Evidence Based Medicine Flashcards
What is meant by a randomised study?
The population receiving the intervention and the population being compared are only difference because of random variation
What is meant by a controlled study?
The intervention is being compared to something else: a placebo, a gold-standard, best practice
What is meant by a trial?
An experimental intervention, with a beginning and an endpoint
How is an odds ratio calculated?
Quantifies the strength of the association between two events
Odds even when exposed/ odds event when not exposed
State the two ways of describing probability
P value
Confidence interval
Define P value
A numerical value indicating the probability that this observation has occurred due to chance
Define confidence interval
A way of indicating a range of values which probably contain the ‘true’ value
e.g. there is a 5% chance of the true value lying outside of these limits
The null hypothesis states that…
There is no relationship between the study variables
There is a clear difference between confidence intervals in the two groups. Why could this be?
The null hypothesis is false
Random chance
Confounding (creates false effect)
There is no clear difference between confidence intervals in the two groups. Why could this be?
The null hypothesis is true
Random chance
Confounding (hides the effect)
The effect size is too small
What is publication bias?
When papers that are published on a topic are an incomplete subset of all the studies that have been conducted
List ways in which publication bias is prevented
Data Protection Act (rules for management of personal information)
Caldicott Guardian Approval (access to patient’s records without consent)
Non-clinical ethics committee (non-NHS)
Clinical research ethics committee (doing things to patients)
What is meant by internal validity?
How good the research methods used by researchers answer the clinical question
Does the study measure what it sets out to measure?
What is meant by external validity?
Will we get the same results in real life settings
Can the results of the study be generalised to a wider population?
What is purposive sampling?
Sample units chosen as they have particular features or characteristics
What is convenience sampling?
Selection made on basis of who is available
What is snowball sampling?
Individuals identifying others to participate
What is opportunistic sampling?
Whoever is there
Define triangulation in terms of research studies
Area under investigation looked at from different perspectives using 2 or more research methods
What is absolute risk?
The number of events in treated or control groups, divided by the number of people in that group
What is ARC?
the AR of events in the control group
What is ART?
the AR of events in the treatment group
What is ARR (absolute risk reduction)?
Probability or chance of an event
Risk in exposed group - risk in non-exposed group
ARC-ART
What is RR (relative risk)?
ART/ARC
What is RRR (relative risk reduction)?
Ratio of the probability of an outcome in an exposed group to the probability of an outcome in an unexposed group
Incidence in exposed/ incidence in not exposed
(ARC-ART)/ARC
What is NNT (number needed to treat)?
1/ARR
Define statistical significance
Likelihood that a relationship between two or more variables is caused by something other than chance
Define cohort study
A longitudinal study that samples a cohort (a group of people who share a defining characteristic, typically those who experienced a common event in a selected period, such as birth or graduation), performing a cross-section at INTERVALS THROUGH TIME
State two ways of making a confidence interval narrower
Lower variability
Larger sample size
What is it called when there is no longer any need to sample more people to reach new conclusions or to back up or challenge existing conclusions?
Saturation
What is qualitative research?
Seeks to analyze data which researchers collect from people in relation to their experiences and circumstances
What is quantitative research?
Describe and measure the level of occurrences within data on the basis of numbers and calculations
What is the difference between a single blind and double blind study?
Single blind = Subjects of the study are left unaware of aspects of the study
Double = both subjects and researchers are unaware of 1 or more aspects being sought
What is a case-control study?
A type of observational study in which two existing groups differing in outcome are identified and compared on the basis of some supposed causal attribute
What is a cross-sectional study?
A type of observational study that analyzes data from a population, or a representative subset, at a specific point in time
Define standardised incidence ratio
Ratio of the observed number of cases to the expected number of cases
Extremely high satisfaction rates suggests…
Inclusion bias
