Table 4

Meta-analyses of overall population estimates of pancreatic cancer risk associated with ST, using estimates for never smokers if overall population estimates are not availablea

Meta-analysis

Estimates Included

Fixed-effect RR (95% CI)

Random-effects RR (95% CI)

Heterogeneity


Chisquared (d.f.)

I2

p



All studies

2,3,9,13,19,20,21

1.04(0.86–1.25)

1.03(0.71–1.49)

15.95 (6)

62.4

0.014


All, preferring snuff to chewingb

2,3,9,13,19,20,24

1.04(0.86–1.27)

1.01(0.68–1.50)

16.35 (6)

63.3

0.012


USA or Canada

2,13,19,20,21

0.92(0.65–1.29)

0.89(0.50–1.60)

8.97 (4)

55.4

0.062


Sweden or Norwayc

3,9

1.09(0.87–1.36)

1.20(0.66–2.20)

6.28 (1)

84.1

0.012


Cohort

2,3,9

1.15(0.93–1.42)

1.31(0.82–2.11)

8.03 (2)

75.1

0.018


Case-controld

13,19,20,21

0.70(0.46–1.05)

0.70(0.43–1.13)

3.44 (3)

12.7

0.329


Chewing tobacco

18,21

0.88(0.55–1.40)

1.27(0.33–4.92)

4.76 (1)

79.0

0.029


Snuff

3,9,19,24

1.03(0.83–1.27)

1.02(0.64–1.64)

9.26 (2)

67.6

0.026


ST unspecified

2,13,20

1.14 (0.70–1.84)

0.89(0.33–2.40)

6.94 (2)

71.2

0.031


Overall estimates exclusively

2,3,9,13,19,21

1.03(0.85–1.25)

1.01(0.67–1.54)

15.93 (5)

68.6

0.007


Excluding Third National Cancer Survey Study[16]

2,3,9,19,20,21

1.07(0.89–1.29)

1.13(0.80–1.59)

11.22 (5)

55.4

0.047


a See Table 3 for the individual study estimates

b Where estimates for both snuff and chewing are available

c There was no significant variation by continent (chisquared = 0.70 on 1 d.f., p = 0.404)

d There was significant variation by study type (chisquared = 4.48 on 1 d.f., p = 0.034)

Sponsiello-Wang et al. BMC Cancer 2008 8:356   doi:10.1186/1471-2407-8-356

Open Data