Skip navigation


We all know what it is, but it is hard to define. Bowel movement patters show that 90% of people in Western countries have between three bowel movements a day to three per week. Constipation is often defined as fewer than three bowel movements a week, though symptoms like straining, passing hard stools and inability to defecate when desired, together with abdominal pain also form part of a diagnosis.

The use of laxatives, both prescribed and non-prescribed, is common, and constipation is thought to be common, especially in older people, women, and people with poor diets. So one might think that there would be a wealth of information from clinical trials on the effectiveness of laxative agents - that we would know that they work, and how well they work.

Bandolier thought so too. So it was a surprise to see from a systematic review [1] that we don't have as much information as we thought, or would like.


This is a good review, with great searching (though some might quibble about the exclusion of reports not in English), and an interesting description about a large number of trials which did not pass muster for various reasons. It makes a good read, though the journal is not readily available and Bandolier had to obtain it through the British Library.

Trials were included if they looked at treatment of constipation in adults for at least two weeks (minimum of one week on treatment or control).


They found 36 trials with 1,815 subjects, of whom 40% were over 60 years, and 70% of whom were women. Many of the trials had poor design. There were 25 different laxative or dietary fibre therapies. Twenty trials compared laxative with placebo or regular diet, and 16 were comparisons with other laxatives.

Bowel movements

The number of bowel movements per week in controls ranged from a mean (or median) of 1.5 to 7.1. Only six of 16 reports had mean bowel movements per week below 3. Laxatives increased the number of bowel movements (Figure) from a control mean (weighted by number of patients) of 3.5 per week to 5.0 per week. The increase was 1.4 bowel movements per week (95% confidence interval 1.1 to 1.8).

Bulk laxatives (six trials) gave an average weighted mean increase of 1.4 (0.6 to 2.2), and other agents (seven trials) gave a weighted mean increase of 1.5 (1.1 to 1.8).

Overall symptom improvement was reported as being significantly better for laxatives than control in nine of eleven trials which measured them.

Results from the trials which compared different laxatives were largely uninterpretable.


Laxatives work. Perhaps there's nothing new in that. What is disappointing is that there is so little evidence to allow us to see which is best. What we can say is that for adult patients with chronic constipation, bran or bulk laxatives work as well as anything. The amount of bran used in the trials ranged from 0.5 gram to 24 grams a day, equivalent to from a quarter of a serving of fruit, vegetables or cereal a day to 12 servings a day. Advising adults with chronic constipation to eat more fruit and vegetables and have some bran seems to be the best evidence available.


  1. SM Tramonte, NB Brand, CD Mulrow et al. The treatment of chronic constipation: a systematic review. Journal of General Internal Medicine 1997 12: 15-24.

previous or next story in this issue