epi midterm

The ages of your subjects are as follows: 15, 17, 18, 18, 19, 19, 19, 20, 21, & 23. Readers will be interested both in the general distribution of age in your study and in the age limits of your data. Choose the best measure of central location, measure of spread (variability), and limits of your data from the list below. Note, you will choose 3 answers.

Range Standard Deviation Mean Response Feedback: Use of mean and SD is appropriate to the apparently normally distributed data.

Upon examination of data on population characteristics of Nebraska residents, you find the following table (Table 2) that shows the occupation-specific census for Nebraska for the time period 1957-1974. Based upon the information from Tables 1 and 2 (repeated below), calculate the occupation-specific leukemia mortality rates (per 100,000) for Nebraska residents in 1957- 1974. These data are used to calculate the answers for Q25-Q31. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was occupation-specific leukemia mortality rate for Other occupations (per 100,000)?

Selected Answer: 88 per 100,000 508/576,785 x 100,000= 88.07 Correct Answer:[None]Response Feedback: 88 per 100,000. Please be sure to express rates as such.

A researcher has conducted an observational study of 1,000 people to compare blood pressures among those who frequent a library (>1 a month) versus those who do not. The blood pressures are approximately normally distributed. This time she is interested in the mean blood pressures, rather than a diagnosis of hypertension. What is the most appropriate statistical test to compare the mean systolic blood pressure between groups?

Selected Answer: T-test Comparing a continuous variable such as blood pressure across two groups is well handled by a t-test. If there had been more than two exposure groups (such as visiting a library never, sometimes but less than weekly, or at least weekly), we may have instead selected an ANOVA approach.

Based on your calculated occupation-specific mortality rates, you now conclude that farmer's are at an increased risk of death due to leukemia.

Selected Answer: True Response Feedback: While we could not reach this conclusion from Table 1 alone, the additional information in Table 2 allowed us to examine what indeed appears to be increased risk in this population group. These data are from a well-known study that linked pesticide exposure to leukemia.

Direct transmission includes which of the following? vehicle-born transmission contact with oral secretions vector-borne transmission airborne transmission contaminated water

contact with oral secretions Response Feedback: Disease transmission dynamics are discussed in Gordis Chapter 2. Along with oral secretions, sexual contact is a common form of direct person-to-person transmission.

Which of the following is commonly used as an indicator of disease severity? cause-specific mortality attack rate secondary attack rate case fatality rate (ratio) proportionate mortality

case fatality rate (ratio) Response Feedback: Case fatality, and more general measures of mortality, are discussed in Gordis Chapter 4.

A small city has a population of 100,000 people (45,000 males and 55,000 females). In that city 1,000 people died in the year 2000 (600 males and 400 females). There were 50 new cases of lung cancer that year (40 males and 10 females); 45 died (36 males and 9 females). Compute the following rates for this city. These data are used to calculate the answers for Q20-Q24. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was crude mortality rate for 2000 (per 100,000)?

1000 per 100,000 1,000/100,000 x 100,000= 1000 Correct Answer: [None] Response Feedback:[None Given]

In a large urban area with a population of 10 million people, 80,000 deaths occurred during the year ending December 31, 2008. These included 1,300 deaths from firearm-related injuries. These data are used to calculate the answers for Q11-Q13. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. Calculate the cause-specific mortality rate from firearm-related injuries in 2008, per 100,000.

13 deaths per 100,000 population 13 per 100,000 1,300/10,000,000 x 100,000= 13 In the future, please record the appropriate units for the metric to be considered correct. (i.e. 13 deaths per 100,000 population). This is important for understanding and interpreting results.

A total of 2,123,323 deaths were recorded in the United States in 1987. The mid-year population was estimated to be 243,401,000. HIV-related mortality and population data by age for all races and for black males are shown in the Table below. These data are used to calculate the mortality rates for Q6-Q9. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What is the HIV-specific mortality rate among 35 to <45 years old?

