Terms in this set (46)
People with and without disease asked about previous exposure to a variable. What is the main measure of association?
Exposure odds ratio
Define exposure odds ratio
Odds of exposure of people with the disease (cases)
Odds of exposure of people without the disease (controls)
Used to compare exposure of people with the disease to the exposure of people without the disease
Case control study
Important measure of the impact of a risk factor
Attributable Risk Percent in exposed (ARP)
100 x [(Risk in exposed - risk in unexposed) / Risk in exposed]
Risk in exposed can be substituted with RR
Cases ie "people with rectal cancer" being MATCHED to controls such as similar age and race, is used to prevent _______
Confounding (inaccurate correlations)
Matching variables should always be potential confounders of the study, such as
Age, race, etc
A group of subjects is selected and their exposure status determined. Cohort then followed for a period of time and observed for development of the outcome.
Selecting pts with a particular disease and pts without the disease and determining previous exposure
Case control study
Also known as prevalence study. Simultaneous measurement of exposure and outcome. Frequently uses surveys.
Predictive value =
PPV = TP / (TP + FP)
NPV = TN / (TN + FN)
Typically used in case-control studies. Odds that the group with the disease was exposed to a risk factor divided by the odds that the group without the disease was exposed.
OR = (a/c)/(b/d) = ad/bc
Typically used in cohort studies. Risk of developing disease in the exposed group divided by the risk in the unexposed group.
Relative risk = (a /a+b) / (c/c+d)
The difference in risk between exposed and unexposed groups, or the proportion of disease occurrences that are attributable to the exposure. ie risk in exposed - risk in unexposed
(a/a+b) - (c/c+d)
The proportion of risk reduction attributable to the intervention compared to a control
Relative risk reduction (RRR)
The difference in risk attributable to the intervention as compared to a control.
Attributable risk reduction (ARR) = (c/c+d) - (a/a+b)
Number needed to treat =
Number needed to harm =
A ratio representing the likelihood of having the disease given a positive result.
Positive likelihood ratio
Positive likelihood ratio formula
LR+ = (sensitivity) / (1-specificity)
A ratio representing the likelihood of having the disease given a negative result
Negative likelihood ratio
Negative likelihood ratio formula
LR- = (1-sensitivity) / specificity
Disease prevalence effect on positive predictive value
When there is greater prevalence, the PPV will have more true positive. When prevalence is lower, the PPV will have a greater number of false positives.
As disease prevalence decreases, NPV will (increase/decrease)
The probability of a type II (Beta) error is related to how much power a study has to detect a difference when a difference actually exists. Power =
Also defined as the probability of rejecting the null hypothesis when it is truly false
1 - Beta
Studies with larger sample sizes have greater _____ to detect differences if they exist
How is false positive rate determined?
If 100/1000 healthy people get a false positive, the FPR is 10%. That is then multiplied by the # of healthy in the population (say 100,000 and 10,000 have the disease. The FPR only applies to those without, so 90,000 without would then show a FPR of 9,000)
In women who have mutation X, 30 develop cancer and 65 do not. In women who do not have mutation X, 20 develop cancer and 75 do not. What is the relative risk?
(30/95) / (20/95) = 30/20 = 1.5
New chemo drug given to 100 with cancer, 40 still alive. normal chemo given to 500, 100 still alive. What is the NNT?
(100-40)/100 = .6 mortality
(500-100)/500 = .8 mortality
.8-.6 = .2
NNT = 1/.2 = 5
This error occurs when researchers fail to reject the null hypothesis when it truly is false, causing them to miss true relationships. (Study finds aspririn doesn't effect platelets when in fact it does)
Type II error
Occurs when a study finds a significant difference between 2 groups when one does not exist
Type I error
alpha is the maximum probability acceptable that a type I error will occur (similar to p)
Crude mortality rate
Number of deaths divided by the total population size
Dividing the number of deaths of a particular disease by the total population size
Cause specific mortality rate
Case fatality rate
Number of deaths from a specific disease divided by number of people with the disease
Calculated by dividing the observed number of deaths by the expected number of deaths
Standardized mortality ratio
Attack rate (incidence measure)
Number of people with incidence/disease divided by population at risk
Knowing the outcome from the beginning and acquiring about past exposures
In case control studies, we calculate the odds of exposure (the chance of being exposed to a particular factor) in case patients (those with disease) and compare it with the odds of exposure in control patients (those without the disease)
Compares the odds of exposure in diseased subjects with the odds of exposure in non-diseased subjects
Exposure odds ratio (a/c) divided by (b/d)
The coefficient of determination expresses the percentage of variability in the outcome factor that is explained by the predictor factor. How is it calculated?
Study has correlation coefficient of -.8 (p<.01). How much variability is explained by variable?
.8^2 = .64
The difference between incidence in exposed and incidence in non-exposed
Cancer in smokers is 30/60 and 10/40 in non-smokers.
Attributable risk is .5 - .25 = .25
The contribution of a given exposure to the incidence of a disease in relative terms. What percent of yearly respiratory disease in smokers is attributable to smoking?
Attributable risk percent
30 of 60 smokers get disease. 10 of 40 non smokers also.
ARP = (30/60 - 10/40) divided by 30/60
Describes the impact of exposure on the entire study population. What percent of the yearly respiratory disease in the population is attributable to smoking?
Population attributable risk. 30 of 60 smokers and 10 of 40 non smokers.
Overall incidence of respiratory disease is 40/100. 30/60 - 40/100 = .5 - .4 = .1 and then divide this by incidence of respiratory disease in smokers: .1/.5 = .2
Confidence interval can be calculated from mean, SD, and z-score. (using standard error of mean).
Mean +- 1.96*SD/sqrt(n) for 95%
Replace 1.96 with 2.58 for 99%
"A positive test result is seen 9 times more frequently in patients with the disease than in patients without the disease"
Positive likelihood ratio
Sensitivity divided by (1-specificity)
A phenomenon whereby a screening test preferentially detects less aggressive forms of a disease and therefore increases the apparent survival time
YOU MIGHT ALSO LIKE...
UWorld Biostatistics Subject Review
Clin Med - Exam 1 Study Guide Part IV - Epidemiology
OTHER SETS BY THIS CREATOR
Step 2 Biostats
Pathoma Ch 10 GI
Pathoma Ch. 17: CNS