Table 2.

Summary risk estimates for the association between cholesterol intake and risk of gastric cancer in adults aged ≥18 years in case-control studies^a.

	#RRb	Pooled RR (95% CI)c	I2 (%)d	P-heterogeneitye	Meta-regression
The highest vs. lowest comparison
Dietary cholesterol intake
Overall	14	1.35 (1.129–1.62)	70.1	<0.001	–
Subgroup analysis
Study location
US	4	1.32 (0.94–1.84)	83.8	<0.001	0.898
Non-US	10	1.37 (1.08–1.74)	63.3	0.004
Study design
Population-based case-control study	9	1.45 (1.17–1.79)	72	<0.001	0.309
Hospital-based case-control study	5	1.13 (0.78–1.64)	64.5	0.024
Study quality
High quality	9	1.56 (1.28–1.89)	53.1	0.029	0.146
Low quality	5	1.01 (0.73–1.41)	76.3	0.002
Number of cases
<400	9	1.37 (1.00–1.89)	69.6	0.001	0.664
≥400	5	1.35 (1.08–1.68)	74.9	0.003
Exposure reporting
Interview	11	1.43 (1.16–1.77)	64	0.002	0.441
Self-report	3	1.16 (0.84–1.61)	77.5	0.012
Adjustment for H. pylori
Yes	4	1.17 (0.71–1.91)	85.6	<0.001	0.471
No	10	1.41 (1.17–1.70)	56.9	0.013
Adjustment for BMI
Yes	8	1.35 (1.14–1.61)	67.6	0.003	0.947
No	6	1.35 (0.77–2.39)	76.9	0.001
Dietary cholesterol intake (per 100 mg/d increase)
Overall	8	1.05 (0.99–1.12)	83.5	<0.001
Subgroup analysis
Study design
Population-based case-control study	5	1.10 (1.04–1.16)	39.7	0.156	0.153
Hospital-based case-control study	3	0.96 (0.83–1.12)	72.8	0.025
Study quality
High quality	6	1.07 (0.99–1.15)	74.8	0.001	0.222
Low quality	2	0.93 (0.64–1.35)	90.7	0.001
Number of cases
<400	5	1.01 (0.90–1.13)	71.2	0.008	0.335
≥400	3	1.10 (1.05–1.15)	30	0.24
Adjustment for H. pylori
Yes	2	0.93 (0.64–1.35)	90.7	0.001	0.407
No	6	1.07 (0.99–1.15)	74.8	0.001
Adjustment for BMI
Yes	5	1.07 (1.00–1.15)	86.7	<0.001	0.421
No	3	0.97 (0.74–1.26)	83.8	0.002

^aBMI, body mass index; CI, confidence interval; RR, Relative Risk; FFQ, food frequency questionnaire; US, United States.

^bNumber of risk estimates.

^cObtained from the random-effects model.

^dInconsistency- the percentage of variation across studies due to heterogeneity.

^eObtained from the Q-test.