14 per 100,000 4,794/34,305,000 x 100,000= 13.97 --> 14

Identify (match) the type of measure of disease that is being described by each of the following statements: QuestionCorrect MatchSelected Match

14.6 births per 1,000 population in 2010 H. crude or annual birth rate Six deaths from motor vehicle-related injuries per 100,000 population in 2010 C. cause-specific mortality rate 120 cases of influenza per 1,000 nurses in Community Hospital in December 2010 E. cause-specific disease incidence rate 867 deaths per 100,000 population in 2009 B.crude (or annual) mortality rate 6.9 infant deaths per 1,000 live births in 2010 I. infant mortality rate Two deaths per 1,000 population aged 10-44 in 2010 F.age-specific mortality rate Fifteen percent of all women with breast cancer in 2010 died from the disease G.case fatality rate (ratio) 724,859 deaths from heart disease among 2,337,256 total deaths in 2010 J.proportionate mortality Proportion of students enrolled in a college who developed influenza during the spring semester of 2010 A.cumulative incidence Percent of men found to have high blood pressure at their yearly physical examination D. prevalence

As of November 2010, the number of deaths from cholera in Haiti was 917. Best estimates of the number who were sickened in the outbreak was 14,600 with a total population of 9,719,932. These data are used to calculate the answers for Q14-Q15. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was the incidence of cholera in this population (per 100,000)?

150 deaths per 100,000 population 14,600/9,719,932 x 100,000= 150.20 --> 150

In Colorado there were 3530 traumatic cases of brain injury in 2000. Of these, 2859 were non-fatal. These data are used to calculate the answers for Q16-Q17. When typing in your answer, present the number as a percentage. Use only rounded whole numbers and do not include decimals, commas, or % sign in your answer. What was the case-fatality rate (ratio) for traumatic brain injury in Colorado in 2000?

19% case-fatality rate 671/3530 x 100= 19.0 --> 19%

In a large urban area with a population of 10 million people, 80,000 deaths occurred during the year ending December 31, 2008. These included 1,300 deaths from firearm-related injuries. These data are used to calculate the answers for Q11-Q13. When typing in your answer, present the number as a percentage. Use only rounded whole numbers and do not include decimals, commas, or % sign in your answer. Calculate the proportionate mortality for firearm-related deaths in 2008.

2% for firearm related deaths in 2008 1,300/80,000 x 100= 1.625 --> 2 1,300 ÷ 80,000 x 100 = 2%

Upon examination of data on population characteristics of Nebraska residents, you find the following table (Table 2) that shows the occupation-specific census for Nebraska for the time period 1957-1974. Based upon the information from Tables 1 and 2 (repeated below), calculate the occupation-specific leukemia mortality rates (per 100,000) for Nebraska residents in 1957- 1974. These data are used to calculate the answers for Q25-Q31. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was occupation-specific leukemia mortality rate for Farmers (per 100,000)?

211 per 100,000 Farmers 433/205,672x 100,000= 210.5 --> 211

In 1995, there were 11,700 reported cases of Lyme disease in the U.S. The estimated population of the U.S, as of July 1, 1995, was 262,755,000. These data are used to calculate the answers for Q18-Q19. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. Please calculate the measure given as your answer to Q18 (per 100,000)?

4 per 100,000 people at risk 11,700/262,755,000 x 100,000= 4.45

Upon examination of data on population characteristics of Nebraska residents, you find the following table (Table 2) that shows the occupation-specific census for Nebraska for the time period 1957-1974. Based upon the information from Tables 1 and 2 (repeated below), calculate the occupation-specific leukemia mortality rates (per 100,000) for Nebraska residents in 1957- 1974.These data are used to calculate the answers for Q25-Q31. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was occupation-specific leukemia mortality rate for Teachers (per 100,000)?

