Behavioral science Flashcards
Lead-time bias
prolonged survival in patients with screening test
- Earlier diagnosis without improved prognosis
Recall bias
Inaccurate patient recall of past exposure
- ex: what caused cancer?
Observer bias
Investigator decision changed by prior knowledge of exposure status
Type 1 error
Reject null hypothesis when it’s true
- Found a difference when there isn’t one)
P-value
Probability of making type 1 error
- Accepted at P< 0.05 (5% chance)
Type 2 error
Failure to reject null hypothesis when it’s false
Beta
Probability of making a type 2 error
Power
1- beta
ex: beta= 20% chance of making type 2 error;
- power= 80% chance of rejecting truly false null hypothesis
Case-control study
Compares those with disease to those without disease
- Measures Odds Ratio (OR):
(exposed with disease/diseased)/ (exposed without disease/ all without disease)
Cohort study
Compares group with exposure/risk factor to group without exposure
- Measures relative risk (RR):
Risk of disease with exposure/ risk of disease without exposure
Clinical trial phases
Phase I= safety, toxicity, PK
Phase II= efficacy, dosing, AEs
Phase III= comparison to standard tx
Phase IV= rare, long-term AE (post-marketing)
Case-fatality rate
Fatal/ (Fatal + nonfatal cases)
Sensitivity
Test detects disease when disease present
- Screening tool for disease with low prevalence
True positives / (TP + False negatives)
Specificity
Test indicates non-disease when disease is absent (ruling in disease)
- Confirmatory test after positive screen
True negatives/ (TN + False positives)
Positive predictive value
Proportion of positive test results that are truly positive: TP/ (TP+ FP)
Negative predictive value
Proportion of negative results that are truly negative
TN/ (FN + TN)