Epidemiology
The questions in Section 1 relate to thearticle:
Argos M et al. Arsenic exposure from drinking water, and all-cause andchronic-disease mortalities in Bangladesh (HEALS): a prospective cohort study. The Lancet 2010;376:252- 258.
Q1. What was the aim of thisstudy? (1mark)
Q2. Identify the criteria for selection into the study and discuss the advantagesand disadvantages of thesecriteria. (5marks)
Q3. a) Were the authors successful at minimising loss-to-follow-up? Provide data to support youranswer. (3marks)
b) Who (which groups) would be most likely to dropout? (2marks)
Q4. What was the primary outcome of the study and how was itascertained? (2marks)
Q5.Commentontheadvantagesanddisadvantagesofthemethodofascertainmentofthe outcome? (4marks)
Q6. a) What were the exposure variables and how were theydefined? (3marks)
b) Comment on the strengths and weaknesses of the exposurevariables. (5marks)
Q7. Is there evidence that BMI is a confounder of the relationship between arsenic exposure andmortality? (2marks)
Q8. On page 255, 3rdparagraph, the authors report the finding that “a one-quartile increasein arsenicconcentrationinwellwaterwasassociatedwitha15%increaseinall-cause mortality (95% CI 1.05-1.26)”. Explain in your own words the interpretation of this95% CI. (2marks)
Q9. On page 255, 5thparagraph, the authors report the finding that“multivariate-adjusted HazardRatio(HR)forcomparisonofhighbaselineexposuretolowbaselineexposure was1.46(95%CI1.14-1.86)fordeathsoccurringafterfollowup1”.Explaininyour own words the interpretation of theHR. (3marks)
Q10.Didtheauthorsobserveachangeinriskofdeathassociatedwithchangesinarsenic concentration in urine over time? Refer to or provide data to support youranswer.
(2marks)
Q11. What is the interpretation of the attributable proportion based upon well waterfor chronic-disease mortality of24%? (2marks)
Q12. Imagine you are designing a RCT to evaluate the impact on urine arsenic ofan interventiontoreducearsenicexposureinwellwater.Theinterventionistheone–offaddition of a chemical to remove arsenic in the water. The follow up will be 6months.
a) Who would be your studypopulation? (2marks)
b) Draw a diagram to illustrate the studydesign. (3marks)
c) Identify and explain one ethical issue of thisstudy. (3marks)
– End of Section 1-
Q13.Anoutbreakofgastroenteritisthatappearstoberelatedtotheconsumptionoffastfood has recently occurred in Mandurah. A case-control study was undertaken andthe following resultsfound.
Fooditem Cases (n=19) Controls (n=17)
Crumbedchicken 14 11
Anychicken 16 16
Eggrolls 14 3
Friedrice 14 9
a) Whichfood(s)doyoususpecttobethecauseofthisoutbreak?Givereasonsforyour answer. (5marks)
b) Was a case-control study the best study design to use here? Explain youranswer.
(4marks)
Q14. In a matched cohort study of oral contraceptive (OC) use and breast cancer, wherethe exposed: non-exposed ratio was 1:1, the matched-pair findings were asfollows.
No history of OCuse
Breast cancer Nobreast cancer
PreviousOCuse Breast cancer 12 102
Nobreast cancer
32
87
a) What is the relative risk of breast cancer for those with a history of OC use? Showall working. (5marks)
b) The 95% confidence interval for the RR is 1.1- 26.5. What does this mean in termsof possible random error in thestudy? (2marks)
Q15. A new screening test for HIV in blood products was administered to 700 peoplewith clinically proven HIV and to 900 people without HIV. The screening test waspositive for 670 of the proven HIV cases and 150 of the people withoutHIV.
a) Calculate the sensitivity and specificity of thetest. (2marks)
b) TheoldscreeningtestforHIVhasasensitivityof80%anditsspecificityis97%. Discuss if you would recommend the old or the new screening test forcommunity based screening of blooddonations? (3marks)
Q16. A number of employees at a factory in WA have recently been diagnosed with cancerand theunionhasdemandedaninvestigationintowhetherexposuretothefactoryworking environmentisplacingtheirmembersatincreasedriskofcancer.Youareaskedto determine whether an increased risk exists and if so provide an estimate of the degreethat factory workers are at risk compared with thecommunity.
A total of 16 cases of the cancer have been diagnosed in factory workers over the last10 years. You collect the following data from those who have had a cancer diagnosiswhile working at the factory over this timeframe.
AGE GROUP inyears PERSON YEARS ATRISK Cases
15-19 22.2 0
20-24 325.8 0
25-29 440.1 1
30-34 505.7 2
35-39 425.1 3
40-44 322.8 4
45-49 243.8 3
50-54 211.7 2
55-59 129.6 0
60-64 79.8 1
65-69 5.0 0
70-74 3.0 0
TOTAL 2714.4 16
Recent research has reported that the incident rate of the cancer in the WA populationis 3330 per 100,000 person-years, with an age distribution in the community asfollows:
AGE GROUP inyears Age-specific IR in WesternAustralia(per100,000PY)
15-19 0.283
20-24 1.646
25-29 11.770
30-34 44.110
35-39 103.474
40-44 196.288
45-49 335.884
50-54 413.418
55-59 503.016
60-64 569.476
65-69 565.352
70-74 585.785
Usingtheaboveinformationabove,determinetheage-standardisedrateratio(showall working)forcancerinthefactoryworkerscomparedwiththecommunityanddescribe (in lay terms) what this resultmeans. (5marks)
End ofAssignment
