|
Outcome of interest
|
No. of studies
|
OR (95 % CI)
|
p
heterogenity
|
I
2 (%)
|
p for interaction
|
|---|
|
Total dietary fiber
|
17
|
0.89 (0.77, 1.01)
|
0.005
|
53.6
| |
|
Study design
|
|
Cohort
|
5
|
0.94 (0.77, 1.11)
|
0.004
|
74.3
|
0.202
|
|
Case-control
|
12
|
0.82 (0.68, 0.96)
|
0.277
|
17.0
| |
|
Study quality
|
|
Low
|
8
|
0.73 (0.56, 0.90)
|
0.335
|
12.2
|
0.033
|
|
High
|
9
|
0.96 (0.83, 1.08)
|
0.04
|
51.7
| |
|
Geographical region
|
|
Europe
|
7
|
0.90 (0.71, 1.09)
|
0.01
|
63.5
|
0.937
|
|
North America
|
6
|
0.90 (0.64, 1.16)
|
0.059
|
53.1
| |
|
Japan
|
2
|
0.95 (0.72, 1.17)
|
0.41
|
0
| |
|
Assessment
|
|
Interview
|
11
|
0.94 (0.79, 1.10)
|
0.313
|
13.8
|
0.931
|
|
Questionnaire
|
4
|
0.93 (0.76, 1.09)
|
0.02
|
69.7
| |
|
Family history
| | | | | |
|
Yes
|
8
|
0.84 (0.62, 1.05)
|
0.002
|
69.4
|
0.44
|
|
No
|
9
|
0.94 (0.81, 1.08)
|
0.187
|
29.0
| |
|
BMI
|
|
Yes
|
8
|
0.87 (0.66, 1.08)
|
0.001
|
70.3
|
0.695
|
|
No
|
9
|
0.92 (0.79, 1.06)
|
0.21
|
26.4
| |
|
Energy intake
|
|
Yes
|
14
|
0.91 (0.78, 1.04)
|
0.007
|
55.1
|
0.507
|
|
No
|
3
|
0.81 (0.54, 1.07)
|
0.14
|
49.2
| |
|
Multiple confoundersa
|
|
Yes
|
6
|
0.82 (0.54, 1.09)
|
0.306
|
14.4
|
0.387
|
|
No
|
11
|
0.95 (0.84, 1.05)
|
<0.001
|
77.7
| |
|
Tumor stage
|
|
Local
|
3
|
0.98 (0.89, 1.08)
|
0.24
|
30.5
|
0.562
|
|
Advanced
|
4
|
0.93 (0.79, 1.07)
|
0.24
|
29.3
| |
|
Source of intake
|
|
Cereal fiber
|
3
|
1.05 (0.94, 1.16)
|
0.52
|
0
|
0.02
|
|
Fruit fiber
|
3
|
0.92 (0.81, 1.03)
|
0.55
|
0
| |
|
Vegetable fiber
|
3
|
0.87 (0.53, 1.21)
|
0.001
|
84.8
| |
|
Legume fiber
|
1
|
0.55 (0.32, 0.95)
|
NA
|
NA
| |
|
Solubility
|
|
Soluble fiber
|
2
|
0.87 (0.52, 1.22)
|
0.13
|
57.2
|
0.777
|
|
Insoluble fiber
|
3
|
0.80 (0.46, 1.13)
|
0.005
|
81.0
| |
-
aMultiple confounders refer to effect estimates adjusted for at least family history, BMI, and energy intake
