Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

BMSc Statistics > Section 7 (Pgs 81-86) > Flashcards

Section 7 (Pgs 81-86) Flashcards

(73 cards)

Start Studying
1
Q

What is the word used to describe a study where the subjects undergo an intervention?

A

Experimental

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is the word used to describe a study where the subjects are monitored?

A

Observational

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What do experimental studies typically do?

A

Compare treatments or assess the outcome of surgery

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What do observational studies typically do?

A

Determine risk factors for disease

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What type of studies are most epidemiological studies?

A

Observational

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the disadvantage of observational studies?

A

Usually less powerful than experimental designs because the conditions have not been controlled in any way by the researcher

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the word used to describe a study where the study is based on a review of existing data e.g. patient records?

A

Retrospective

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the word used to describe a study is based on data which has to be collected at some time in the future?

A

Prospective

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Describe a typical retrospective study? (2)

A

Review of a patients’ lifestyles to identify risk factors for a disease or review of medical records to determine the effect of a previous intervention

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Describe a typical prospective study? (2)

A

Trials for comparing treatments, or assessing the input of an intervention on a patient’s condition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the word used to describe a study where each subject is observed once providing a snapshot of a situation?

A

Cross-sectional

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Give examples of a typical cross-sectional study? (3)

A

Surveys
Investigation of the incidence of a disease in a population
Laboratory experiments

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the word used to describe a study that monitors a situation over time?

A

Longitudinal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the word used to describe a study where the subjects are unaware of which treatment they haven been given?

A

Blind

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the word used to describe a study in which neither the subjects nor the person monitoring their progress has knowledge of the treatment given?

A

Double blind

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Are blind and double blind trials observational or experimental?

A

Experimental

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is the word used to describe a study where the subjects are selected at random?

A

Randomised

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What effect does randomisation have on confounding factors?

A

Randomisation balances out confounding factors, both known and unknown

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Compared to randomisation, what is a more powerful way of dealing with known confounding factors?

A

To use them to stratify the population and select subjects randomly from each stratum

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What is called when known confounding factors are used to stratify the population and select subjects randomly from each stratum?

A

Randomised stratified trial

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is the word used to describe observations when variables are measured more than once on each subject or subjects are matched for factors critical to the outcomes?

A

Dependent

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is the word used to describe observations when variables measured are not related?

A

Independent

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Are designs based on independent or dependent data more powerful?
Why?

A

Dependent data

Some of the greatest variation (between subjects) has been removed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is the word used to describe a study which includes a comparison group of subjects who are not given the treatment?

