Skip to main content

Table 15 Inverse-variance weighted regression analyses – ever smokers

From: Indirectly estimated absolute lung cancer mortality rates by smoking status and histological type based on a systematic review

 

Males

Females

 

All

Squamous

Adeno

All

Squamous

Adeno

Overall deviance (d.f.)

42483 (128)

3257 (42)

2618 (42)

6005 (90)

644 (37)

1228 (37)

Drop in deviance (d.f.) and associated p valuea from including:

      

Trendb (1 d.f.)

14470***

927***

1257***

2425***

205***

602***

Continentc (3 d.f.)

6245***

454

594*

914**

38

224(*)

Trend given continent (1 d.f.)

17056***

1117***

1169***

2278***

196***

514***

Continent given trend (3 d.f.)

8831***

644**

505***

767***

29

136*

Residual devianced (d.f.)

19181 (124)

1687 (38)

855 (38)

2813 (86)

410 (33)

490 (33)

Deviance explained

54.9%

48.2%

67.3%

53.2%

36.3%

60.1%

Means (SEs)d:

      

Constant

4.80 (0.16)

4.57 (0.30)

2.36 (0.41)

3.58 (0.27)

3.12 (0.41)

1.20 (0.60)

Trend

0.27 (0.03)

0.21 (0.04)

0.62 (0.09)

0.51 (0.06)

0.37 (0.09)

0.84 (0.14)

Asia (not China)

-0.22 (0.18)

-0.51 (0.33)

-0.14 (0.30)

-0.58 (0.34)

-0.72 (0.49)

-0.52 (0.49)

Europe

0.67 (0.14)

0.31 (0.28)

-0.44 (0.29)

-0.37 (0.17)

-0.14 (0.29)

-0.68 (0.39)

America

0.44 (0.14)

0.23 (0.27)

0.31 (0.25)

0.14 (0.14)

-0.19 (0.22)

0.18 (0.24)

Drop in deviance and p valuea from adding:e

      

10 level regionf (6 d.f.)

6944***

368

336**

620**

114

84

5 level periodg (3 d.f.)

9002***

666***

279**

477**

54

15

Interactions of trend and continent (3 d.f.)

3825***

78

46

603***

84(*)

80

  1. a p values are coded as *** p < 0.001, ** p < 0.01, * p < 0.05, (*) p < 0.1. If no code is shown p ≥ 0.1. d.f. = degrees of freedom.
  2. b Linear trend in period scoring 1 = 1930–60, 2 = 1961–70, 3 = 1971–80, 4 = 1981–90, 5 = 1991–99.
  3. c Continent defined as China, Asia (not China), Europe or America.
  4. d For model including linear trend in period and continent. Note that estimates for Asia (not China), Europe and America are relative to rates for China.
  5. e Drop in deviance from adding the specific variable to the model including linear trend in period and continent.
  6. f Levels as shown in Tables 10, 11, 12 and 13.
  7. g This tests for departure from linear trend.