Table 6.

 Association between diet group and intake of dietary fibre and risk of diverticular disease stratified by sex

	All participants				Men				Women
	Cases	Relative risk (95% CI)*	P value		Cases	Relative risk (95% CI)*	P value		Cases	Relative risk (95% CI)*	P value
Diet group (P=0.699 for interaction†):
Meat eater	633	1.00	0.003‡		155	1.00	0.125‡		478	1.00	0.025‡
Fish eater	77	0.91 (0.71 to 1.16)			17	1.04 (0.61 to 1.75)			60	0.88 (0.66 to 1.16)
Vegetarian or vegan	102	0.69 (0.55 to 0.86)			26	0.65 (0.41 to 1.02)			76	0.71 (0.55 to 0.92)
Englyst fibre intake (g/day) (P=0.223 for interaction†)§:
1 (lowest fifth)	184	1.00	<0.001¶		42	1.00	0.263¶		142	1.00	<0.001¶
2	177	0.86 (0.69 to 1.06)			48	1.04 (0.67 to 1.60)			129	0.82 (0.64 to 1.04)
3	164	0.76 (0.61 to 0.96)			32	0.63 (0.38 to 1.04)			132	0.81 (0.63 to 1.04)
4	160	0.72 (0.57 to 0.92)			42	0.83 (0.51 to 1.34)			118	0.70 (0.53 to 0.92)
5 (highest fifth)	127	0.59 (0.46 to 0.78)			34	0.77 (0.45 to 1.31)			93	0.56 (0.41 to 0.76)

*Stratified by sex, method of recruitment, and region of residence and adjusted for smoking, education level, Townsend deprivation index, hyperlipidaemia, receiving long term medical treatment, ever used oral contraceptives, ever used hormone replacement therapy, and BMI, where appropriate and using separate models for each subset.

†Tests of interaction performed with data for all participants by adding diet group or fifths of dietary fibre intake × sex interaction to model as appropriate.

‡χ² test.

§Further adjusted for sex specific fifths of energy intake.

¶P for trend obtained by replacing fifths of dietary fibre by continuous variable equal to sex specific median in each fifth.