Ratio of n-3/n-6 PUFAs and risk of breast cancer: a meta-analysis of 274135 adult females from 11 independent prospective studies

Table 3 Sensitivity analysis for the relationship between intake n-3/n-6 PUFAs ratio and breast cancer risk

Sensitivity analysis	Diet ratio of n-3/n-6				Serum PL ratio of n-3/n-6				Intake ratio of n-3/n-6
Sensitivity analysis	N	Pooled RR (95% CI)	Heterogeneity		N	Pooled RR(95% CI)	Heterogeneity		N	Pooled RR(95% CI)	Heterogeneity
	N	Pooled RR (95% CI)	I ² Degree	P	N	Pooled RR(95% CI)	I ² Degree	P	N	Pooled RR(95% CI)	I ² Degree	P
Exclusion of studies with potential selection bias [20, 32]	5	0.87 (0.80, 0.96)	0.00% (Low)	0.5s9	4	0.73 (0.51, 1.01)	0.00% (Low)	0.67	9	0.86 (0.79, 0.94)	0.00% (Low)	0.72
Exclusion of studies without covariates adjusted [18]	6	0.90 (0.82, 0.99)	17.30% (Low)	0.30	4	0.82 (0.56, 1.22)	36.50% (Low)	0.17	10	0.89 (0.81, 0.99)	19.10% (Low)	0.26
Contour-enhanced funnel plots of fixed effect model	6	0.91 (0.84, 0.98)	17.30% (Low)	0.30	5	0.90 (0.68, 1.19)	21.50% (Low)	0.27	11	0.91 (0.84, 0.98)	11.40% (Low)	0.33
Contour-enhanced funnel plots of random effect model	6	0.90 (0.82, 0.99)	17.30% (Low)	0.30	5	0.87 (0.63, 1.20)	21.50% (Low)	0.27	11	0.90 (0.82, 0.98)	11.40% (Low)	0.33

ISSN: 1471-2407