A

Controlled

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
What is the word used to describe a study in which the control is a placebo?
Placebo-controlled
26
Describe a parallel group design?
one in which the treatment and control groups are observed at the same time
27
What is the name for a parallel group study in which each individual subject who received the treatment is matched for factors critical to the outcomes with a subject in the control group?
Case-controlled
28
What is the name for a parallel group trial which continues until a difference between the treatment and control groups becomes apparent?
Sequential
29
What is the name for a controlled trial in which each subject acts as their own control?
Crossover design
30
What must be ensured when carrying out a crossover study?
Ensure there is no carry-over effect from one treatment to another
31
Why are case-controlled and crossover designs efficient?
Because some of the between-subjects variation has been removed
32
What happens in a prospective cohort trial?
A group of subjects who are initially disease-free are followed over time (factors subjects are exposed to are recorded over time)
33
What are the disadvantages of prospective cohort studies?
They are large and costly and take many years to provide results
34
What kind of disease are prospective cohort studies unsuitable for?
Rare diseases
35
What happens in a retrospective cohort study?
Patients without disease are looked back on to see what risk factors they were exposed to
36
What is a case control study?
A retrospective study of diseased subjects looking at range of factors they have been exposed to to establish risk of disease
37
What is the key difference between a case control and a cohort study?
A cohort study starts with the risks looking at whether disease develops (retro. = data exists, prospective = follow up subjects) where as a case control study start with patients who have the disease and looks back for risks
38
What is an advantage of case control studies?
Cheap
39
What is a disadvantage of a case control study?
Subject to bias because subjects are not chosen randomly
40
What is the name for non-randomised studies?
Quasi-experimental studies
41
What is the main disadvantage of quasi-experimental studies?
Not randomised and therefore susceptible to bias
42
Does a cohort or case control study start with the disease?
Case control
43
Does a cohort or a case control study start with the exposure?
Cohort
44
What is at the bottom for the pyramid of evidence? (3)
Cross-sectional, case reports, expert opinion
45
What is the 2nd level up on the pyramid of evidence?
Observational studies such as case-control and cohort studies
46
What is the middle layer in the pyramid of evidence?
RCT's
47
What is the 2nd top level in the pyramid of evidence?
Systematic review
48
What is the top level of the pyramid of evidence?
Meta-analysis
49
What should you do before starting a study to test its design?
A pilot study
50
What are the 3 types of experimental studies?
RCT's Crossover studies Quasi-experimental studies
51
What are the 4 types of observational studies?
Case-control Cohort (retrospective and prospective) Cross-sectional
52
What is the main advantage of a crossover trial over a parallel group design?
The principal source of variation between subjects is removed entirely as each subject is their own control - therefore if there are real treatment effects, a crossover trial is more able to detect them than a parallel group study
53
Are crossover trials or parallel group trails more powerful?
Crossover trials (therefore fewer subjects are needed for these)
54
Does trial designs in which between subject variation is removed more or less powerful? What does this mean
More powerful | Require fewer subjects
55
What is the principle weakness of the crossover trial?
Danger of one treatment having a carry-over effect
56
How can the risk of having a carry-over effect in crossover trials be removed? (2)
By having a period between successive treatments (a wash out period) which is sufficiently long for the first treatment to be fully out of a subject's system Randomising the order of the allocation of the treatments
57
Give a circumstance when a parallel group trial would be more appropriate than a crossover study?
When participants can only be exposed to one treatment e.g. in a study to compare the impact of 2 training programmes on performance
58
Give the 3 main disadvantages of cohort studies?
They are large, costly and take a long time to produce results, particularly if the disease under consideration is rare
59
Give the main advantage of cohort studies?
the subjects are randomly chosen and therefore free from bias
60
What are the 2 main advantages of case control studies?
Cheap | Provide speedy results
61
What is the main disadvantage of case control studies?
Subjects are not randomly chosen (any trial that is not randomised is vulnerable to bias and therefore the conclusions that can be drawn are limited)
62
What is the main indication for a cohort study over a case control study?
Occurrence being investigated is common
63
What is the main indication for a case control study over a cohort study?
Investigation of the risk factors for a rare disease
64
What are the 6 factors that determine what sample size will be sufficient to detect a real treatment effect?
``` Significance level of the test Power of the test required Size of the effect to be identified Standard deviation of the measurements Study design Practical issues e.g. what is realistically available, for example, financially ```
65
What is the symbol for significance level?
Alpha
66
What is the symbol for power of the test?
Beta
67
What is the significance level usually set at?
5%
68
What is the power of the test almost always set at?
80%
69
What happens to sample size required as the variation in a measure increases?
Sample size also increases
70
What is the easiest way to estimate the standard deviation of the measurements involved in a study? What if this information is not available?
To review the results from previous similar investigations | Need to carry out a small pilot study
71
What feature about study design will affect sample size?
How much between-subject variation is removed e.g. it is removed totally in a crossover and partially in a case-control = fewer subjects required
72
Equation for sample size in a two group parallel trial, with sig. level 5% and power 20%?
n = 16s squared/ d squared
73
What happens to sample size needed as the size of the effect to be identified decreases?
Decreases

BMSc Statistics (9 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions