Reading a claim

How Constipation Is Measured: Rome IV, The Bristol Scale, And What 'Better' Means In A Trial

The seller's page talks about healthy digestion, everyday metabolic wellness and a lighter feeling. The carton itself prints none of those phrases. Whichever wording you meet, a sensible reader eventually asks how anyone would know. Research on constipation has a small standard toolkit for that question: a symptom-based diagnosis called Rome IV, a seven-picture stool chart called the Bristol scale, and a trial endpoint called the complete spontaneous bowel movement. This article explains what each of them captured, and why none of them is what a label word such as regular or lighter means.

Sourced from the product label and PubMed-indexed research · Last updated September 27, 2026 · 12 min read · Reading a claim

Three things called constipation

In conversation, constipation is a feeling. In a clinic it is a diagnosis. In a trial it is a number. The three are related, but they are not the same thing, and confusion starts when a word from one setting is used to imply a result in another. A product page can borrow the everyday word, and a trial abstract can use the number, and a reader is left to guess whether the two describe the same change.

The carton for SodaMelt is a quiet document. It prints the panel, a suggested use, a storage line, a caution and the federal disclaimer, and it does not say what the capsule is for. The phrases healthy digestion, everyday metabolic wellness and a lighter feeling are the seller's wording, and the benefits page quotes them as such. So the fair question is what evidence in this field would look like if someone wanted to support wording of that kind. The answer is a short list of instruments, and this article goes through them one by one.

Rome IV: a diagnosis made of symptoms

The Rome criteria come from the Rome Foundation, which has revised its classification of functional gastrointestinal disorders roughly once a decade. The fourth edition, Rome IV, was published in 2016. Its introductory article describes it as a compendium of what was learned since Rome III appeared ten years earlier (the Rome IV introduction). The chapter on the bowel sorts functional bowel disorders into five categories: irritable bowel syndrome, functional constipation, functional diarrhoea, functional abdominal bloating or distension, and an unspecified group. It adds a sixth, opioid-induced constipation, which it treats as distinct from the functional disorders (the Rome IV bowel disorders chapter).

For an adult, functional constipation is defined by a checklist, as reproduced in a 2021 review. The criteria must be met for the last three months, with symptom onset at least six months before diagnosis, and must include two or more of the following:

• straining during more than a quarter of defecations;
• lumpy or hard stools, Bristol types 1 or 2, in more than a quarter of defecations;
• a sensation of incomplete evacuation in more than a quarter of defecations;
• a sensation of anorectal obstruction or blockage in more than a quarter of defecations;
• manual manoeuvres to help defecation, such as digital evacuation or support of the pelvic floor, in more than a quarter of defecations;
• fewer than three spontaneous bowel movements a week.

Two further conditions apply: loose stools must rarely be present without the use of laxatives, and the criteria for irritable bowel syndrome with constipation must not be fully met (the 2021 diagnostic review). That last condition is the important boundary. Irritable bowel syndrome with constipation requires recurrent abdominal pain, on average at least one day a week over three months, linked with two or more of: relation to defecation, a change in stool frequency, or a change in stool form. Constipation with pain is a different diagnosis from constipation without it.

Three features of this checklist matter for what follows. First, it is made of symptoms and thresholds, not of a test result, so it measures what a person reports. Second, it borrows the Bristol scale for one of its items, which is why the next section matters. Third, it carries laxatives inside its own definition. Loose stools rarely present without laxatives is a criterion, and the same review notes that the irritable bowel syndrome subtypes can only be confidently established when the person is evaluated off the medications used to treat bowel habit abnormalities. In other words, the diagnostic system already assumes that laxatives distort what it is trying to observe.

A 2024 consensus guideline makes a candid remark about all this. It says the criteria are useful for building robust clinical trials, but that their complexity limits their usefulness in everyday practice (the 2024 consensus on constipation in adults). Rome IV is a research and diagnostic tool. It was never a definition of what a healthy bowel feels like.

The Bristol scale: seven pictures and what they track

The Bristol Stool Form Scale is a chart of seven stool types, running from separate hard lumps at one end to watery stool with no solid pieces at the other. Types 1 and 2 are the constipated end, the ones Rome IV counts as lumpy or hard, and types 6 and 7 are the loose end. It began as a research tool in Bristol, and its best-known validation is a 1997 paper by Lewis and Heaton.

That paper is worth knowing in some detail. Sixty-six volunteers had their whole-gut transit time measured with radio-opaque marker pellets, had their stools weighed, and kept a diary of stool form on a seven-point scale and of how often they went. At baseline, transit time correlated with defecation frequency (r = 0.35), with stool output (r = -0.41), and best of all with stool form (r = -0.54). Then the researchers changed transit on purpose. When 44 volunteers took senna, transit time fell while frequency, form score and stool output all rose. When 43 took loperamide, everything moved the other way. The change in transit correlated best with the change in stool form (r = -0.65), and the authors concluded that a stool form scale can be used to monitor change in intestinal function (the Lewis and Heaton study).