41 per 100,000 teachers. 9/22076 x 100,000= 40.7 --> 41

A small city has a population of 100,000 people (45,000 males and 55,000 females). In that city 1,000 people died in the year 2000 (600 males and 400 females). There were 50 new cases of lung cancer that year (40 males and 10 females); 45 died (36 males and 9 females). Compute the following rates for this city. These data are used to calculate the answers for Q20-Q24. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was cause-specific mortality rate for lung cancer in 2000 (per 100,000)?

45 per 100,000 45/100,000 x 100,000= 45

A small city has a population of 100,000 people (45,000 males and 55,000 females). In that city 1,000 people died in the year 2000 (600 males and 400 females). There were 50 new cases of lung cancer that year (40 males and 10 females); 45 died (36 males and 9 females). Compute the following rates for this city. These data are used to calculate the answers for Q20-Q24. When typing in your answer, present the number as a percentage. Use only rounded whole numbers and do not include decimals, commas, or % sign in your answer. What was proportionate mortality ratio (PMR) for lung cancer in 2000?

5% PMR for lung cancer in 2000 45/1,000 x 100= 4.5-- > 5

Upon examination of data on population characteristics of Nebraska residents, you find the following table (Table 2) that shows the occupation-specific census for Nebraska for the time period 1957-1974. Based upon the information from Tables 1 and 2 (repeated below), calculate the occupation-specific leukemia mortality rates (per 100,000) for Nebraska residents in 1957- 1974. These data are used to calculate the answers for Q25-Q31. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was occupation-specific leukemia mortality rate for Engineers and scientists (per 100,000)?

50 per 100,000 engineers and scientists 25/50,456 x 100,000= 49.5 -->50

As of November 2010, the number of deaths from cholera in Haiti was 917. Best estimates of the number who were sickened in the outbreak was 14,600 with a total population of 9,719,932. These data are used to calculate the answers for Q14-Q15. When typing in your answer, present the number as a percentage. Use only rounded whole numbers and do not include decimals, commas, or % sign in your answer. What was the case-fatality rate (ratio) for cholera in Haiti in the 2010 outbreak?

6% case-fatality rate 917/14,600 x 100= 6.28 --> 6

Upon examination of data on population characteristics of Nebraska residents, you find the following table (Table 2) that shows the occupation-specific census for Nebraska for the time period 1957-1974. Based upon the information from Tables 1 and 2 (repeated below), calculate the occupation-specific leukemia mortality rates (per 100,000) for Nebraska residents in 1957- 1974. These data are used to calculate the answers for Q25-Q31. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was occupation-specific leukemia mortality rate for Laborers (per 100,000)?

69 per 100,000 65/94,572 x 100,000= 68.7 --> 69

A small city has a population of 100,000 people (45,000 males and 55,000 females). In that city 1,000 people died in the year 2000 (600 males and 400 females). There were 50 new cases of lung cancer that year (40 males and 10 females); 45 died (36 males and 9 females). Compute the following rates for this city. These data are used to calculate the answers for Q20-Q24. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was Sex-specific mortality rate for women in 2000 (per 100,000)?

727 per 100,000 400/55,000 x 100,000= 727.27--> 727

A total of 2,123,323 deaths were recorded in the United States in 1987. The mid-year population was estimated to be 243,401,000. HIV-related mortality and population data by age for all races and for black males are shown in the Table below. These data are used to calculate the mortality rates for Q6-Q9. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What is the HIV-specific mortality rate among black males 35 to <45 years old?

73 per 100,000 1,212/166,3000 x 100,000= 72.8 --> 73 Correct Answer: [None] Response Feedback:[None Given]

In a large urban area with a population of 10 million people, 80,000 deaths occurred during the year ending December 31, 2008. These included 1,300 deaths from firearm-related injuries. These data are used to calculate the answers for Q11-Q13. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. Calculate the crude death rate for this population in 2008, per 100,000.

