Quantitative Study Design Flashcards

1
Q

Sampling/selection bias

A

sample does not represent population of interest (may colllect more extreme views)

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

Recall bias

A

inaccurate recall of past events/exposures/behaviours

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

Information bias

A

incorrect measurement eg miscalibrated machine

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

The Hawthorne effect

A

participants change their behaviour when they know they are being observed

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

Attrition bias

A

differential dropout from studies eg sicker participants drop out so our outcome is only measured on healthier participants

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

Experimental study design

A

researchers have intervened in some way (prospective)

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

Observational study design

A

the researchers have not intervened, merely observed. Can be:
• retrospective- looking back into the past
• Cross-sectional- a single snapshot of time
• Prospective- following up over time

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

Types of data

A

• categorical variables: binary, ordinal, nominal
• numeric variables: discrete, continuous

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

Binary

A

Only 2 categories

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

Ordinal

A

categories with natural order eg stage of cancer

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

Nominal

A

categories with no natural order eg blood group, ethnicity

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

Discrete

A

observations can only take certain numerical values eg number of children

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

Continuous

A

observations can take any value within a range eg age, body temperature
Restriction is precision of measurement tool

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

Proportion

A
  • the number with a characteristic or outcome divided by the total number. Used to describe the probability or risk (scale 0 to 1)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Risk

A

probability of event occurring - probably occurring divided by total

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

Odds

A

the number with an exposure or outcome divided by the number without. The ratio of the probability of an event occurring to the probability of it not occurring- occurring divided by not occurring

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

Rate

A

incidence of health-related events or outcomes. Allows account for variation if follow-up time or time at risk of an outcome

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

Risk difference

A

absolute risk difference (subtraction)- no difference = 0

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

Risk ratio

A

relative risk- risk in one group divided by the risk in the other

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

Ratios

A

No difference = 1
Ratios > 1 indicate higher risk/odds in group of interest
Ratios < 1 indicate lower risk/odds in group of interest

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

Numbers needed to treat (NNT)

A

1 divided by absolute risk difference
• always round up- has to be a whole number

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

Prevalence

A

number of existing cases in a population at a defined time point

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

Incidence

A

number of new cases in a population over a defined time period

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

PICO framework

A

• Intervention studies: PICO
• Observational studies: PECO → ‘exposure’ rather than ‘intervention’
• Non-comparative studies (e.g. in qualitative research): PEO
• Example research question framed using PICO:
Population
In middle-aged women (>40 years old) with raised cholesterol (>5 mmol/L)
Intervention
…does new statin x
Comparator
…compared to current statin y
Outcome
…provide greater reduction in cholesterol?
(Ideally with clinically meaningful reduction in cholesterol defined)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

9 Bradford Hill criteria

A
  1. Strength of association – the stronger an association, the more likely it is to be causal
  2. Consistency – association shown across different studies in different locations, populations, using different methods, etc.
  3. Specificity – specific exposure-outcome relationship, e.g. asbestos and asbestosis
  4. Temporality – exposure must precede outcome
  5. Biological gradient – dose-response, i.e. increase in exposure = increase in outcome
  6. Plausibility – biological mechanism that would explain outcome development
  7. Coherence – compatible with existing theories
  8. Experiment – outcome altered with experimentation, e.g. reversible
  9. Analogy – similar cause-effect relationships established
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

When designing studies must consider

A

• Ethics: are the rights, safety and wellbeing of participants protected?
• Feasibility: will it give the best quality evidence given our resources (time, money, personnel, etc.)?
• Efficiency: is there a faster/cheaper way to address the question without compromising quality?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

Which type of study can only use odds ratio

A

Case-control study

28
Q

Absolute risk

A

Probability of event within time period

29
Q

Relative risk

A

Probability of event relative to exposure

30
Q

Proportion

A

Outcome/ total number
Scale : 0-1

31
Q

Odds

A

Outcome with exposure/ outcome without exposure
Probability / (1-probability)

32
Q

Odds ratio

A

Odds that case was exposed/ odds that control was exposed

33
Q

Ratio >1

A

Higher risk

34
Q

Ratio <1

A

Lower risk

35
Q

Risk ratio/ relative risk

A

Disease incidence in exposed/ disease incidence in non-exposed

36
Q

NNT (numbers needed to treat)

A

1/ |relative difference|
Number needed to take drug for one success

37
Q

NNH (number needed to harm)

A

1/|relative difference|
Number needed to take drug for one adverse effect

38
Q

Parametric

A

Make distributional assumptions

39
Q

95% reference range

A

Mean +/- 1.96*SD

40
Q

What percentage of values lie within 1SD of the mean

A

68%

41
Q

What percentage of values lie within 2SD of the mean

A

95%

42
Q

What percentage of values lie within 3SD of the mean

A

99.7%

43
Q

What is the y axis height of normal distribution determined by

A

SD variation

44
Q

What is the x axis location of a normal distribution determined by

A

Mean

45
Q

Normal distribution

A

Bell-shaped curve
Mean = median = mode
Report mean +/- 2 SD

46
Q

Negative skewed distribution

A

Skewed negative (left): median> mean

left skewed distribution: same mean but median is higher

47
Q

Positively skewed distribution

A

right skewed distribution: same mean but median is lower
Mean > median

48
Q

Skewed distribution

A

Report median and interquartile range (centiles)

49
Q

Meta-analysis

A

Combining systematic reviews to address a question

50
Q

Ecological fallacy

A

Can’t make individual-level inference

51
Q

What does PICO stand for

A

Patient
Intervention
Control
Outcome

52
Q

Systematic sampling

A

Selected at equal intervals

53
Q

Advantages and disadvantages of systematic sampling

A

+ve - easy
-ve - need list, large standard error, periodicity pattern

54
Q

Random sampling

A

Sampling error can be managed

55
Q

Cluster random sampling

A

Cluster population then sample whole clusters

56
Q

Advantages and disadvantages of cluster sampling

A

+ve - list not needed, cheap
-ve - increases standard error

57
Q

Simple random sampling

A

Computer generated numbers

58
Q

Advantages and disadvantages of simple sampling

A

+ve - easy, quick
-ve - poor representation of minorities, bigger standard error than stratified

59
Q

Stratified random sampling

A

Samples taken from each group

60
Q

Advantages and disadvantages of stratified sampling

A

+ves - increases representation, decreases standard error
-ves - expensive, require prior population info

61
Q

Types of random sampling

A

Systematic
Cluster
Simple
Stratified

62
Q

Advantages and disadvantages of Non random sampling

A

+ve - convenient
-ve - sampling error can’t be measured, high potential for bias

63
Q

What is bias affected by

A

Sample size
NOT population size

64
Q

Observer bias

A

Researcher

65
Q

Confounding variables

A

Relate to exposure and outcome but not on causal pathway or with intervening variable

66
Q

Preventing confounding

A

Randomisation: generates comparable groups
Restriction: eliminates variation in confounders eg only recruiting females
Matching: control group matches confounders in case group

67
Q

What is the Most appropriate study design to investigate an infectious outbreak

A

Case-control study