Notice what senna did there. Senna is a stimulant laxative from the same broad chemical family as four of the eleven entries in the SodaMelt blend, though senna itself is not on this panel. The scale registered what a stimulant does to transit, and it did so reliably. That is precisely why the scale cannot tell you whether taking a stimulant is a good idea. It tells you the stimulant is acting like a stimulant.

The scale has limits, and the literature is frank about them. A 2010 multicentre study recorded stool form and frequency while measuring transit in two ways, with a wireless capsule and with markers, in 110 people, 46 of whom had chronic constipation. In the constipated adults there were moderate correlations between stool form and whole-gut transit, r = -0.61 by capsule and -0.45 by markers, and with colonic transit, r = -0.62. A Bristol value below 3 predicted delayed whole-gut transit with a sensitivity of 85 percent and a specificity of 82 percent. In healthy adults no correlation was found. And stool frequency was a poor surrogate for transit in everyone, even in constipated adults with fewer than three bowel movements a week (the multicentre transit study). Frequency, the number people find easiest to count, is the weakest of the three.

The scale's reliability has also been tested. In 169 healthy volunteers, the scale type they chose correlated with measured stool water content (Spearman's rho 0.49), and when 86 volunteers classified 26 stool models, 81 percent were correctly classified. Accuracy fell below 80 percent for types 2, 3, 5 and 6, which happen to be the types around the clinical decision points, and repeated classification of duplicate models agreed 76 percent of the time (the validity and reliability study). That study used healthy adults and 19 patients with diarrhoea-predominant irritable bowel syndrome, so it says nothing directly about constipation. It does show that even a good instrument is fuzzy exactly where a decision would be made.

What regular turns out to be

Product language often uses the word regular, so it is worth asking what the population really does. The most cited answer is a 1992 survey of the East Bristol population. The researchers questioned 838 men and 1,059 women, 72.2 percent of a random stratified sample, and most kept records of three consecutive defecations, including stool form on a validated six-point scale. The most common habit was once a day, but that was a minority practice in both sexes. A regular 24-hour cycle was apparent in only 40 percent of men and 33 percent of women. A third of women went less often than daily, and 1 percent went once a week or less. Normal stool types, defined as those least likely to evoke symptoms, made up only 56 percent of all stools in women and 61 percent in men. The authors concluded that conventionally normal bowel function is enjoyed by fewer than half the population (the East Bristol survey).

That is not an argument that bowel habits do not matter. It is an argument that regular is not a benchmark most people meet, and that a bottle promising regularity is promising something the general population does not reliably have. Rome IV, notice, does not set the bar at daily. Its frequency criterion is fewer than three spontaneous bowel movements a week, which leaves a wide range of ordinary habit outside the diagnosis.

The trial endpoint: the complete spontaneous bowel movement

When researchers test a treatment for constipation, the workhorse endpoint is the complete spontaneous bowel movement, usually abbreviated CSBM. Definitions vary in their fine print, but the core is consistent. A spontaneous bowel movement is one that happens without laxatives, enemas, suppositories or other aids, and a complete one is also accompanied by a feeling of complete evacuation. One recent trial states it that way (a 2026 probiotic trial in functional constipation), and another specifies that no laxative was used in the preceding 24 hours (a 2026 retrospective study of neuromodulation). Patients record each movement in a diary, and the trial counts them per week.

A pair of pivotal drug trials shows how demanding the endpoint is. In two 12-week, double-blind, placebo-controlled trials of linaclotide in 1,276 patients with chronic constipation, the primary endpoint was three or more CSBMs a week plus an increase of at least one CSBM over baseline, sustained for at least 9 of the 12 weeks. On the lower dose, 21.2 percent and 16.0 percent of patients reached it in the two trials, and on the higher dose 19.4 percent and 21.3 percent, against 3.3 percent and 6.0 percent on placebo (the two linaclotide trials). A drug that clearly works, with results that were statistically decisive, got about one patient in five over that line. This is what a responder threshold is: a strict bar, applied identically to everyone, and counted as a proportion.

The Rome Foundation's guidance on trial design says the same in general terms. The accepted standard is a double-masked, placebo-controlled, parallel-group design. The primary analysis should be the proportion of patients meeting a responder definition, or a prespecified clinically meaningful change in a patient-reported outcome. Investigators are told to be aware of, and to minimise, expectancy effects, which it lists as placebo, nocebo and precebo (the Rome guidance on treatment trials).

