EBM Flashcards
Case Control Study
Retrospective, observational study that helps answer questions about aetiology and potential harms of interventions.
Allows odds ratios to be calculated but not absolute risks. Difficult to control for confounding variables.
E.g. Is mobile phone use associated with brain tumours?
Cohort Study
Prospective, observational study that helps answer questions about aetiology and prognosis.
Identifies cohort who receive an exposure of interest and a matched control who don’t and these are both then followed up over time to see if there is a difference in the outcome of interest.
e.g. Group of doctors who smoked vs groups of doctors who didn’t and observe the incidence of lung cancer between the groups
Confounding variable
Factors that you are not interested in but will affect the result. Can be difficult to control.
Cross sectional study
Observational study that observes a defined population at a single point in time or time interval and measures the prevalence of risk factors. Cheap and simple but is subject to bias.
Sensitivity
Probability of a positive test amongst patents with the disease. Very sensitive test will have very few false negatives and be good at picking up disease.
SnNOUT means that if a test is very Sensitive (Sn) and Negative test rules the diagnosis out.
Specificity
Probability of a negative test among patients without the disease. A very specific test will have few false positives and be good at ruling a disease out.
SpPIN means if a test is highly Specific, a Positive result rules the diagnosis in.
Odds Ratio
Odds of an event happening vs it not happening.
p-value
Probability that no difference exists between interventions for a given endpoint. Can take any value between 0 (no chance) and 1 (certainty).
If less than 0.05 then the effects of two interventions are said to be statistically significantly different.
Prevalence
Probability of a disease in a population at any one point in time.
E.g. if the prevalence of diabetes in the population is 2% = 2% of the population at the time of the study have diabetes.
Publication Bias
Negative trials are just as valid as positive ones but are less likely to be published.
Type 1 error
The incorrect rejection of a true null hypothesis (also known as a “false positive” finding).
Type 2 error
Incorrectly retaining a false null hypothesis (also known as a “false negative” finding).
Type 1 error
The incorrect rejection of a true null hypothesis (also known as a “false positive” finding).
Result is statistically significant but it is a chance finding and in fact there is no real difference.
Type 2 error
Incorrectly retaining a false null hypothesis (also known as a “false negative” finding).
Study finds no significant difference when there is a real treatment difference.