Statistics Flashcards
(2011)
Explain and discuss the statistical concepts of linear regression and correlation.
.
(2012)
Give an examples of the types of data that can be analyzed using the Student’s “t-test”? Under what circumstances should the “paired” and “unpaired” t-test be used?
.
(2010)
What is “95% confidence interval”? Explain the use of this and how it compares with p values.
.
(2009)
Mean arterial blood pressure has been measured in 2 groups of patients 1 hr after administration or placebo or antihypertensive drug. Explain how these data could be analysed.
.
(2009)
Define confidence interval (95%). Explain the advantages of using confidence intervals over standard deviation in various statistical analyses.
.
(2007)
Define “numbers needed to treat (NNT)” and outline its relation to risk reduction. Outline advantages and limitations of using NNT to describe effectiveness of a new analgesic.
.
(2006)
What is a meta-analysis? Briefly explain the advantages and problems of meta-analysis.
.
(2001/05)
Describe how you would devise a clinical trial to determine the efficacy of a new drug dolasetron as an anti-emetic in the post-operative period.
.
(2004)
The following table shows the results of a study comparing the incidence of heparin induced thrombocytopenia (HIT) in unfractionated heparin and low molecular weight heparin (LMWH) use.
The following table shows the results of a study comparing the incidence of heparin induced thrombocytopenia (HIT) in unfractionated heparin and low molecular weight heparin (LMWH) use.
HIT No HIT Heparin 8 92 100 LMWH 2 98 100
i) What statistical analysis is appropriate? Give your reasons.
ii) Explain the terms:
a) Relative risk reduction
b) Absolute risk reduction
c) Numbers needed to treat (NNT)
.
(2003)
Define the following terms: Sensitivity, Specificity, Positive predictive value, Negative predictive value.
How can the above terms be used to assess the performance of a new diagnostic test?
.
(2000)
Define the following terms and explain the effect of a left skew on the terms defined: Normal distribution, Mean, Median, Mode, Standard deviation, Standard error.
.
