case-fatality rate
of cause specific deaths
of cases of disease
incidence (risk, attack rate)
of new cases of illness
of people at risk of illness (or in population)
prevalence
of existing cases of disease (old and new)
of total population
Absolute Risk Reduction
subtraction of 2 risks
Relative Risk Reduction
absolute risk reduction
Risk of unexposed
number needed to treat
1
ARR
Hills Criteria:
- 5.
- Strength: size of measure of association
- Consistency: reproducibility
- Temporality: cause precedes effects
- Biologic Gradients: presence of a gradient of risk (dose-response) assocaited with degree of exposure
- Plausibility: biological feasibility to the assocaiton
bioethics:
- 4.
- Autonomy: self-rule/self-determination ; no coercion ; understanding of risks
- Beneficence: benefit individual patients (not society)
- Justice: fair treatment regardless of patient characteristics
- Nonmaleficence: do no harm (researchers cannot withhold information)
NHANES
National Health and Nutrition Examination Survey
assess the health and nutritional status of adults and children
NHIS
National Health Interview Survey
health of civilian, non-institutionalized population
NAMCS
National Ambulatory Medical Care Survey
based on sample of visits to non-federal, non-institutinoalized (office-based) physicians
NHCS
National Hospital Care Survey
non-federal hospital-based settings
BRFSS
Behavioral Risk Factor Surveillance System
telephone health survey that collects information of health risk behaviors, preventative helath practices, and health care access related to chronic diseases and injury
Sensitivity
how accurately a test can correctly detect presence of disease when in fact disease is actually (known to be) present
True Positives
all diseased
Specificity
how accurately a test can correctly detect absence of disease when the disease is actually (known to be) absent
True Negative
all non-diseased
Positive Predictive Value
how accurately a positive test correctly predicts the presence of disease
True Positive
all positive tests
Negative Predictive Value
how accurately a negative test correctly predicts the absence of disease
True Negative
all negative tests
Diagnostic Accuracy/Precision
percentage of all correctly-identified patients (True Positives and True Negatives) out of the total number of screened patients
True Positives + True Negatives
total screened
Nominal Data:
- Prediction
- correlation
- Proportion of events (survival)
- more than 3 groups
- related
- independent
- ____ test of Inequality
- 2 groups
- related
- independent
- logistic regression
- contingency coefficient
- log-rank
- cochran
- chi-square, fisher’s exact
- Bonferroni
- McNemar
- Pearson’s Chi-Square
ordinal data:
- prediction
- correlation
- proportion with event (survival)
- 3 or more groups
- related
- independent
- post-hoc
- 2 groups
- related
- independent
- multinomial logistic regression
- spearman correlation
- cox proportional hazards
- friedman
- kruskal-wallis
- student-newman-keul, dunnet, dunn
- wilcoxon signed rank
- mann-whitney
interval data:
- prediction
- correlation
- proportion with event (survival)
- 3 or more groups
- related
- independent
- post-hoc tests
- 2 groups
- related
- independent
- linear regression
- pearson correlation
- kaplan-meier
- repeated measures ANOVA (repeated measures ANCOVA when cofounder present)
- ANOVA (ANCOVA when confounders present)
- Bonferroni, Tukey, Scheffe, Dunn, Dunnet, Student-Newman-Keul
- paired T test (or repeated measures ANOVA)
- student t test (or ANOVA)
regression
provide a measure of the relationship between variables by allowing the prediction about the dependent, or outcome, variable (DV) knowing the value/category of independent variables (IVs)
correlation
r = -1
r = 0
r = 1
provides a quantitative measure of the strength and direction of a relationship between variables
strong negative
no association
strong positive
proportion of events (survival)
compares the proportion of events over time, or time-to events between groups
commonly represented by kaplen-meier curve
