Fermented milk containing Bifidobacterium lactis DN-173 010 improved self-reported digestive comfort amongst a general population of adults. A randomized, open-label, controlled, pilot study

Authors


  • Addresses of the institution at which the work took place
    Two clinical centres (4-Front Research, Cheshire and Widnes, UK)

Correspondence to: Denis Guyonnet, Danone Research, Route Départementale 128, 91767 Palaiseau Cedex – FRANCE. Email: denis.guyonnet@danone.com

Abstract

AIM:  Some probiotics improve digestive comfort of people with Irritable Bowel Syndrome, but this needs confirmation in a healthy population. The objective of this pilot study was to investigate the effect of consuming fermented milk containing the probiotics Bifidobacterium lactis DN-173010 and yoghourt strains (test product) on digestive comfort and symptoms amongst adults without diagnosed gastrointestinal disorders.

METHODS:  The study was designed to approximate a real-life situation, by using a branded product in the intervention groups. In an open-label, randomized, controlled trial, 371 adults reporting digestive discomfort were randomized into three groups who had a daily consumption of either one or two pots of test product over 2 weeks, or to follow their usual diet. Digestive comfort and bother from digestive symptoms were assessed by questionnaire at baseline and follow-up (per protocol population n = 360). Self-reported change in digestive comfort and computed change between baseline and follow-up for each of 20 items were compared between groups (Cochran-Mantel-Haenszel test).

RESULTS:  A higher percentage of participants consuming the test product reported improved digestive comfort (1-pot group 82.5%; 2-pot group 84.3%), than controls (2.9%). Their self-reported change scores differed significantly (P < 0.001). For both test product groups, almost all symptom scores improved significantly more than controls (P < 0.001). There were no significant differences between 1-pot and 2-pot groups.

CONCLUSIONS:  This pilot study shows that daily consumption of a probiotic food in real-life conditions may be useful in improving digestive comfort and symptom experience of adults from general population. Further double-blind randomized controlled studies are required to confirm these health benefits.

INTRODUCTION

Digestive comfort or gastrointestinal (GI) well-being is something to which everyone aspires.1 Gut-derived symptoms are part of the normal physiological digestive process and GI symptoms are still common in the populations of developed countries.2 More than half of the population frequently experiences one or more unpleasant gut-related sensations,3 or change in stool consistency.4,5 Most available measures focus on symptoms rather than comfort and the boundary between normal and what individuals consider abnormal or troublesome is unclear. A German study showed that both healthy participants and patients with Irritable Bowel Syndrome (IBS) experienced GI symptoms, the only difference being the frequency of symptoms.6 The diagnosis of IBS is therefore based on the characterization of GI symptoms (duration, frequency and severity) associated with altered bowel function. The importance of the presence of GI symptoms from the population's perspective has been highlighted by the impaired health related quality of life of people with IBS.7,8

Probiotics are defined as ‘live microorganisms that, when ingested in adequate amounts, exert a health benefit to the host’.9 Some probiotic foods exert positive effects on digestive health.10 The interest of using probiotics to improve digestive dysfunction has been shown in a variety of GI conditions including IBS11,12 and inflammatory bowel disease.13 Few clinical trials have assessed the effects of probiotics in healthy populations on GI parameters and most of these studies showing positive effects have investigated the effect on bowel habits in small samples (30–80 participants).14–18 Other studies have investigated the effect of probiotics on GI functions such as GI transit time, with inconsistent results: improvement19,20 or no effect.21,22 The probiotic strain Bifidobacterium lactis DN-173 010 is recovered alive in stools in large quantities which demonstrates its potential to exert a beneficial effect throughout the GI tract.23–25 In healthy adults as well as elderly people, probiotic fermented milks containing Bifidobacterium lactis DN-173 010 alone or in association with the yoghurt symbiosis have been shown to decrease whole GI26 and/or colonic transit time.19,20 The effect of this fermented milk (FM) on colonic transit time was also demonstrated in IBS patients with predominant constipation.27 Interestingly, FM consumption in this IBS population was also associated with an improvement in GI symptoms and digestive comfort.28

The purpose of this pilot study was to assess whether consumption of FM containing B. lactis DN-173 010 and yoghurt symbiosis in real-life conditions may improve digestive comfort in a target population of healthy people having self-reported digestive discomfort. This study was also used to pilot an early draft of a new questionnaire to assess digestive comfort outside clinical populations.

