Wynder and Graham (1) studied 684 cases of microscopically confirmed bronchogenic carcinoma (lung cancer); these included 605 male and 25 female patients with epidermoid, undifferentiated and unclassified carcinomas and 39 male and 15 female patients with adenocarcinoma.(These are different kinds of lung cancer with microscopic evaluation).

A standard questionnaire form was used which included questions on history of other lung disease, smoking, occupation, exposure to dusts and fumes, alcohol intake, residence, education, and cause of death of parents and siblings.Personal interviews were obtained in 634 cases, a mailed questionnaire was used in 33 cases, and the information was obtained in 17 cases from a person who had known the patient intimately throughout his/her adult life.The physician authors of the report apparently conducted these interviews.

In addition, two nonmedical investigators interviewed 780 male and 552 female patients without cancer of the lung on the general surgical and medical services at three hospitals in St. Louis: Barnes Hospital, Jefferson Barracks Veterans Hospital, and St. Louis City Hospital.Table 1 shows the ageadjusted data on smoking habits for the 605 male lung cancer patients and the 780 male controls.The data are age-adjusted based on the age distribution of the male lung cancer patients.

Table 1:AGE-ADJUSTED DATA ON SMOKING HABITS

IN MALE LUNG CANCER PATIENTS AND CONTROLS

Percentage Distribution

Amount of Cigarette

Smoking for 20+ Years^*

Lung Cancer

Patients

Lung Cancer

Controls

None (less that 1 per day)

1.3

14.6

Light (1-9 per day)

2.3

11.5

Moderately Heavy (10-15 per day)

10.1

19.0

Heavy (16-20 per day)

35.2

35.6

Excessive (21-34 per day)

30.9

11.5

Chain (35+ per day)

20.3

7.6

^*Pipe and cigar smokers were included by counting one cigar as five cigarettes and one pipe as two and onehalf cigarettes.Among the lung cancer patients, 4% were pipe smokers (controls, 12.4%) and 3.5% were cigar smokers (controls, 7.8%).

Question :Convert the percentages into numbers and pretend that data are not ageadjusted.Compared to the non-smokers, what are the odds ratios for each smoking category?

Case

Control

Light Smoker

Non-Smoker

Total

Odds Ratio =

Case

Control

Moderate Smoker

Non-Smoker

Total

Odds Ratio =

Case

Control

Heavy Smoker

Non-Smoker

Total

Odds Ratio =

Case

Control

Excessive Smoker

Non-Smoker

Total

Odds Ratio =

Case

Control

Chain Smoker

Non-Smoker

Total

Odds Ratio =

Question 3:Would you conclude from these data that an association exists between smoking and lung cancer?Why or why not?