Science of Practice

Question 1

Describe a case-control study and what is more powerful than it.

Answer 1

Retrospective. Not conducted over time.
Researchers choose individuals with a particular characteristic (the cases) and compare features of interest with a control group of individuals who do not have the characteristic. This type of study is required for questions around causation.
Cohort, RCTs ,Crossover

Question 2

Describe a cohort study and what is more powerful than it

Answer 2

follow individuals with a particular condition or who receive a particular treatment and they are compared with another group of people who are not affected by the condition or who did not receive the treatment.
Crossover, RCTs

Question 3

Describe a cross-sectional study and what is more powerful than it

Answer 3

defined population is observed at a single point in time or at certain time intervals
case-control, Cohort, crossover RCTs

Question 4

Define Selection bias

Answer 4

is linked to the recruitment of patients; a cohort of individuals deciding not to participate in the study could lead to under- or overestimation of a particular effect or risk. Patients lost to follow-up, self-selection into a study and lack of randomisation are further examples.

Question 5

Define information bias

Answer 5

Occurs when patients are misclassified, for example, when individuals with a particular disease or exposure are classified as nondiseased or nonexposed due to an inaccuracy of diagnostic tests. Such a problem will invalidate outcomes.

Question 6

Define confounding variable

Answer 6

are variables that influence both the dependent variable and independent variable, causing a spurious association. Confounding variables are those that may compete with the exposure of interest (e.g. treatment)
in explaining the outcome of a study. For example, in a study about whether a lack of exercise (independent variable) leads to weight gain (dependent variable) in children, the number of calories consumed would be a confounding variable. Identifying possible confounders early on and controlling for them will assist in eliminating bias.

Question 7

Define number needed to treat

Answer 7

number of patients that need to be treated in order for one to benefit. For example, an NNT of 20 would be interpreted as ‘20 patients need to be treated to avoid one additional death’.

Question 8

Give an example of when a Chi squared test would be used

Answer 8

Chi-squared tests are commonly performed on 2 x 2 data, for example, to investigate if there is an association between maternal smoking (yes or no) and child’s asthma status (asthmatic or not asthmatic) in a given population

Question 9

What is the null hypothesis?

Answer 9

That we accept that there is no significant difference between results.
IF P value is <0.05, then we say it is significant and thus we REJECT the null hypothesis.

Question 10

What does a 95% confidence interval (95% CI) mean?

Answer 10

there is a 5% chance that the ‘true’ population mean lies outside the quoted range.

Question 11

How do you determine if there is significant result in confidence intervals following a given intervention?

Answer 11

95% CI do not overlap
e.g daily steps undertaken by the teenagers increased from 12.3k (95% confidence intervals (CI) 7.5 to 14.1) to 17.7k (95% CI 15.2 to 20.3) after the exercise programme

Question 12

When interpreting confidence intervals between means, the confidence interval crossing 0 means what?

Answer 12

the studied effect is not significant as zero indicates no difference between means

Question 13

When interpreting confidence intervals between means, the confidence interval being entierly negative or positive numbers means what?

Answer 13

the studied effect is significant as the 95% confidence interval does not cross zero

Question 14

When Interpreting a 95% CI for ratios (OR, RR etc), what does it mean if the CI crosses 1?

Answer 14

There is no difference between arms of the study and the outcome of the study can be interpreted as not statistically significant

Question 15

When Interpreting a 95% CI for ratios (OR, RR etc), what does it mean if the CI is <1 but does not cross it?

Answer 15

the intervention is likely to be less effective and, as the ratio does not cross 1, the findings are statistically significant

Question 16

When Interpreting a 95% CI for ratios (OR, RR etc), what does it mean if the CI is >1 and does not cross it?

Answer 16

the intervention is likely to be more effective and, as the ratio does not cross 1, the findings are statistically significan

Question 17

What does an odds ratio do?

Answer 17

Determine whether a particular exposure is a risk factor for a particular outcome and to compare the magnitude of various risk factors for that outcome.

Question 18

What does OR = 1 mean?

Answer 18

Observed outcome is the same in the two groups

Question 19

What does OR = <1 mean?

Answer 19

Intervention provides a poorer outcome

Question 20

What does OR = >1 mean?

Answer 20

Intervention provides a better outcome

Question 21

Describe relative risk

Answer 21

the probability that an event will occur, whether it be a desirable or undesirable event. It explains how many times more likely that an event will occur in the treatment/intervention/exposure group, relative to the control group.

Question 22

What does RR = 1 mean?

Answer 22

No difference in given event occuring between two groups

Question 23

What does RR <1 mean?

Answer 23

Given event less likely to occur with exposure

Question 24

What does RR > 1 mean?

Answer 24

Given event more likely to occur with exposure

Question 25

Define risk difference (RD)

Answer 25

The difference in risk of a condition, such as a disease, between an exposed group and an unexposed group.
If RR1 is the disease risk in an exposed population and RR2 is the disease risk in a nonexposed population then the risk difference (RD) equals RR1–RR2. RD can therefore be a number between -1 and +1.

Question 26

What does RD = >0 mean?

Answer 26

Increased risk in exposed group

Question 27

What does RD = 0 mean?

Answer 27

No difference in risk between groups

Question 28

What does RD =<0 mean?

Answer 28

Decreased risk in exposed group

Question 29

Define Specificity

Answer 29

Probability test is negative when disease is absent

Question 30

Define Sensitivity

Answer 30

Probability test is positiive when disease is present

Question 31

Define positive predicitive value

Answer 31

Probability Test is postitive when disease is absent

Question 32

Define negative predicitive value

Answer 32

Probability test is negative when disease is absent

Question 33

Describe prognostic biomarker

Answer 33

used to stratify patients according to the prognosis of their disease subtype.
Example! In childhood acute lymphoblastic leukaemia, cytogenetic analysis identifies patients at a higher risk of treatment failure, e.g. t(9;22) Philadelphia chromosome.

Question 34

Describe Predicitve biomarker

Answer 34

predict a patient’s response to a particular treatment. Mutations of Kir6.2 causing infant-onset diabetes insipidus predicts sensitivity to sulphonylureas

Question 35

Describe response biomarker

Answer 35

provide a surrogate measure of a patient’s disease status and response to the chosen therapy – fever or C-reactive protein in infection

Question 36

Describe pharmokinetic biomarker

Answer 36

used to assess the therapeutic or toxic effects of a drug. Antibiotics such as gentamicin or vancomycin will commonly have drug levels monitored.

Question 37

Describe imaging biomarker

Answer 37

non-invasive imaging, e.g. CT or MRI, may provide prognostic or response biomarker information.

Question 38

Define P value

Answer 38

The probability that agiven outcome will occur by chance, i.e. that the null hypothesis is true.

Question 39

What is meant by intention to treat?

Answer 39

Patients are analysed according the group they were origninally randomised into

Question 40

How do you calculate NNT?

Answer 40

1/ (option a)- (option b)

Question 41

How do you calculate sensitivity?

Answer 41

Proportion that have the condition, that test positive
The number of true positives/ (number of true positives + number of false negatives)

Question 42

How do you calculate specificity?

Answer 42

Proportion that do not have condition, that test negative
The number of true negatives / (number of true negatives+ false positives)

Question 43

How do you calculate Positive predictive value?

Answer 43

Proportion with positive test that have the condition
True positive/ ( True positive +false positive )

Question 44

How do you calculate Negative predictve value?

Answer 44

Proportion with negative test who don’t have condition
True negative/ (True negative + false negative)