MI 580: Principles of Epidemiology
Fall 2023 – Homework 2
Due: September 27th at the start of class; 95 points
Instructions:
For all calculations be as explicit as possible by showing you set up all calculations related to the
final answer. Showing your work is a component of the homework grade. Please submit the
homework electronically. All calculations can be rounded to one decimal place.
Question 1 [10 points]: As an epidemiologist working for the health department in
Philadelphia, you conduct a study looking at the impact of Covid-19 on mortality. You sample
100 individuals with Covid-19 and 100 individuals without Covid-19. Among the 100 with
Covid-19, 23 die where Covid-19 is thought to be the cause of the death. An additional 5 die
but the death is due to another cause. Among the 100 individuals without Covid-19, 14 die.
1a. Calculate the case-fatality rate of Covid-19. (3 points)
Case-fatality rate of Covid-19 = (number of people died from Covid-19/number of people who
have Covid-19)*100 = 23/100 = 23%
1b. Calculate the relative survival rate. (3 points)
Relative survival rate = (observed survival of individuals with Covid-19/expected survival of
individuals without Covid-19)*100
observed survival of individuals with Covid-19 = 100 - 28 = 72
expected survival of individuals without Covid-19 = 100 – 14 = 86
Relative survival rate = (72/86)*100 = 83.7%
1c. For relative survival rate, what other considerations would you make when recruiting the
patients into your study? (2 points).
When recruiting patients into a study for relative survival rate analysis, I will consider factors
such as age, gender, comorbidities, and other relevant characteristics that may affect survival
outcomes. This is important to ensure that the comparison group without Covid-19 is
comparable to the group with Covid-19 in terms of these factors.
1d. What additional information would you need to calculate 5 year survival rate? (2 points).
I would need information on the long-term survival outcomes of individuals with and without
Covid-19 over a period of 5 years. This would require follow-up data on the survival status of
individuals in both groups at 5 years after their initial diagnosis or enrollment in the study.
Question 2: [15 points] You’re researching the link between alcohol consumption and heart
disease. You come across an article that presents the average total alcohol sales per adult
person capita for 10 countries. The article also presents the prevalence of heart disease for
those 10 countries. The authors notice a correlation such that the countries with the highest
alcohol sales are the same ones with the highest prevalence of heart disease. They make the
conclusion that alcohol consumption likely causes heart disease.
, MI 580: Principles of Epidemiology
Fall 2023 – Homework 2
2a. Which study design did the authors select [2 points]
The authors selected an observational study design.
2b. Is there conclusion correct? Please provide justification to support your answer. [3 points]
The conclusion made by the authors that alcohol consumption likely causes heart disease based
on the observed correlation between alcohol sales and heart disease prevalence is not
necessarily correct. Correlation does not imply causation. There could be other factors or
confounding variables that are responsible for the observed association. It is important to
consider other potential explanations and conduct further research, such as experimental or
longitudinal studies, to establish a causal relationship between alcohol consumption and heart
disease.
2c. Select another epidemiologic study design and provide details around how you would
define the exposure and outcome, how you would sample your patients and any strengths and
limitations of the study design [10 points]
Another epidemiologic study design that could be used to investigate the link between alcohol
consumption and heart disease is a cohort study. In this design, a group of individuals without
heart disease would be identified and followed over time to assess their exposure to alcohol
consumption and the development of heart disease.
Exposure: Alcohol consumption would be defined as the amount and frequency of alcohol
consumed by each individual in the cohort.
Outcome: The outcome would be the occurrence of heart disease among the individuals in the
cohort during the follow-up period.
Sampling: A representative sample of individuals without heart disease would be recruited
from the general population or specific settings (e.g., hospitals, communities) and followed over
time to assess their exposure and outcome status.
Strengths: Cohort studies allow for the assessment of temporal relationships between exposure
and outcome, provide information on multiple outcomes, can study rare exposures, and can
estimate incidence rates.
Limitations: Cohort studies can be time-consuming, expensive, and subject to loss to follow-up.
They may also be influenced by confounding variables if not properly controlled for.
Question 3 (30 points)
You work for a small pharmaceutical company that is pushing the development of a new pain
medication. In order to get approved, you know there is a fairly rigorous process. Your
company decides that at some point around Phase 3, at least 2 interventional studies will need
to be conducted.
3a. What is the most likely study design that the company is considering for Phase 3? [2 points]
The most likely study design that the company is considering for Phase 3 would be Randomized
Controlled Trial.
