In: Biology
Quiz 9 STUDY QUESTIONS
1 For each of the following statements, indicate whether the statistical association is likely the result of chance, confounding. bias. or A case-control study showed that a strong association exists between birth order and Down syndrome.
b. A case-control study lound a positive association between self-reported chest radiographs during pregnancy and breast cancer. A randomized clinical trial found that drug A versus placebo did not significantly improve 10-year survival (RR = 0.35; 95% confidence interval, 01455014 A cohort study found no statistical association between smoking and pancreatic cancer (RR = 1: = P value = 0.85). A hospital-based case-control study identified a strong association between oral contraceptives and thromboembolism. Many doctors suspected the association and hospitalized some women who used oral contraceptives for evaluation
2 Match the following methods for minimizing chance, bias, and confounding in an experimental study Chance Bias Confounding a Randomization b. Blind c. Increase sample size
3. Recall the causal criteria presented by Sir Austin Brad- ford Hill in 1965 Discuss these criteria in the context of smoking and lung cancer.
4.Suppose you suspect based on descriptive epidemiol ogy that college students who perform better aca demically are more likely to have an office job and be obese 10 years after graduation. You decide to select 500 graduating seniors randomly and classify them according to grade point average as high versus low (where the cut point is at the median of the GPAs for these students). The resulting 2 x 2 contingency table is as follows: Below is a table GPA High Low Total Obese at 10 Years Yes 60 40 100 No 190 210 400 Total 250 250 500 Apply this data to the six steps of hypothesis.
5. Match the following Predisposing factors Enabling factors Precipitating factors Reinforcing factors a Facilitate manifestation of a disease le housing) Associated with definitive onset of disease (eg. toxin Increase level of susceptibility m a host of age) d. Aggravate presence of disease leg repeated exposure)
6. Compare a direct causal association with an individual causal association. Use specific examples
7. Define and compare the difference between static inference and causal inference. 8. Why might studying a sample be preferred population? A component cause is also called which of the following?
a. Risk factor b. Web of causation c. Epidemiologic triangle
d. Each of the above are component causes 10 Webs of causation play a more useful role when one is trying to describe disease etiology for which type of disease? a. Acute b. Infectious C. Chronic d. Two of the above.
1) Confounding bias occurs when a third variable erroneously disturbs the relationship between the association and causation. Statistical association can also occur by chance when there is no biological plausibility.
a) case-control study showed that a strong association exists between birth order and Down syndrome - confounding bias (age).
b) A case-control study found a positive association between self-reported chest radiographs during pregnancy and breast cancer - statistical chance
c) A clinical trial found that drug A versus placebo did not significantly improve 10-year survival (RR = 0.35; 95% confidence interval, 01455014 - statistical chance
d)A cohort study found no statistical association between smoking and pancreatic cancer (RR = 1: = P value = 0.85) - confounding bias (age, alcohol)
e)A hospital-based case-control study identified a strong association between oral contraceptives and thromboembolism. Many doctors suspected the association and hospitalized some women who used oral contraceptives for evaluation - confounding bias (age, comorbidities)