800 deaths per 100,000 population 80,000/10,000,000 x 100,000= 800

In Colorado there were 3530 traumatic cases of brain injury in 2000. Of these, 2859 were non-fatal. These data are used to calculate the answers for Q16-Q17. When typing in your answer, present the number as a percentage. Use only rounded whole numbers and do not include decimals, commas, or % sign in your answer. What was the survival rate for traumatic brain injury in Colorado in 2000?

81% survival rate 2859/3530 x 100= 80.9--> 81

Upon examination of data on population characteristics of Nebraska residents, you find the following table (Table 2) that shows the occupation-specific census for Nebraska for the time period 1957-1974. Based upon the information from Tables 1 and 2 (repeated below), calculate the occupation-specific leukemia mortality rates (per 100,000) for Nebraska residents in 1957- 1974. These data are used to calculate the answers for Q25-Q31. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was occupation-specific leukemia mortality rate for Clerks and salespersons (per 100,000)?

85 per 100,000 Clerks and salespersons 84/98,832 x 100,000= 84.9 --> 85

A total of 2,123,323 deaths were recorded in the United States in 1987. The mid-year population was estimated to be 243,401,000. HIV-related mortality and population data by age for all races and for black males are shown in the Table below. These data are used to calculate the mortality rates for Q6 - Q9 . When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What is the crude mortality rate (per 100,000)?

872 deaths per 100,000 2,123,323 /243,401,000x 100,000= 872.36--> 872

A researcher has conducted an observational study of 1,000 people to compare blood pressures among those who frequent a library (>1 a month) versus those who do not. The blood pressures are approximately normally distributed. She would like to find out if hypertension (assume a systolic blood pressure > 140 mm Hg) is less prevalent in the library group. What is the most appropriate statistical test to compare the prevalence of hypertension between groups?

Chi-squared test This analysis has two exposure groups (library, no library) and two outcome groups (hypertension, no hypertension). A Chi-squared test is appropriate to see whether the two categorical variables are associated.

The following information is used for Q16 & Q17 A researcher writes, "Before the study we set our alpha (α) level at 0.01. The main comparison we conducted had a p-value = 0.04." True or False. This is a statistically significant result.

FALSE Response Feedback: This result is not significant, because the p-value is greater than α.

A study of mortality among residents of Nebraska used death certificates over the time period 1957-1974. The study observed a high number of cases of leukemia among farmers. Leukemia deaths by the decedents' occupations are shown in Table 1. These data are used to calculate the answers for Q25-Q31. True or False: From the data shown in Table 1, farmers are at increased risk of death from leukemia.

False Response Feedback: Crucially missing from the provided table is the number of persons in each occupational group who were at risk of death from leukemia. While 38.5% of the deaths are among farmers, that is only suggestive of increased risk if that is more than we would expect. It could be that approximately 38.5% of the population is in this occupational group.

In 1995, there were 11,700 reported cases of Lyme disease in the U.S. The estimated population of the U.S, as of July 1, 1995, was 262,755,000. These data are used to answer Q18-Q19. What would be most appropriate measure of the risk of getting Lyme disease if you lived in the U.S. that year?

Incidence Incidence specifically provides information on the fraction of the population being infected with new cases of Lyme disease over a period of time, which is the most relevant to individual risk. While incidence being high will also tend to increase prevalence and mortality rate, prevalence is additionally affected by how long the infection lasts and mortality rate is additionally affected by the severity and available treatment.

For another variable, you find a mean of 50, a median of 30, and a range from 0 to 200. Choose the best measure of central location and measure of spread (variability) from the list below. Note, you will choose 2 answers.

Median Interquartile Range Response Feedback: This variable is not normally distributed so we would use the median and IQR. Note that we could also use the range for the measure of spread, but the IQR would better characterize the bulk of the data, leaving out any outlying values.

The following information is used for Q16 & Q17. A researcher writes, "Before the study we set our alpha (α) level at 0.01. The main comparison we conducted had a p-value = 0.04." In this case, what type of error do you think applies here?

