You are on page 1of 3

Seminar 4 Questions

EPIDEMIOLOGY AND BIOSTATISTICS 1 (HSH205)


SEMINAR 4: ANALYTICAL STUDIES (2)

Read
• Chapter 8 of the Carneiro et al (2011) textbook Cohort Studies.
• Chapter 3 of the Carneiro et al. (2011) textbook Measures of association and impact.
• Doll et al (2004) Mortality in relation to smoking: 50 years' observations on male
British doctors paper (available on CloudDeakin)
• Yiannakoulias, N. (2009) Using population attributable risk to understand geographic
disease clusters (available on CloudDeakin)

1. In recent years researchers have become interested in whether there is an association


between regular physical activity and risk of ovarian cancer. A study investigating this
relationship found that 40 out of 4,500 women who engaged in regular physical activity
developed ovarian cancer during 10 years of follow-up, while 55 women out of 4,500
who did not engage in regular physical activity developed ovarian cancer during the same
period.

Using this information answer the following questions


a. What is the study factor in this study?
b. What is the outcome factor?
c. If you were to use these data to compute the relevant measure of association (ie
relative risk, risk ratio, or rate ratio) to assess whether there is an association
between physical activity and ovarian cancer, which measure would you calculate?
d. Calculate that measure
e. State in words your interpretation of your result
f. If there was no association between regular physical activity and ovarian cancer,
what would be the numeric value of the measure of association (ie relative risk,
risk ratio, or rate ratio)?

2. All 10,000 couples who registered with an infertility treatment clinic in Kelseyvale during
the period 1985 to 1995 provided details of their history of infertility and of any prior
pregnancies or births. Only half of the couples eventually underwent any infertility
Seminar 4 Questions

treatment. Researchers recontacted the group who remained untreated in 2005. The
following data were recorded for this group. Of the 1,000 couples who had reported at
least one pregnancy prior to registering with the infertility clinic, 250 reported at least one
live birth by 2005. There had been 300 live births among those who had not been
pregnant prior to their registration at the infertility clinic.

a. What type of study is described here?


b. What methods would you use to investigate whether there is an association
between pregnancy prior to registration at the infertility clinic and the likelihood of
a live birth within the next 10 years?
c. Calculate an appropriate measure of association, and briefly interpret your result.
d. Compute the risk difference for a live birth for a couple who had a pregnancy prior
to registration at the infertility clinic and those who had no prior pregnancy.
e. State in words your interpretation of your result

3. Indicate which measure (prevalence, cumulative incidence, or incidence rate) you could
calculate from the information provided in each of the following examples.
a. In 2006, a survey of 300 14 to 16 year old school high school students found that
20% had smoked marijuana during the previous six months.
b. Of 1800 students who graduate from Deakin University in 2006, 700 were parents.
c. Before the discovery that infants sleeping on their front (prone sleeping position)
increased the risk of ‘cot death’, the number of babies who died of sudden infant
death syndrome (SIDS) during their first year of life was 27 per 100,000 baby
years of follow-up.

4. Read the introduction and methods sections of the paper by Richard Doll and colleagues
(available on CloudDeakin).

Lung cancer death rates (per 100,000 persons per year) for smokers and non-smokers for
male and female doctors are shown in the following table. Calculate appropriate measures
to assess the association for each level of smoking and lung cancer for males and females
separately. Briefly interpret your results.
Seminar 4 Questions

Lung cancer death rates per 100,000 persons per year


Sex of doctors Non-smokers Smoking 1-14 Smoking 15-24 Smoking 25+
cigarettes per day cigarettes per day cigarettes per day
Male 10 78 127 251
Female 7 9 45 208

You might also like