data analysis

Question 1

definition of confidence interval

Answer 1

95% confident that true value lies between these 2 particular values

Question 2

definition and interpretation of risk ratio

Answer 2

Ratio of probability of an event occurring in an exposed group to the probability of the event occurring in a comparison, non-exposed group
i.e. 7.7 risk ratio = 7.7x more likely for event to occur w exposure vs w/o

Question 3

sensitivity and specificity definitions

Answer 3

Specificity: proportion of true negative results that are correctly identified as negative by the test
100% specificity means that all those with a positive test have the disease

Sensitivity: proportion of true positive results that are correctly identified as positive by the test
100% sensitivity means that all those with a negative test do not have the disease

Question 4

positive and negative predictive values; definition

Answer 4

Positive predictive value: proportion of truly positive patients in relation to the total number of test positive patients

Negative predictive value: proportion of truly negative patients in relation to the total number of test negative patients

Question 5

how are likelihood ratios used

Answer 5

Pre-test probability X likelihood ratio gives post-test probability
Use pre and post-test probability to calculate likelihood ratio
Clinically useful tests have a LR of > 10 or <0.1
1. LR<0.1=‘strongly negative result’ i.e. good at ruling out diagnosis
2. LR>10 = ‘strongly positive result i.e. good at diagnosing
3. LR 1 = not informative