The basic problem your body is solving

Every day you take in compounds your body has no use for. Some come from food: plants make a great many chemicals, and a fair number of them are biologically active. Some come from the air, from packaging, from medication, from the by-products of your own metabolism.

Most of these share an inconvenient property: they dissolve in fat rather than water. That matters because the main exit routes (urine, and to a lesser extent bile) are water-based. A fat-soluble compound entering the bloodstream will tend to partition into tissue and stay there. It cannot easily leave.

So the body's solution is not to remove these compounds directly. It is to chemically rebuild them into something water-soluble enough to be excreted. That process is called biotransformation, and it is what "detoxification" means in a biology textbook.

It happens mainly in the liver, but not only there. The lungs are metabolically active: enzymes that process foreign compounds are found in bronchial lining cells, in type II pneumocytes, and in alveolar macrophages. The gut and its resident bacteria are involved too, in ways we come to shortly.

The part that breaks the popular model

Biotransformation is usually described in two stages, and the naming convention is standard enough that the US National Library of Medicine uses it as a formal indexing term.

Phase I adds or exposes a reactive chemical handle, usually an oxygen atom. This is mostly the work of a large family of enzymes called cytochrome P450.

Phase II attaches something bulky and water-loving to that handle: a sugar derivative (glucuronidation), a sulfate group, or glutathione. The result is heavier, much more water-soluble, and far easier to excrete.

Almost no consumer article mentions this: Phase I frequently produces something more chemically reactive, and often more toxic, than what it started with.

That is not a fringe finding. One review states it directly: in some instances, the biotransformation of a given compound can result in the generation of more reactive, and frequently more toxic, metabolites. Another describes the metabolic activation of drugs by cytochrome P450 into chemically reactive electrophiles as a major initiating event in a form of drug-induced liver injury. A third notes that two specific P450 enzymes play critical roles in converting certain combustion-related compounds and cooked-meat compounds into reactive intermediates.

The intermediate is the dangerous moment. Phase II exists largely to catch it, grabbing the reactive handle before it reacts with something you would rather keep intact, like DNA.

Even Phase II is not uniformly protective. A review of herbal compound metabolism notes that certain naturally occurring plant constituents can be bioactivated by sequential hydroxylation and sulfation (a Phase II reaction) into intermediates capable of forming DNA adducts.

So the tidy model, in which Phase I "breaks toxins down" and Phase II "flushes them out," is not what happens. What happens is a two-step chemical relay in which the first step is often the risky one and the second step is the rescue.

Then the gut can undo it

There is a second complication that dismantles the "removal" metaphor entirely.

Many conjugated compounds leave the liver not in the blood but in bile, which empties into the small intestine. That looks like the end of the journey. Often it is not.

Gut bacteria produce enzymes (β-glucuronidase, among others) that can cleave the conjugate back off. The compound is returned to its original, fat-soluble form, in the intestine, where it can be reabsorbed and travel back to the liver to start again. This loop is called enterohepatic circulation, and its consequences are measurable: it produces multiple peaks in a drug's blood concentration over time and a longer apparent half-life.

The clearest documented example is a chemotherapy drug, irinotecan, which gut microbial β-glucuronidase actively reactivates.

Elimination, in other words, is not a one-way valve: it is a cycle with a leak in it, and your gut bacteria control the leak.

So what is happening, right now

All of this runs continuously, and there is no retrievable evidence that it pauses, falls behind, or builds a backlog waiting for an intervention.

That matters, because the entire commercial category rests on the opposite premise: that things build up, that your system falls behind, and that something must be done periodically to catch up. We looked for authoritative support for that premise and could not find any. It is not a scientific claim that has been tested and confirmed; it is a frame.

What changes these pathways in people

This is where the article could easily become an advertisement, so let us be precise about what is established.

Some things do alter these enzymes in humans, and the best-documented examples are mostly cautionary rather than beneficial.

Grapefruit juice inhibits an intestinal P450 enzyme responsible for the first-pass metabolism of many medications, which can raise the blood concentration of numerous drugs and lower a few others. This is a real, clinically managed interaction.

St John's wort works the other way. One of its constituents binds a receptor that switches on the same enzyme family, and clinical studies show it increases the metabolism of various drugs, including combined oral contraceptives, ciclosporin and indinavir. Speeding up drug clearance sounds harmless until the drug in question is the one keeping a transplant or a contraceptive working.

Cruciferous vegetables (broccoli, Brussels sprouts, cabbage, cauliflower, radish, watercress) do measurably change enzyme activity. A systematic review and meta-analysis of 23 human dietary intervention trials found they raised the activity of one P450 enzyme by 20 to 40% and a glutathione transferase by 15 to 35%.

That is a real finding from real human trials. Now read the authors' own next sentence: there is contention between reports, and the clinical significance of potential diet-drug interactions remains unclear.

