Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

EBM > Observational Studies > Flashcards

Observational Studies Flashcards

1
Q

What are the two types of observational study?

A

Case controls
Cohort studies

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

What is the difference between the types of observational study?

A

Case-controls - selected based on OUTCOME
(Retrospective - looks at exposures)

Cohort - selected based on EXPOSURE
(looks at outcomes)

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

How do we divide epidemiological studies?

A

Descriptive (ecological, cross-sectional, case reports/series)

Analytic (observational, intervention)

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

What are descriptive studies?

A

Used to generate hypotheses

  • measure variation of disease
  • assist DM
  • suggest aetiological factors
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is a cross sectional study?

A

Point prevalence/looking at a specific point in time (ie - smoking rates after legislation)

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

What are the advantages & disadvantages of cross-sectional studies?

A

ADVANTAGES
- cheap
- quick

DISADVANTAGES
All arise from the ‘slice of time’ aspect:
- cannot examine causality (as disease & exposure examined at one time)
- cohort effect (variation in participant characteristics over time)
- incidence impossible (don’t know rate of NEW cases, only number of cases)

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

What is an ecological study?

A

Units of observation = groups rather than individuals (ie - ‘white British’ vs ‘Bangladeshi’ in RFs for LC)

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

What are the advantages & disadvantages of ecological studies

A

ADVANTAGES
- cheap
- quick (as use secondary data usually - ONS etc)
- useful (hypothesis generation, prevalence, incidence - especially in rare diseases)

DISADVANTAGES
- data from group cannot be accurately applied to individuals within that group
- unclear time between exposure & disease

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

What is the main benefit of observational studies?

A

They can accurately determine cause-effect relationships!

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

What are the advantages of case-control studies?

A
  • quick
  • cheap(er than prospective cohort)
  • effective for rare diseases (as uses OR as an approximation of RR, which is more accurate in smaller cohorts)
  • can study many EXPOSURES
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are the disadvantages of case-control studies?

A
  • no incidence data (start with patients WITH condition already)
  • can’t estimate ABSOLUTE risk (approx RR via OR)
  • selection bias (difficult to find selection controls ie - never had BrCa)
  • recall & recording bias
  • poor quality when investigating specific exposures (as may have too small data pools)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are the advantages of cohort studies?

A
  • produces risk data! Accurate
  • multiple OUTCOMES from one exposure
  • useful for rare exposures
  • causality (good for working out)
  • can potentially eliminate some biases (recall bias, recording bias as in controlled environment of study)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are the disadvantages of cohort studies?

A
  • expensive! can span many years, require a lot of work & funding
  • larger sample size required
  • not good for rare outcomes
  • not good for conditions with long latency periods (study must be longer)
  • LOSS TO FOLLOW UP BIAS (attrition)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What sources of bias do you need to account for in case-control studies?

A
  • recall bias
    Extent of recall differs btw cases & controls
  • recording bias
    Certain pts will have more illness & thus more detailed notes
  • interviewer bias
    Non-neutral approach (different btw cases & controls)
  • response bias
    Response rate low/different btw cases & controls
  • sampling bias
    (Minor) way people selected for comparison group may be skewed = changes OR & size of estimated effect

ALSO: as no incidence data, cannot work out RR, can only use OR as approx (more accurate with rarer diseases)

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

What is a nested case-control?

