words, the same portion taken on an empty stomach and after lunch represents two different situations from the body's perspective.

The conclusion for daily use is simple and does not require changing the portion. Take the product under the same conditions, on an empty stomach or with a meal, but consistently. If you change this condition during observation, you do not know whether the change in feelings comes from the portion or from breakfast.

Na czym polega zasada start low, go slow?

This recommendation comes from the review of clinical pharmacology of cannabinoids, formulated directly as a response to the lack of data. The authors state that the limited availability of pharmacokinetic and pharmacodynamic information necessitates starting with a low dose and increasing it slowly, carefully observing both desired and undesired effects in the patient (Lucas, British Journal of Clinical Pharmacology, 2018).

It is worth noting where this principle comes from. It is not the result of a study that demonstrated the superiority of slow increases over rapid ones. It is a consequence of ignorance: since it is impossible to predict what concentration a specific portion will yield in a specific person, the only safe strategy is to approach it from below.

Pharmacokinetics also explains why assessment after one day makes no sense. In a phase one study, steady state with twice-daily dosing was established after about two days, and accumulation reached from 1.8 to 2.6 times. The effective half-life was estimated at 10-17 hours, and the terminal half-life at about 60 hours. Therefore, the concentration after the first dose is not the concentration at which the body ultimately operates.

Practically, this means one thing: take a break long enough between changes to assess a stable state, not a single episode. A step-by-step outline of such an investigation is detailed in a separate text about increasing the dose of CBD. Consulting a doctor before starting is a requirement, not a formality.

Jakie dawki podawano w badaniach klinicznych?

The range is much broader than guides suggest, and studies rarely involve healthy individuals taking supplements. The following summary provides only what has been realistically applied: who participated, how many people, what dose, and for how long. None of these values are recommendations for the reader, and none should be applied to oneself without a doctor.

Note the distance between these numbers and supplementation. Devinsky's study concerned severe childhood epilepsy and a registered drug, not an over-the-counter product. Taylor's study examined tolerance, not efficacy for any indication. Transferring these values to one's own plan is an abuse, even when a table in a guide does it.

Why are doses from epilepsy studies not a model for supplements?

Because they concern a different product, different people, and different supervision. In a 2017 study, 120 children and young adults with Dravet syndrome were randomly assigned to an oral cannabidiol solution of 20 mg per kilogram of body weight per day or to a placebo, in both cases as an adjunct to existing antiepileptic treatment. The median of seizure attacks decreased from 12.4 to 5.9 per month, compared to a decrease from 14.9 to 14.1 in the placebo group (Devinsky, New England Journal of Medicine, 2017).

Three elements of this description are lost when the number reaches the guide. The first is the basic treatment, which cannabidiol did not replace but only supplemented. The second is Dravet syndrome, a severe childhood epilepsy with high mortality. The third is the blinded study with a control group, in which the difference was measured against placebo, not against well-being from a month ago.

The cost of this efficacy is also not accounted for. In the group receiving cannabidiol, diarrhea, vomiting, fatigue, fever, drowsiness, and abnormal liver function tests occurred more frequently than in the placebo group. More participants withdrew from the study than from the control group. This data comes from the same work from which the figure of 20 mg per kilogram originates.

Separately, it is worth distinguishing between efficacy studies and tolerance studies. The work in which healthy volunteers took from 1500 to 6000 mg at once examined safety and pharmacokinetics, not action for any indication. The statement "such an amount was well tolerated" is not the same as "such an amount works," and in product descriptions, these two meanings blend into one.

Why does a larger dose not mean a stronger effect?

Because in the only study that directly tested this in humans, the largest dose acted like a placebo. Fifty-seven healthy men were randomly assigned to one of four groups and subjected to a simulated public speaking event. Anxiety during the speech was significantly reduced only in the middle group. The groups with the smallest and largest doses did not differ from placebo (Linares, Revista Brasileira de Psiquiatria, 2019).

The authors interpreted this arrangement as an inverted U-shaped curve and noted that it is consistent with previous observations in animals. They also pointed out a limitation that guides remain silent about: the study involved only healthy men, one indication, and one administration. It is unknown whether the shape of the curve and the position of its peak are the same in women, in sick individuals, and with chronic use.

The conclusion that the authors themselves formulate is: optimal therapeutic doses of cannabidiol must be determined rigorously for the research results to be transferable to clinical practice. This statement comes from a 2019 study, and in the following years, no study has replaced this work in this regard.

