INTERNATIONAL JOURNAL OF FOOD SCIENCES AND NUTRITION 5
parallel group design, it was calculated that a sample size of 30 evaluable patients per group would have 80 % power, with an estimated standard deviation of 2.0, to detect a true treatment difference of 1.5 bowel movements per week. To allow for dropouts, 5 more subjects were recruited per group. The calculations assumed a 2-sided significance test carried out at the 5 % level. Subjects were randomised by a computer generated schedule to either the treatment or placebo groups (1:1) and were balanced using a block size of 6. All the statistical analyses were carried out using SPSS Version 22 for Windows (SPSS Inc., Chicago, IL). All the tests were two-sided and performed at the 5 % level of significance. Due to sphericity not being assumed, the Greenhouse – Geisser test was used to determine whether or not a statistical differ- ence existed in the number of bowel movements per week using the treatment group as a fixed factor and the number of bowel movements in pre-treatment as the baseline covariate. Cohen ’ s d was used to deter- mine the strength of any statistically significant result (Small effect: 0.2 d < 0.5; Medium effect: 0.5 d < 0.8; Large effect: d 0.8). Similar analysis was used to determine whether or not a statistical difference existed between groups in the average distance a reported Bristol stool type was away from type 1 each week. Subscores obtained from the PAC-QOL and PAC- SYM questionnaires were analysed using ANCOVA or an appropriate non-parametric test with the pre-study scores used as the covariate. The overall incidence of treatment-related adverse events were analysed by the Fisher ’ s Exact test.
randomised subjects completed the study and were included in data analysis (Figure 1).
Clinical response Primary outcome
There was an improvement in the number of bowel movements per week in the Lepicol group versus the placebo group of 1.082 in week 1 ( p ¼ .021) and 1.079 in week 2 ( p ¼ .049), with the baseline being the num- ber of bowel movements in the pre-treatment (Table 4, Figure 2). There was a trend towards an improvement in the Lepicol group from the pre-treatment to week 3 (increase of 0.87 bowel movements, p ¼ .112) and week 4 (increase of 0.69 bowel movements, p ¼ .259) versus placebo, but this was not significant. Secondary outcomes Stool consistency – Subjects reported the consistency of each bowel movement in their daily diaries using the Bristol Stool Form Scale (stool types 1 – 7, with the hardest stool being type 1 and softest type 7). Figure 3 indicates that there was a trend in both groups for the stool to become softer (i.e. move away from type 1), with the greatest improvement seen in the Lepicol group in week 2, when the average stool type was type 3, while the majority of bowel movements in the pla- cebo group were reported as closer to type 2. But no statistical difference ( p ¼ .217) was found between groups from week to week (with the pre-treatment taken as the baseline) in the distance the reported stool type was away from type 01. Quality of life Subjects ’ overall quality of life percentage score (across all 5 subcategories combined) was 12.033 % better in the Lepicol group versus placebo (Table 5). When the QOL subcategories were analysed there was a statistic- ally significant improvement of 15.2 % in the Lepicol versus placebo for “ intensity of symptoms ” ( p ¼ .030). There was an improvement of 10.5 % for the “ effects on daily life ” , but this was not significant ( p ¼ .055). Subjects in the Lepicol group were 8.6 % more content with their “ feelings on constipation ” and 13.1 % more content with “ life with constipation ” , but neither were significant (feelings p ¼ .155, life with constipation p ¼ .067). For “ satisfaction with their quality of life ” , a statistically significant increase of 28.5 % was found between the groups at the end of the study ( p ¼ .003
Results Subjects
From a total of 217 initial telephone interviewees, 92 were suitable for invitation for a screening visit (exclu- sions – 106 reported > 3 bowel movements/week, 3 IBS diagnoses, 8 disallowed medication, 3 chronic unstable disease, 2 had no symptoms in the previous 3 months, 3 unable to attend visits/communication issues). Twenty-three subjects did not meet the inclu- sion/exclusion criteria after the 2-week screening period (9 > 3 bowel movements/week following a 2 week diary, 9 screen fails, 5 chose not to enter study). Sixty-nine subjects were randomised to receive either test product ( n ¼ 35) or placebo ( n ¼ 34) and were included in the intent-to-treat population. All
Powered by FlippingBook