METHODS

Study population

Between 5 April and 3 May, 2005, a total of 371 adults were recruited in the UK from two clinical centers (4-Front Research, Cheshire and Widnes). Healthy men and women aged 18–65 years with self-reported digestive discomfort and a self-reported stool frequency of 3–21 times per week were selected from amongst volunteers. Those rating their overall well-being (when considering digestive problems experienced in the previous 2 weeks) as either ‘well’ or ‘very well’ at the screening visit, or with GI symptoms requiring treatment were excluded, as were pregnant or breast-feeding women, women planning to become pregnant during the study, individuals with known lactose intolerance or known allergy to milk proteins, those having used laxatives or remedies to promote digestion and those who had consumed probiotics or the test FM in the previous two weeks. Digestive discomfort was assessed at the screening visit using a questionnaire exploring the frequency of four digestive problems each on a 4-point scale (‘regularly’, ‘often’, ‘from time to time’, ‘never’). ‘Discomfort’ was defined as the presence of two or more symptoms occurring ‘regularly’ or ‘often’.

Study Design

The study was a two-center, open-label, randomized, controlled trial, in parallel with three groups stratified by sex. Two product groups, each of 150 adults, assessed the effect of daily consumption of a FM containing B. lactis DN-173 010 on self-reported digestive comfort. A group of 70 formed a control group. After a screening visit to check their medical history and inclusion/exclusion criteria, eligible volunteers were randomized into one of the three groups: the 1-pot group consuming one pot of test FM daily for 14 days; the 2-pot group consuming two pots of test FM daily for 14 days; and the control group continuing with their normal diet without intervention. All were encouraged to maintain all other aspects of their dietary and physical exercise habits. Participants in the two test groups could choose the time at which they consumed the product, but members of the 2-pot group were instructed not to consume both at the same time. The study protocol was conducted in accordance with the Declaration of Helsinki and approved by an independent Ethical Committee. All recruits provided written informed consent before inclusion.

Products

The test FM was fermented milk (Activia, Danone), containing B. lactis DN-173 010 (1.25 × 1010 colony forming unit (cfu) per pot) together with the two classical yoghurt starters, S. thermophilus and L. bulgaricus (1.2 × 109 cfu/pot). It was supplied as multipacks (8 pots in mixed fruit flavors by pack, i.e., 2 strawberry, 2 apricots, 2 rhubarb and 2 figs). Each serving (one pot) contained 125 g. The product was not specially prepared for the study and was supplied by Danone in branded packaging, as marketed and sold to UK consumers.

Assessment of digestive comfort

Self-reported change in general digestive comfort was evaluated at the end of the study with one question, ‘Think back over the last 2 weeks, how has your digestive comfort changed?’, using a 5-point Likert scale (feel ‘much worse’; ‘slightly worse’; ‘neither better nor worse’; ‘better’; ‘much better’). For the purposes of summarizing the data, these scores were regrouped into three categories (‘worse’, ‘no change’ or ‘better’) and presented graphically.

Digestive comfort was assessed by a self-completed, 20-item questionnaire (Table 1) at baseline (screening visit, the day before the two test FM groups started consuming the product) and again at 2-week follow-up (after the 14 days in which the two test groups consumed the FM product). Participants answered each question by ticking the answer that best described how they had felt over the past 2 weeks (5-point Likert scales with response options depending on item wording). Thirteen questions assessed the bother from different digestive symptoms or troubles (Q1–Q4; Q7–Q15). Six questions explored the impact of digestive symptoms on other domains such as energy, vitality, daily life or the future (Q5–Q6; Q16–19). One other (Q20) assessed general well-being due to digestive problems.

