quantitate study design - critial numbers 1 Flashcards
what do descriptive statistics do
describe a sample
what do inferential statistics do
make inferences about our population
how can a sample be biased
when certain subgroups from the target population are over/under represented
it arises when imperfections in the research process cause our findings to deviate from the truth
what makes results generalisable
when the sample represents the population of interest
what does bias impact
the validity and reliability of study findings
what are the 5 types of bias
sampling bias
recall bias
information bias
the ‘hawthorne’ effect
attrition bias
define sampling bias
when the sample does not represent the population of interest
define recall bias
when there is inaccurate recall of past events/exposures/behaviours
define information bias
when there is incorrect measurement eg miscalibrated machine
define the ‘hawthorne’ effect
when participants change their behaviour when they know they are being observed
define attrition bias
when there is differential dropout from studies eg sicker participants drop out so our outcome is only measured on healthier participants
what are confounding variables
they obscure the real-life effect of exposure on an outcome
a confounder is related to both exposure and outcome
but is not on the causal pathway
what are the 2 types of study designs
experimental
observational
what is an experimental study design
when the researchers have intervened in some way
what is an observational study design
when the researchers have not intervened, and have merely observed
what are the 3 types of observational studies
retrospective - looking back into the past
cross-sectional - a single snapshot in time
prospective - following up over time
what are randomised controlled trials
when participants are randomly allocated to different interventions and follow up
what are 4 variations on the standard randomised controlled trial design
cluster randomised trials
crossover trials
multi arm and factorial trials
adaptive design trials
what are cluster randomised trials
when participant sare randomised in grooups rather than indivusally
what are crossover trials
when participants receive both interventions in a randomised order
what are multi arm and factorial trials
when two or more interventions are evaluated in a single study
what are adaptive design trials
when accruing information is used to inform planned design adaptations
what are 4 variations on the standard randomised controlled trial design
cluster randomised trials
crossover trials
multi arm and factorial trials
adaptive design trials
give 4 pros of randomised control trials
- gold standard
- reduces potential for confounding
- reduces bias via control and blinding
- can determine causal effects
give 3 cons of randomised control trials
- can be unfeasible or unethical
- require expert management
- expensive
what are cohort studies
non randomised
observational
typically prospective
give 3 pros of cohort studies
- useful when random allocation not possible
- can work for rare exposures - select participants on the basis of exposure
- can examine multiple outcomes
give 3 cons of cohort studies
- may require long follow up
- can be expensive
- not ideal for rare outcomes
what are case control studies
non randomised
observationall
retrospective
give 3 pros of case control studies
- faster because use past data so don’t need long follow up
- useful for rare outcomes - select participants on the basis of outcome
- cheaper
give 3 cons of case control studies
- more prone to bias or poor quality data
- harder to show causal relationship
- not ideal for rare exposures
what are cross-sectional studies
non randomised
observational
single time point
how do cross-sectional studies work
1 sample, 4 outcomes –>
unexposed, outcome
unexposed, no outcome
exposed, outcome,
exposed, no outcome
give 3 pros of cross sectional studies
- relatively quick
- cheap
- can assess multiple exposures/outcomes
give 3 cons of cross sectional studies
- susceptible to bias
- cannot prove causality
- not ideal for rare exposures/outcomes
what are ecologolical studies
when the unit of observation is a group rather than an individual
give 2 pros of ecological studies
- large scale comparisons
- can quantify geographical or temporal trends
give 2 cons of ecological studies
- ecological fallacy
- cannot make inferences at the individual level
what is the hierarchy of evidence from best to worst
1) systematic review/meta-analysis
2) randomised control trials
3) cohort studies
3) case control studies
4) cross sectional studies
5) case studies/ expert opinion/ anecdote
how do cross-sectional studies work
1 sample
4 outcomes –>
how do case control studies work
1 sample –> divides into cases and controls
cases –> exposure
controls –> exposure
how do cohort studies work
1 sample –> divides into exposed and unexposed
exposed –> outcome
unexposed –> outcome
how do randomised controlled trials work
1 sample –> divided into intervetion an comparator
intervention –> outcome
comparator –> outcome
