EBP Flashcards
Which type of observational study gives the strongest evidence?
cohort
Which type of study gives the strongest evidence?
Randomised Control Trial
What are the 5 types of study?
Randomised Controlled Trial, Cohort Study, Case-control Study, Cross-sectional Study, Ecological Study
What does the P stand for in PECO?for observational studies
Patient/Problem/Population
What does the E stand for in PECO for observational studies?
Exposure
What does the C stand for in PECO for observational studies?
Control
What does the O stand for in PECO for observational studies?
Outcome
What does the P stand for in PICO for interventional studies?
Patient/problem/population
What does the I stand for in PICO for interventional studies?
Intervention
What does the C stand for in PICO for interventional studies?
comparison/control/comparator
What does the O stand for in PICO for interventional studies?
Outcome
What is a risk ratio?
an indication of strength of association between exposure and outcome
What is risk ratio also known as?
relative risks
How do you calculate risk ratio?
risk in exposed group/risk in unexposed group
How do you calculate risk?
number of new cases of outcome in time period/number without outcome at onset
What should always be presented with risk ratios?
confidence intervals and p-values
What does it mean if risk ratio is 1?
the risk is the same in the exposed and unexposed group
What does it mean if the risk ratio is less than 1?
there is a reduction in risk
What does it mean if the risk ratio is more than 1?
there is an increase in risk
How can you calculate risk difference?
risk in exposed group - risk in unexposed group
What are some strengths of cohort studies?
exposure precedes the outcome, measurement bias for the exposure is minimised, outcome incidence and risk can be measured directly
What are some weaknesses of cohort studies?
time consuming and expensive, large sample sizes, long follow up periods needed for rare outcomes, loss to follow up bias possible, measurement bias for the outcome possible
What is an odds ratio?
an indication of the strength of association between exposure and outcome
How can you calculate odds ratio?
odds in exposed group/odds in unexposed group
How can you calculate odds?
number of cases/number of controls
What should odds ratio always be presented with?
confidence intervals and p-values
Can you calculate risk in case-control studies?
no
What are some strengths of case control studies?
relatively quick as no follow up, efficient for rare outcomes, no loss to follow up bias, measurement bias for the outcome should be minimised
What are some weaknesses of case control studies?
Can’t be sure exposure precedes outcome, can’t calculate risk, selection bias may be likely, measurement bias for the exposure is possible
What is a confounder?
third factor that provides an alternative explanation for the observed association between the exposure and outcome
What is an adjusted estimate?
when you have adjusted the association between the exposure and outcome to account for confounders
What does it mean if the adjusted and unadjusted estimates are similar?
no confounding
What does it mean if the adjusted and unadjusted estimates are different?
confounding
What is subgroup analysis?
stratified analysis
When is subgroup analysis appropriate?
if there is evidence of main effects and an interaction/effect modification
When is confounding present?
when the observed effect is distorted by the influence of another variable
When is an interaction/effect modification present?
if the estimate of association differs according to categories of another variable
What is a systematic review?
scientific investigations of primary research studies
Do systematic reviews provide strong evidence?
yes
Why should you perform a systematic review?
to collate results of primary studies into a single estimate, to increase precision of data, to potentially resolve discrepancies between existing studies, establish conclusions
What are the five steps of a systematic review?
establish a research question, determine the inclusion and exclusion criteria, define a search strategy and screening of papers, assess the quality and risk of bias, collate results of studies to provide one estimate
What is a meta analysis?
results from each study are collated into a single estimate
Do all studies in a meta-analysis have the same weighting?
no, the larger a sample size, the more weight that study is given
What does a fixed effects model assume?
we are measuring the effect of the same exposure or intervention on the same outcome and the only reason for differing effect sizes between studies is due to sampling and random error
What does a random effects model assume?
we are measuring the effect of the same exposure or intervention on the same outcome and effect sizes differ between studies due to sampling and random error and other differences between studies
What does the I squared statistic describe?
the percentage of the variability in effect estimates that is due to heterogeneity rather than chance
Which effects model should be used when the I squared statistic is greater than 50%?
random effects model
Which effects model should be used when the I squared statistic is less than 50%?
fixed effects model
What is publication bias?
when journals may be more likely to publish studies which find an association in comparison to studies which find no association
How can you assess publication bias?
funnel plots and statistical tests
How do you graphically display a meta-analysis?
forest plot
How can you calculate heterogeneity?
I squared statistical test
What is a natural experiment?
a study design that occurs in nature
What is the newcastle ottawa scale used to asses?
the quality of the studies
What is CASP?
Critical Appraisal Skills Programme
What are the 7 principles of quality in qualitative research?
credibility, transferability, dependability, trustworthiness, integrity, transparency, conformability
What is credibility?
whether the research method is appropriate for the research question
What does it mean if a research question is appropriate?
the findings will accurately reflect the viewpoints and experiences of the participants
What is transferability?
whether the findings of the study can be applied to other settings or sample populations
What is dependability?
whether the method is explained to a sufficient detail to allow others to replicate the study
What is trustworthiness?
whether there is any conflict of interest by the researchers, whether there are any ethical issues
What is integrity?
whether it is clear who is checking the data at the inputting, analysis and write up stages
What is transparency?
whether there is full disclosure of all the procedural aspects of the research and if it complies with the reporting standards
What is conformability?
whether attempts have been made to ensure that the findings represent the thoughts of the participants and not the researcher
How can you ensure conformability and avoid researcher bias?
let participants be able to read the transcripts