Table 1.  Items of the questionnaire used
No.Item wordingResponse optionsImprovement
  1. Response: 5 response options (5-point Likert scale) were available for each item; both extreme possibilities of answers are indicated. Improvement: the sense (↓ for at least –1 point or ↑ for at least +1 point) of score change associated with an improvement is indicated by an arrow for each item.

Q1Have you been bothered by general digestive discomfort?not at all – extremely
Q2Have you been bothered by stomach ache or stomach pains?not at all – extremely
Q3Have you been bothered by your stomach gurgling when you don't feel hungry?not at all – extremely
Q4Have you been bothered by feeling bloated?not at all – extremely
Q5Have you had to loosen your belt or even lie down after meals?not at all – extremely
Q6Have your clothes or your waistband felt uncomfortably tight?not at all – extremely
Q7Have you been bothered by excessive or trapped wind (flatulence)?not at all – extremely
Q8Have you been bothered by a swollen stomach?not at all – extremely
Q9Have you been bothered by feeling heavy because of your digestive problems?not at all – extremely
Q10Have you been bothered by feeling lethargic because of your digestive problems?not at all – extremely
Q11Have you been bothered by feeling uncomfortable because of your digestive problems?not at all – extremely
Q12Have you been satisfied with your digestion?not at all – extremely
Q13Have you been bothered by diarrhoea?not at all – extremely
Q14Have you been bothered by constipation?not at all – extremely
Q15Have you been bothered by sluggish bowels?not at all – extremely
Q16Despite my digestive problems, I think that over the next few years, I will be able to achieve the things that matter to me (career, family life, retirement ... ).totally disagree – totally agree
Q17I attach little importance to my digestive pains, even if they can bother me in everyday life.totally disagree – totally agree
Q18Despite my digestive problems I can live a normal life.totally disagree – totally agree
Q19I feel that nothing I can do will change my digestive problems.totally disagree – totally agree
Q20Think back over the last 2 weeks, how would you score your general well-being due to any of the digestive problems described above?fell ill/not at all well – feel very well

Participants in the two test FM groups recorded their consumption of study products and their bowel movements daily, in a diary provided. The investigator recorded medications taken during the study and any adverse events declared by participants.

Statistical methods

The sample size calculation was based on the self-reported change score, administered at follow-up. It was anticipated on the basis of previous unpublished studies in a general population that 150 per test group would allow detection of a difference of 15% in improved digestive comfort between the three treatment groups. It was expected that about 65% of participants in the 1-pot group and 80% of the 2-pot group would claim to feel better on the follow-up self-reported change score. Fifty in the control group would guarantee a power of 80% if only 50% or fewer claimed any improvement in digestive comfort, with at least 65% of members of the 1-pot group asserting improvement.

After screening visits, 371 volunteers were accepted in the study and assigned to three groups (151 in 1-pot group, 150 in 2-pot group, 70 in control group) stratified by sex. All analyzes were performed using SPSS (V 15.0.1.1) and SAS Enterprise Guide (v 4.1) Statistical Software on the per protocol (PP) population. Baseline demographic and digestive comfort characteristics were compared using Pearson's chi-square tests to ensure similarity between the three groups.

The computed difference between the baseline and follow-up score was calculated for each of the 20 items, yielding a nine-point scale for each item (–4 to +4). Then, in order to summarize results for a given question, a change in response of between one and four points in the direction of improvement was interpreted as feeling ‘better’, no change (score of zero) as feeling ‘the same’, and a change of between one and four points in the direction of deterioration as feeling ‘worse’. The direction associated with an improvement (e.g., lower score for a symptom) is indicated for each item in Table 1. This coding is in agreement with recommendations determining a meaningful improvement or worsening of digestive symptoms as at least 1 step on a 7-step ordinal scale.29

The scores of the three groups on the self-reported digestive comfort change item, as well as the computed change scores for each of the 20 items were compared using non-parametric tests, since the data were not normally distributed. The Cochran-Mantel-Haenszel (CMH) test with modified ridit scores was used for both multiple comparisons between the three groups and posthoc pair-wise comparisons. A lower significance level (P < 0.001) than the one defined in the protocol (P < 0.05) was applied using Bonferroni's inequality, to take into account the multiplicity of tests of significance for posthoc tests comparing pairs of groups (1-pot or 2-pot groups vs. control; 1-pot vs. 2-pot group).

