Coffee Intake and Gastric Cancer Risk: The Singapore Chinese Health Study

Cheryl E. Ainslie-Waldman; Woon-Puay Koh; Aizhen Jin; Khay Guan Yeoh; Feng Zhu; Renwei Wang; Jian-Min Yuan; Lesley M. Butler

doi:10.1158/1055-9965.EPI-13-0886

DOI: 10.1158/1055-9965.EPI-13-0886 Published April 2014

Article Figures & Data

Tables

Additional Files

Table 1.

Distribution of selected baseline characteristics stratified by sex and coffee intake frequency

	Men				Women
	Coffee intake				Coffee intake
Characteristic	<1 cup/day	1 cup/day	2–3 cups/day	≥4 cups/day	<1 cup/day	1 cup/day	2–3 cups/day	≥4 cups/day
Person years of follow-up	108,838	118,655	136,914	22,002	155,252	207,347	140,411	13,742
Mean age, y (SD)	57 (8)	57 (8)	56 (8)	55 (7)	56 (8)	56 (8)	56 (8)	56 (8)
Body mass index, kg/m² (%)
<20	13.6	15.6	17.0	20.1	15.9	14.1	14.4	15.3
20–<24	51.8	53.0	54.2	54.2	54.6	54.7	55.2	56.2
24–<28	27.5	25.3	23.1	20.5	21.5	23.3	22.5	20.3
≥28	7.1	6.2	5.8	5.2	8.0	7.9	7.9	8.2
Education, ≥secondary level (%)	43.1	37.8	34.9	33.9	26.1	19.4	17.0	15.3
Smoking status (%)
Never	53.4	44.1	35.0	19.4	94.7	91.6	88.4	76.7
Former	23.5	22.4	19.9	14.4	2.1	2.7	2.6	3.4
Current	23.1	33.5	45.1	66.2	3.2	5.8	9.1	20.0
Alcohol use, ≥1 drink/week (%)	18.5	24.1	24.8	24.2	3.6	4.5	5.9	8.5
History of ulcer (% yes)	6.6	4.1	3.7	5.6	3.3	1.8	1.8	3.0
History of diabetes (% yes)	10.9	8.9	7.0	6.0	10.8	9.3	7.5	5.0
Additions to coffee or tea (% yes)
Sugar	48.1	71.7	78.5	83.2	36.9	64.5	74.5	84.7
Milk	50.9	50.7	48.1	38.3	42.3	47.4	45.1	34.6
Mean daily intake (SD)
Total energy, kcal	1,697 (584)	1,702 (599)	1,816 (618)	1,929 (674)	1,369 (465)	1,366 (451)	1,466 (493)	1,568 (558)
Caffeine, mg/1,000 kcal	45.6 (46.9)	81.3 (43.2)	151.4 (59.3)	255.8 (105.0)	33.8 (42.8)	86.1 (40.7)	173.3 (65.4)	306.1 (114.8)
Fruit, g/1,000 kcal	132.8 (96.8)	129.0 (91.6)	110.5 (83.1)	92.6 (75.3)	146.8 (108.7)	143.0 (102.8)	121.9 (92.6)	97.3 (93.2)
Vegetables, g/1,000 kcal	67.3 (32.5)	65.9 (30.3)	61.5 (29.5)	56.5 (27.6)	82.5 (39.0)	80.9 (36.0)	74.4 (34.2)	70.2 (34.3)

Table 2.

