HWWWWW
PUBH 380: Intervention Studies/Randomized Trials
Homework 4 (33 points + BONUS!)
NAME:___________________________________
Objectives
· Understand the design and analysis of cohort studies
· Define and know how to calculate and interpret risk ratios, risk differences, attributable risk percent and rate ratios
· Understand the timing of measurements, i.e. prospective, longitudinal, retrospective
· Appreciate the circumstances in which cohort studies are appropriate
Procedure
· Read the paper by Fontham et al. (1991). Lung cancer in nonsmoking women: A multicenter case-control study. Cancer Epidemiology, Biomarkers, and Prevention, 1, 35-43. You will use it to answer question 3.
For questions 1&2: based on the table and formulas below, calculate, label, AND state the meaning of your answer (interpret) in a sentence:
|
Status as Adolescents |
Number of Participants |
CHD Deaths |
Person-years of observation |
|
Overweight |
238 |
40 |
9,329 |
|
Not overweight |
270 |
30 |
10,980 |
Rate Ratio = a/b
c/d
Risk Ratio = a/(a + b)
c/(c + d)
1. Complete the 2x2 table based on the information above: (6 points)
|
Status as Adolescents |
CHD Deaths |
Alive |
Total |
|
Overweight |
a |
b |
238 |
|
Not overweight |
c |
d |
270 |
|
Total |
a+c |
b+d |
508 |
a =
b =
c =
d =
a+c =
b+d =
a) Calculate the risk of coronary heart disease (CHD) death in participants who were overweight as adolescents and in participants who were not. (8 points)
b) Calculate the risk ratio of CHD death associated with having been overweight in adolescence. (4 points)
c) The rate ratio of CHD death for individuals overweight at adolescence compared to lean adolescents. (BONUS 4 points)
2. Based on your calculations, what can you conclude about the effect of being overweight during adolescence on the future risk of coronary heart disease? (2 point)
3. Use the data on "Education" in Table 4 of the Fontham et al. (1991) article to calculate the odds ratio for the association between less than high school education vs. (some college + college + graduate education) and lung cancer in nonsmoking women. Consider "exposed" as women with less than high school education; consider "nonexposed" as the combination of women with some college, college, and graduate education. Use the population controls as the control group.
a. What are the values of cells (A, B, C, and D) of the 2 by 2 table? (4 Points)
b. What is the value of the odds ratio? (5 Points)
c. Interpret this odds ratio in a sentence. Can this odds ratio (OR) be used as an estimate of the risk ratio (RR) (4 Points)
1