A

Identify cases of a disease in a predefined cohort & equal number of controls from same cohort (who were selected based on exposure & not)

TAKES PLACE IN A COHORT STUDY (selected for exposure) BUT SELECTED within that population based on OUTCOME

E.g. - a group of smokers is the cohort, selected for the case-control based on whether they have lung cancer or not, retrospectively assessed for an occupational exposure to asbestos. (This may be done to remove the effect of confounding from the RF of smoking)

ADVANTAGES
- cheaper, easier than cohort as less effort to collect & analyse data
- minor relative loss in statistical efficiency
- good for biological precursors of disease

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

What are the Bradford Hill criteria for causality? (9)

TSBCPCRSA
Tarantulas Scare Boys and Carnivorous Piranhas Come with the Rain on Saturday Afternoons

A

(After bias, confounders & random chance excluded):

1. Temporality
- exposure before effect
2. Strength
- strong association (10x increased risk)
3. Biological response (dose-response relationship)
- greater exposure = greater effect
4. Consistency
- consistent with other studies, different populations
5. Plausibility
- biological mechanisms that support association (limited by understanding & beliefs)
6. Coherence
- does not contradict current evidence
7. Reversibility
- risk of effect reduced once remove exposure
8. Specificity
- causality more likely if cause very specific
9. Analogy
- more likely if analogy exists with other disease/species/setting

17
Q

What is ‘matching’ in observational studies?

A

To reduce confounding, researchers may ‘match’ cases with controls on the basis of potential confounders
(ie - match cases of DS & controls with similar maternal age if investigating the causal link wit birth order)

18
Q

What is the key benefit of retrospective cohort studies?

A

Advantages
. Shorten study time as aim for halfway through long latency period
. Less likely to LTF as shorter
. Cheaper
Disadvantage
. Retrospective element - if already developed outcome then lose temporal relationship guarantee

19
Q

What is the difference between matching and stratification

A

Matching - observational analytic studies when randomisation is impossible (often with age) effectively eliminates = STUDY DESIGN

Stratification - study without matching, group individuals in analysis & compare (sub-group analysis) — within group can eliminate as confounders = ANALYSIS

20
Q

What is the difference between cross-sectional & ecological studies?

A

Cross-sectional - unit of measurement is INDIVIDUAL
Ecological - unit of measurement is GROUP/POPULATION

21
Q

Why are incident cases preferable to prevalent cases?

A

Prevalent cases are affected by survival - those who have had the disease longer are more likely to have severe disease & thus may have had a greater exposure to the exposure (ie - smoked more heavily/suffered with COPD longer). They will be underrepresented as they are more likely to die of other causes during the observation period (and not develop the outcome).

Produces an errant lessening of the strength of association btw E & O

22
Q

Why is a prospective study preferable to retrospective studies?

A
  • can be sure E preceded O
  • researcher more in control over data collection
23
Q

Ecological fallacy

A

An ecological study’s applicability is limited as one cannot assume relationships at pop level hold true at individual level

24
Q

What are the advantages of case control studies?

A
  • multiple exposures (retrospective)
  • quick, inexpensive, easy
  • Efficient for rare diseases
25
Q

What are the disadvantages of case-control studies?

A
  • Can’t look at multiple outcomes for one exposure
  • Can’t estimate Absolute Risk as no incidence data (OR Instead)
  • Recall, recording, interviewer, response & sampling bias
26
Q

What are the advantages of cohort studies?

A
  • multiple outcomes (prospective=follow)
  • absolute & RR
  • Less doubt about cause/effect
27
Q

What are the disadvantages of cohort studies?

A
  • Can’t reasonably compare/draw conclusions from multiple exposures for one outcome
  • Expensive if long or large
  • Rare outcome - need large populations
  • Long time to develop - need long follow up time (time consuming & expensive, more likely to LTF = attrition bias)
  • Healthy worker effect
28
Q

What is the difference between closed & open cohorts?

A

Closed cohort - recruit at same time, follow up for set time (can calculate incidence)

Open - people enter & leave at any time — ppl in different times, need to calculate person years - total time/no. Outcomes)

29
Q

What are the advantages and disadvantages of descriptive studies?

A

ADVANTAGES
- hypothesis generation
- Quick, easy & inexpensive
Cross sectional - multiple outcomes & exposures
Ecological - study population exposures

DISADVANTAGES
- no control (eco - no test intervention)
- high risk bias
- no causality

EBM (8 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions