Study Designs Flashcards
Ecological
Population-level data
eg electoral role
Ecological advantages
Fast
Inexpensive
Few ethical issues
Data collection fairly easy
Can look at trends between regions or over time
Ecological Disadvantages
Cannot determine individual-level associations
Cannot demonstrate cause and effect
Lack of control over variables collected
Cross-sectional
Outcome and exposure status measure simultaneously
Cross-sectional advantages
Can look at associations- hypothesis generating
Data collection fairly easy
Can study multiple exposures and outcomes
Fast
Inexpensive
Cross-sectional disadvantages
Cannot demonstrate cause and effect
Prone to bias and confounding
Not useful for rare exposures or outcomes
Case-control
Participants selected on the basis of outcome
Case-control advantages
• Can look at association between outcome and prior exposures
• Fast
• Inexpensive
• Good for studying rare outcomes
• Can study multiple exposures
Case-control disadvantages
• Cannot determine incidence/risk of outcome
• Limited control over data quality – poor historic records or recall bias
• Retrospective nature limits ability to determine causality
• Not useful for rare exposures
Cohort
Participants selected in the basis of exposure
Cohort advantages
• Can look at incident cases and associations with exposure
• Good for studying rare exposures
• Can study multiple outcomes
• Control over data collected
• Exposure determined before outcome occurs so can demonstrate temporality for potential cause and effect
Cohort disadvantages
• Mostly prospective which can be time consuming
• Risk of loss to follow-up
• Expensive
• Not useful for rare outcomes
Randomised controlled trial
Participants randomly allocated to interventions then followed up to compare outcomes
Experimental study with random allocation into treatment groups (NOT random sampling)
RCT advantages
• Can study intervention effects on outcome(s)
• Random allocation means confounding factors should be evenly distributed
• Control over variables collected
• Comparator group means ability to account for placebo/temporal effects
• Less prone to bias, particularly where blinding and objective outcome assessment used
• Gold standard for establishing causality- can show cause-effect relationship
RCT disadvantages
• Time consuming
• Expensive
• Require expertise to run
• Can only be used where ethics and participant willingness permit randomisation to intervention
• Overly strict eligibility criteria may render sample not fully representative of population
Affected by non-compliance
Not practical for long-term effects
Case control studies can only use…
Odds not risks
Three-arm trial
a 3-arm trial consists of the intervention, the active control, and a placebo arm. This design is particularly attractive when the efficacy of the active control has changed, is volatile, or is in doubt
Cluster randomisation
a randomised controlled trial in which pre-existing groups, called clusters, of individuals are randomly allocated to treatment arms. For example, clusters may be clinical practices or schools where the individuals are patients and school children, respectively.
2 characteristics a placebo treatment should have
- Inactive form of ingredients
- Identical look and administration of treatment
Blinded
Participants do not know whether they receive treatment or placebo
Random assignment
Assignment is unpredictable and equal probability of ending up in either group
Hierarchy of evidence
- Systematic review of RCTs
- RCT
- Cohort
- Case control
- Cross-sectional
- Case series
- Anecdote
2 types of observational study
Descriptive and analytical
Descriptive study : individuals
Case report
Case series
Descriptive study : populations
Ecological study
Cross-sectional study
Single time point analytical study
Cross-sectional study- population: see how many have disease and how many have symptoms
Exposure <-> outcome
Retrospective analytical study
Case-control study- compare prevalence of symptoms in control vs disease group
Outcome —> exposure
Prospective analytical study
Cohort study- sample of unaffected and measure exposure to risk factor over time
Exposure —> outcome
Experimental intervention study
Randomised controlled trial
Types of RCT
Cluster
Crossover
Multi-arm + factorial
Adaptive
Cluster RCT
Randomise groups rather than individuals
Crossover RCT
Receive both interventions in randomised order
ONLY if incurable or chronic
Multi-arm + factorial RCT
> 1 interventions compared
Adaptive RCT
Accruing information informs study eg early termination I’d clear treatment difference
