Condition: Constipation

The joint AGA and ACG guideline makes a strong recommendation for polyethylene glycol in adults with chronic idiopathic constipation, alongside sodium picosulfate, linaclotide, plecanatide and prucalopride. Fiber supplements, senna and lubiprostone carry conditional recommendations on low-certainty evidence, and magnesium oxide and lactulose on very-low-certainty evidence. The moderate-certainty evidence supports the strong recommendations for PEG, linaclotide, plecanatide and prucalopride. Measured in: Adults with chronic idiopathic constipation, across the trials underpinning a GRADE-based guideline. A strong recommendation says the direction is settled, not that the effect is large. The guideline covers idiopathic constipation only, so it does not speak to opioid-induced constipation, secondary causes, or defecatory disorders, where a laxative is treating the wrong mechanism.

In patients who had already failed fiber and suppositories, five weekly biofeedback sessions gave major symptom improvement in 43 of 54 (80%) at six months against 12 of 55 (22%) on polyethylene glycol plus matched counseling, sustained at 24 months. A separate three-arm trial found biofeedback better than sham feedback and better than diet, exercise and laxatives on correcting dyssynergia, balloon expulsion time and complete spontaneous bowel movements. Measured in: 109 adults with chronic severe pelvic floor dyssynergia (Chiarioni) and 77 adults with dyssynergic defecation, 69 of them women (Rao). This applies only to people with a confirmed defecatory disorder, which needs anorectal manometry and a balloon expulsion test to establish. Both trials ran in specialist motility centers with experienced therapists, and results with a less experienced provider are unlikely to match. Access, not efficacy, is the limiting factor.

The American Gastroenterological Association recommends laxatives as first-line treatment for opioid-induced constipation, recommends naldemedine and naloxegol over no treatment for people who do not respond, and suggests methylnaltrexone. These peripherally acting antagonists block the opioid effect in the gut without reducing pain relief. Measured in: Adults on opioids with opioid-induced constipation, across the trials underpinning a GRADE-based guideline. The guideline covers medical management once opioid-induced constipation exists, and does not address whether the opioid itself is the right prescription. Unlike sedation and nausea, the bowel effect does not fade with continued use, so this is an ongoing problem rather than an early one.

311 of 473 participants (66%) responded to fiber against 134 of 329 (41%) on control. Stool frequency improved with a standardized mean difference of 0.72 and consistency by 0.32. Effects appeared only above 10 g a day and after four weeks or more. Psyllium and pectin were the fiber types with significant effects. Measured in: 1,251 adults with chronic constipation across 16 randomized controlled trials. Flatulence was consistently higher in the fiber arms, and it is the usual reason people stop before the four weeks the effect needs. The dose and duration thresholds mean most self-directed fiber attempts are below the level at which anything was measured.

Psyllium 10 g gave adequate symptom relief in 57% during the first month against 35% on placebo, and symptom severity fell 90 points against 49 at three months. Wheat bran 10 g did not clearly separate from placebo (it reached significance at three months in the primary analysis but not on stricter accounting), and the bran group had the highest early dropout, mostly people leaving because symptoms worsened. Measured in: 275 primary-care patients aged 18 to 65 with irritable bowel syndrome, randomized to psyllium (n=85), bran (n=97) or rice flour placebo (n=93) for 12 weeks. This was irritable bowel syndrome rather than chronic constipation alone, so it transfers to constipation-predominant IBS more securely than to slow-transit constipation. The bran finding is a non-separation from placebo plus a dropout pattern, which is weaker evidence of harm than a measured worsening would be.

Of 230 patients with chronic constipation who underwent anorectal manometry, balloon expulsion testing and whole-gut transit scintigraphy, 22% had dyssynergic defecation, 55% had slow transit constipation, 13% had both, and 36% were normal on both measures. Measured in: 230 consecutive patients at a tertiary motility center, 89% women. What could explain it instead: Referral bias dominates: these are people who reached a specialist motility unit, which selects hard for severity, for failed prior treatment, and for having insurance or access. The proportions in an unselected community population would be very different, and slow transit in particular is estimated at only 2 to 4% of the general population.. Retrospective, single center, and 89% female, so the numbers describe a referral stream rather than constipation at large. The finding that survives is the direction: most people still constipated after standard treatment have something specific that testing identifies.