RESULTS

Demographics and baseline characteristics

From 371 screened participants, 362 completed the study (ITT population). Two were excluded for major protocol deviation, leaving a per protocol population of 360 participants (1-pot 144; 2-pot 147; control 69). Demographics and baseline characteristics and baseline questionnaire responses are presented in Tables 2 and 3, respectively. No significant differences between groups were observed for age, sex ratio and digestive symptom frequency, nor for baseline questionnaire responses.

Table 2.  Baseline characteristics of participants in per protocol population (n = 360)
 1-pot group
(n = 144)
2-pot group
(n = 147)
Control group
(n = 69)
  1. Age and digestive symptom frequency are expressed as means ± SD. No significant (P > 0.05) differences were found between groups in all the variables tested.

Age (years)39.5 ± 12.338.5 ± 11.038.9 ± 10.8
Sex ratio (female/male)111/33110/3755/14
Digestive symptom frequency10.4 ± 5.3 9.9 ± 5.7 9.3 ± 5.1
Table 3.  Baseline and follow-up responses for the questionnaire items in the per protocol population (n = 360)
 1-pot group (n = 144)2-pot group (n = 147)Control group (n = 69)
BaselineEndBaselineEndBaselineEnd
  1. All data are expressed as means ± SD. No significant (P > 0.05) differences were found between groups on any item for baseline responses.

Q1. general digestive discomfort3.49 ± 0.812.07 ± 0.773.29 ± 0.882.03 ± 0.753.28 ± 0.892.61 ± 0.75
Q2. stomach ache or pains2.77 ± 0.981.61 ± 0.752.84 ± 1.101.59 ± 0.762.90 ± 1.092.25 ± 0.86
Q3. stomach gurgling2.85 ± 1.041.60 ± 0.692.67 ± 1.141.71 ± 0.812.71 ± 0.872.28 ± 0.89
Q4. feeling bloated3.74 ± 0.811.66 ± 0.683.53 ± 1.011.71 ± 0.743.49 ± 1.022.80 ± 1.07
Q5. loosen belt after meals3.21 ± 0.871.62 ± 0.753.04 ± 0.911.65 ± 0.742.96 ± 0.982.59 ± 0.98
Q6. clothes uncomfortably tight3.38 ± 0.731.81 ± 0.743.26 ± 0.761.85 ± 0.773.29 ± 0.842.83 ± 0.89
Q7. excessive or trapped wind3.31 ± 0.961.97 ± 0.953.35 ± 1.112.24 ± 1.143.17 ± 1.032.62 ± 0.93
Q8. a swollen stomach3.29 ± 1.051.58 ± 0.733.20 ± 1.131.73 ± 0.733.16 ± 1.112.58 ± 1.02
Q9. feeling heavy3.20 ± 1.001.56 ± 0.743.14 ± 1.091.56 ± 1.123.10 ± 1.032.67 ± 1.00
Q10. feeling lethargic2.90 ± 1.061.43 ± 0.672.82 ± 1.171.37 ± 0.632.86 ± 1.082.41 ± 0.93
Q11. feeling uncomfortable3.38 ± 0.951.69 ± 1.203.16 ± 0.961.74 ± 0.733.30 ± 0.942.71 ± 0.94
Q12. satisfied with your digestion2.06 ± 1.003.60 ± 1.022.15 ± 0.963.58 ± 1.042.12 ± 2.232.23 ± 0.88
Q13. diarrhoea1.97 ± 0.921.24 ± 0.461.97 ± 1.041.46 ± 0.771.93 ± 0.931.45 ± 0.72
Q14. constipation2.74 ± 1.221.49 ± 0.722.54 ± 1.181.48 ± 0.732.52 ± 1.201.97 ± 1.04
Q15. sluggish bowels2.81 ± 1.141.63 ± 0.792.53 ± 1.081.53 ± 0.722.52 ± 1.022.17 ± 0.91
Q16. impact on future3.92 ± 0.874.10 ± 1.053.88 ± 0.944.12 ± 0.983.57 ± 1.223.61 ± 1.09
Q17. importance of digestive pains2.99 ± 1.003.09 ± 1.262.93 ± 1.163.11 ± 1.282.75 ± 1.143.00 ± 1.14
Q18. impact on daily life3.78 ± 1.004.15 ± 1.013.71 ± 1.054.16 ± 0.983.36 ± 1.203.81 ± 1.00
Q19. Feeling about things that could change digestive problems2.54 ± 0.951.97 ± 0.862.59 ± 1.022.11 ± 0.992.48 ± 0.992.61 ± 1.05
Q20. general well-being2.62 ± 0.503.92 ± 0.722.63 ± 0.513.94 ± 0.692.68 ± 0.473.22 ± 0.64

