Kent State University Biostatistics Public Health Questions

Question 11 pts

Five epidemiologic studies were conducted on the same exposure-disease relationship. Assume that there is no bias and confounding in these studies. The following results were reported:

StudySample SizeRisk RatioP value95% Confidence IntervalA1002.0.100.8-4.2B5002.0.060.9-3.3C10003.5.022.6-4.5D20003.0.0152.2-3.5E25003.2.0013.8-3.6

1. Interpret each study result. Include interpretations of the risk ratio, P value, and confidence interval. Jot down your interpretation to compare against the information provided when you submit your self-check.
2. What is the relationship between the sample size and the width of the confidence interval? Choose the best answer below and enter the associated roman numeral here. 
   1. Sample size increases, the width of the confidence interval widens (sampling variability or uncertainty around the confidence interval increases)
   2. Sample size increases, the width of the confidence interval narrows (sampling variability or uncertainty around the confidence interval decreases)
3. What is the relationship between the sample size and P value? Choose the best answer below and enter the associated roman numeral here. 
   1. Sample size increases, the width of the P value widens (sampling variability or uncertainty around the confidence interval increases)
   2. Sample size increases, the width of the P value interval narrows (sampling variability or uncertainty around the confidence interval decreases)
4. Which give more information, the P value or the confidence interval? 
5. Is there a relationship between the sample size and the risk ratio? Enter “yes” or “no”. 
6. Are the five study results consistent based on statistical significance? Enter “yes” or “no”. 
7. Are the five study results consistent based on the point estimates and confidence intervals? Enter “yes” or “no” for each.
   • Point Estimates: 
   • Confidence Intervals: 

Flag question: Question 2Question 21 pts

In a prevalence study of smoking and lung cancer, 20,000 people in Ohio were surveyed in 2009. Of the total population surveyed 8,000 were smokers and 12,000 were non-smokers. A total of 80 people in the survey were identified as having lung cancer; 60 of those with lung cancer smoked.

1. What is the Prevalence Ratio (PR) and 95% confidence interval for lung cancer among smokers and non-smokers? Interpret the result. Do not use a computer program for your calculations (i.e. use formulas). Jot down your interpretations to compare against the information provided when you submit your self-check.
   • Prevalence Ratio: Round value to two decimal places. 
   • 95% Confidence Interval: Round values to two decimal places and separate with a comma. 
2. What is the Prevalence Odds Ratio (POR) and 95% confidence interval for lung cancer among smokers and non-smokers? Interpret the result. Do not use a computer program for your calculations (i.e. use formulas). Jot down your interpretations to compare against the information provided when you submit your self-check.
   • Prevalence Ratio: Round value to two decimal places. 
   • 95% Confidence Interval: Round values to two decimal places and separate with a comma. 
3. Are the PR and POR similar or different? Explain. Jot down your explanation to compare against the information provided when you submit your self-check. Enter only “similar” or “different” here.  

Flag question: Question 3Question 31 pts

In a study of oral contraceptive (OC) use and bacteriuria, a total of 2390 women age 16 to 49 years were indentified who were free from bacteriuria. Of these, 482 were OC users at the initial survey in 1973, while 1908 were not. At a second survey in 1976, 27 of the OC users had developed bacteriuria, as had 77 of the OC nonusers (Evans et al. Oral Contraceptives and bacteriuria in a community-based study. N Engl J Med 1978;299:536).

1. What is the cumulative incidence ratio (CIR) and 95% confidence interval for bacteriuria among OC users versus non-OC users? Interpret the result. Jot down your interpretations to compare against the information provided when you submit your self-check. 
   • Cumulative Incidence Ratio (CIR): Round value to two decimal places. 
   • 95% confidence interval: Round values to two decimal places and separate with a comma.  
2. What is the Risk Odds Ratio (ROR), 95% confidence interval for bacteriuria among OC users versus non-OC users?  Interpret the result. Jot down your interpretations to compare against the information provided when you submit your self-check.
   • Risk Odds Ratio (ROR): Round value to two decimal places. 
   • 95% confidence interval: Round values to two decimal places and separate with a comma. 

Flag question: Question 4Question 41 pts

Consider the following data from a study: Cohen et al. Prevention of HIV-1 Infection with Early Antiretroviral Therapy. N Engl J Med 2011;365: 493-505.

The clinical trial had two arms: immediate anti-retroviral (ART) therapy vs. deferred ART arm
Tuberculosis cases in immediate anti-retroviral (ART) therapy arm = 17

Person-years at risk of a clinical event in immediate ART arm = 1661.9
Tuberculosis cases in deferred ART arm= 33
Person-years at risk of a clinical event in deferred ART arm= 1641.8

Is the immediate ART associated with tuberculosis infection? What is the magnitude of the association and is chance a likely explanation of any observed association? Jot down your answer and compare it with the sample answer given when you submit your self-check.

Before submitting your self-check, enter rate ratio and its 95% confidence interval: (all values rounded to two decimal places; use leading zeros when appropriate; enter rate ratio followed by lower and upper limits in parentheses, separate by a comma) 

Flag question: Question 5Question 51 pts

Consider the following data from a study: Lannoy et al. Tuberculosis incidence and risk factors among patients living with HIV/AIDS in public health service institutions in Brasilia, Federal District. Rev Soc Bras Med Trop 2008;41: 549-555.

The incidence rate in people who started anti-retroviral therapy (ART) with baseline CD4 counts < 200 cells/µL was 0.60 cases / 100 person-years of observation (95% CI, 0.15 to 2.37).
The number of cases of Tuberculosis in this group was: 2

The incidence rate in people off ART with baseline CD4 counts < 200 cells/µL was 5.47 cases / 100 person-years of observation (95% CI, 2.73 to 10.94).
There number of cases of Tuberculosis in this group was: 8

Is ART therapy associated with Tuberculosis incidence? What is the magnitude of the association and is chance a likely explanation of the observed association? Jot down your answer and compare it with the sample answer given when you submit your self-check.

Before submitting your self-check, enter rate ratio and its 95% confidence interval: (all values rounded to two decimal places; use leading zeros when appropriate; enter rate ratio followed by lower and upper limits in parentheses, separate by a comma)