Epidemiology Case Studies and Questions

Scenario 1 (for questions 1-7) You are interested in studying the relationship between physical activity and pancreatic cancer. This cancer is rarely detected in individuals under the age of 40, and is currently found in both women and men throughout the world. You identify 35 doctor’s offices in the United Kingdom, and invite any patients at these doctor’s offices who are between the ages of 40 and 74 to participate in your study. You obtain informed consent from those who agree, and they are asked to complete an extensive questionnaire. In that questionnaire, they are asked if they have ever been diagnosed with pancreatic cancer; those who report that they were diagnosed are excluded from the study. This leaves you with a total study population of 25,242 participants at the start of your study in 2017.
These participants are also asked a series questions about their physical activity levels, and from the results of those questions, you characterize each participant as either demonstrating “Low to medium activity”(n = 14,175) or “Medium to high activity”(n=11,067). You track the participants for 10 years; for anyone who dies during this time, you obtain death certificate information on cause of death. If they died of pancreatic cancer, you record that information. After 10 years, you invite all living participants into the clinic and collect a blood sample to test for pancreatic cancer.
Of those in the low-to-medium activity group, 2961 stayed in the study for 5 years; 962 stayed in the study for 1 year; all the rest participated for the full study duration. Of those in the medium-to-high activity group, 1987 quit the study after 7 years; 1432 quit the study after 5 years; 522 quit the study after 1 year; and the rest participated for the full study duration. The total number of pancreatic cancer cases recorded in the high activity group was 120.The total number of pancreatic cancer cases recorded in the low activity group was 387.

1. Identify the following:
a. The overall population
b. The study population
c. The disease of interest
d. The case definition
e. The risk factor under study

2. What kind of study is this (Choose one)?
a. Prospective, population-based cohort study
b. Prospective, exposure-based cohort study
c. Retrospective, population-based cohort study
d. Retrospective, exposure-based cohort study

3. Was participant recruitment conducted using random selection?
4. Does this study include an element of random assignment?
5. What measure of disease should you calculate in this study? Be as specific as possible.
6. What measure of association should you calculate in this study? Be as specific as possible.
7. Calculate and interpret the appropriate measure of association. Consider the “high activity group” to be your exposed participants.

Scenario 2 (for questions 8-14): It’s 2017, and major traffic revisions are planned in Boise, including the addition of a new highway. This new highway will be located near several elementary schools, and community members have voiced concerns about potential air pollution exposures to kids attending these schools. You have been asked to conduct a study of the relationship between living close to highways and asthma among children aged 5-10, because if children who live near highways are more likely to have asthma, then it may not be a good idea to locate highways near schools, either. You review birth certificates from 2007-2012 and locate 100 children whose address at their time of birth was within 500 m of a major roadway and 100 children whose address at birth was not within 500 m of a major roadway. You then track down the parents of these children, ask them to participate in your study, and make sure that they haven’t moved since their kids were born (conveniently, none had). You also ask them if their child has received a diagnosis of asthma from their doctor. You finds that, of the 100 children whose address at birth was within 500 m of a major roadway, 15 have been diagnosed with asthma. Of those who did not live within 500 m of a major road at birth, 9 have been diagnosed with asthma.

8. Identify the following:
a. The overall population
b. The study population
c. The disease of interest
d. The case definition
e. The risk factor under study

9. What kind of study is this (Choose one)?
a. Prospective, population-based cohort study
b. Prospective, exposure-based cohort study
c. Retrospective, population-based cohort study
d. Retrospective, exposure-based cohort study

10. Was participant recruitment conducted using random selection?
11. Does this study include an element of random assignment?
12. What measure of disease should you calculate in this study? Be as specific as possible.
13. What measure of association should you calculate in this study? Be as specific as possible.
14. Set up a 2×2 table, and calculate and interpret the appropriate measure of association.

Scenario 3 (for questions 15-21): You are a fan of the sport of thoroughbred horse racing but are concerned about what you perceive to be a high rate of injuries among professional jockeys in the United States. In particular, you are concerned that the new, lightweight helmets you have seen on some jockeys may not be sufficiently protective. It’s 2017, and you work with the American insurance companies who ensure all professional jockeys to get a complete list of all of the professional jockeys in the country who have never had a work-related injury claim. You use a computer program to randomly generate a subset of 900 jockeys from this list and invite them to participate in your study. All 900 agree and provide informed consent. You ask them to complete a questionnaire about the type of helmet they use and classify each jockey as either using a traditional helmet or a new lightweight helmet. You learn that 317 of them are using the lightweight helmets and the other 583 are using traditional helmets. You then follow these jockeys through 2 racing seasons. The insurance agency reports to you any injuries for which jockeys file work-related insurance claims during this time. At the end of 2 seasons, they have reported to you 94 injuries among the lightweight helmet wearers and 42 injuries among the traditional helmet wearers.

15. Identify the following:
a. The overall population
b. The study population
c. The disease of interest
d. The case definition
e. The risk factor under study

16. What kind of study is this (Choose one)?
a. Prospective, population-based cohort study
b. Prospective, exposure-based cohort study
c. Retrospective, population-based cohort study
d. Retrospective, exposure-based cohort study

17. Was participant recruitment conducted using random selection?
18. Does this study include an element of random assignment?
19. What measure of disease should you calculate in this study? Be as specific as possible.
20. What measure of association should you calculate in this study? Be as specific as possible.
21. Set up a 2×2 table, and calculate and interpret the appropriate measure of association.