Assessments

Self-reported change in general digestive comfort. Response to the single change item administered at follow-up is shown in Fig. 1. The percentage reporting an improvement in digestive comfort was higher in both test product groups (82.5% and 84.3% in 1-pot and 2-pot groups, respectively) compared with the control group (2.9%). The between-group differences were confirmed using the CHM test, each of the two product groups differing from the control group (P < 0.001). There was no significant difference between 1-pot and 2-pot groups.

Figure 1.

Change in perceived digestive comfort (single self-reported change question completed at follow-up) in per protocol population (n = 360) at the end of the study. Scores are expressed as percentage of participants per category (worse, no change or better digestive comfort).

Digestive symptoms. Table 4 shows the percentage of participants reporting a decrease in the bother from each digestive symptom and the computed change in the score for each item. Significantly greater improvements (P < 0.001) for both test product groups vs. the control group were observed for computed change between end of study and baseline for almost all symptoms. However, using the Bonferroni correction accepting P < 0.001 for posthoc tests, Q2 (stomach ache/pains), Q13 (diarrhea) and Q14 (constipation) did not differ between the three groups. Moreover, the beneficial effect for some symptom items was greater for the 1-pot group than the 2-pot group. Computed change for Q1 (general digestive discomfort) differed (P < 0.001) only between the 1-pot group and control group whilst the 2-pot group did not differ significantly (P < 0.002) from the control group. The same was true of Q3 (stomach gurgling) and Q15 (sluggish bowels). Computed change scores ranged from –0.51 (Q13 diarrhea) to –2.08 (Q3 feeling bloated) in the test product groups and from –0.35 (Q15 sluggish bowel item) to –0.70 (Q3 feeling bloated) in control group.

Table 4.  Change in bother from digestive symptoms in the per protocol population (n = 360)
 1 pot group
(n = 144)
2 pots group
(n = 147)
Control group
(n = 69)
  1. Scores are expressed as percentage of participants reporting an improvement and then as mean change ± SD in brackets. Analysis used CMH test for overall difference between the 3 groups and pair-wise comparisons (* P < 0.001 between test group vs. control group). No significant (P > 0.001) differences were found between 1-pot and 2-pot groups on any item.