For the reader, this means changing the question. Instead of "how much to take for it to work stronger," it is more sensible to ask "how to check if an increase actually makes a difference." This is the focus of the text about adjusting the dose instead of endlessly increasing it.

What daily dose does EFSA consider safe?

The EFSA Panel on Nutrition and Novel Foods established a provisional safe dose in 2026 of 0.0275 mg per kilogram of body weight per day, which is about 2 mg daily for a person weighing 70 kg. This value was obtained using the benchmark dose method from sub-chronic studies compliant with good laboratory practice, applying an uncertainty factor of 400 (EFSA, 2026).

This number applies under narrowly defined conditions. It pertains exclusively to dietary supplements with a purity of cannabidiol of at least 98%, without nanoparticles, produced by a process deemed safe and excluding genotoxicity. Outside of this scope, the panel does not comment on any value.

The panel also states what cannot be determined. The safety of cannabidiol cannot be established for individuals under 25 years of age, for pregnant and breastfeeding women, and for those taking medications simultaneously. This is not a precautionary formula, but a conclusion from a review in which animal studies showed consistent liver toxicity, and human studies indicated liver risk, especially when combined with medications.

It is worth comparing this value with the number that circulates in product descriptions. The statement about safety up to 1500 mg daily has no backing in the document it is often referenced. Doses of this magnitude appeared in a tolerance study involving healthy volunteers over seven days, under supervision, and not as a recommendation for consumers. The difference from the EFSA position is several hundredfold and is not a matter of interpretation.

Kiedy dawkowanie CBD wymaga rozmowy z lekarzem?

Always when you are taking any medications regularly. Cannabidiol is metabolized in the liver and can inhibit the enzymes and transporters responsible for the metabolism of other substances, altering their concentration in the blood. A documented example is the inhibition of clobazam metabolism (Lucas, British Journal of Clinical Pharmacology, 2018).

The same review indicates situational contraindications: significant psychiatric, cardiovascular, renal, or liver disease. It also mentions that older individuals may experience symptomatic benefits but are more susceptible to adverse effects. This is a group where self-adjusting doses is the most risky.

The safety and adverse effects review of cannabidiol lists fatigue and diarrhea as the most common, alongside changes in appetite and body weight. The authors also note that the impact of cannabidiol on liver enzymes and drug transporters remains insufficiently studied, as does its potential effect on hormonal balance (Iffland, Cannabis and Cannabinoid Research, 2017).

In the registration study of children with Dravet syndrome, among the adverse effects occurring more frequently than in the placebo group were diarrhea, vomiting, fatigue, fever, drowsiness, and abnormal liver function test results. The last item is the reason why the EFSA position on the inability to determine safety for individuals taking medications should be read literally, not as a precautionary formula.

How to keep your own records when adjusting the dose?

Start with one goal and one measure. "Better sleep" is not a measurable goal, while "time to fall asleep in minutes" and "number of awakenings at night" are. Without a previously chosen measure, any change in feelings can be explained in several ways, and after a month, you won't be able to distinguish the effect of the product from changes in the weather and work stress.

Record four things with each intake: the amount in milligrams, the time, the presence of a meal, and the chosen measure. The first three change the body's exposure, as shown by studies on the impact of meals and steady-state conditions. The fourth is the only reason you are keeping records. A note without a measure is a memory, not data.

Keep records for a long enough period to encompass the natural variability of the symptom. For conditions with a fluctuating course, a week is too short to draw any conclusions, as improvement could occur without any intervention. Comparing several weeks before and several during gives a result that can be discussed with a doctor.

Do not change two things at once. Simultaneously increasing the amount and switching from capsules to drops results in an inconclusive outcome, as both changes affect exposure. This is the most common reason someone cannot say after two months what worked for them and what did not. One change at a time, assessment after establishing concentration, note with a measure.

What mistakes most often spoil dose selection?

The first and most costly is counting drops instead of milligrams. A drop depends on the pipette, temperature, and viscosity, so the same number of drops practically means different masses of the substance. The result is an observation that cannot be repeated or compared with anything, including your own note from a week ago.

• Assessment after one day. Steady state is established only after about two days of regular intake, and accumulation reaches nearly two and a half times the first exposure (Taylor, 2018).
• Changing the conditions of intake during observation. A high-fat meal increases exposure several times, so comparing a fasting day with a day after lunch does not measure the dose.
• Treating doses from studies as recommendations. Values from studies on epilepsy and tolerance pertained to different individuals, a different product, and medical supervision.
• Taking numbers from product descriptions as data. Bioavailability values repeated in marketing materials have no backing in a systematic review of studies in humans.
• Ignoring drug interactions. With ongoing treatment, the decision about cannabidiol belongs to the doctor, as the change in concentration pertains to the medication, not the supplement.

