Digestion Supplement Guide: Six Families, Six Different Scales

Six families. The third column is what the published work used, not a recommendation, and not a prediction about any particular bottle.

Read down the right-hand column and the arithmetic of this aisle becomes obvious. A capsule holding a few hundred milligrams of powder can reach the third family and cannot reach the first two, whatever it prints on the front.

That is not a criticism of capsules. It is a statement about volume: a gram of psyllium is roughly the contents of two large capsules, and the trials asked for ten.

A family with no established scale is treated carefully in both directions here. Nobody has settled what it does in people, so a label promising a result from one is promising beyond the literature.

Anthranoids are the plant compounds behind senna, cascara sagrada, Chinese rhubarb, the latex of the aloe leaf and buckthorn bark. They reach the colon largely unchanged, bacteria there cleave the sugar off, and the freed compound reduces water reabsorption and increases propulsive contractions (a review of anthranoid laxatives, a review of chronic stimulant laxative use).

Two consequences follow from that bacterial step. The effect arrives six to twelve hours later rather than immediately, and its strength varies between people because gut flora vary.

The scale is the reason this family turns up in capsules. Anthranoid glycosides are active in tens of milligrams, so a few hundred milligrams of plant material is enough to matter (a survey of anthraquinones in laxative plants, a toxicity assessment of buckthorn bark and emodin). Nothing else on this shelf works at that weight.

• What they do when they work: looser stool, urgency, and cramping. Those are the mechanism rather than side effects, which is the single most useful thing to know before taking one.
• Duration: every over-the-counter product in this class prints a limit on consecutive days. A supplement containing the same compounds often does not.
• Long-running use: the literature describes melanosis coli, a benign pigmentation of the colonic lining that reverses after stopping (a study of melanosis coli at colonoscopy, a prospective colonoscopy study), and a bowel that has adapted to stimulation (a review of the cathartic colon question, a review of long-term stimulant laxative use).
• The rarer signals: a case report of hepatitis associated with a cascara supplement (a case report of cascara sagrada liver injury), colonic changes in rats given cascara (a rat-colon assay of senna and cascara glycosides), and cytotoxicity work on the compound class (a structure-activity comparison of anthranoids).

Aloe deserves separating out, because the word covers two completely different materials. The inner gel is the one sold as a drink and the one with the negative and mixed irritable bowel trials behind it (a placebo-controlled trial that found no benefit, a randomised trial of an aloe extract, a systematic review of aloe in IBS, a placebo-controlled pilot study). The latex just under the leaf skin is the anthranoid, it is the reason the whole-leaf material is a laxative, and it is what the rodent toxicology programme examined (the National Toxicology Program's rat study, the full NTP technical report, a review of aloe toxicity, NCCIH).

A label saying only aloe vera leaf has not told you which. A label saying aloe vera gel, or naming a decolourised extract, has.

Bitters are the older tradition on this shelf. Gentian is the type specimen, its chemistry is well described and its principal compound has a substantial pharmacological literature (a review of Gentianae Radix, a review of gentiopicroside). What is hard to find is a randomised human trial of an outcome anybody buys it for.

Adsorbents are the newer arrival. Bentonite is a clay, and the usable literature treats it as a pharmaceutical excipient and an adsorbent rather than as a benefit (a review of bentonite in drug delivery). That framing matters practically rather than academically: an adsorbent can take up other things swallowed at the same time, which makes it a spacing question for anyone on a prescription.

Neither family is being dismissed here. The honest position is that their human outcome evidence does not exist yet, and a label promising a result from either is promising past what has been established.

There is a third quiet category worth naming, which is the ingredient whose evidence belongs to a different part of the plant or a different preparation. Oat is the clean example: every beta-glucan result used bran or oatmeal at 3 g of beta-glucan a day (a meta-analysis of oat beta-glucan, a systematic review of oat beta-glucan and LDL), and a label naming the straw is naming a different material. Alfalfa is another: the cholesterol work used 40 g of heat-prepared seeds three times a day (the alfalfa seed trial, an earlier study of alfalfa seeds), and the fact most relevant at supplement scale is the canavanine story rather than the lipid one (the L-canavanine finding).

Most constipation is a matter of fluid, fibre from food, movement and routine, and reviews of the condition put diet first for good reasons (a 2024 meta-analysis of foods, drinks and diets, a 2024 Cochrane review). A capsule that bypasses all four will keep needing to be taken.

Some of it is not. Blood, fever, pain, a sudden change in habit that persists, or unexplained weight change are signals that belong in an appointment rather than in a shopping basket, and reviews aimed at recognising them exist for exactly that reason (a review of what makes a fibre supplement effective).

In between sits the ordinary case, where an over-the-counter option chosen knowingly is entirely reasonable (a systematic review of over-the-counter options). What makes it knowing is understanding which family you are buying and at what scale, which is what this guide is for.

Six rows applied to one label. It reaches one family of the six, and that is the honest reading of what the capsule does.

A guide that never touched the product beside it would be evasive, so it is applied here and the result is what it is. One family of six.

That is not the same as saying the product does nothing. A stimulant laxative that works is a product that works, and it is a narrower thing than a twelve-ingredient panel implies.

The ingredients page runs the same analysis name by name, with the trial dose beside each label position.