One of that review's authors was affiliated with a pharmaceutical company's R&D arm, which is worth knowing alongside the finding.

But that hedge is the whole problem in miniature.

The strongest evidence that exists, and why it still doesn't get you there

The best human research on deliberately changing these pathways comes from a series of trials in Qidong, China, using a broccoli sprout beverage providing glucoraphanin (the precursor the body converts to a compound called sulforaphane) plus a smaller amount of sulforaphane itself.

They are properly conducted. One randomised 291 people for 12 weeks and found increased urinary excretion of glutathione-derived conjugates: benzene up 61%, acrolein up 23%, and crotonaldehyde not significantly changed. A later dose-response trial in 170 people found the full dose raised excretion of one marker by 63.2%, while the half dose (+11.3%) and the fifth dose (−6.4%) were both statistically null.

Three things follow.

Dose is everything. The same ingredient did nothing measurable at half and at one-fifth of the effective amount. "Contains broccoli extract" tells you nothing.

These were biomarkers, not health. Every one of these trials measured a chemical in urine. None measured whether anyone was healthier. The authors say only that the approach "may" attenuate long-term risk.

The population was not a general one. These were rural Chinese adults with substantial airborne pollutant exposure, drinking a whole-food beverage.

On the broader question of whether inducing these enzymes helps anyone, a review put it about as plainly as a review can: no agents have yet demonstrated clear benefit in human cell systems, or in clinical trials.

There is one further nuance, and it runs against the naive "more enzyme is better" model. Researchers studying one P450 enzyme family ended up concluding that it has an essential protective role against a common environmental carcinogen: in knockout mice, removing it made things worse. That is animal work and should not be read as a human finding, but it illustrates the point: these enzymes are not a dial labelled "more is better."

What the research on detox programmes found

Two systematic reviews are worth knowing about.

The first examined commercial detox diets. Its conclusion, verbatim: "To the best of our knowledge, no randomised controlled trials have been conducted to assess the effectiveness of commercial detox diets in humans." It adds that the handful of studies that do exist are "hampered by flawed methodologies and small sample sizes," and that the evidence for individual foods comes mostly from animals.

The second examined colonic cleansing. It found no methodologically rigorous controlled trials supporting the practice for general health promotion and, in contrast, multiple case reports and case series describing adverse effects. Its conclusion was that the practice "cannot be recommended at this time."

There is also one randomised trial of an actual commercial detox supplement, and it deserves attention precisely because it is the closest thing that exists to a test of the category. Twenty-eight days, 14 people in the intervention group and 18 controls. The result: no significant changes in the urinary hepatic detoxification biomarkers, and none in cellular glutathione or the glutathione ratio. A superoxide dismutase measure rose 23%, but at p = 0.06 (not statistically significant), and was nonetheless reported as a positive finding. Five of the paper's seven authors were employed by the manufacturer.

The study's title says the product "supports Phase II detoxification enzymes." Its results do not show that.

What this article is not going to do

We are not going to end this by telling you what to take.

There is a version of this page that walks through the physiology and then arrives at a product, and it would be easy to write. The reason we are not writing it is straightforward: the evidence that would be needed does not exist. Not for this site's products, and not for anyone's.

What that evidence would look like is well defined. The Federal Trade Commission's guidance for health products requires substantiation by competent and reliable scientific evidence (research conducted and evaluated objectively by relevant experts and generally accepted as yielding reliable results) for every objective claim conveyed, expressly or by implication. In practice that means testing the actual finished product at its actual dose, in people resembling the buyer, measuring an outcome that answers the claim as an ordinary reader would understand it, and having it replicated independently.

That does not exist here. The same guidance is explicit that a disclaimer does not fix the gap: the DSHEA disclaimer or similar statements won't cure an otherwise deceptive ad, particularly where the deception concerns health-related benefits.

There is also a logical trap worth naming. If your detoxification system is working normally (and in a healthy adult, it is), then there is nothing missing for a product to add. If a claim implies the system isn't working properly, it has stopped being a claim about normal function and become a claim about an abnormal one, which is a different and much more heavily regulated kind of statement.

The useful part

The best-documented ways to change your biotransformation enzymes are ways that affect your medications.

Grapefruit juice can raise drug levels. St John's wort can lower them, including for oral contraceptives. These are not theoretical. If you take prescription medication and you are also taking a botanical supplement, a large daily dose of anything, or a "detox" product whose full ingredient list you have not read, that is worth raising with your pharmacist, who can check the specific combination in about a minute and does this all day.

A safety point follows directly from the review evidence above: products in this category are not presumptively safe because they are sold without a prescription. The colonic cleansing review found documented harms. Some plant constituents are themselves bioactivated into liver-damaging intermediates, which is the same Phase I chemistry described at the top of this article, working in the direction nobody wants.

If a product makes you feel unwell, that is a reason to stop and to ask a clinician. It is not evidence that anything is working.