KEY (Case-Control Studies)

(1) DOLL1950

or = 14.0 (95% confidence interval for OR : 3.5, 122.2 by the exact method computed by STATCALC (Mehta et al., 1985).
Data indicate a strong positive association between smoking and lung cancer. Smokers have ~14x the lung cancer rate of non-smokers.

(2) ESOPH_CA: Esophageal cancer and tobacco consumption (dichotomized at less than 80 gms./day)
or = 2.0 (95% confidence interval for OR: 1.4, 2.8). In testing H₀: OR = 1, c²_Yates 14.11 with 1 df (p = .00017).
High tobacco consumption doubles the risk of esophageal cancer.

(3) ESOPH_CA.REC: Esophageal cancer and age (dichotomized at 55 years of age)
or = 4.0 (95% confidence interval for OR: 2.8, 5.7). In testing H₀: OR = 1, c²_Yates = 68.00 with 1 df (p < .000001).
The older age-group has about 4 times the risk of esophageal cancer.
 

(4) BD2.REC (In utero X-ray Exposure and Childhood Leukemia/Lymphoma)
or = 1.7 (95% confidence interval: 1.5, 1.9; p < .00000005 by the c² uncorrected).
This represents a weak but statistically significant association between in utero irradiation and childhood leukemia and lymphatic cancers (an average 70% increase in risk).