Relationship between bladder cancer and total fluid intake: a meta-analysis of epidemiological evidence

Table 3 Stratified pooled OR and 95% CIs for the highest vesus lowest level of total fluid intake and bladder cancer risk

Subgroup	Number of studies	Pooled OR (95% CI)	P values for interaction or moderation effect between the stratified factor and fluid consumption
All studies	21	1.06 (0.88-1.27)	<0.001
Study design
Cohort studies	4	0.98 (0.85-1.13)	0.480
Case-control studies	17	1.07 (0.86-1.33)	<0.001
Gender
Female	11	1.00 (0.74-1.35)	0.012
Male	10	1.22 (0.95-1.57)	<0.001
Geographical Region
Europe and America	17	1.20 (1.02-1.42)	<0.001
Asia	4	0.27 (0.10-0.72)	<0.001
Asia*	2	0.78 (0.58-1.06)	0.234
Publication time
After 2000	14	0.92 (0.77-1.10)	<0.001
Before 2000	7	1.51 (0.93-2.47)	<0.001
Type of control
Hospital	8	0.71 (0.44-1.15)	<0.001
Population	13	1.21 (1.01-1.45)	<0.001
Smoking status
Never	6	0.93 (0.73-1.19)	0.471
Ever	6	0.80 (0.62-1.02)	0.032
Adjustments #
≥ 3 factors	14	1.12 (0.94-1.33)	<0.001
< 3 factors	7	0.88 (0.45-1.70)	<0.001
Type of beverage
Water	11	1.05 (0.88-1.25)	<0.001
Coffee	14	1.17 (1.03-1.33)	0.005
Tea	7	0.96 (0.77-1.21)	0.044
Green tea	5	0.76 (0.66-0.95)	0.317
Black tea	4	0.80 (0.65-0.97)	0.291
Alcoholic beverage
Wine	5	0.90 (0.61, 1.32)	<0.001
Beer	6	1.09 (0.79-1.49)	0.001
Liquor	5	0.98 (0.77-1.24)	0.094
Number of beverages
≥ 5	16	1.19 (0.97-1.46)	<0.001
< 5	4	0.60 (0.34-1.06)	<0.001

*except papers that did not adjust for smoking habits and that calculated total fluid intake based on 3 beverages only.
#adjusted for at least three of the following factors: age, gender, smoking, BMI, education,energy intake and place of residence.

ISSN: 1477-7819