Q1. general digestive discomfort84.0%78.9%55.1%
(–1.42 ± 1.01*)(–1.26 ± 1.00)(–0.67 ± 0.82)
Q2. stomach ache or pains73.6%76.2%52.2%
(–1.16 ± 1.16)(–1.24 ± 1.11)(–0.65 ± 1.12)
Q3. stomach gurgling when not hungry72.7%64.4%39.1%
(–1.24 ± 1.11*)(–0.95 ± 1.23)(–0.43 ± 1.04)
Q4. feeling bloated93.8%87.0%53.6%
(–2.08 ± 0.98*)(–1.82 ± 1.13*)(–0.70 ± 1.12)
Q7. excessive or trapped wind74.3%63.9%52.2%
(–1.34 ± 1.33*)(–1.11 ± 1.35*)(–0.55 ± 0.87)
Q8. a swollen stomach81.1%81.6%45.0%
(–1.71 ± 1.16*)(–1.48 ± 1.10*)(–0.58 ± 0.90)
Q9. feeling heavy because of your digestive troubles79.9%83.7%46.4%
(–1.64 ± 1.20*)(–1.56 ± 1.12*)(–0.43 ± 0.98)
Q10. feeling lethargic because of your digestive troubles77.8%73.5%46.4%
(–1.47 ± 1.13*)(–1.45 ± 1.23*)(–0.45 ± 1.02)
Q11. feeling uncomfortable because of your digestive troubles84.7%76.2%50.7%
(–1.69 ± 1.20*)(–1.41 ± 1.09*)(–0.59 ± 1.05)
Q12. satisfied with your digestion76.4%74.1%27.6%
(1.53 ± 1.36)(1.43 ± 1.40)(0.12 ± 0.92)
Q13. diarrhoea56.6%44.9%37.7%
(–0.73 ± 0.90)(–0.51 ± 1.04)(–0.48 ± 0.76)
Q14. constipation72.9%61.9%43.5%
(–1.25 ± 1.24)(–1.05 ± 1.20)(–0.55 ± 1.01)
Q15. sluggish bowels71.6%65.3%40.6%
(–1.17 ± 1.27*)(–1.00 ± 1.12)(–0.35 ± 0.91)

Table 5 provides the participants’ assessment of the impact of digestive symptoms on seven other domains. The percentage of test group participants with improved scores was greater than for the control group for four items (Q5, Q6, Q19 and Q20), and computed change scores for these items differed significantly (P < 0.001) between groups on posthoc tests, with the two test groups having greater improvement than the control group. However, the differences for Q19 did not survive the Bonferroni correction. Two of the significant items (Q5, Q6) both concerned tightness of clothes around the waist. There was no significant difference between 1-pot and 2-pot groups for any item.

Table 5.  Change in the impact of digestive symptoms on other domains in the per protocol population (n = 360)
 1 pot group
(n = 144)
2 pots group
(n = 147)
Control group
(n = 69)
  1. Scores are expressed as percentage of participants reporting an improvement and then as mean change ± SD in brackets. Analysis used CMH test for overall difference between the 3 groups and pair-wise comparisons (* P < 0.001 between test group vs. control group). No significant (P > 0.001) differences were found between 1-pot and 2-pot groups on any item.

Q5. loosen belt or even lie down after meals86.1%82.2%43.5%
(–1.59 ± 1.00*)(–1.40 ± 0.91*)(–0.36 ± 0.86)
Q6. clothes or waistband felt uncomfortably tight88.9%83.0%43.5%
(–1.57 ± 0.95*)(–1.41 ± 0.95*)(–0.45 ± 0.85)
Q16. impact of digestive problems on future32.9%33.3%26.1%
(0.19 ± 1.01)(0.23 ± 1.19)(0.04 ± 1.49)
Q17. importance of digestive pains30.5%34.7%37.7%
(0.10 ± 1.24)(0.18 ± 1.40)(0.25 ± 1.21)
Q18. impact of digestive problems on daily life36.8%43.5%36.2%
(0.38 ± 1.16)(0.44 ± 1.14)(0.45 ± 1.30)
Q19. Feeling about things that could change digestive problems53.5%46.9%27.6%
(–0.57 ± 1.14)(–0.48 ± 1.09)(0.13 ± 1.06)
Q20. general well-being linked to digestive troubles83.2%85.7%44.9%
(1.29 ± 0.87*)(1.31 ± 0.82*)(0.54 ± 0.74)