Bisacodyl, sodium picosulfate, prucalopride and velusetrag all beat placebo on reaching three or more complete spontaneous bowel movements a week. On network meta-analysis no drug was superior to any other on the primary endpoints, and bisacodyl produced the largest change in spontaneous bowel movements per week of anything compared, prescription agents included. Measured in: 9,189 patients across 21 randomized controlled trials of prucalopride, lubiprostone, linaclotide, tegaserod, velusetrag, elobixibat, bisacodyl and sodium picosulfate. Indirect comparison through placebo rather than head-to-head trials, so the ranking is model-derived. The bisacodyl evidence rests on a single trial against nine for prucalopride, which makes its top position the least stable number in the analysis.

At eight weeks the complete response rate was 68% for Ma Zi Ren Wan, 57.7% for senna and 33.0% for placebo. Eight weeks after treatment stopped, Ma Zi Ren Wan held 47.4% while senna fell to 20.6% and placebo sat at 17.5%. Measured in: 291 patients meeting Rome III criteria for functional constipation, recruited from 8 clinics in Hong Kong between 2013 and 2015, randomized 1:1:1. Recruitment was to a specific Chinese medicine pattern of excessive syndrome, so the result does not transfer to a person whose constipation is Qi-deficient or cold in type. Single region, and the comparator senna dose of 15 mg daily is at the lower end of what is used clinically. The durability gap over senna is the more interesting finding and it rests on one trial.

Over weeks 1 to 8, electroacupuncture increased mean weekly complete spontaneous bowel movements by 1.76 against 0.87 for sham, a difference of 0.90. In weeks 9 to 20, after treatment ended, the gap widened to 1.09. 31.3% of the electroacupuncture group reached three or more per week during treatment, against 12.1% of the sham group. Measured in: 1,075 adults with chronic severe functional constipation across 15 hospitals in China, 536 electroacupuncture and 539 sham, 28 sessions over 8 weeks. A difference of roughly one extra bowel movement a week is measurable but small, and two thirds of the treated group did not reach the three-per-week threshold. Sham acupuncture at non-acupoints still involves needling, so it is not an inert control. All sites were in China, where trials of acupuncture report larger effects than trials run elsewhere.

Dried plums at 50 g twice daily improved complete spontaneous bowel movements and stool consistency more than psyllium at 11 g twice daily, with both arms delivering an identical 6 g of fiber a day. Measured in: 40 adults with mild to moderate constipation, 3 men and 37 women, mean age 38, single-blind crossover with 3 weeks per arm and a 1 week washout. Forty participants, 37 of them women, single center, and single-blind only, since a prune cannot be disguised as psyllium. Both arms matched on fiber grams, so the difference points at sorbitol and the polyphenols rather than fiber content. The crossover design means carryover between arms cannot be fully excluded despite the washout.

In a random stratified population sample keeping prospective records, a regular 24-hour defecation cycle appeared in only 40% of men and 33% of women. Once daily was the most common single habit and remained a minority practice in both sexes; a further 7% of men and 4% of women had a regular twice or thrice daily habit. Measured in: 838 men and 1,059 women from a random stratified sample of the East Bristol population, recording three consecutive defecations with stool form on a validated scale. What could explain it instead: Self-recording changes behavior and recall, and people who agree to keep a stool diary are not a random slice of the population even within a stratified sample. Diet and physical activity in 1980s Bristol also differ from most readers' circumstances, and both move stool frequency.. This describes a single English city in the late 1980s, so the exact percentages are local. What travels is the shape: daily is one pattern among several and not the standard the phrase 'regular' implies.

Overall symptom improvement in 70.6% on magnesium oxide 0.5 g three times daily against 25.0% on placebo over four weeks, with improvements in spontaneous bowel movements, Bristol stool form, colonic transit time and quality of life. Measured in: 34 Japanese adults with chronic constipation, all female, 33 completing (17 magnesium oxide, 16 placebo). Thirty-four participants is a small trial, single country, and every participant was a woman. Magnesium accumulates in impaired kidney function and hypermagnesemia from oral magnesium laxatives has caused serious harm in that group.