The same measures can be, and are, used for supplements. The 2026 trial mentioned above enrolled 104 adults meeting the Rome IV criteria for functional constipation and gave them either a probiotic strain, Lactiplantibacillus plantarum Probio87, or placebo for eight weeks with four more weeks of follow-up. The primary endpoint was weekly CSBMs, secondary outcomes included Rome IV symptom scores and a constipation quality-of-life questionnaire, and the probiotic group had significantly more CSBMs than placebo at weeks 8 and 12. That is a different organism from the Lactobacillus acidophilus at the bottom of the SodaMelt blend, so it says nothing about this product. It is included to show what evidence in this format looks like, and to make one point clear: the format exists and is used, and this article cites no trial of the SodaMelt capsule in that format.

What better means depends on the question asked

One more result explains why the choice of endpoint matters as much as the result. A 2020 systematic review and meta-analysis pooled placebo arms from 46 randomised trials in chronic constipation, 5,992 patients in all, and found a pooled placebo response rate of 28.75 percent. The interesting part is the split by how better was defined. Where the endpoint was subjective improvement, the placebo response was 41.40 percent. Where it was an improvement in complete spontaneous bowel movements, it was 18.31 percent, and for a composite endpoint 20.35 percent (P below 0.001). The authors did not recommend subjective improvement as the endpoint for future trials (the meta-analysis of outcome measures).

The lesson for a reader is direct. If the only evidence offered for a product is that people said they felt better, you are looking at the yardstick that responds most readily to a sugar pill. That does not mean the people are wrong about how they feel. It means that feeling better and a measured change in bowel function are different claims, and the first is much easier to produce than the second.

Where a laxative capsule falls outside the definitions

Put the pieces together and a structural problem appears for any capsule containing stimulant laxatives, which four of the eleven blend entries on this label are, as the ingredients page and the article on how long a stimulant is meant to be taken explain.

• The diagnosis assumes no laxative. Rome IV counts loose stools without laxatives, and it evaluates bowel-habit subtypes off medication. Someone already taking a stimulant does not fit cleanly into the classification.
• The endpoint excludes assisted movements. A spontaneous bowel movement is one without laxatives or other aids, so a movement produced by the capsule is by definition not the thing trials count. An honest tally of movements on a stimulant is a tally of induced movements, and it cannot be laid beside CSBM results.
• The scale will move for the wrong reason. In the 1997 study senna raised stool form scores, shortened transit and increased frequency. A better Bristol number on a stimulant shows the stimulant working as a stimulant. It does not show that the underlying tendency to constipation has changed.

None of that says a stimulant does nothing. The senna result shows the opposite: it moved things along, reliably, in ordinary volunteers, and the senna trial covered in the article on the 2023 guideline found a clear benefit in people with long-standing constipation. It says that the instruments researchers use were built to measure a bowel left to itself, and they lose their meaning when a stimulant is doing the work. The honest description of a stimulant capsule is that it can produce a movement. Whether the bowel is better in the sense Rome IV or a CSBM count would recognise is a different question, and it can only be answered after the capsule is stopped.

Setting the label's wording against the instruments

The carton prints the last row and none of the others. The stool-weight question behind a lighter feeling is taken up in what a bowel movement weighs.

The table is not a trap. Sellers of every kind of product use everyday words, and shoppers are entitled to notice which of them a measurement could ever support. Healthy digestion is a description of a state, and no criteria set exists for it in the sources above. Regular can be operationalised, but not to mean daily. A lighter feeling maps most closely onto the sensations Rome IV asks about, which are real and can be scored, but those are separate from weight.

A record that would mean something

If you want your own before-and-after that resembles the research, the useful trick is to record what the trials record. A 2021 review describes a constipation stool diary, kept prospectively for one to two weeks, as a way of reducing recall bias and improving the monitoring of treatment response (the 2021 diagnostic review). Its content is not exotic. For each bowel movement, write down:

• the day and time;
• the Bristol type, 1 to 7;
• whether it felt complete;
• whether you used anything to bring it on in the previous 24 hours, a laxative, a suppository, an enema, or the capsule;
• whether you strained, or felt blocked.

Two weeks before starting gives a baseline. Keeping it going during a course gives a comparison. Then count separately: movements with nothing taken, and movements after the capsule. If you stop, keep the diary going, because the unassisted count after stopping is the closest thing you can measure to the bowel left to itself. It is the number a trial would call spontaneous.

Two honest caveats. A diary is not a diagnosis, and if your entries look like the Rome IV list above, or you have any of the warning signs discussed in the article on when constipation is a question for a doctor, the useful next step is a clinician rather than more entries. And none of this measures the thing the seller's wording gestures at, which is a feeling. If a lighter feeling is what you are after, write that down too, on a scale of your own, with the date. It will not be evidence for anyone else, but it will keep you honest with yourself.

References

Sourced from the product label and PubMed-indexed research. The label facts on this page are transcribed from the printed SodaMelt panel, and every study is linked so you can read the abstract yourself. This page is general information, not medical advice.