Circulating Resistin Levels and Risk of Colorectal Cancer: A Meta-Analysis

Table 3

Subgroup analysis of circulating resistin levels and colorectal cancer.

Stratification group	Data points ()	Cases		Random effects	Heterogeneity
		CRCs	Controls	WMD (95% CI)	(%)	value
All studies	11	965	1325	1.47 0.78, 2.16	72	<0.0001
Study location
Western	4	178	159	0.93 −1.20, 3.05	83	0.0005
Asian	6	331	332	1.70 1.03, 2.37	57	0.04
USA	1	456	834	0.50 −0.52, 1.52	—	—
Study design
Case-control	10	509	491	1.84 1.51, 2.18	70	0.0004
Case-cohort	1	456	834	0.50 −0.52, 1.52	—	—
Sample size
<100 CRC cases	8	294	276	1.48 0.38, 2.57	75	0.0002
≥100 CRC cases	3	671	1049	1.42 0.55, 2.28	74	0.02
Blood sample
Plasma	4	673	1043	0.59 −0.39, 1.57	50	0.11
Serum	7	292	282	1.97 1.18, 2.75	68	0.004
Assay method
ELISA	9	442	431	1.68 0.95, 2.41	72	0.0004
Multiplex immunoassay	2	523	894	0.45 −0.50, 1.40	0	0.78
Study quality
Low (NOS score ≤ 5)	7	329	312	1.69 0.98, 2.41	62	0.01
High (NOS score > 5)	4	636	1013	1.22 −0.33, 2.78	79	0.002
Mean age, years
<60	2	66	67	1.15 0.19, 2.11	38	0.2
≥60	9	879	1238	1.56 0.65, 2.47	79	<0.0001
Not mentioned	1	20	20	1.53 0.45, 2.61	—	—
Mean BMI, kg/m^2
<25	1	115	115	1.40 0.64, 2.16	73	0.01
≥25	6	665	1026	1.83 0.54, 3.12	79	0.0002
Not mentioned	4	185	184	1.03 −0.14, 2.21	73	0.01