Table 3. Measures of compliance.
| MODIFIED INTENTION-TO-TREAT | PER-PROTOCOL | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Placebo (n = 27) | POM (n = 28) | Absolute Effect Size1 | Relative Effect Size2 | SE2 | t/χ2 | P3 | Placebo, Metabolite -ve (n = 15) |
POM, Metabolite +ve (n = 17) |
Absolute Effect Size1 | Relative Effect Size2 | SE2 | t/χ2 | P3 | |
| Days of consumption, median (IQR) | 23 (14, 40) |
18.5 (7, 30.5) |
6.8 | 0.44 | 4.17 | 1.6 | 0.11 | 23 (15, 40) |
18 (7, 24) |
9.7 | 0.71 | 4.89 | 2.0 | 0.06 |
| Maternal UA or DMEAG at enrollment, n (%) | 3 (11) | 2 (8)a | -3.1% | 0.64 | 0.08 | 1.00 | 1 (7) | 1 (7) | 0.0% | 1.00 | 0.14 | 1.00 | ||
| Maternal UA or DMEAG at delivery, n (%) | 11 (42)b | 17 (68)c | 25.7% | 1.49 | 0.13 | 3.4 | 0.07 | 0 (0) | 17 (100) | 100.0% | 31.11 | 0.17 | <0.001 | |
| Cord UA or DMEAG at delivery, n (%) |
8 (33)d | 14 (61)e | 27.5% | 1.69 | 0.13 | 3.6 | 0.06 | 0 (0) | 14 (93) | 93.3% | 25.78 | 0.17 | <0.001 | |
| Positive for metabolites in cord blood or maternal blood at delivery | 11 (42)f | 17 (68)g | 25.7% | 1.49 | 0.13 | 3.4 | 0.07 | 0 (0) | 17 (100) | 100.0% | 31.11 | 0.17 | <0.001 | |
DMEAG; dimethylellagic acid glucuronid; IQR–interquartile range; POM–pomegranate; SE–standard error; UA–urolithin A
1 Absolute effect size calculated as the mean difference for continuous variables, and the risk difference (%) for categorical variables (Placebo = reference)
2 Relative effect size calculated as Cohen’s d for continuous variables, and the relative risk for categorical variables (Placebo = reference). For analyses with zero in one or more cells, 0.5 was added to each cell prior to calculation of the relative risk and its standard error (SE). Corresponding SE are reported.
3 Fisher’s exact test (2-sided) used to compare proportions by group where expected cell size < 5.
a n = 25
b n = 26
c n = 25
d n = 24
e n = 23
f n = 26
g n = 25