Epidemiology Homework 2 PDF

Preview

Summary

Epidemiology Homework 2 Professor Quinlan....

Description

Department of Epidemiology Milken Institute School of Public Health Homework Assignment #2 – Measuring Health PubH 3131: Epidemiology: Measuring Health and Disease Spring 2021 Due Date: Monday, February 8 at the start of class (11:10AM Eastern). Please upload an electronic copy of this assignment to Blackboard by the specified due date. Purpose: The purpose of this second homework assignment is to give you some practice computing and interpreting key measures of health and disease. We will look at measures of incidence, prevalence, and mortality to see what they tell us about the health of populations of interest. Skills: This second homework assignment will provide an opportunity:  Compute and interpret measures of prevalence and incidence  Consider what prevalence and incidence measures tell you about a population’s health  Identify ways we might be able to measure a disease of interest  Compute and interpret measures of mortality Instructions: The second homework assignment contains 18 questions and 90 possible points and is worth 7.5% of your course grade. Work individually! For each question please show your work as partial credit may be awarded according to the following rubric: 0% 50% 75% 100% Question approached Question approached Question attempted, Question blank, correctly and all parts and completed but minor error in incorrect with no correctly with no answered but final approach or only work provided, or answer incorrect due errors partially completed major error in to minor error approach. Tools: To complete the second homework assignment you may use:  Any relevant textbooks or other lecture materials  Online resources STUDENT NAME:____Abigail Alpert_____________________

Schools and COVID-19 We are trying to assess how two different schools have done in addressing the COVID-19 pandemic. Use information below to answer the questions that follow. At the start of the week School A has 500 students and staff. Testing reveals that 55 of them have COVID-19 and 30 are still sick by the end of the week. During the week, 35 students/staff develop new cases of COVID-19 and all are still sick by the end of the week. In addition, 50 healthy students without COVID-19 leave the school in the middle of the week. At the start of the week School B has 250 students and staff. Testing reveals that 35 of them have COVID-19 and 10 are still sick by the end of the week. In addition, 25 staff members have been vaccinated and are presumed to be immune from COVID-19. During the week, 10 students/staff develop new cases of COVID-19 and all are still sick by the end of the week.

1. At which school is the burden of COVID-19 greater at the start of the week? You must provide a calculation to justify your answer. (10 points) Point prevalence start of week school A: number who are sick 55 = =0.11∗100=11 % ¿ population∈school 500 Point prevalence start of week school B: number who are sick 35 = =0.14∗100=14 % ¿ population∈school 250 School B has a higher burden of Covid-19 based on the calculation. Prevalence is a good measurement to calculate the burden and we use point prevalence because we are interested on a specific date. 2. At which school is the risk of developing COVID-19 during the week greater? You must provide a calculation to justify your answer. (10 points) Cumulative Incidence School A: number of new cases during time period number at risk at start of time period 35 ¿ =0.0786∗1,000=78.6 per 1,000 500−55 ¿

Cumulative Incidence School B: number of new cases during time period ¿ number at risk at start of time period ¿

10 10 = =0.0526∗1,000 =52.6 per 1,000 250 −35−25 190 2

We calculate cumulative incidence because it is a good indication of risk. The denominator for A is determined by subtracting those already who were sick because they are no longer at risk from the beginning of the time period considered. We use the population from beginning of the week, so it does not affect our calculation that 50 students left the school in the middle of the week. Same is done for school B where we do not include teachers who are vaccinated and those who were sick in the beginning of the week. Based on our calculations, School A has higher risk of developing COVID-19 during 19. 3. At which school is the burden of COVID-19 higher at the end of week? You must provide a calculation to justify your answer. (10 points) Point prevalence end of week school A: 65 number who are sick = 30 + 35 = =0.144∗100=14.4 % ¿ number still ∈school 500−50 450 Point prevalence end of week school B: number who are sick = 10+ 10 =0.08∗100=8 % ¿ number still ∈school 250 Point prevalence is a good indication of burden for the end of the week. Based on the calculation, school A has a higher burden of Covid-19 at the end of the week because it has a higher point prevalence. Impact on Prevalence 4. The prevalence of diabetes is increasing in the Country X. Based on what you know about prevalence, which of the options below could explain this increase? Provide a justification for your choice. (4 points) a. The case definition for diabetes has become stricter, requiring laboratory tests. b. The population of Country X has increased in size. c. Better management strategies for diabetes allow people to live longer. d. Improvements in diet reduce the number of new cases each year.

The correct answer is C. This answer increases the duration of diabetes. The calculation for prevalence is incidence * duration. Therefore, by increasing duration, prevalence increases as well. Another way to think about this is that people with this disease are living longer so they are staying in the “bathtub” a greater period of time whereas usually people leave the bathtub because they die sooner.

3

A, B, and D would all decrease prevalence. In the case for B, we are unsure if the increased population includes more people who have diabetes or if they are healthy so there is not enough information to conclude it can increase prevalence. 5. Regulators ban flavored e-cigarette pods nationwide and this leads to a reduction in the number of new cases of vaping-related lung disease diagnosed each month. The duration of the disease remains unchanged, however. What impact will this have on the long-term point prevalence of vaping-related lung disease? Justify your choice in 1-2 sentences. (4 points) A reduction of new cases means a lower incidence rate of vaping related lung disease (less “water” entering the “bathtub”) which means a decrease in prevalence. Because prevalence = incidence * duration, prevalence will decrease because incidence is decreasing. 6. Being a smoker has been shown to increase the time it takes for a person to recover from the flu. Assume in Town A no one smokes. In Town B, half of the population smokes. If the number of new cases per week is that same in the two towns, which town will have the higher point prevalence of the flu by the end of the week? Justify your choice in 1-2 sentences. (4 points) Town B will have a higher point prevalence of the flu by the end of the week because recovery time will take longer for citizens of Town B who smoke compared to citizens of Town A where no one smokes. Since duration is shorter in town A, and because prevalence = incidence * duration, town A will have a smaller prevalence.