DISCUSSION

The present pilot study shows that an intervention protocol assessing the effect of dairy food product on digestive comfort in a general population is feasible. The results also show that the consumption of a commercially available FM containing specific probiotic strains may exert positive effects on self-reported digestive comfort in a general population of healthy adults in real-life conditions. Beyond this global improvement, the bother from most of the digestive symptoms assessed was reduced in a high and significant percentage of participants. No significant differences were observed between 1 or 2 servings per day. The intervention did not result in adverse digestive comfort and therefore the consumption of the tested FM is safe and well tolerated. The results of this study are consistent with other studies conducted in double-blind placebo-controlled trials, documenting the beneficial effects of the same product on digestive functions such as GI transit time in healthy populations19,20 and IBS,27 as well as improvements in digestive comfort in an IBS population.28

The key features of the open-label design were compatible with the purpose of the study, which was to assess the potential benefit of consumption of a probiotic food in real-life conditions, in a group recruited from the general population as having self-reported digestive problems not confirmed by clinical diagnosis. The study was adequately controlled for this purpose, as it was clearly not designed to compare the effect of the tested FM vs. a control product that had already been done in double-blind controlled studies.20,27,28 This type of open design has been used in IBS trials for other probiotics.30,31 We acknowledge that the main limitation of this study design is that it may overestimate the percentage of participants experiencing better conditions, but it should be noted that not all the items showed an improvement and the improvement was in some cases more positive for the 1-pot group than for the 2-pot group. Moreover, advertising may have affected participant's expectations of the product. No studies have been conducted to determine the part of advertising on the overall benefit perceived by participants in this type of subjective assessment study. A recent meta-analysis of the placebo effect in IBS trials, using complementary and alternative medicine, including dietary intervention as probiotics, showed a mean placebo rate of 42.6%.32 The use of a control group, even without a specific control product treatment, allows determination of the regression to the mean and the natural variation (i.e., improvement or worsening) of these parameters within this population.33 In the present study, the control group improved in almost every scale with a range from 3 to 55% (3% for digestive comfort; from 28 to 55% for bother from symptoms). The improvement that could be compared to a placebo effect could be linked to a modification of eating behavior, participants in control group behaving more healthily due to their participation in a trial. The features of the design associated with well-documented ‘known placebo effect’ in the field of digestive comfort, as highlighted in IBS, may explain in part both the high percentage of participants who experienced better digestive comfort and the high difference observed between test product groups and control group.