Fall 2023 – Homework 2
Due: September 27th at the start of class; 95 points
Instructions:
For all calculations be as explicit as possible by showing you set up all calculations related to the
final answer. Showing your work is a component of the homework grade. Please submit the
homework electronically. All calculations can be rounded to one decimal place.
Question 1 [10 points]: As an epidemiologist working for the health department in
Philadelphia, you conduct a study looking at the impact of Covid-19 on mortality. You sample
100 individuals with Covid-19 and 100 individuals without Covid-19. Among the 100 with
Covid-19, 23 die where Covid-19 is thought to be the cause of the death. An additional 5 die
but the death is due to another cause. Among the 100 individuals without Covid-19, 14 die.
1a. Calculate the case-fatality rate of Covid-19. (3 points)
Case-fatality rate of Covid-19 = (number of people died from Covid-19/number of people who
have Covid-19)*100 = 23/100 = 23%
1b. Calculate the relative survival rate. (3 points)
Relative survival rate = (observed survival of individuals with Covid-19/expected survival of
individuals without Covid-19)*100
observed survival of individuals with Covid-19 = 100 - 28 = 72
expected survival of individuals without Covid-19 = 100 – 14 = 86
Relative survival rate = (72/86)*100 = 83.7%
1c. For relative survival rate, what other considerations would you make when recruiting the
patients into your study? (2 points).
When recruiting patients into a study for relative survival rate analysis, I will consider factors
such as age, gender, comorbidities, and other relevant characteristics that may affect survival
outcomes. This is important to ensure that the comparison group without Covid-19 is
comparable to the group with Covid-19 in terms of these factors.
1d. What additional information would you need to calculate 5 year survival rate? (2 points).
I would need information on the long-term survival outcomes of individuals with and without
Covid-19 over a period of 5 years. This would require follow-up data on the survival status of
individuals in both groups at 5 years after their initial diagnosis or enrollment in the study.
Question 2: [15 points] You’re researching the link between alcohol consumption and heart
disease. You come across an article that presents the average total alcohol sales per adult
person capita for 10 countries. The article also presents the prevalence of heart disease for
those 10 countries. The authors notice a correlation such that the countries with the highest
alcohol sales are the same ones with the highest prevalence of heart disease. They make the
conclusion that alcohol consumption likely causes heart disease.
, MI 580: Principles of Epidemiology
Fall 2023 – Homework 2
2a. Which study design did the authors select [2 points]
The authors selected an observational study design.
2b. Is there conclusion correct? Please provide justification to support your answer. [3 points]
The conclusion made by the authors that alcohol consumption likely causes heart disease based
on the observed correlation between alcohol sales and heart disease prevalence is not
necessarily correct. Correlation does not imply causation. There could be other factors or
confounding variables that are responsible for the observed association. It is important to
consider other potential explanations and conduct further research, such as experimental or
longitudinal studies, to establish a causal relationship between alcohol consumption and heart
disease.
2c. Select another epidemiologic study design and provide details around how you would
define the exposure and outcome, how you would sample your patients and any strengths and
limitations of the study design [10 points]
Another epidemiologic study design that could be used to investigate the link between alcohol
consumption and heart disease is a cohort study. In this design, a group of individuals without
heart disease would be identified and followed over time to assess their exposure to alcohol
consumption and the development of heart disease.
Exposure: Alcohol consumption would be defined as the amount and frequency of alcohol
consumed by each individual in the cohort.
Outcome: The outcome would be the occurrence of heart disease among the individuals in the
cohort during the follow-up period.
Sampling: A representative sample of individuals without heart disease would be recruited
from the general population or specific settings (e.g., hospitals, communities) and followed over
time to assess their exposure and outcome status.
Strengths: Cohort studies allow for the assessment of temporal relationships between exposure
and outcome, provide information on multiple outcomes, can study rare exposures, and can
estimate incidence rates.
Limitations: Cohort studies can be time-consuming, expensive, and subject to loss to follow-up.
They may also be influenced by confounding variables if not properly controlled for.
Question 3 (30 points)
You work for a small pharmaceutical company that is pushing the development of a new pain
medication. In order to get approved, you know there is a fairly rigorous process. Your
company decides that at some point around Phase 3, at least 2 interventional studies will need
to be conducted.
3a. What is the most likely study design that the company is considering for Phase 3? [2 points]
The most likely study design that the company is considering for Phase 3 would be Randomized
Controlled Trial.
