This abstract was presented as the Public Health Nutrition Theme highlight.

Prunes (dried plums) are perceived to help maintain healthy bowel function⁽Reference Muller-Lissner, Kaatz and Brandt¹⁾ and preliminary research suggests they improve stool frequency and consistency in mild constipation⁽Reference Attaluri, Donahoe and Valestin²⁾. The high fibre and sorbitol content may aid in laxation⁽Reference Stacewicz-Sapuntzakis³⁾, however the effects of prunes on gastrointestinal (GI) health are not well researched and mechanisms of action are unclear.

We conducted a three-arm, parallel group, randomised controlled trial to investigate the dose dependent effect of prunes on GI outcomes. 120 healthy people with 3–6 stools/wk (mean age 35 years, 60% female) were randomised to high-dose prunes (120 g/d + 300 ml water/d), low-dose prunes (80 g/d + 300 ml water/d) or control (300 ml water/d) for 4 weeks. The primary outcome was stool weight, assessed by 7-day total stool collection. Other outcomes included stool frequency and consistency (stool diary), whole gut transit time (WGTT, radio-opaque markers) and GI symptoms (GI symptom rating scale). Outcome assessors were blinded to group assignment. Intervention effects were compared using analysis of covariance with baseline measurements as covariates, or Kruskal–Wallis test for non-parametric variables.

In the per-protocol analysis (n = 104), compared with control, 120 g prunes increased stool weight and 80 g prunes increased the number of bowel movements (BMs). 80 g prunes also increased spontaneous BMs, but prunes had no effect on complete BMs, complete spontaneous BMs, WGTT, stool consistency, stool water and stool pH.

WGTT, whole gut transit time; BMs, bowel movements; BSC, Bristol Stool chart (1–7)

Means are significantly different from control (ANCOVA with baseline as covariate): * P 0·035 **P 0·027 *** P 0·032

Of 14 GI symptoms recorded, incidence of flatulence was higher in the 120 g group (4·9 (SD 2·2) d/wk) and 80 g group (4·8 (SD 2·6) d/wk), compared to control (0·5 (SD 0·6) d/wk, P 0·009) and severity of flatulence was higher in the 120 g group (1·1 (SD 0·7)) and 80 g group (1·0 (SD 0·7)) compared with control (0·5 (SD 0·6), P 0·001) although actual severity was relatively low (0 = absent, 1 = mild, 2 = moderate, 3 = severe). Incidence of acid reflux was higher in the 120 g group, compared to control (0·7 (SD 1·4) d/wk vs 0·1 (SD 0·2) d/wk, P 0·02) as was its severity (0·1 (SD 0·2) vs 0·0 (SD 0·0), P 0·02).

In healthy individuals, prunes seem well tolerated and significantly increase stool weight and bowel movements, but do not decrease WGTT. Stool weight is inversely related to colon cancer risk⁽Reference Cummings, Bingham and Heaton⁴⁾. Therefore prunes have potential health benefits in the UK where average stool weight is low⁽Reference Cummings, Bingham and Heaton⁴⁾. These findings support the existing health claim for prunes in the maintenance of normal bowel function.