How to evaluate a cannabis claim

When you read a claim about medicinal cannabis in New Zealand, the first question is not “is this true?” It is “what is this claim actually saying, and who is saying it?” A single sentence can mix three different kinds of truth: what a regulator has authorised, what a study shows, and what a company wants you to believe. The method below separates them.

How to evaluate a cannabis claim: Start with the regulatory frame

New Zealand’s Medicinal Cannabis Scheme sits under the Misuse of Drugs Act 1977 and is administered by the Medicinal Cannabis Agency (MCA), part of Medsafe. Under this arrangement, a preparation is only “medicinal cannabis” if it is prescribed by a medical practitioner, meets Good Manufacturing Practice (GMP) standards, and has been notified to the Ministry of Health. That is the legal baseline. As of 2024, the framework does not permit a general licence to use cannabis, and it does not make raw plant material available outside a prescription and pharmacy dispensing (New Zealand Ministry of Health, 2024).

The first check in any claim is therefore regulatory: does the item sit inside the Medicinal Cannabis Scheme? If it does not, the claim is commercial, not medicinal.

What the law actually requires

Under the Misuse of Drugs Act, medicinal cannabis is a controlled drug. Access is only through a prescription from a medical practitioner. The Medicinal Cannabis Agency licenses growers, manufacturers, and importers, and it sets quality standards. Any item that is not a “controlled drug” under the framework, or that is not manufactured to GMP standards, is outside the legal pathway.

The first test of any claim is therefore not “does it sound scientific?” but “does it fit the regulated framework?” A preparation that is not GMP-certified, not authorised by the health ministry, or not dispensed through a pharmacy is not medicinal cannabis in New Zealand, regardless of what the label says.

A decision rule for legal status

Ask: Does the claim say “authorised by Medsafe” or “registered with the Medicinal Cannabis Agency”? If not, it is outside the regulated pathway. Medsafe does not authorise individual cannabis items for general sale; it licenses preparations for prescription under the framework. A claim that something is “Medsafe approved” is misleading unless it is a specific medicine authorised under the Misuse of Drugs (Medicinal Cannabis) Regulations 2019.

If a claim says “legal” or “approved” without naming the Medicinal Cannabis Scheme, Medsafe, and a prescription, treat it as unverified. Legal medicinal cannabis in New Zealand requires a prescription from a medical practitioner under the Misuse of Drugs Act 1975 (as amended), and items must meet GMP standards set by the health ministry.

The core question: what is the claim actually saying?

Every cannabis claim contains three separate layers. Pull them apart before you evaluate anything.

Layer 1, The commercial promise. What is the seller or advocate trying to get you to do? Buy a preparation, book a consultation, sign a petition, believe a brand is “medical grade”. This layer is not proof; it is persuasion.

Layer 2, The regulatory status. Is the item authorised under the Medicinal Cannabis Scheme? Does the claim match what Medsafe and the health ministry actually permit? Under the Misuse of Drugs Act, medicinal cannabis is a prescription medicine, not a general licence. That distinction is not optional.

Layer 3, The supporting data. What does the peer-reviewed clinical guidance actually say about the condition, the cannabinoid, and the dose? Methodical reviews from 2023 and 2024 (e.g., those cited by the NZ Formulary) are the baseline. If a claim contradicts the review, it is not “emerging science”; it is marketing.

Why claims fail

Most cannabis claims fail for one of five reasons. Use this checklist when a claim looks plausible.

The claim is too vague to test

1. The claim is not tied to a specific preparation. “Cannabis treats pain” is not a claim about a named item. A claim about “THC 25 oil for neuropathic pain” is testable. The more specific, the more you can check.
2. The endpoint is vague. “Feels better,” “improves wellbeing,” or “reduces stress” are not measurable. A useful claim names a specific outcome (e.g., “reduces pain scores by 30% over 4 weeks”) and a population.
3. The claim is about a product, not the active ingredient. “This oil works” is meaningless unless the THC:CBD ratio, dose, and indication are specified. A claim about “cannabis” generally often collapses when you ask which cannabinoid, which ratio, and which route.
4. The claim conflates “cannabis” with a specific item. “Cannabis works” is not the same as “this specific GMP flower at this THC:CBD ratio works for this condition.”
5. The claim ignores the dose. Many effects are dose-dependent; a claim that “cannabis treats pain” is meaningless without stating the dose, product form, and route.

