Collecting Evidence

Question 1

What are the 4 things you have to consider when constructing a clinical question?

Answer 1

• patient
• intervention
• comparison
• outcome

Question 2

What are the 3 points that make up a good clinical question?

Answer 2

• define precisely whom the question is about (how would I describe a similar group of patients)
• define which option you are considering (eg. drug treatment) and possible comparison (eg. placebo/standard therapy)
• define the desired (or undesired) outcome (eg. reduced mortality, better QOL)

Question 3

What are the steps you would go through to find information to answer your clinical question?

Answer 3

1. first use:
   - Cochrane Reviews
   - NICE and SIGN guidelines
2. then use:
   - MedLine (Ovid, PubMed)

Question 4

Define hypothesis

Answer 4

Explanation for a scientific problem that can be tested by further investigation

Question 5

What is a null and alternative hypothesis?

Answer 5

• null: 2 sets of data from same population and are not different
• alternative: 2 sets of data from different populations and are different

Question 6

What are the different types of quantitative data?

Answer 6

• discrete: can only have certain numerical values (number of children)
• continuous: do not have discrete steps (weight and height)

Question 7

What are the different categorical variables?

Answer 7

• nominal (unordered categories) eg. male/female
• ordinal (ordered categories):
• objective eg. heavy/moderate/light drinkers
• subjective eg. health questionnaire where people will have different perceptions of their health

Question 8

How would you test your hypothesis?

Answer 8

• assume null hypothesis
• determine probability that null hypothesis is correct (P-value)
• P value close to 0 is in favour of alternative hypothesis
  (a P-value of 0.1 means 10% probability)

Question 9

What does a P of <0.05 indicate?

Answer 9

• cut off indicating that the null hypothesis can be reasonably rejected (1 in 20 chance of it happening)
• statistically significant difference

Question 10

Describe the significance of the P value

Answer 10

P > 0.1 = not significant (data consistent with null hypothesis)
P > 0.05 = not significant (in favour of alternative)
P < 0.05 = significant ( in favour of alternative)

Question 11

What are type I and type II errors and how do they arise?

Answer 11

• they arise from our interpretation of P values
• type I: rejecting the null hypothesis when it is true (false positive) - concluding there is an effect when there isn’t (P is small)
• type II: not rejecting null hypothesis when it is false (false negative) - concluding there is no effect when there is (P is large)

Question 12

What is the power of a test?

Answer 12

• its ability to reject the null hypothesis when it is false

- capacity to detect an effect if one is present

Question 13

How can you increase the power of a test?

Answer 13

• making sure the sample size is large enough

- making sure the variation between individuals is small enough