. 2023 Jun 25;2(1):e000339. doi: 10.1136/bmjmed-2022-000339

Table 1.

Linear dose-response meta-analyses of body mass index and mortality outcomes: summary of main results

	No of studies	No of deaths/total No of men	Summary hazard ratio (95% CI) for random effects	Summary hazard ratio (95% CI) for fixed effects	Heterogeneity			Small study effects: Egger’s P value, Debray’s P value	τ²
	No of studies	No of deaths/total No of men	Summary hazard ratio (95% CI) for random effects	Summary hazard ratio (95% CI) for fixed effects	I² (%) (95% CI)	95% PI	Q value, P value	Small study effects: Egger’s P value, Debray’s P value	τ²
All cause mortality
Linear dose- response meta-analysis (per 5 unit increase in body mass index)*	25	13811/58 574	0.98 (0.93 to 1.04)	1.02 (0.99 to 1.04)	77 (66 to 84)	0.77 to 1.26	103,<0.01	0.13, 0.49	0.01
Mortality specific to prostate cancer
Linear dose- response meta-analysis (per 5 unit increase in body mass index)	16	3412/55 457	1.08 (1.00 to 1.17)	1.06 (1.01 to 1.10)	54 (19 to 74)	0.84 to 1.38	33,<0.01	0.23, 0.32	0.01

CI=confidence interval; PI=prediction interval.

*Number of deaths in an included publication was not given or could not be estimated, or both.