HR and 95% CIs for baseline characteristics in relation to gastric cancer risk

	Overall		Men		Women
	Cases, n	HR (95% CI)	Cases, n	HR (95% CI)	Cases, n	HR (95% CI)
N	647		394		253
Education
No formal education	224	1.00^a	64	1.00^a	160	1.00^a
Primary level	309	0.87 (0.72–1.05)	243	1.11 (0.84–1.47)	66	0.61 (0.45–0.82)
≥Secondary level	114	0.63 (0.49–0.81)	87	0.74 (0.53–1.04)	27	0.59 (0.38–0.90)
P_trend		<0.001		0.032		0.001
Body mass index, kg/m²
<20	80	1.00^a	45	1.00^a	35	1.00^a
20–<24	367	1.26 (0.99–1.60)	224	1.46 (1.06–2.02)	143	1.00 (0.69–1.45)
24–<28	144	1.24 (0.94–1.63)	93	1.51 (1.06–2.16)	51	0.92 (0.59–1.42)
≥28	56	1.71 (1.21–2.42)	32	2.24 (1.41–3.53)	24	1.20 (0.71–2.02)
P_trend		0.012		0.002		0.774
Self-reported diabetes
No	588	1.00^a	366	1.00^a	222	1.00^a
Yes	59	1.01 (0.77–1.33)	28	0.84 (0.57–1.23)	31	1.27 (0.86–1.85)
History of ulcer
No	619	1.00^a	375	1.00^a	244	1.00^a
Yes	28	1.03 (0.71–1.51)	19	0.93 (0.58–1.48)	9	1.42 (0.73–2.77)
Smoking status
Never	351	1.00^b	134	1.00^b	217	1.00^b
Former	90	1.16 (0.84–1.62)	79	1.00 (0.70–1.44)	11	2.32 (0.99–5.42)
Current	206	1.70 (1.33–2.17)	181	1.64 (1.25–2.15)	25	1.82 (0.95–3.47)
Alcohol use
Nondrinker	547	1.00^a	305	1.00^a	242	1.00^a
<7 drinks/week	68	1.21 (0.94–1.57)	60	1.25 (0.95–1.65)	8	1.06 (0.52–2.15)
≥7 drinks/week	32	1.13 (0.78–1.62)	29	1.16 (0.79–1.71)	3	0.99 (0.32–3.09)
P_trend		0.208		0.169		0.936
Fruit intake
Quartile 1	209	1.00^c	115	1.00^c	94	1.00^c
Quartile 2	175	1.02 (0.83–1.25)	109	1.10 (0.84–1.43)	66	0.91 (0.66–1.26)
Quartile 3	140	0.88 (0.70–1.10)	86	0.88 (0.66–1.18)	54	0.90 (0.63–1.28)
Quartile 4	123	0.80 (0.63–1.03)	84	0.83 (0.60–1.13)	39	0.78 (0.52–1.19)
P_trend		0.053		0.128		0.265
Vegetable intake
Quartile 1	198	1.00^c	108	1.00^c	90	1.00^c
Quartile 2	167	0.98 (0.80–1.21)	103	1.10 (0.83–1.44)	64	0.86 (0.62–1.19)
Quartile 3	151	0.98 (0.78–1.23)	101	1.18 (0.89–1.57)	50	0.76 (0.52–1.09)
Quartile 4	131	0.98 (0.72–1.21)	82	0.98 (0.70–1.37)	49	0.91 (0.60–1.37)
P_trend		0.627		0.845		0.408

↵^aHRs are adjusted for age (years), gender (in overall model only), interview year (1993–1995, 1996–1998), dialect (Hokkien, Cantonese), cigarette smoking status (never, former, current), number of cigarettes smoked per day (never, 1–12, ≥13), years smoked (never, 1–39, ≥40), and body mass index (<20, 20–<24, 24–<28, ≥28 kg/m²).
↵^bHRs are adjusted for same covariates as listed above (a), excluding cigarette smoking status.
^cHRs are adjusted for same covariates as listed above (a), including total energy intake (kcal/day).

Table 3.

Coffee and caffeine intake and gastric cancer risk

	Overall		Men		Women
	Cases, n	HR (95% CI)	Cases, n	HR (95% CI)	Cases, n	HR (95% CI)
Coffee intake
Never/monthly	132	1.00^a	75	1.00^a	57	1.00^a
Weekly	61	1.06 (0.78–1.44)	32	0.95 (0.62–1.43)	29	1.22 (0.78–1.92)
1 cup/day	199	0.84 (0.66–1.07)	118	0.98 (0.72–1.33)	81	0.66 (0.45–0.97)
2–3 cups/day	223	1.00 (0.71–1.40)	147	1.12 (0.74–1.69)	76	0.81 (0.45–1.47)
≥4 cups/day	32	0.93 (0.49–1.79)	22	1.06 (0.48–2.32)^b	10	0.76 (0.23–2.53)^b
P_trend		0.449		0.719		0.068
Nondaily	193	1.00^a	107	1.00^a	86	1.00^a
Daily	454	0.85 (0.69–1.04)	287	1.03 (0.79–1.34)^c	167	0.63 (0.46–0.87)^c
Caffeine intake^d
Q1 (lowest)	161	1.00^d	85	1.00^e	76	1.00^e
Q2	134	0.78 (0.62–0.98)	86	0.89 (0.66–1.21)	48	0.66 (0.46–0.94)
Q3	157	0.86 (0.69–1.07)	108	1.02 (0.76–1.36)	49	0.65 (0.45–0.94)
Q4 (highest)	195	0.98 (0.79–1.22)	115	1.02 (0.75–1.37)^f	80	0.98 (0.71–1.36)^f
P_trend		0.869		0.685		0.902

