5 Written questions
5 Matching questions
- External Validity
- Attributable Risk
- Baye's Theorem of Conditional Probabilities
- Regression to the Mean
- a The more sensitive a test is, the better will be its NPV. The more specific a test it, the better will be its PPV.
- b the proportion of persons in a defined population at a given point in time having the condition in question
- c "degree to which the results of an observation hold true in other settings" "generalizability" "Assuming that the results of a study are true, do they apply to my patients as well?" p. 10
- d The absolute difference in the rate of events between the control and the intervention group. ( same as ARR)
- e "Patients selected because they represent an extreme value in a distribution can be expected, on average, to have less extreme values on subsequent measurements..This occurs for purely statistical reasons, not because the patients have necessarily improved." This is the reasoning behind repeating lab tests: "Subsequent values are likely to be more accurate estimates of the true value." p. 32
5 Multiple choice questions
- (Dr. Wheat) author? journal?....i think...Kerr White New England Journal of Medicine, Nov. 2, 1961 "Ecology of Medical Care"
- Proportion of patients who die of a disease
- How much of the risk is reduced in the experimental group compared with the control group. RRR=ARR/(rate of control group)
- 11,700. Of these 11,700 known to have HIV, ~4,500 have AIDS. (This number with AIDS will probably not be asked since the term AIDS is somewhat subjective. The viral load limit that a person must have to be considered suffering from AIDS has been lowered in recent years.)
5 True/False questions
How are laboratory abnormalities defined? → As a fraction of the actual distribution (Page 29)
Cohort Study → "incidence study" Sample of people either free of outcome of interest and then followed over time as it emerges, or all diseased then followed for outcomes of that disease.. p. 64, All participants in a cohort study could experience the outcome of interest. 81 See Table 5.2 p. 84 "Advantages and Disadvantages of Cohort Studies"
Internal Validity → "degree to which the results of a study are correct for the sample of patients being studied...determined by how well the design, data collection, and analyses are carried out and..threatened by..biases and random variation" p. 10
Approximately how many people live in the US? → ...300,000,000
Relative Risk → A general term for measures of association calculated from data in a 2x2 table. = Risk of disease in exposed (a/(a+b)) / Risk of disease in unexposed (c/(c+d))