Statistical Methods in Epidemiology (401176)
Added on - 23 Feb 2020
Trusted by +2 million users,
1000+ happy students everyday
1000+ happy students everyday
Showing pages 1 to 2 of 4 pages
1Western Sydney UniversitySchool of MedicineStatistical Methods in Epidemiology (401176)ASSIGNMENT 1Spring Semester, 2017Due date: 4 September, 2017This assignment is based on the learning objectives and concepts for ‘Descriptive statistics’ asdescribed in the Unit Learning Guide. To answer question 4, SAS codes are available in vUWS inorder to analyse the given data forthat question. For answering question 5, you are required tocalculate the test statistic using a scientific calculator or Microsoft Excel based on the formulaprovided in class lecture which is also available in vUWS under ‘Lectures’. For answering questions 1through 3, no computation is required. There are a total of 100 marks and this assignment willcontribute 25% towards the total assessment for this subject. The marking matrix that will be usedfor this assignment is also attached.This assignment is limited to 10pages.See specific instructions below relating to the pageformatting of Assignment 1.Assignments that do not comply with these specifications will not bemarked.Your assignment should be typed, with adequate space left between questions. Assignments shouldbe submitted via vUWS. Be to the point in your answers, and use the number of marks allocated toeach question as a guide for how much to write.Please notethis is an individual exercise.Late assignments will not be accepted withoutpriorapproval.
2Please answer all questionsQ.1(i)Which is a better graphical method for describing frequency distribution for a continuousvariable-histogram or Kernel density plot ?(ii) Considerastudy involvingarandomsample of 36setsof twin infants.Onememberofeachtwin-pairisrandomlychosentoreceivedietAandtheotherdietB.Theonemonth weight gainismeasuredforall 36x2=72infants.Wewish tocomparethe mean one-month weightgainforthetwodiets,i.e.AandB.Describe the statistical test you would perform? How would youconstructa95%confidenceintervalfor the difference?(iii)If you were to compare the spread/variation for two variables with different measurementscales, which statistical measure would you use ?Q2.(i) Explain confounding and effect modification. Which has a stronger effect ?(ii) What are the consequences of not adjusting for confounding in its presence while assessing adisease-exposure relationship ? What is residual confounding bias ?(iii)Which is a more efficient method for adjusting confounding bias-stratification or matching ?(iv) What are the main limitations of stratification as a method for adjusting confounding bias ?(iv) When isthe relative odds equivalent to therisk ratio? Whichis the preferred statistical measureof association in a (a) case-control study, (b) cross-sectional study, (c ) prospective randomizedcontrolled study and (iv) retrospective cohort study.Q3.A variable can be a confounder, effect modifier, both or none of the two. There are statistical testsfor detecting effect modification. But, there is no statistical test for detecting an operationalconfounder. For example, if a test for comparing unadjusted and adjusted odds ratios show nosignificant difference, but one is considerably larger than the other, then one would still adjust for theconfounder. However, if a test for comparing unadjusted and adjusted odds ratios shows significantdifference, but one is not considerably larger than the other, one would not have to adjust for theconfounder.Let us consider a study for assessing the association between smoking &lung cancer. Is sex aconfounder or effect modifier (quantitative or qualitative)?We have 4 different scenarios, such as:OR (Men) OR (Women) Crude OR Adj OR2.51 2.15 2.32 2.351.06 0.95 2.02 1.014.40 3.41 4.02 2.632.15 0.65 1.42 1.29The following table presents unadjusted and age-adjusted coronary event rates and death subsequentto a coronary event, for men in north Glasgow, 1991. The exposure of interest is social deprivation. Is