Stats and ethics Flashcards
What is sensitivity?
Proportion of patients with the condition who have a positive test result
TP / (TP + FN )
What is specificity?
Proportion of patients without the condition who have a negative test result
TN / (TN + FP)
What is the positive predictive value?
The chance that the patient has the condition if the diagnostic test is positive
TP / (TP + FP)
What is Negative predictive value?
The chance that the patient does not have the condition if the diagnostic test is negative
TN / (TN + FN)
What is the likelihood ratio for a positive test result?
How much the odds of the disease increase when a test is positive
sensitivity / (1 - specificity)
What is the likelihood ratio for a negative test result?
How much the odds of the disease decrease when a test is negative
(1 - sensitivity) / specificity
What is a problem with case control studies?
Recall bias
What is a Cohort study?
Observational and prospective. Two (or more) are selected according to their exposure to a particular agent (e.g. medicine, toxin) and followed up to see how many develop a disease or other outcome.
The usual outcome measure is the relative risk.
Examples include Framingham Heart Study
What are some advantages and disadvantages of cohort studies?
Advantages:
- Can follow-up group with a rare exposure
- Good for common and multiple outcomes
- Less risk of selection and recall bias
Disadvantages:
- Take a long time
- Loss to follow up
- Need a large sample size
What is a case-control study?
Observational and retrospective. Patients with a particular condition (cases) are identified and matched with controls. Data is then collected on past exposure to a possible causal agent for the condition.
The usual outcome measure is the odds ratio.
Inexpensive, produce quick results
Useful for studying rare conditions
Prone to confounding
What are the advantages and disadvantages of a case-control study?
Advantages:
- Good for rare outcomes
- Quicker than cohort or intervention
- Can investigate multiple exposures
Disadvantages:
- Difficulties finding controls of match with cases
- Prone to selection and information bias
What is a cross-sectional study?
Provide a ‘snapshot’, sometimes called prevalence studies
Provide weak evidence of cause and effect
What are some advantages and disadvantages of a cross-sectional study?
Advantages:
- Cheap and quick
- Provide data on prevalence at a single point in time
- Large sample size
- Good for public health planning
Disadvantages:
- Risk of reverse causality (don’t know whether outcome or exposure came first)
- Cannot measure incidence
- Risk of recall bias and non-response
What are the advantages and disadvantages of RCT?
Advantages:
- Low risk of bias and confounding
- Can infer causality
Disadvantages:
- Time consuming
- Expensive
- Specific inclusion/exclusion criteria may mean the study population is different from typical patients
What are the different factors that can explain associations?
- Chance
- Bias
- Confounding
- Reverse causality
- A true association
What is selection bias and what can cause it?
Systematic error in the selection of study participants of the allocation to different study groups
- Non- response
- Loss to follow up
- Are those in the intervention group different to those in the control group
What are the different types of information bias?
Measurement (e.g. different equipment used to measure the outcome in the different groups)
*Observer (e.g. the researcher knows which participants are cases and which are controls and subconsciously reports/measures the exposure or outcome differently depending on which group they are in)
- Recall (e.g. events that happened in the past are not remembered and reported accurately)
- Reporting (e.g. respondents report inaccurate information because they are embarassed)
What is a type 1 error?
Two types of errors may occur when testing the null hypothesis
type I: the null hypothesis is rejected when it is true
What is a type 2 error?
type II: the null hypothesis is accepted when it is false - i.e. Failing to spot a difference when one really exists, a false negative.
How do you work out relative risk?
Relative risk (RR) is the ratio of risk in the experimental group (experimental event rate, EER) to risk in the control group (control event rate, CER).
The term relative risk ratio is sometimes used instead of relative risk.
EER/CER