Statistics/ Trials

Question 1

What is selection bias in clinical research?

Answer 1

An error in assigning individuals to groups, leading to systematic differences that influence the outcome. It means the study sample is not representative of the target population

Question 2

Name two common subtypes of selection bias.

Answer 2

Volunteer Bias: When individuals who choose to participate are systematically different from those who do not.
Non-responder Bias: When individuals who do not respond to a survey or study differ systematically from respondents.

Question 3

What is Berkson’s bias (admission bias)?

Answer 3

A type of selection bias in hospital-based case-control studies, where the combination of having a disease and being exposed to a risk factor increases the chance of hospital admission, skewing the sample.

Question 4

What is recall bias, and in which study type is it a particular problem?

Answer 4

A difference in the accuracy of recollections between study groups, often because participants with a disease search their memories more thoroughly. It is a major problem in case-control studies.

Question 5

What is publication bias, and why is it a critical concern for meta-analyses?

Answer 5

The failure to publish valid study results, often because they are negative or uninteresting. It distorts the evidence base, as meta-analyses may then only include positive studies, overestimating an effect.

Question 6

What is work-up bias (verification bias)?

Answer 6

A bias in diagnostic test studies where the gold-standard test (e.g., an invasive biopsy) is only performed if the new test is positive. This can falsely inflate estimates of the new test’s accuracy (sensitivity/specificity).

Question 7

What is the Hawthorne effect?

Answer 7

A change in a group’s behavior simply because they know they are being studied, which can alter the study outcome.

Question 8

What is lead-time bias?

Answer 8

A bias that occurs when a new screening test diagnoses a disease earlier than an old test, making survival appear longer, even if the time of death is unchanged and no life is actually extended.

Question 9

what is a clinical audit?

Answer 9

Clinical audit seeks to improve patient care and outcomes through systematic review of care against explicit criteria and the implementation of change

Question 10

how to work to variance

Answer 10

Variance = square of standard deviation (SD^2)

Variance is a measure of the spread of scores away from the mean.

Question 11

how to work out standard deviation

Answer 11

Find the mean.
Subtract the mean from each value.
Square each result.
Find the mean of the squared values.
Take the square root.
Formula (population):
SD = √[ Σ(x − x̄)² ÷ n ]
Formula (sample):
SD = √[ Σ(x − x̄)² ÷ (n − 1) ]

Interpretation:
Low SD → values close to mean
High SD → values spread out

​

​

Question 12

how is specificity calculated

TN= true negative
FP= false positive

Answer 12

The specificity is calculated as TN / (TN + FP)

Question 13

what is an assay

Answer 13

A laboratory method used to detect or quantify a substance or measure its biological or chemical activity.

Question 14

A scientist is developing an assay whereby he heats up a DNA specimen at 95ºC then cools it to 55ºC whilst adding primers to specific sequences. Following each primer, heat-stable DNA polymerase is added and the cycle is repeated.

Which molecular technique has been described?

Answer 14

Polymerase Chain Reaction (PCR) uses denaturation, annealing and elongation to amplify a desired fragment of DNA

Question 15

Expectation bias (Pygmalion effect)

Answer 15

Only a problem in non-blinded trials. Observers may subconsciously measure or report data in a way that favours the expected study outcome.

Question 16

Late-look bias

Answer 16

Gathering information at an inappropriate time e.g. studying a fatal disease many years later when some of the patients may have died already

Question 17

Procedure bias

Answer 17

Occurs when subjects in different groups receive different treatment

Question 18

What is the purpose of a Phase 0 clinical trial?

Answer 18

To assess how a drug behaves in the human body (pharmacokinetics and pharmacodynamics) and determine whether further trials are feasible.

Question 19

How many participants are typically involved in a Phase 0 trial?

Answer 19

A very small number of participants.

Question 20

What is the main aim of a Phase I clinical trial?

Answer 20

To assess safety and identify side-effects before larger studies.

Question 21

Who typically participates in Phase I clinical trials?

Answer 21

Healthy volunteers.

Question 22

What is the main aim of a Phase II clinical trial?

Answer 22

To assess efficacy in patients with the target disease.

Question 23

Who participates in Phase II clinical trials?

Answer 23

A small number of patients affected by the disease being studied.

Question 24

What does a Phase IIa clinical trial assess?

Answer 24

Optimal dosing.

Question 25

What does a Phase IIb clinical trial assess?

Answer 25

Efficacy.

Question 26

What is the main aim of a Phase III clinical trial?

Answer 26

To assess effectiveness of a treatment compared to established treatments.

Question 27

How many participants are typically included in Phase III trials?

Answer 27

Hundreds to thousands of participants.

Question 28

What study design is commonly used in Phase III trials?

Answer 28

Randomised controlled trials.

Question 29

What is the purpose of a Phase IV clinical trial?

Answer 29

Postmarketing surveillance to monitor long-term effectiveness and side-effects.