Selected Answer: Type I error Feedback: Setting alpha to 0.01 is designed to prevent a type I error (the p-value must be less than alpha to be considered 'significant'). The risk of rejecting the null falsely (Type I error) is reduced even more with this smaller alpha than with the conventional 0.05. However, the tradeoff is that the risk of concluding that there is no difference when there is an underlying true difference (Type II error) is increased.

A small city has a population of 100,000 people (45,000 males and 55,000 females). In that city 1,000 people died in the year 2000 (600 males and 400 females). There were 50 new cases of lung cancer that year (40 males and 10 females); 45 died (36 males and 9 females). Compute the following rates for this city. These data are used to calculate the answers for Q20-Q24. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What was cause-specific mortality rate for lung cancer among males in 2000 (per 100,000)?

The question is asking for the cause-specific mortality rate for lung cancer among males, so the denominator should be the total number of males in the population. = (36/45000) * 100000 = 80 per 100000 males

A researcher reports, "We found no evidence for a difference between treatment and control, p = 0.30." In this case, what type of error do you think applies here?

Type II error The p-value of 0.3 does not allow us to exclude chance as an explanation for the observed difference between treatment and control groups. Thus, the researcher may concluding there is a difference (rejecting the null) even if there is no underlying difference (the null hypothesis of no difference in the underlying population is true). A type I error is of the most concern here. In contrast, a type II error arises when a researcher fails to reject a null hypothesis even though there truly is a difference in the population. Of note: that the observed difference is large and yet the p-value suggests this could be due to chance points to a lack of statistical power; the sample size may be too small.

A total of 2,123,323 deaths were recorded in the United States in 1987. The mid-year population was estimated to be 243,401,000. HIV-related mortality and population data by age for all races and for black males are shown in the Table below. These data are used to calculate the mortality rates for Q6-Q9. When typing in your answer, present the number adjusted per 100,000. Use only rounded whole numbers and do not include decimals or commas in your answer. What is the HIV (cause)-specific mortality rate for the entire population?

[13,468 ÷ 243,401,000] x 100,000 = 6 HIV-related deaths per 100,000

An epidemiologist with an interest in injury prevention wants to examine death rates in the United States from firearm-related injuries from 1900 through 2000. In examining rates in the same population over such a lengthy period of time, what kind of rates are the most appropriate to explore and compare? proportionate mortality rates for firearm-related injuries age-adjusted firearm-specific mortality rates case-fatality rates from firearms injuries age-specific firearm-related mortality rates

age-adjusted firearm-specific mortality rates Response Feedback: Age-adjustment is helpful in this scenario because the age distribution of the US population changed over the time period, and because age is strongly related to firearm-related injuries. Demographic change may have explained a change in the overall rate of firearm-related injuries, but a change in age-specific rates would be more useful to inform injury prevention efforts.

Deaths from HIV & AIDS among 20-24 year-olds in a population is an example of which kind of rate? a disease-specific mortality rate an age- and cause-specific death rate an age-adjusted rate an age-specific incidence rate

an age- and cause-specific death rate Response Feedback: Age- and cause-specific are discussed in Gordis Chapter 4.

In the definition of epidemiology, the terms distribution and determinants together refer to:

frequency, pattern, and causes of health events Response Feedback: For the definition of epidemiology, see the section "What is Epidemiology" in Chapter 1 of the Gordis Epidemiology.

In the epidemiologic study of Kawasaki syndrome (a syndrome of unknown origin that affects the skin, mucous membranes, and the immune system of infants and young children) the mean serum protein levels of children with Kawasaki syndrome was lower than the mean serum protein levels of children without Kawasaki syndrome. The difference was statistically significant at the 5% level (p < 0.05). This means that

the difference between mean serum protein levels is unlikely to have occurred by chance alone While additional information would be needed before inferring causation (to evaluate alternate explanations such as reverse causation bias, confounding bias), the p-value does provide some information that the observed difference is unlikely to arise due to chance alone if the null hypothesis of no difference is true.