Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

EBM > Observational > Flashcards

Observational Flashcards

1
Q

If feasible, which study is the best design to examine the relationship between environmental exposures and disease outcome?

A

Cohort

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

What do descriptive studies be used for?

A

Describe pattern of disease in populations and can be used to generate hypotheses

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

What are analytical studies used for?

A

To uncover cause of disease and test hypotheses

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

What stays the same in cross-sectional studies?

A

Time

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

How should you determine a causal association in observational studies?

A
  1. Consider bias, confounding and chance

2. If none of these explain, consider the criteria for causality

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

What is bias?

A

Systematic deviation from the truth, or process leading to such deviation

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

What are the 2 main classes of bias?

A

Selection bias and information bias

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

What is selection bias?

A

The way in which groups are selected for a study means they cannot be validly compared with each other (or general population)

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

What are the 3 types of selection bias?

A

Sampling bias, response bias, follow-up bias

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

What is sampling bias?

A

If each potential member of the population being studied does not have an equal chance of selection into the study

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

What is response bias?

A

Some people more likely to take on part, and these people likely to differ systematically thus the sample cannot represent the population

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

What is follow up bias?

A

Especially in cohort studies, there can be a selective and systematic loss of follow up

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

What type of bias is the healthy worker effect?

A

Selection bias - sampling type

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

What is information bias?

A

The way data is obtained from groups differs systematically

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

What is recall bias?

A

A type of information bias. One of the groups has consistently higher or lower levels of recall errors e.g. rumination of risk factors if have illness

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

What is recording bias?

A

A type of information bias. One group has more complete information.

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

What is interviewer bias?

A

If the interviewer knows the state of the individual, they might probe more or less for evidence of exposure to factor of interest

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

What type of bias is social acceptability bias and what is it?

A

A type information bias when subjects give socially desirable answers that do not accurately reflect the truth

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

How can you combat social desirability bias?

A

Compare different records, use anonymous self-completion questionnaires

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

What is confounding?

A

When an extraneous factor is not adequately controlled and correlates (inversely or directly) to both IV and DV. Thus, it could give rise to an association between these 2.

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

What does confounding damage?

A

The internal validity of an experiment

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

What can confounders do to the association between exposure and disease?

A

Exaggerate, cause or mask

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

How can you measure disease?

A

Incidence rate, incidence risk, point prevalence

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

What is the incidence?

