Biostatistics Flashcards
Crush this shit
Number of “new” cases/number of people at risk of acquiring that disease
Incidence
Total number of cases (new + old)/whole population
Prevalence
What is the importance of incidence?
Acute cases, e.g., incidence of Zika virus infection in August
Attack rate
is the number of people with disease over the number of people at risk for the disease.
What is the significance of prevalence?
Chronic cases only,a e.g., prevalence of diabetes in United States
Identifies all patients with the disease, and if negative, can virtually rule out the diagnosis.
Sensitivity
When you lower the cut off point you are ______ more patient with the disease? Another way of describing this is if a cutof of DM hba1c went from 6.5 to 5.5 how would that affect the amount of people with disease?
including
When you increase the cut off you are ______ more patient
excluding
Sensitivity and (PPV or NPV) go with one another?
NPV, and this is because as we lower the cut off this increases the amount true positives lowering the FN (increasing the NPV) and vice versa if we increase the cut off the TP go down and FN increases (decreasing the NPV)
Specificity and (PPV or NPV) go together?
Specificity goes with PPV, as the cut off increases the TN go up and the false positives go down (PPV increases) and vice versa if the cut off were to decrease
This is a rare disease reporting with details of the presentation of disease, images, treatment, and follow-up.
Case report
This study is Snapshot of present time uses surveys to identify current health problems.
Cross sectional
Study that starts with a risk factor and observes for disease development. The group is divided into exposed and unexposed.
cohort
Starts with disease and identifies risk factors. Group is composed of patients with and without disease and mutual risk factors
Case control
What is the weakness of a cohort?
Attrition (loss of study participants) or migration, selection bias is built in, confounding bias
What is the weakness of a case control study?
Recall bias, confounding bias
Is the graph positively or negatively skewed?
Positive skewed the mean will be the greatest value the median is ALWAYS the middle and the mode will be the smallest value
Would the mean or the mode have the higher value?
The mode
A pneumonic that will always help you find the mode.
Fat mode- Mode is the nearest to the fat portion of the graph.
_____ % of the population lies in one standard deviation.
68
______ % of the population lies within two standard deviations.
95
_____ % of the population lies within three standard deviations?
99.7
This statistical test has two numerical variables and example of this test (ie comparing height with income).
Pearson correlation
The statistical test that has 3 or more categorical variables with a comparison of a mean value (ie the mean hba1c black, hispanic, and Indigenous and the
ANOVA
The statistical test that with two categories and a dependent variable. (ie mean height between males and females)
T-test or Z-test
A statistical test with two categorical variables.
Chi square
The number needed to treat?
1/ARR Is the number of persons that need to be treated to prevent an adverse event.
The number needed to harm?
1/ARI, how many treatments before an adverse event.
What is the p value?
The probability that the null hypothesis is true. ie if a person has a p value of <0.05 that indicates that there is a less than 5% chance that the null hypothesis would be true.
What is a type 1 error
You said there is a relationship when in reality there isn’t (false positive)
What is a type 2 error
You said there is no relationship when there is a relationship
Relative risk
the risk of disease in exposed group/ the risk of disease in the unexposed group
Odd ratio
AD/ BC or (A/B)/(C/D)
absolute risk reduction
CER (c/c+d) - EER (a/a+b)
Attributable risk
EER (a/a+b) - CER (c/c+d)
This bias prolongs the survivor rate because disease detection is determined earlier?
Lead time bias
Confounding variables
Both the independent variable and other external factors can influence the dependent variable (outcome)
What prevents cofounding varibales?
Matching
What is case fatality?
The number of persons with the disease over the number of people that die from the disease.