Incidence by Any Name We conduct a small study of two different dietary supplements to prevent peanut allergy. In each case, we enroll 5 people who are free of peanut allergy and then follow them forward to see what happens over the 5 years of our study. Our goal is to see which supplement is superior in preventing peanut allergy. The data from our study is below.

Person

Supplement A Outcome

1 2 3

Followed 5 years, no allergy Followed 5 years, no allergy Developed allergy after 3 years

Perso n 1 2 3

4

Developed allergy after 4 years

4

Supplement B Outcome Developed allergy after 1 year Died at the end of year 1 Left the study allergy free after 1 year Left the study allergy free after 2 years 4

5

Left the study allergy free after 4 years

5

Followed 5 years, no allergy

7. Calculate the cumulative incidence (risk) of peanut allergy over the 5-year study for Supplement A and Supplement B. Based on these results, which supplement appears superior? (5 points) Cumulative incidence Supplement A: number of new cases during time period ¿ number at risk at start of time period 2 ¿ ∗1,000=400 per 1,000 5 Cumulative incidence Supplement B: number of new cases during time period ¿ number at risk at start of time period 1 ¿ ∗1,000=200 per 1,000 5 Based on the cumulative incidence over the 5-year study, Supplement B appears superior to Supplement A. 8. Calculate the incidence density rate of peanut allergy over the 5-year study for Supplement A and Supplement B. Based on these results, which supplement appears superior? (5 points) Incidence Density Supplement A number of new casesduring time period 2 =0.28 per person− year = total person−timeat risk duringtime period (3+ 4) Incidence Density Supplement B 1 number of new casesduring time period = =1 per person− year ¿ total person−timeat risk duringtime period 1 ¿

Based on these results, Supplement A appears superior to Supplement year because less people developed a peanut allergy each year.

5

A Rough Flu Season The 2017-2018 flu season was especially rough. By the end of the season, an estimated total of 49 million people in the United States had contracted the flu, of whom an estimated 79,000 died from the flu. Assume that the United States population was 300 million during the 2017-2018 flu season and that during the season a total of 2.9 million Americans died from all causes (including flu). Assume the population of the United States remained the same at the midpoint of the season. 9. What was the proportionate mortality from flu during the 2017-2018 flu season? (5 points) number of deaths number of deaths Proportionate Mortality Rate=¿ cause∈time period ¿ all causes∈time period ¿ ¿ ¿

79,000 =0.027∗100=2.7 % 2,900,000

10. What was the case fatality rate from flu during the 2017-2018 flu season? (5 points) number of people who die ¿ Case fatality Rate=¿ a disease ∗100 number of people who havethe disease ¿

79,000 =0.0016∗100=0.16 % 49,000,000

11. What was the flu-related mortality rate during the 2017-2018 flu season? (5 points) totaldeaths during time period ¿ Mortality Rate=¿ flu total population that could have died

¿

79,000 ∗100,000=26.33 per 100,000 people 300,000,000

12. In a country with an age distribution that is much younger than in the United States the proportionate mortality from flu in 2018 was 5%. Based on this information alone, can we determine that the risk of dying from flu in 2018 was higher in this other country compared to the United States? Why or why not? (5 points)

6

Comparing mortality between two countries can be misleading due to the different age structure between the populations. Therefore, we cannot based on this information alone determine the risk of dying from flu in 2018 was higher in this country compared to the United States. In order to make comparisons we would need data that is standardized between the two populations.

13. You are tasked with designing a prevention program to target one of two different infectious diseases. You only have funding to target one of them, so you must chose. The first infectious disease (Disease A) has a case fatality rate of close to 75%, but a very low cause-specific mortality rate. The other (Disease B) has a case fatality rate less than 0.1%, but a relatively high cause-specific mortality rate. Which do you choose? Provide a brief (1-2 sentences) justification. (5 points) I would choose a prevention program to target disease A because the case fatality rate is very high (as seen in rabies). Since 75% of those infected die, compared to less than 1% from disease B, I believe it is more effective to research and come up with a vaccine for disease A.

Maternal and Infant Mortality In 2018, the total US population was approximately 305 million people. During that year, there were approximately 2.7 million deaths from all causes. In addition, during the year there were 3.9 million live births. Over the course of the year, 850 women died due to pregnancy-related complications, and 52,000 children under 18 years of age died, of which 22,258 children were less than one year of age. 14. What was the infant mortality rate for the US in 2018? (5 points)

Infant Mortality Rate= ¿

deaths of childrenless than1 year of age total number of live births

22,258 ∗1,000=5.71 per 1,000 live births 3,900,000

15. What was the maternal mortality rate for the US in 2018? (5 points) deaths related ¿ Maternal Mortality Rate=¿ childbirth total number of live births 850 ∗100,000=21.79 per 100,000 live births ¿ 3,900,000

7

Name that Measure For each statement below, indicate the measure of morbidity or mortality being referenced.

16. Recent estimates indicate that globally, approximately 2% of people who develop symptomatic COVID-19 will die from it. (1 point) Case fatality rate 17. Approximately 3.5 million Americans are current suffering from gout. (1 point) Point Prevalence 18. Fewer people are dying of cancer each year, with the estimate for 2020 being 149 deaths per 100,000 people. (1 point) Cancer mortality rate

8...