What the Research Does and Does Not Show

“Research shows” may be the most trusted phrase in kratom marketing, and the easiest one to write. It borrows the authority of a study the reader never sees. This lesson teaches you to find that study and test the claim against it yourself.

The last lesson showed what a lab report can and can’t prove. Research is the other thing sellers lean on, and it can be checked the same way. It takes five questions and usually a few minutes. By the end you will have watched them take apart one real claim, and you’ll be able to do the same to the next one you read.

Why Does “Research Shows” Work So Well?

Because most people can’t check it, and the phrase counts on that.

A claim that cites research sounds finished. It suggests someone careful already did the hard part. But the sentence you read is rarely the study’s own finding. It’s someone’s summary of it, and the person writing the summary usually has a reason to write it the way they did.

The fix isn’t to distrust every study. It’s to put the claim back next to the study and see whether the two match. That’s worth doing, because far less human research exists on kratom than most people assume¹. When we searched the federal registry of studies in people in September 2026, it listed 15 studies of kratom and its alkaloids, and four of them were finished.

First, Find the Study

A claim that points to no study is an opinion, however it’s dressed.

Look for a named paper, a journal or a registration number. ClinicalTrials.gov is the federal registry where studies in people are recorded. A study registered before it starts states what it will measure, so its published paper can be checked against the plan, and so can a study that finished and never reported back.

A link to a news story, a seller’s blog or another company’s press release is not the study. Follow it until you reach the paper itself, or its record on the registry.

If a claim names no source, ask for one. If nobody can produce it, write that down. It is the most common answer you will get, and it is an answer.

Where Do You Look in a Paper?

Not only at the top.

The abstract is the short summary at the start of a paper, and it’s where most claims get their words. It is also where the qualifications get squeezed out. The methods section tells you what was given, to whom and for how long. The limitations paragraph, usually near the end, is where the authors say what their own study can’t show. The funding and conflicts statement tells you who paid and who stands to gain.

Read the limitations before the conclusion. A peer-reviewed paper was checked by other researchers before a journal printed it. That is a check on the method, not a stamp of truth.

The Five Questions

Once you have the study, ask these in order.

What did it give people? Leaf powder, an extract, or a single purified alkaloid made in a laboratory. Those are three different subjects. Much of the lab work on how kratom acts in the body uses one purified compound, and a study of that compound is not a study of the powder in a package.

How many people, and for how long? Fifteen people over one afternoon is a signal worth watching, not a finding about everyone. Every figure needs its denominator too. In the 2017 to 2018 Salmonella outbreak from the last lesson, 199 people in 41 states fell ill over sixteen months. Fifty were hospitalized, out of the 132 whose records showed whether they had been. Fifty alone sounds one way. Fifty out of 132 tells you something.

Who was in it? Volunteers who had never taken kratom, people with opioid experience and long-term takers recruited online will answer the same question differently. The third level showed that a history with opioids goes with more reported withdrawal from kratom, so a result from one of those groups may not describe the others. A result describes the kind of person it studied.

Did it watch, or did it assign? A survey asks people about their own experience, and it records what they chose to answer and what they remember. An observational study watches a group without giving them anything. Both can show that two things turn up together, and neither can show that one caused the other. A randomized controlled trial assigns people by chance to the substance or to a placebo, a dummy that looks the same. When it’s double-blind, neither the volunteers nor the staff know who got which. That design is the best tool there is for showing cause, and kratom has very few trials built that way.

Who paid, and where is it registered? A sponsor doesn’t make a result wrong. It makes the sponsor a fact you weigh beside the result, the same way you ask whose name sits on a lab report. Look for the funding and conflicts statement, and read past the abstract to find it.

One Claim, Taken Apart

In January 2026, a journal published the largest controlled trial of kratom leaf so far. A week later, a company selling a drink made with kava and kratom put out a press release. It cited the trial as evidence for its own drink, counted it among 19 clinical studies, and said the trial examined the same leaf its drink contains. Here is that claim run through the five questions.

What did it give people? One well-characterized leaf powder, in capsules, in a clinic. It didn’t give anyone the drink.

How many, and for how long? Forty-nine people took kratom and 67 took a placebo. They took one serving, then one a day for fifteen days, and were watched for a few weeks after.

Who was in it? The methods section lists who could take part: healthy nonsmokers aged 18 to 55 who had never taken kratom or hadn’t for a year, screened to exclude several variations in how the body processes drugs.

Did it watch, or assign? It assigned. It was randomized, double-blind and placebo-controlled, the strongest design there is, and that deserves credit.

Who paid, and where is it registered? The paper names who funded the trial and says its authors were paid for their work. That sentence sits partway through the discussion, in the middle of a paragraph rather than under a heading of its own, which is exactly why claims skip it. The trial was approved in Canada and doesn’t appear on the American registry. The drink company that cited it didn’t run it.

Now put the claim back next to the study. The trial is real and well built. The authors concluded the leaf powder was safe and well tolerated at the amounts tested. They also wrote that long-term effects might not have been detected and that more study is needed in people who take kratom regularly. A second paper on the same trial added that its findings may not carry over to regular takers or to people who use other substances.

The press release carried that conclusion from one capsule product, taken by 49 healthy newcomers for fifteen days, to a different product bought by anyone, taken for any length of time. The research did not make that trip. The claim did.

The same trial has been stretched in other ways. Some pages online called it an FDA-funded trial. It was not. The FDA’s own study was a separate, smaller pilot with different volunteers, and pages that merge the two are passing along a study that never happened.

What a Good Study Can’t Tell You

What will happen to you.

A trial reports what happened to the people in it, under its conditions, for as long as it ran. That is its strength and its limit. The people who actually buy kratom look far more like the long-term takers in the federal field studies than like the volunteers in that trial.

So a claim that passes all five questions describes someone else. That’s why we don’t say kratom is safe², and why no study, however well built, can say it about you.

Three Moves to Watch For

Most stretched claims make one of three swaps.

The product swap: a study of one product, or of one purified alkaloid, quoted as if it covered another. The people swap: a result from healthy newcomers or screened volunteers applied to everyone. The time swap: fifteen days quoted as if it said something about years. Once you know the three, you’ll start seeing them in headlines, product pages and forum posts.

When There Is No Study

Sometimes the honest answer to “what does the research show” is nothing, and that’s a finding too.

No study has compared stopping all at once with stopping gradually. No controlled study has followed people taking kratom for more than a few weeks. No systematic data backs the claim that small amounts and large amounts act in opposite directions. When a confident claim lands in one of those gaps, the confidence came from somewhere other than research.

A headline about a purified alkaloid³ belongs here too. It may describe a real study. It does not describe leaf.

Keep the five questions in reach and use them on anything, including what we write. If you’d like the safety side of this course on paper, the Print Edition⁴ summarizes it.

Key Takeaways

• “Research shows” is someone’s summary of a study. Find the study before you trust the summary.
• A claim with no study behind it is an opinion. Ask for the source, and write it down when nobody can give one.
• Ask what was given, to how many people for how long, who they were, whether the study watched or assigned, and who paid.
• A study of one purified alkaloid is not a study of leaf, and a study of one product is not a study of another.
• A strong trial describes only the people in it, for as long as it ran. It can’t tell you what will happen to you.