↵^aHRs are adjusted for age (years), gender (in overall model only), interview year (1993–1995, 1996–1998), dialect (Hokkien, Cantonese), education (less than secondary, secondary or greater), cigarette smoking status (never, former, current), number of cigarettes smoked per day (never, 1–12, ≥13), years smoked (never, 1–39, ≥40), body mass index (<20, 20–<24, 24–<28, ≥28 kg/m²), caffeine (mg/day), and total energy intake (kcal/day).
↵^bP for sex interaction = 0.733.
↵^cP for sex interaction = 0.184.
↵^dThe following are the cutpoints for first through fourth quartiles of caffeine intake (mg/day): <71.2, 71.2–114.0, 114.1–224.1, >224.1.
↵^eHRs are adjusted for same covariates as listed above (^a), excluding caffeine and with the addition of alcohol consumption (never, ever).
↵^fP for sex interaction = 0.948.

Table 4.

Coffee intake and gastric cancer risk by smoking status, alcohol use, and body mass index

	Overall		Men		Women
Coffee intake	Cases, n	HR (95% CI)	Cases, n	HR (95% CI)	Cases, n	HR (95% CI)
Never smokers
Nondaily	132	1.00^a	48	1.00^a	84	1.00^a
Daily	219	0.71 (0.54–0.92)	86	1.06 (0.69–1.63)	133	0.54 (0.38–0.76)
Ever smokers
Nondaily	61	1.00^b	59	1.00^b	2	1.00^b
Daily	235	1.10 (0.79–1.52)	201	1.00 (0.72–1.41)	34	3.36 (0.75–15.03)
Alcohol nondrinkers
Nondaily	167	1.00^a	83	1.00^a	84	1.00^a
Daily	380	0.84 (0.67–1.05)	222	1.09 (0.80–1.47)	158	0.61 (0.44–0.84)
Alcohol drinkers
Nondaily	26	1.00^a	24	1.00^a	2	1.00^a
Daily	74	0.84 (0.5–1.42)	65	0.80 (0.46–1.38)	9	1.41 (0.24–8.18)
BMI (<23.1 kg/m²)
Nondaily	87	1.00^c	51	1.00^c	36	1.00^c
Daily	226	0.91 (0.68–1.23)	163	1.13 (0.78–1.64)	63	0.59 (0.35–0.99)
BMI (≥23.1 kg/m²)
Nondaily	106	1.00^c	56	1.00^c	50	1.00^c
Daily	228	0.79 (0.60–1.05)	124	0.90 (0.62–1.32)	104	0.66 (0.43–0.99)

↵^aHRs are adjusted for age (years), gender (in overall model only), interview year (1993–1995, 1996–1998), dialect (Hokkien, Cantonese), education (less than secondary, secondary or greater), cigarette smoking status (never, former, current), number of cigarettes smoked per day (never, 1–12, ≥13), years smoked (never, 1–39, ≥40), body mass index (<20, 20–<24, 24–<28, ≥28 kg/m²), caffeine (mg/day), and total energy intake (kcal/day).
↵^bHRs are adjusted for same covariates as listed above (^a), excluding cigarette smoking status.
↵^cHRs are adjusted for same covariates as listed above (^a), excluding body mass index.

Table 5.

Coffee intake and gastric cancer risk among cases and controls nested within the Singapore Chinese Health Study

Coffee intake	Cases/controls, n	Multivariable-adjusted OR (95% CI)^a	H. pylori–adjusted OR (95% CI)^b
Overall
Nondaily	45/107	1.00	1.00
Daily	88/282	0.54 (0.31–0.91)	0.50 (0.29–0.86)
Men
Nondaily	25/72	1.00	1.00
Daily	60/178	0.84 (0.43–1.65)	0.76 (0.38–1.52)
Women
Nondaily	20/35	1.00	1.00
Daily	28/104	0.22 (0.08–0.56)	0.21 (0.08–0.56)

↵^aORs are adjusted for age (years), gender (in overall model only), interview year (1993–1995, 1996–1998), dialect (Hokkien, Cantonese), education (less than secondary, secondary or greater), cigarette smoking status (never, former, current), number of cigarettes smoked per day (never, 1–12, ≥13), years smoked (never, 1–39, ≥40), body mass index (<20, 20–<24, 24–<28, ≥28 kg/m²), caffeine (mg/day), total energy intake (kcal/day), and date of biospecimen collection.
↵^bORs are adjusted for the same covariates as in model (a), in addition to H. pylori serology status.

Additional Files

Tables

Supplementary Data

Files in this Data Supplement:

Supplementary Tables 1 and 2 - 84K, Table S1. Distribution of selected baseline characteristics stratified by coffee intake frequency. Table S2. Distribution of Helicobacter pylori (H. pylori) seroprevalence and atrophic gastritis status stratified by coffee intake among controls.