RESS Flashcards

1
Q

are audits and service evaluations called research in the NHS

A

No

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

Difference between an audit and service eval?

A

Audit- are standards being met, using existing standards and existing management data
SE- what factors are responsible- information gathering.

Both look at existing practice (research looks at new practice)

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

What does PECOS stand for?

A
Patient/participant/people
Exposure/event/experimental intervention
Comparison
Outcome
Study design
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is MeSH?

A

Medical subject headings- all papers should be indexed with their relevant MeSH terms to be looked up.

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

3 competing priorities in studies?

A
  • correct answer
  • managing resources
  • complying with regulations
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is selection bias

A
  • non-representative sample

- selection influences exposure and outcome (confounding)

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

What are 3 types of measurement bias

A

Information
Observer
Recall

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

What is information bias

A

Mis-classified or mis-collected data

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

What is observer bias

A

the observers (or researcher team) know the goals of the study or the hypotheses and allow this knowledge to influence their observations during the study.

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

What is recall/prestige bias

A

Response is influenced by prior knowledge or belief

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

3 types of analytical bias

A

Loss to follow up
Omitted variable
Attributional

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

What is loss to follow up

A

Specific participants excluded

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

What is omitted variable bias

A

Imprecise adjustment for confounding variables

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

What is attributional bias

A

Misperception/incorrect interpretation of why something caused something else

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

What is publication bias?

A

exciting research is published more than boring research/research with no interesting outcome

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

Cross sectional studies provide evidence of association ____ a sample

A

Within

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

Case-control studies provide evidence of association _______ samples

A

Between

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

Cohort studies provide evidence of ____________

A

Directionality of associations

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

Trials provide evidence of _____

A

Causality

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

Meta-analyses provide evidence of _________

A

Reproducibility/generalisability

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

What is STROBE

A

Strengthening the reporting of observational studies in epidemiology

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

Observation –> pattern –> tentative hypothesis –> theory is what sort of reasoning?

A

Inductive/descriptive

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

Theory –> hypothesis –> testing –> observation –>reject/don’t reject is what sort of reasoning?

A

Deductive/analytical

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

Case studies involve what sort of reasoning?

A

Inductive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Analytical reasoning is observational or experimental. What is the difference?
Observational- selective sampling e.g. cross sectional or cohort study Experimental- selective exposure e.g. a trial
26
What 3 things did the Belmont report decide?
Ethical concepts: 1. Respect for persons- informed consent 2. Justice- equitable distribution of burdens and benefits 3. Beneficence- risks to human subjects must be justified by the value of the knowledge the study is expected to generate
27
What 6 things are required for participation to be voluntary?
1. Informed participants 2. Consented participants 3. Reward-free 4. Free to decline 5. Free to withdraw 6. Rights protection
28
What are the 4 vulnerable groups?
1. limited agency (children) 2. diminished autonomy (dementia) 3. with needs (unwell) 4. unable to consent (unconscious)
29
What is meant by ethics
Ensuring projects maximise benefits and minimise risks
30
what is meant by governance
Appropriate permissions
31
Does this require ethical approval: | data/material that is already publicly available
No
32
Does this require ethical approval: | projects collecting new info only on existing service delivery
No
33
Does this require ethical approval: | Non human subjects covered by the animals act 1986
yes
34
Does this require ethical approval: | Projects on non-human subjects/materials
No
35
Does this require ethical approval: | Projects involving a new intervention
Yes
36
Does this require ethical approval: | Service evaluations involving vulnerable participants or ethically sensitive data
Yes
37
Does this require ethical approval: | Projects using only existing service management data
No
38
Does this require ethical approval: | Projects collecting new information on more than only existing service delivery
Yes
39
What 3 things must a project protocol include?
Why the project is necessary What it will involve How it will avoid ethical, legal, governance issues
40
When must ethical approval be re-applied for after already having been granted?
Substantial changes | Unexpected deleterious effects
41
3 types of sampling
Complete Unstratified random Stratified random
42
What is a p value
Probability that the results are by chance if the null hypothesis is true (i.e. your exposure has no effect)
43
What p value is normally accepted for something to be significantly different from the null hypothesis?
P<0.05
44
What is an odds ratio
the odds that an outcome will occur given a particular exposure, compared to the odds of the outcome occurring in the absence of that exposure.
45
a 95% confidence interval will......?
Contain the true value that we are trying to estimate 95% of the time
46
A confidence interval of what value means that results are inconclusive?
If the interval crosses odds ratio of 1.
47
increasing/decreasing sample size will increase/decrease confidence interval
Increase, decrease
48
What is power?
The probability of rejecting the null hypothesis when the null hypothesis is false (I.E. finding an association, if it is indeed there to be found)
49
What makes a power greater?
Mean effect is bigger Little variation Large sample size
50
When do you calculate power?
Before conducting a study to test design and to be able to justify. After conducting an inconclusive study- to demonstrate that you were sufficiently well-powered to detect an association
51
What value of power do we normally aim for?
80%
52
What is odds???
The probability that something occurs divided by the probability that it doesn't occur (exposed/unexposed)
53
What is the odds ratio if no effect?
1 i.e. the null effect. Don't want this to be in the confidence intervals!
54
Do you need to adjust for confounders?
Yes because they suggest a false statistical relationship
55
Do you need to adjust for mediators?
No because they are part of the causative pathway
56
Do you need to adjust for competing exposures?
You can if you want, might make the association easier to detect
57
Give an example of a functional causal relationship
no contraception results in teenage mother
58
e.g. of empirical causal relationship
Based on previous statistical analysis
59
e.g. of theoretical causal relationship
teenage grandmother = teenage mother
60
e.g. of speculative causal relationship
teenage grandfather = teenage mother
61
What is a latent variable?
Missing data
62
How can you cope with a latent variable?
If it is the exposure or outcome- change the study | If it is a confounding variable could use a proxy e.g. how many takeaways in place of saturated fat.
63
How can you cope with missingness?
Compare to other participants Create a 'missing' category Interpolate (i.e. statistically estimate) the measurement
64
When do you do a logistic regression?
When the outcome is binary
65
What does an R^2 of 1 mean?
Perfect correlation
66
Is a higher or lower R^2 better?
Higher
67
R^2 is between what numbers
0-1
68
When is a pseudo R^2 used?
Logistic regression
69
When is R^2 used?
Linear regression
70
What is COREQ
Consolidated criteria for reporting qualitative research- a 32 item checklist for interviews and focus groups. Useful tool when carrying out qual research.
71
In qualitative research what is induction?
Generating a universal statement from individual cases
72
In qualitative research what is deduction?
Using theories to explain individual cases
73
In qualitative research what is purposive sampling?
Choosing participants who have the potential to provide rich, relevant and diverse data relevant to the research question
74
In qualitative research what is maximum variability sampling?
Wide range of experiences, no preconceptions. Exploratory.