Exercise improved constipation symptoms with a relative risk of 1.97 (95% CI 1.19 to 3.27). Aerobic exercise alone gave a relative risk of 2.42 (95% CI 1.34 to 4.36). Eight of the nine trials used aerobic exercise, including qigong, walking and general physical movement. Measured in: 680 participants across 9 randomized controlled trials. The authors name a high risk of bias across the included trials and call for more rigorous work before the effect is treated as established. Wide confidence intervals, small trials, and no blinding is possible in an exercise study. The interventions were heterogeneous enough that no dose can be read off the result.

Both groups on a standardized 25 g fiber diet improved over two months. The group instructed to drink 2 liters of mineral water a day (achieving 2.1 liters) gained more stool frequency and cut laxative use further than the group drinking to thirst (achieving 1.1 liters). Measured in: 117 adults aged 18 to 50 with chronic functional constipation, randomized to ad libitum fluid or 2 liters of mineral water daily for two months. A 1998 single-center trial with no blinding, and the intervention was mineral water specifically, so its mineral content is not separable from the volume. It shows fluid adding to fiber, and it does not show that extra fluid alone helps someone who is already drinking normally.

Probiotics reduced whole-gut transit time by 12.4 hours, increased stool frequency by 1.3 bowel movements a week, and improved stool consistency compared with placebo. Measured in: Adults with functional constipation across 14 randomized controlled trials. The authors flag a high risk of bias across the included trials. The effect was strain-specific and most of the signal came from Bifidobacterium lactis, so a different strain on a shop shelf is a different intervention from the one that was measured.

Compared with bulk-forming laxatives, stimulant laxatives improved constipation more in pregnancy on moderate-quality evidence, with more diarrhea (moderate quality) and more abdominal discomfort (low quality), and no difference in women's satisfaction with treatment. Measured in: Pregnant women with constipation, across the randomized trials available to a Cochrane review. The review found few trials and called for more, so the comparison rests on a thin base. It compares two laxative classes with each other rather than either against no treatment, and it does not settle the safety question that most pregnant readers are actually asking.

After two weeks of no fiber, participants chose their own level. At six months the 41 on no fiber had gone from one bowel movement every 3.75 days to one a day, with bloating and straining reported by none. The 16 on reduced fiber went from every 4.19 days to every 1.9 days. The 6 who resumed high fiber stayed at roughly one every 6.83 days, all with bloating and straining. Measured in: 63 adults with idiopathic constipation, 16 men and 47 women, median age 47 (range 20 to 80), organic causes excluded by colonoscopy. Participants sorted themselves into the three groups rather than being randomized, so the people who felt better without fiber are the people who stayed without it. Single center, no control arm, no blinding. What it establishes is that some people with real constipation improve on less fiber, not how many or which ones.

Across 1,119 recorded bowel movements, using a defecation posture modification device raised the odds of complete emptying (odds ratio 3.64, 95% CI 2.78 to 4.77) and lowered the odds of straining (odds ratio 0.23, 95% CI 0.18 to 0.30). Without the device, bowel movements took about 25% longer. Measured in: 52 healthy volunteers, mean age 29, 40.1% female, two weeks without the device and two weeks with it. Healthy volunteers rather than people with constipation, so it does not tell you whether posture helps a clinical problem. No blinding is possible with a footstool, and the outcomes were self-reported by people who knew which condition they were in. The order was not randomized.

Rectal bleeding in primary-care patients aged 50 or over carried a pooled positive predictive value of 8.1% for colorectal cancer, and anemia 9.7%. Adding constipation to rectal bleeding gave a positive likelihood ratio of one or less, while adding a change in bowel habit gave 1.8 and weight loss 1.9. The highest-risk features were a palpable rectal or abdominal mass, and rectal bleeding combined with weight loss. Measured in: 23 diagnostic studies of symptomatic adults in primary care, each with at least 100 participants, searched to February 2010. Constipation adding nothing on top of rectal bleeding is not the same as constipation being safe to ignore: this measures its incremental value in patients who already have bleeding. The studies predate widespread fecal immunochemical testing and bowel screening programs, both of which change what reaches a GP.