The supporting data is weak or missing

1. The data is missing or mislabelled. A claim may cite “studies” but not name the study design, sample size, or whether it was peer-reviewed. A single case report is not a methodical review.
2. The data is not cited. A claim without a source is an anecdote; a claim with a source you cannot check is a marketing statement. In this country, the NZ Formulary and Medsafe are the reference points.
3. The data is a single case or small series. Methodical reviews and peer-reviewed clinical guidance carry more weight than a single patient story; this is especially true for pain and epilepsy, where placebo responses are high.
4. The claim is based on a case report or anecdote. Case reports are hypothesis-generating, not proof of efficacy; demand a methodical review or controlled trial.
5. The claim is extrapolated from animal or lab studies. In vitro and animal data are not human proof; the claim should state the species and model.
6. The claim cites a single study. A single study is a data point, not a consensus. Look for methodical reviews, meta-analyses, or at least two independent trials with consistent results. The NZ Formulary and Healthify are good starting points.
7. The claim lacks a denominator. “80% of patients improved” is meaningless without knowing how many were treated, whether there was a control group, and how improvement was measured.
8. The claim confuses association with causation. Observational surveys cannot prove that cannabis caused an improvement; there may be placebo effects, co-interventions, or natural recovery.
9. The claim ignores the placebo response. Pain trials routinely show 20-30% placebo improvement; a product claim must beat that, not merely match it.
10. The claim cites a “study” but no journal, year, or sample size. Peer-reviewed clinical guidance (e.g., methodical reviews in the NZ Formulary) is the standard; anecdotes are not proof.

The source or framing is misleading

1. The source has a conflict of interest. A company selling a product is not an impartial source, even if it cites research. Check who funded the study, who wrote the guideline, and whether the authors have financial ties to the manufacturer.
2. The claim uses “natural” or “herbal” as a quality marker. In the NZ framework, “natural” is not a regulated claim; GMP certification and health ministry authorisation are the only quality markers.
3. The claim cites overseas authorisations. This country has its own regulatory pathway; overseas authorisation does not mean the item is authorised or subsidised locally.
4. The claim relies on anecdote or “my doctor said”. Personal testimony is not proof; the NZ Formulary and methodical reviews are the baseline.
5. The claim says “clinically proven” without a named condition or endpoint. Ask: proven for which condition, and which outcome (pain, sleep, anxiety, spasticity)? The word “proven” is meaningless without a measurable endpoint.
6. The claim says “clinically proven” without a source. No product in this country is universally “proven”; supporting data varies by condition and item. The NZ Formulary lists products but does not rank them by efficacy.
7. The endpoint is surrogate, not clinical. A product that “increases endocannabinoid levels” is not the same as “reduces pain”; the first is a lab measurement, the second is a patient outcome. Always ask what was actually measured.
8. The claim uses “natural” or “plant-based” as a synonym for safe. Natural does not mean safe. Raw cannabis flower can contain mould, pesticides, and variable THC/CBD levels; GMP certification is the only guarantee of quality locally. The Medicinal Cannabis Scheme requires items to meet GMP standards and be authorised by the health ministry. (Source: New Zealand Ministry of Health, Medicinal Cannabis Scheme, 2020.)
9. The regulatory status is absent. In this country, medicinal cannabis products must be authorised under the Medicinal Cannabis Scheme and prescribed by a medical practitioner. If a claim implies you can buy it over the counter or without a prescription, it is outside the legal pathway.

Where to verify a claim

When a claim passes the initial checks, move to these sources.

• NZ Formulary. Lists authorised products and summarises clinical guidance by condition.
• Medsafe and the Medicinal Cannabis Agency. Publish the current regulatory requirements and GMP standards.
• Healthify. Provides patient-facing summaries of supporting data for cannabis and specific conditions.
• Methodical reviews. For chronic pain, look for the 2024 review by Solmi et al. and the 2017 US National Academies report; for epilepsy, the 2018 GW Pharmaceuticals trial data.

A claim that cannot be traced to at least one of these sources, with a named study design and population, is not yet ready to act on.

Driving and cognitive impairment: an urgent caution

Cannabis impairs driving. THC affects reaction time, coordination, and judgement for hours after use, and impairment can persist beyond the period a person feels “high”. Do not drive after using any cannabis product. The law sets a zero-tolerance approach for drivers who test positive for THC, and prescription status does not provide a defence. If you are prescribed a THC-containing medicine, ask your doctor explicitly about driving restrictions and document the advice.

How to read a product label or website claim

When you encounter a claim on a label, website, or consultation, work through these four steps in order.

1. Isolate the exact wording. Write down the claim verbatim. “Supports sleep” is different from “treats insomnia”. The first is a structure-function statement; the second is a therapeutic claim that requires authorisation.
2. Check the regulatory pathway. Does the item appear on the Ministry of Health’s list of verified medicinal cannabis products? If not, stop. The claim is outside the legal framework.
3. Ask for the citation. Request the journal, year, study design, and sample size. A refusal or a link to a blog post is a red flag.
4. Cross-check the condition. Look up the condition on the NZ Formulary and Healthify. If the claim goes beyond what the summaries say, treat it with scepticism.

A distinct claim for the material below

No other consumer resource in this country combines the Misuse of Drugs Act requirements, the Medicinal Cannabis Agency’s GMP standards, and a step-by-step claim-separation method into a single checklist that starts with the regulatory frame rather than the science. Most guides begin with “what the research says”. This one begins with “is it even a legal medicine?” because if the answer is no, the rest of the claim does not matter.