Reference no: EM132395160
Assignment
The Western Collaborative Group Study (WCGS) was a cohort study that followed people for 10 years to see who died of coronary heart disease (CHD). Apart from CHD, all variables were measured at the start of the study (baseline). You should answer the following questions which use techniques/analyses. You should present your results in report format. Summary statistics however should not be cut and pasted but presented in table format.
Question 1
The variable ‘chd' indicates whether someone dies of CHD over the 10 years of follow-up. First consider the variables systolic blood pressure (SBP), body mass index, and smoking status. You should summarise the distribution of each variable in those who die of CHD and those who don't. Present appropriate graphs (one graph for each variable is enough) and summary statistics. You should explain your choice of summary statistic (mean or median for example).
Question 2
Using a test-test compare the mean level of SBP in those who die of CHD and those who don't. You should report the results in table format like that below.
Title
|
Survived
|
Died of CHD
|
CHD vs Survived
|
|
|
Mean (SD)
|
Mean (SD)
|
Difference in means (95% CI)
|
p-value
|
Systolic blood pressure (mmHg)
|
|
|
|
|
SD = standard deviation. CI = confidence interval
Question 3
Using a chi-square test compare the probability of CHD mortality among those who smoke and those who don't at baseline.
Question 4
You want to compare the probability of CHD mortality in those who have arcus senilis (AS) and those who don't. Arcus senilis is a whitening of the eye. However, you suspect that the observed association may be confounded by age. You also want to examine whether the association between AS and CHD varies with age (effect modification).
You should create a variable ‘over50' that is 0 is someone aged under 50 and 1 if someone is aged 50 or over. Using methods described in weeks 5-8 examine whether there is any evidence of confounding or effect modification. You should report odds ratios (OR) by age-group along the Mantel Haenszel common OR and a test of homogeneity. What do you conclude and how would you report your results?
Question 5
Repeat question 4 but now look at the association between behaviour pattern (type A or type B) and CHD mortality. You should report odds ratios (OR) by behaviour pattern along the Mantel Haenszel common OR and a test of homogeneity. What do you conclude and how would you report your results?
Attachment:- Assignment File.rar