This study used a self-administered questionnaire comprising general questions about digestion, including self-reported change in digestive comfort, and item-by-item assessment of the bother from digestive symptoms. Most symptom items demonstrated sensitivity to change, but it is possible that increasing the range of response options to seven (rather than five) may increase the chances of detecting improvements, especially in symptoms such as stomach aches and pains, diarrhea, constipation and sluggish bowels, which were not considered particularly bothersome at baseline (Table 3). Among the questions asking about other domains of life (Table 5), closer examination of items Q16, 17, 19 and 20 revealed wording that might be confusing to respondents. Furthermore, the results from these items were hard to interpret. Q16 referred to achieving in the domains of ‘career, family life and retirement’ over the next few years, though one or more of these areas of life were likely to be irrelevant to many respondents. Q17 had complex wording, asking about the ‘importance’ attached to digestive problems, ‘even if they can bother me in daily life’, and there is no reason to suppose that the level of importance would change as a result of eating the test product. Q19 assumed that the respondent had digestive problems, which may not have been true at follow-up, making this question hard to answer. Finally, Q20 asked ‘Think back over the last 2 weeks, how would you score your general well-being due to any of the digestive problems described above?’ was lengthy and the clumsy causal phrasing of the final clause may have been confusing. These items have been modified or excluded from a more recent version of the measure (now called the Digestive Comfort Questionnaire (DCQ)), which also has a better balance between positively and negatively worded items than in the present study, in which items most were negatively worded. Validation of the new DCQ in an undergraduate student population indicated that subscale scores can be computed, which will reduce the need for so many tests of significance.34 In the absence of relevant biological marker to assess digestive comfort or digestive symptom relief, Patient Reported Outcomes (PROs) are recognized as the most relevant tools for the assessment of drug efficacy in IBS trials.29 However, at the time of the present study, there was no validated PRO to assess digestive comfort or GI well-being in populations without a clinical diagnosis of bowel disorders, which may explain the dearth of published trials investigating the effect of dietary intervention in this area. Another specific health related quality of life questionnaire (Food and Benefit Assessment Quality of life questionnaire (FBA)), has been developed recently to assess the effect of dietary intervention, such as consumption of specific probiotic foods, on digestive comfort and other concepts such as vitality.35 This study showed that even a population of adults without clinical diagnosis of GI disorders experienced digestive discomfort and bother from digestive symptoms which is in agreement with previous observational studies, demonstrating that people from the general population frequently report digestive troubles.2,6,36 Future studies using a validated PRO such as the DCQ or FBA will provide data that will allow normative values for the general population to be generated. Investigations on the effects of probiotics in healthy adults have focused on the measurement of GI transit or bowel function demonstrating some positive effects.10,14–17,19,20 Unfortunately, the benefit perceived by participants in these trials was not assessed and therefore, the real end-benefit of such functional changes for the consumers can be open to question. The trial reported here showed that participants consuming the FM containing B. lactis DN-173010 strain experienced an improvement in their perceptions of digestive comfort, even though they did not have diagnosed IBS. Interestingly, these results are consistent with other positive effects on GI functions in healthy populations19,20,26,37 and on digestive comfort in IBS.27,28 Positive results on ‘bloated feelings’ are particularly interesting because they confirm the effect of the tested product on bloating, which had already been observed in an IBS population.27,28 This interest is reinforced by the observation that ‘feeling bloated’ is the most bothersome symptom reported by participants in this study. The high prevalence of bloating/′feeling bloated′ in the general population as well as in IBS has been already described.3,38 In contrast, no effect was observed on diarrhea, which was the least bothersome digestive trouble in this study at baseline and therefore, the scores may have been subject to a ceiling effect, so that no improvement could be detected. Importantly, no diarrhea side-effect associated with the consumption of Activia has been reported since the start of the marketing of this product 20 years ago, so it is likely that these scores reflect no change one way or the other. Notably, general digestive discomfort, stomach gurgling and sluggish bowels improved significantly more than the control group in the 1-pot group but not the 2-pot group. It is possible that, from the point of view of consumers, one pot of the FM product a day could exert a stronger effect than two pots. Moreover, when consumers are buying the product, rather than receiving it as part of a trial, the ‘one-pot a day’ message may be more appealing.

Together these results point towards an overall positive effect for this probiotic food on gut flora and symptoms in different conditions (i.e., in people in the general population with digestive discomfort or an IBS population). This is in line with the importance of maintaining a well-balanced gut flora to maintain normal GI functions.39 Recent studies have shown that unbalanced gut flora is involved in IBS, especially in some related symptoms such as bloating or altered bowel functions.11,40,41 It could be hypothesized that an alteration in gut flora, even in non-diseased individuals, could lead to some GI symptoms or digestive discomfort. The use of some specific probiotic foods could help to avoid these disturbances and related digestive symptoms or to facilitate the restoration of a ‘normal or well-balanced’ gut flora. Recent development of new molecular method of gut flora analysis in stools will allow an investigation of these potential alterations in the coming years.41,42

This pilot study provides first evidence of the interest of testing the benefit of the tested FM on digestive comfort in a general population. Findings of this trial show that fermented milk containing a specific probiotic strain may improve digestive comfort in a general population when consumed in real-life conditions. Future double-blind, randomized, controlled studies are required to confirm the beneficial effects observed in the present study.

ACKNOWLEDGMENTS

The authors thank Françoise Tondu for her excellent review of the statistical analysis and also Anne-Laure Berthet and Fazia Tebah for technical assistance.

This study was supported by a grant from Danone. Denis Guyonnet and Blandine Stefani are employees of Danone group; Alison Woodcock and Claire Hall were paid consultants to Danone Research.

Ancillary