Table 1. Infections Where PAFs Were Estimated Using the Prevalence of Infection in the Population and Measures of Association.

Infection, cancer(s)	Data used to estimate measure of associationa	OR (95% CI)	Source of prevalence data	Range of prevalence estimates by sex and age group, y	PAF, % (95% CI)b
Hepatitis B virus
Hepatocellular carcinoma	Pooled ORs from 4 US case-control studies	24.2 (14.5 to 40.3)	NHANES data collected, 1999-2010	Males: 0.46% overall, from 0.31% (≥60 y) to 0.96% (50-59 y) Females: 0.23% overall, from 0.08% (15-29 y) to 0.35% (30-39 y)	9.4 (2.9 to 14.5)
Intrahepatic bile duct	Pooled ORs from 4 US case-control studies	3.4 (1.2 to 9.4)			0.9 (0.0 to 1.8)
HCV
Hepatocellular carcinoma	Pooled ORs from 5 US case-control studies	29.8 (11.9 to 74.6)	NHANES data collected, 1999-2010	Males: 1.76% overall, from 0.26% (15-29 y) to 4.45% (45-49 y) Females: 0.82% overall, from 0.10% (15-29 y) to 2.43% (45-49 y)	32.1 (9.0 to 45.2)
Intrahepatic bile duct	Pooled ORs from 4 US case-control studies	4.5 (3.5 to 5.7)			4.9 (2.2 to 7.4)
Burkitt lymphoma (age, ≥50 y only)	ORs from 5 studies (from Australia, Canada, Europe, and the US) assessing HCV seropositivity in the InterLymph Non-Hodgkin Lymphoma Subtypes Project	4.1 (1.1 to 15.4)			3.7 (0.0 to 7.1)
Chronic lymphocytic leukemia or small vessel lymphoma		2.08 (1.23 to 3.49)			1.9 (0.0 to 3.7)
DLBCL		Males: 2.17 (1.44 to 3.26) Females: 1.98 (1.18 to 3.34)			1.5 (0.2 to 2.8)
Lymphoplasmacytic lymphoma		2.51 (1.03 to 6.17)			2.0 (0.0 to 3.9)
Marginal zone lymphoma		3.04 (1.65 to 5.60)			2.7 (0.1 to 5.2)
Helicobacter pylori
Gastric, noncardia	Pooled ORs from nested case-control studies from the US, Europe, and Australia: 5 studies that used ELISA or EIA corrected for measurement error and 3 studies that used immunoblot	12.8 (8.5 to 19.2)	NHANES data collected, 1999-2000	Males: 29.2% overall, from 14.6% (10-14 y) to 51.0% (80-84 y) Females: 28.3% overall, from 9.3% (10-14 y) to 49.6% (70-74 y)	75.4 (67.8 to 78.8)
Gastric, MALT, and DLBCL	One US study of 20 cases matched to 82 controls	7.9 (1.6 to 38.1)			70.8 (0.8 to 90.5)
Esophageal adenocarcinoma, protective effect	Pooled ORs from US case-control and nested case-control studies: 3 studies used ELISA and 1 study used immunoblot	0.73 (0.55 to 0.95)			−10.9 (−1.8 to −20.8)c

Abbreviations: DLBCL, diffuse large B-cell lymphoma; EIA, enzyme immunoassay; ELISA, enzyme-linked immunosorbent assay; HCV, hepatitis C virus; MALT, mucosa-associated lymphoid tissue; NHANES, National Health and Nutrition Examination Survey; OR, odds ratio; PAF, population attributable fraction.

^a The characteristics of included studies are reported in Supplement 1 under their respective infection and cancers.

^b The PAF was calculated by summing the number of attributable cases obtained after applying sex- and age-specific PAFs, then dividing the sum by the associated cancer cases.

^c To account for H pylori’s protective effect in esophageal adenocarcinoma, cases attributable to esophageal adenocarcinoma were subtracted from the total cases attributable to H pylori.