The common denominator of these mistakes is one: replacing measurement with impression. A product whose contents you have not calculated, taken under variable conditions and assessed after a day, provides no information that could be used for the next decision.

Frequently Asked Questions

How many milligrams of CBD are in one drop of 5, 10, and 20 percent oil?

Assuming 30 drops per milliliter, 5% oil gives about 1.67 mg per drop, 10% oil about 3.33 mg, and 20% oil about 6.67 mg. This is an arithmetic conversion based on concentration and bottle volume, not a dosage recommendation. The actual volume of a drop depends on the pipette, temperature, and viscosity of the preparation.

What daily dose of CBD does EFSA consider safe?

The EFSA panel derived a provisional safe dose of 0.0275 mg per kilogram of body weight per day in 2026, which is about 2 mg for a person weighing 70 kg, with an uncertainty factor of 400. This value applies only to supplements with a purity of cannabidiol of at least 98% and without nanoparticles.

Is there a safe dose of CBD during pregnancy or with medications?

There is no established value. EFSA states that the safety of cannabidiol cannot be determined in individuals under 25 years of age, in pregnant and breastfeeding women, and in those taking medications simultaneously. In these situations, the decision to use it is made by the attending physician, not a guide.

Ile CBD z olejku trafia do krwiobiegu?

A systematic review of pharmacokinetics in humans established the absolute bioavailability only after smoking, which was 31%. For sublingual and oral administration, no study has measured it. Values repeated in product descriptions, such as 6% orally, are not supported by this review.

Does a meal affect the absorption of CBD?

Yes, significantly. In healthy volunteers, a high-fat meal increased the maximum concentration by 4.85 times and the area under the curve by 4.2 times (Taylor, 2018). In eight adults with drug-resistant epilepsy, the maximum concentration was on average fourteen times higher after a meal than on an empty stomach (Birnbaum, 2019).

Does a higher dose of CBD work stronger?

Not in every range. In a randomized study involving 57 healthy men, anxiety before public speaking was reduced only by a dose of 300 mg. Doses of 150 mg and 600 mg did not differ from placebo. The authors described this relationship as a U-shaped curve.

How long does it take for CBD to reach a steady concentration in the blood?

In a phase one study, steady state with twice-daily dosing was established after about two days, with accumulation ranging from 1.8 to 2.6 times. The effective half-life was estimated at 10-17 hours. Therefore, assessing the effect after a single dose measures something different than steady state.

When should dosage selection be discussed with a doctor?

With any ongoing treatment, in cases of liver, heart, kidney diseases, and mental disorders. Cannabidiol inhibits the metabolism of some medications, a documented example being clobazam. Abnormal liver function tests were also noted in a registration study in children with Dravet syndrome.

Podsumowanie: co o dawkowaniu CBD wiadomo, a czego nie

It is known how to convert the contents of a bottle into milligrams, as this is arithmetic. It is known that a high-fat meal increases exposure several times and that steady state is established after about two days. It is finally known that in the only study comparing three doses in humans, the middle one worked, not the highest.

However, it is not known what guides provide most readily. Bioavailability has not been measured after sublingual or oral administration. There is no study that would establish an effective range in supplementation for healthy individuals. There are also no data allowing predictions of what amount will yield what concentration in the blood for a specific person.

In this situation, the only value with official status is the provisional safe dose from EFSA: 0.0275 mg per kilogram of body weight per day, about 2 mg for a person weighing 70 kg, and this is only for supplements with a purity of at least 98%. Safety cannot be established for individuals under 25 years of age, for pregnant and breastfeeding women, and for those taking medications.

The practical conclusion is therefore not "take this much," but "count in milligrams, change one thing at a time, record the measure, and talk to your doctor if you are taking anything regularly." The conversion from this text is sufficient to compare any two products in the category oils na tej samej skali.

This article is for informational and educational purposes and does not constitute medical advice. Before starting to use cannabis or CBD for therapeutic purposes, consult a doctor, especially if you are taking other medications, are pregnant, or breastfeeding.

Author: Michał Waluk · Opublikowano: 2026-05-11 · Aktualizacja: 2026-08-10