(2) Clinical treatments for heart disease

Question 1

What is an intervention study?

Answer 1

• It is an experiment in which an intervention is administered in order to evaluate its efficacy and safety
• Intervention – allocation of participants to treatment or control group (untreated, standard therapy,placebo)
• We observe/measure how participants respond to the intervention
• Also called experimental study, clinical trial, RCT

Question 2

What is random allocation?

Answer 2

• Random allocation (randomization) means that individuals were allocated randomly to each treatment/control group
• Assignment of patients to a group is independent of the allocation of other patients
• Avoids allocation bias
• Difference in health outcomes between intervention & control groups is attributed to intervention.
• Randomization is not same meaning as random sample of people!

Question 3

What is the placebo effect?

Answer 3

• A cure/alleviation of symptoms following an intervention which the patient believes to be effective against the ailment, but which, in fact, is completely neutral (e.g. a sugar pill)
• Such effects seem to based on the person’s own self-healing capacity, which is triggered by the belief that they are receiving an active medication

Question 4

What is a double blind trial?

Answer 4

Where both the researcher and the participants know whic group they’ve been allocated

Question 5

What is a single blind trial?

Answer 5

Where either the participant or the researcher doesn’t know which group the patients are in. But not both

Question 6

What is an intervential study with parallel groups?

Answer 6

When there a random assortment of patients into either the treatment or control group. After some time has passed measure the outcome of the two groups.

Question 7

What is a cross over study?

Answer 7

Where patients are randomly assined into a treatment/control group and after some time they swap groups and the outcome is measured after some time has passed

Question 8

What is the null hypothesis?

Answer 8

There is no asscoiation between treatment and outcome

Question 9

What is the alternate hypothesis?

Answer 9

There is an association between treatment and outcome

Question 10

What is a contingency table?

Answer 10

A table showing the distribution of one variable in rows and another in columns, used to study the correlation between the two variables.

Question 11

What is the Chi squared test?

Answer 11

• Test compares observed and expected (under Ho) counts between groups •Two categorical variables
• Each variable can have 2 or more levels
• Data tabulated in a contingency table as counts of individuals by level of each variable
• Validity: 80% expected frequencies > 5 and all > 1
• Test gives a P-value

Question 12

How to interprate P values?

Answer 12

The P value tells you the probability the results came about by chance. In order to reject H0 over H1 the P value must be less than 0.05.

Question 13

What is the number needed to treat?

Answer 13

How many patients do you need to treat to prevent one additional outcome?
If 100 people were treated using propranolol rather than placebo, 4 less people would die within follow-up period => NNT = 25

Question 14

What is the link between the confidence interval and the P value?

Answer 14

If the 95% confedence interval contains the null value then p>005 and there is no association. If the CI excludes the null value then p<0.05 and there is an association

Question 15

How do you work out the CI?

Answer 15

Mean +/- 2x standard devation

Question 16

What is intention to treat?

Answer 16

It includes all original participants regardless of whether the dropped out or not

Question 17

What is on treatment analysis?

Answer 17

Comparing the people who only stayed on the course