A Meta-Analysis of Statin Use and Risk of Hepatocellular Carcinoma

Table 4

Sensitivity analysis to examine sources of heterogeneity observed in summary estimate.

Sensitivity analysis	No. of studies	Adjusted OR	95% CI	Tests of heterogeneity		Heterogeneity between groups ()
					I^2 (%)
Sensitivity analysis (to examine source of heterogeneity seen in observational studies)
Study quality
High quality	15	0.58	0.48–0.70	<0.00001	94	0.86
Low quality	13	0.56	0.45–0.70	<0.00001	79
Study design
Cohort	18	0.59	0.48–0.72	<0.00001	91	0.63
Case-control	10	0.54	0.42–0.70	<0.00001	92
Sensitivity analysis (to examine source of heterogeneity seen in high-quality observational studies)
Study location
Asian	5	0.50	0.42–0.59	0.13	44	0.07^a
Western	10	0.64	0.52–0.79	<0.00001	92

, source of heterogeneity between groups.