A

The number of new cases arising in a population over specified period of time in a defined population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
How do you calculate the incidence rate?
Number of new cases / (number of person-years accumulated)
26
How do you calculate incidence risk?
Number of new cases / number persons at risk at beginning of observation period
27
What is the prevalence of a disease?
The total number of individuals who have the disease at a particular time Incidence x Duration
28
What is the point prevalence?
The proportion of people with a disease at a specific point in time
29
How do you calculate the point prevalence?
Number of persons with disease at specific time point / total population at risk of disease at same time point
30
How do you calculate period prevalence?
Number of persons with disease at any time over specific period / total population seen over period of time
31
What are the 2 types ways to measure effect?
Ratios and differences
32
What to ratios measure?
The strength of an association between exposure and disease occurrence
33
What do differences measure?
The magnitude of an effect of risk on rate
34
How do you calculate relative risk?
Incidence risk in exposed group / incidence risk in unexposed group
35
What are the two types of differences of effect?
Risk difference and attributable risk
36
What type of studies is risk difference used in?
Therapy/intervention studies
37
Which type of studies is attributable risk used in?
Aetiological/cohort
38
What method is used to assess causality?
Bradford-Hill Method
39
Which 3 things do you need to exclude first to assess causality?
Bias, chance and confounding
40
What are the criteria for assessing causality?
1. Biological plausibility 2. Time 3. Strength of association (RR) 4. Biological gradient or dose-response relationship 5. Consistency: with other studies 6. Specificity 7. Coherence 8. Experiments 9. Analogy
41
How do you look at specificity?
If supposed cause is associated with one disease only, or disease is associated with one cause only this adds weight to causal interpretations
42
What shows coherence of a study?
If it does not contradict what is already known about natural history and biology of disease
43
How is the impact of chance shown in a study?
P values and CIs
44
How is the impact of bias shown in a study?
Study design, selection bias and information bias
45
How is the impact of confounding shown in a study?
Reference to all other known risk factors being accounted for
46
What is a dose-response relationship?
The more exposure, the higher risk or vice versa
47
Which type of study can not imply as strongly reliable causality due to the temporal sequence?
Case control as disease known before exposure
48
How do descriptive observational studies characterise disease?
Time, place, person
49
What are the aims of descriptive observational studies?
1. Alert medical community to demographics of people most/least susceptible to disease 2. Assist Evidence based planning of health and medical care facilities 3. Hypothesis generation: Provide suggestions concerning disease aetiology for further investigation using analytic studies
50
What are the 3 main types of descriptive observational studies?
1. Case report/ series 2. Cross sectional studies 3. Ecological studies
51
What are cross-sectional studies?
Taking a snapshot of a cross section at a particular point in time
52
What must you first clearly define in cross-sectional studies?
A target population to whom study is generalisable
53
What quality of a study does sampling affect?
Generalisability
54
What is a sampling frame?
The way a sample is drawn from the target population
55
Should sampling for cross sectional study be random or non-random?
Either
56
What are types of random sampling?
1. Simple random sampling eg random number from register 2. Cluster sampling (GP/schools) 3. Stratified sampling (compare small groups)
57
What are examples of non-random methods is sampling?
1. Systematic (every 10th patient) 2. Snowball (ask patient to recommend friends) 3. Street survey (pick people that look the part)
58
What are advantages of cross-sectional studies?
1. Quick and inexpensive
59
What are disadvantages of cross-sectional studies?
1. Time sequence and causation (disease and exposure measured simultaneously) 2. Cohort effects with age relationships 3. Problems with interpreting prevalence (mix of incidence and survival) e.g. only good for less aggressive diseases with higher survival
60
What test would you use to compare means of a cross-sectional study?
T-test
61
What test would you use to compare proportions of a cross-sectional study?
Chi-squared
62
What test would you use to analyse correlation of a cross-sectional study?
Pearson
63
What test would you use to analyse strength of association of a cross-sectional study?
``` Prevalence ratio = prevalence of outcome in those exposed / unexposed Odds ratio (approx the same as PR) < 10% is a rare outcome ```
64
What is the unit of observation for ecological studies?
Groups of individuals
65
Name some advantages of ecological studies
1. Quick 2. Inexpensive 3. Hypothesis-generating
66
Name some disadvantages of ecological studies
1. Subject to confounding 2. Prone to sampling bias 3. Prone to information bias (recording could differ) 4. Group of people described by disease data are not often the same as those described by exposure data 5. Time relationship between measurement of exposure and disease is often unclear
67
What is ecological fallacy?
Assuming the relationship found amongst groups in an ecological study holds at an individual level (can’t because do not know the individuals are the same and the time frame)
68
How can you visually analyse ecological study results?
Scatter plot
69
How can you numerically analyse ecological study results?
Pearson’s r
70
Which 2 features distinguish observational from descriptive studies?
The unit of observation is the individual and the risk period is prolonged
71
In which type of observational analytic study is exposure status known at the start?
Cohort
72
In which type of observational analytic study is disease status known at the start?
Case control
73
What are advantages of cohort studies?
1. Can eliminate recall bias 2. No limit to number of diseases can research 3. Temporal aspect of cause and effect is present 4. Useful for rare exposures 5. Measures absolute and cumulative risk
74
What are disadvantages of cohort studies?
1. Follow-up bias 2. Not useful for rare diseases 3. Expensive and difficult 4. Unnecessarily expensive if disease is rare and only need relative risk associated with different exposure levels to risk factor
75
Why would you choose a prospective over a retrospective cohort study?
Exposure can be accurately measured and confounders can be assessed
76
Why would you chose a retrospective over a prospective cohort study?
Cheap and short time scale
77
How should you choose cohorts for a cohort study?
Free of disease or researcher unaware of disease status. Exposed and comparison should be comparable.
78
What is the downside of using occupational groups as cohorts?
Healthy worker effect: workers have lower overall death rates than general population as severely ill and disabled are excluded from this group
79
What are the 3 ways to refine measure of exposure in cohort studies?
1. Dichotomous (ever vs never exposed) 2. Categorical: none/low/medium/high 3. Quantitative: continuous variable
80
What are important sources of bias to note in cohort studies?
Lost to follow-up and healthy worker
81
Which type of risk analysis determines causality?
Attributable risk
82
How can cohort study results be adjusted for confounding?
1. Standardisation 2. Stratification 3. Regression models: linear, logistic, cox, poisson
83
How can cohort results be indirectly standardised?
Adjust for mortality/disease incidence compared to population using SIR and SMR, adjust for age and sex
84
Which statistical test is used to analyse survival curves?
Log rank
85
What are SMRs?
Standardised mortality ratio - comparing observed number of deaths in study with what would expected if mortality rates from general population were applied according to age/sex profile of study population
86
Why are population-based case control studies better?
Well defined cohort so easier to select controls e.g. from national registry
87
What are some advantages of case-control studies?
1. Good for rare diseases 2. Can explore numerous risk factors 3. Can study diseases with long latency 4. Can estimate relative risk of disease associated with difference exposure levels (odds ratio approximation to relative risk)
88
What are disadvantages of case-control studies?
1. Retrospective so temporal relationship may not be clear 2. Specific to one disease 3. Selection bias 4. Recall bias 5. Cannot determine absolute risk
89
Why should we use incident cases for case-control studies?
Can be more confident about temporal relationship of risk and disease, and cases are not impacted by risk of survival (as prevalent cases are)
90
What does bias mean in relation to selecting controls for case-control studies?
Sampling controls has over or under represented individuals with respect to exposure than in the underlying cohort
91
What is the key principle of selecting controls?
Should be selected from subjects who would have become cases in the study if they developed the disease
92
What is the advantage of matching controls to cases?
Forces cases and controls to be similar on possible confounding factors
93
What must you remember when matching controls to cases?
Use methods of analysis which account for matching, otherwise the results will be biased
94
Which types of bias can impact case-control studies?
Recall, recording, interviewer, sampling and response
95
Which type of analysis should you use in case control studies?
Odds ratio
96
What does odds ratio approximate in case control studies?
Relative risk
97
Why can you not use relative risk in case control methods?
Do not have incidence rates
98
How do you adjust for confounding in an unmatched case control study analysis?
Logistic regression
99
How do you adjust for confounding in a matched case control study analysis?
Conditional logistic regression
100
What is selective survival bias?
If patients die before interview they are excluded from study, but the reason for death could be the disease studied
101
What is a nested case control study?
Case control study takes place within a pre-defined cohort study

EBM (8 decks)

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