Obtaining Data in a Cohort Study

Module Epidemiologic Study Designs Worksheet

Name: _______________________________________

Instructions: To complete the assignment, review the module’s Learning Resources, read the details for each section carefully, and provide brief answers to the questions below.

Part 1: Study Designs and Measures of Association (32 pts)

Read the following research scenarios. Identify the study design and justify your choice. Then determine the appropriate measure of association related to it.

An online survey was emailed to every 10th student in a university directory. Each student was asked about their age, sex, race/ethnicity, and alcohol consumption during the preceding 7 days. Over 50% of the students completed the survey and 71% reported drinking at least once.

1. Identify the study design (2 pts)

Cross-sectional

2. Justify your choice (4 pts)

Takes data from a single point in time from a large population, looking at numerous characteristics all at once

3. Determine the appropriate measure of association (2 pts)

Prevalence ratio

Using electronic medical records at a large research hospital, 500 patients diagnosed with sepsis are matched to 500 patients who do not have sepsis based on sex, age, and reason for admission. Researchers assess length of stay in the hospital and intensive care unit (ICU) admission for possible association to sepsis.

4. Identify the study design (2 pts)

Matched-case control

5. Justify your choice (4 pts)

Allows for comparison of patients with sepsis to patients without sepsis who match on other variables; it is also retrospective in that it looks at outcomes after exposure

6. Determine the appropriate measure of association (2 pts)

Odds ratio

A random sample of men from four hospital systems were given a colonoscopy. Half of the men diagnosed with colorectal cancer were assigned to a novel drug treatment and the other half were assigned to the gold standard treatment (surgery and radiation). Both groups were followed for 5 years to monitor recurrent cancer and survival.

7. Identify the study design (2 pts)

Randomized controlled trial

8. Justify your choice (4 pts)

Participants are randomly assigned to a new intervention or to the control (standard intervention) as a way to test efficacy of new intervention compared to standard over 5 years

9. Determine the appropriate measure of association (2 pts)

Risk ratio

Data and biologic materials have been collected from 5,000 participants coming from 250 families in Jackson, Mississippi since 2000. Participants provided extensive medical and social history and had an array of physical and biochemical measurements and diagnostic procedures at baseline and two follow-up periods. Annual follow-up interviews and surveillance monitor incidence of heart failure, stroke, and coronary heart disease.

10. Identify the study design (2 pts)

Prospective cohort

11. Justify your choice (4 pts)

Follows over time a single group of participants; obtains baseline data and then later at two follow-ups; has the goal of observing heart issues over a defined period of time

12. Determine the appropriate measure of association (2 pts)

Incidence rate ratio

Part 2: Case-Control Study Design (20 pts)

Review the following research question and consider issues related to the design of a case-control study to test the hypothesis. Answer in complete sentences.

Is there an association between eating cantaloupe from the farmer’s market and salmonella infection?

13. How will you define a “case”? Are there any special considerations? (4 pts)

A case would be a person diagnosed with salmonella; special considerations would be distinguishing between sources of infection.

14. Where would you find the cases? (4 pts)

Hospital records

15. Describe a suitable control population and where you would find them. (4 pts)

People without a salmonella infection who are similar in demographic characteristics (e.g., age, sex) and from the same geographic area. Controls could be found in the same hospitals, through general practice records.

16. What factors/characteristics would you attempt to measure similarly from each study participant? (4 pts)

Consumption of cantaloupe from the farmer’s market

17.What difficulties might you encounter in measuring these factors/characteristics? (4 pts)

Potential confounding factors (i.e., other sources of salmonella)

Part 3: Cohort Study Design (20 pts)

Review the following research question and consider issues related to the design of a cohort study to test the hypothesis. Answer in complete sentences.

Does occupational exposure to radiation increase risk for cancer mortality among healthcare workers?

13. How will you define a “cohort”? Are there any special considerations? (4 pts)

Cohort would be healthcare workers with documented occupational exposure to radiation. Special considerations would be excluding workers with non-occupational exposure.

14. Where would you find the cohort? (4 pts)

Hospital employee records

15. How would you measure exposure to radiation? Were would you find this data? (4 pts)

Dosimetry records; can be found in hospital records and/or personal dosimetry logs

16. How would you measure the outcome of cancer mortality? Where would you find this data? (4 pts)

Cancer mortality can be measured using hospital records, death certificates, cancer registries, obtained from hospital databases and vital statistics offices

17. Describe how long you would follow-up with the cohort and any difficulties you might encounter. (4 pts)

Follow up should be 10 to 20 years so as to observe potential cancer development; difficulties would be in maintaining contact with participants for that long and accounting for confounding variables

Part 4: Randomized Controlled Trial Study Design (20 pts)

Review the following research question and consider issues related to the design of a randomized controlled trial (RCT) to test the hypothesis. Answer in complete sentences.

What is the efficacy of a new vaccine to prevent respiratory syncytial virus (RSV) as compared to placebo?

13. Describe the inclusion and exclusion criteria you might use for this research scenario. (4 pts)

Inclusion criteria would be healthy people between 18 and 65 years of age, no prior RSV vaccine. Exclusion criteria would be pregnant women, people with vaccine-related allergies, and immunocompromised people.

14. What is the importance of randomization in an intervention trial? What does it accomplish? (4 pts)

Randomization is important at intervention trial because it eliminates selection bias and allows intervention and control groups to be comparable at baseline. Confounding variables are distributed equally among the groups, so that assessment of the intervention’s effect can be seen as more accurate

15. How might you implement randomization in this research scenario? (4 pts)

Use computer-generated random numbers and assign them to participants in both groups, blinded so that neither participants nor researchers know the identities (or which they group they are in) of the participants behind the numbers

16. To minimize opportunities for bias, how might you implement blinding in this research scenario? (4 pts)