study designs 2 Flashcards
what type of study is a case control study
analytical, observational type of study
case control steps
- they work backwards, we start with participants with a known outcome status
1) identify a source population
2) identify the cases (people with outcome) and controls (people without outcome)
3) assess their prior exposure lvl in cases and controls
4) calculate your odds ratio (measure of association)
what measures of occurrence and association can we calculate in a case control study?
occurrence: none
association:
odds ratio= odds of exposure in cases/ odds of exposure in controls
= (exposed cases/ unexposed cases)/ (exposed controls/unexposed controls)
odds ratio > 1 means that exposure may be a risk factor for the disease
case control important points
- index dates can be used for transient exposures in controls who didn’t have an associated event, the exposure is measured on the same date as the case eg: texting while driving (exposure) and having a car accident, see whether controls used their phone on the date the case had the car accident (index date)
- when selecting cases, there needs to be a clear definition of the outcome and cases should be newely incident
- the purpose of having a control group is to estimate the prevalence of exposure in the population from which the cases come from
case control important points pt 2
- controls must also be capable of becoming a case, and often we select multiple controls per one case, for better statistical power
- control selection, there are several ways we can select the controls, one is from the hospital
- using hospital based controls can cause problems as they aren’t necessarily representative of the population which the cases came from
- we need to ensure that the exposure is measured before the outcome occurred
- when measuring exposure level in a case control study, there is a potential for information bias (recall and interviewer bias)
case control studies strengths and limitations
strengths:
- good for rare outcome & transient exposures
- can assess multiple exposures
- temporal sequencing
- quick and inexpensive compared to other studies
limitations:
- can only study one outcome
- difficult to select one appropriate control group
- prone to selection and recall bias
what type of study is a randomised control trial
analytical, interventional type of study
steps of a randomised control trial
- similar to a cohort study, however instead of measuring an exposure we randomise an intervention (eg: medication)
1) identify a source population
2) randomly select the sample population who don’t have the outcome of interest
3) randomise the sample to either the intervention or control group
4) follow up over time and see who develops the outcome
5) calculate your measures of association and occurrence
relative risk equation
RR= incidence (intervention)/ incidence (control)
randomisation (random allocation)
- when we randomly allocate the participants into either the control group or the intervention group
purpose of randomisation
- purpose of randomisation is to control confounding (when another variable like age or sex distorts the relationship between exposure and outcome)
- randomisation does this as if done correctly, there should be the same proportion of known and unknown confounders in each group (so groups are comparable)
other variants of randomisation
cluster randomisation- where subgroups are randomised instead of individuals (eg: whole GP practice is randomised to control group)
stratified (block) randomisation- where participants are randomised within blocks
how do we protect randomisation and its benefits?
1) concealment of allocation
- allocation concealment is a different concept to blinding
- it means that the person randomising the participants does not know what the next treatment allocation will be, it is concealed and unpredictable
- as well as preserving the benefits of randomisation, concealment of allocation prevents selection bias, from participants or their doctors from selecting the treatment they want
2) intention to treat analysis
- this is when we analyse as we randomise, means we aren’t swapping people between groups (& therefore maintaining the randomisation)
- intention to treat analysis also gives us a ‘real world effect’, people don’t take interventions perfectly in real life
3) per protocol analysis
- analyse participants who fully complied to the study protocol
- can show the efficacy of the treatment, but we lose the benefits of randomisation
Bias in randomised control trials
1) lack of blinding
- if people involved in the study know whether they were in the intervention or control group this may influence them
- studies can be single blinded (participants) or double blind (participants & researchers) however its better just to be specific who is blinded in the study
2) loss to follow up
- if people leave the study (or are lost to follow up) this can cause confounding and bias, people who left may be different to those who stayed
3) non- adherence
- participants are sometimes not good at following instructions
to much adherence = bias
randomised control trials strengths and limitations
strengths:
- gold standard study design to test an intervention and cause and effect associations
- eliminate confounding and bias (if done well)
- calculate incidence and measures of association
limitations:
- many exposures cannot be randomised and we need to have clinical equipoise (genuine uncertainty about benefit/harm of intervention)
- expensive
- sometimes participants aren’t representative of the general population, therefore are not generalisable
- not good for rare outcomes
