Question 1 Which type of study design is each of the following, briefly describe why?

i The smoking histories of patients entering a hospital with lung cancer are compared with the smoking histories of patients entering for other conditions requiring surgery.

ii Inductees into the army are asked about their smoking history and current habits. Smokers and non-smokers are subsequently compared in relation to the development of lung cancer and other chronic diseases.

iii In a study of the relationship between reproductive abnormalities and in utero exposure to diethylstilbestrol (DES), the incidence rates of reproductive abnormalities in individuals whose mothers were exposed to DES when they were pregnant 20-30 years previously are compared to the incidence rates of reproductive abnormalities in individuals who were not exposed to DES 20-30 years previously.

iv Estrogen levels are measured in the blood of a sample of women in the age group 50-59 years. At the same examination, their bone mineral density is also measured. The proportion of women with low bone mineral density is compared in relation to estrogen level.

Question 2 Sputum samples of 5776 patients were examined by direct smear (screening test) and culture examination (gold standard for diagnosis) for the presence of tuberculosis bacilli at the laboratory of New Delhi Tuberculosis Centre. The direct smear screening test identified 1169 positives out of a total of 1889 sputum positive cases (that were confirmed by culture or the gold standard test) and it tested negative to 3843 sputum negative cases for the presence and absence of tuberculosis bacilli respectively. Calculate the sensitivity, specificity, positive and negative predictive values of the screening test. Interpret your results for each measure.

Question 3 Pan, S.Y., K.C. Johnson, A.M. Ugnat, et al., Association of obesity and cancer risk in Canada. Am J Epidemiol 2004; 159:259-68. The authors conducted a population-based, case-control study of 21,022 incident cases of 19 types of cancer and 5,039 controls aged 20-76 years during 1994-1997 to examine the association between obesity and the risks of various cancers. Compared with people with a body mass index of less than 25 kg/m2, obese (body mass index of >30 kg/m2) men and women had an increased risk of overall cancer (multivariable adjusted odds ratio = 1.34, 95% confidence interval (CI): 1.22, 1.48). Overall, excess body mass accounted for 7.7% of all cancers in Canada-9.7% in men and 5.9% in women. This study provides further evidence that obesity increases the risk of overall cancer, non-Hodgkin''s lymphoma, leukemia, multiple myeloma, and cancers of the kidney, colon, rectum, breast (in postmenopausal women), pancreas, ovary, and prostate.

i State the null hypothesis for this study.

ii Explain and interpret the odds-ratios for the risk of overall cancer.

iii Explain and interpret the confidence interval for increased risk of overall cancer. Are the results statistically significant?

iv Has chance been ruled out from this result? Explain briefly.

v Would you accept or reject the null hypothesis? Explain briefly.

Question 4 The following table shows data from a random sample of middle-aged men. A binge drinker is defined as someone who usually drinks six or more bottles of beer per drinking session. This was recorded at the outset of the study; mortality was recorded from death certificates over an average of 7.7 years' follow-up. Binge Drinker Cardiovascular death. Total Yes. No Yes. 7 63. 70 No. 52 1519 1571 Total. 59 1582 1641 i Estimate the relative risk of cardiovascular death for bingers compared to nonbinge drinkers. ii Estimate the odds ratio for cardiovascular death for bingers compared to nonbinge drinkers. iii Estimate the risk of cardiovascular death attributable to beer binging for the group. iv What percentage of deaths could have been prevented if there were no bingers at all?

Question 5 Please read the interesting article available on BB under Assignment and answer the following questions.

1. How would you write the study hypothesis? Use your own words (and thoughts).

2. What is meant by ‘Teaspoon years'? How were these calculated?

3. Explain the study results in your own words; briefly describe findings as well as their significance.

4. Do you think the study was appropriate in reference to ethics? Yes/No and why?

5. Do you think the response rate and data collection could have been improved? Briefly describe how?