Practical exercise 10

Table 1 gives the observed number of lung cancer cases in the male population in four Danish cities in the period 1968-1971 distributed over five age groups. Table 2 gives the corresponding approximate numbers of inhabitants.

Table 1: Observed number of lung cancer cases in the male population between 1968 and 1971 in four Danish cities distributed over age groups.

 City Age group Fredericia Horsens Kolding Vejle 40-54 11 13 4 5 55-59 11 6 8 7 60-64 11 15 7 10 65-69 10 10 11 14 70-74 11 12 9 8

Table 2: Approximate number of male inhabitants in four Danish cities in the period 1968-1971, distributed over

age groups.

 City Age group Fredericia Horsens Kolding Vejle 40-54 3059 2875 3142 2520 55-59 800 1083 1050 878 60-64 710 923 895 839 65-69 581 834 702 631 70-74 509 634 535 539

You may read the data from the tables into R by the command:

path="http://www.uio.no/studier/emner/matnat/math/STK4080/h14/lungcancer.txt"

The data are organized with one line for each combination of city and age group, and with the following variables in the four columns:

city:         city (1: Fredericia, 2: Horsens, 3: Kolding, 4: Vejle)

age:                          age group  (1: 40-54 years, 2: 55-59 years, 3: 60-64 years, 4: 65-69 years, 5: 70-74 years)

cancer:                   number of lung cancer cases

population:         approximate number of male inhabitants

In this exercise we will consider lung cancer incidence for all cities combined (the effect of city will be considered in a later exercise).

a) Compute the total number of lung cancer cases in and the approximate number of inhabitants for all cities combined. Convert the approximate number of inhabitants to person years by multiplying the number of inhabitants by four. (Why?) For help to this and the following questions, see the R-commands to the lectures from week 43.

b) Compute occurrence/exposure rates of the lung cancer incidence in the five age groups with standard errors, and make a plot of the occurrence/exposure rates with standard 95% confidence limits. Give an interpretation of the plot.

c) Compute log-transformed 95 % confidence limits and add them to the plot in b. Comment on the difference between the two sets of confidence limits.