Biostatistics Flashcards
How is a cross-sectional study performed?
What information can be derived from it?
An observational study that collects data from a group of people at a single point in time.
Used to assess disease frequency and associated risk factors.
Measures disease prevalence and show risk factor association with disease - no causality. “What is happening?”
How is a case-control study performed?
What information can be derived from it?
What statistic is used to compare the groups?
An observational and retrospective design.
Compares a group of people with disease to a group without disease. Then, looks for prior exposure or risk factor.
Uses an odds ratio
“What happened?”
How is a cohort study performed?
What information can be derived from it?
What statistic is used to compare the groups?
An observational and prospective or retrospective design.
Compares a group with a given exposure or risk factor to a group without exposure. Does the exposure increase likelihood of disease?
Use relative risk
How is a twin concordance study performed?
What information can be derived from it?
Compares frequency with which both monozygotic twins or both dizygotic twins develop same disease. Measures heritability and influence of environmental factors.
How is an adoption study performed?
What information can be derived from it?
Compares siblings raised by biological versus adoptive parents with respect to disease. Measures heritability and environmental factors.
What is the purpose and typical study sample for a Phase I clinical trial?
Phase 1: “Is it safe?” Assess safety, toxicity, and pharmocokinetics.
Study sample is small number of healthy volunteers.
What is the purpose and typical study sample for a Phase II clinical trial?
“Does it work?” Assess treatment efficacy, optimal dosing, and adverse effects.
Study sample is a small number of patients with disease of interest.
What is the purpose and typical study sample for a Phase III clinical trial?
“Is it as good or better?” Compare the new treatment to the current standard of care.
Study sample is a large number of patients randomly assigned to either treatment under investigation or best available treatment (or placebo if none exists).
What is the purpose and typical study sample for a Phase IV clinical trial?
“Can it stay?” Detects rare or long-term adverse effects of treatment. Can result in removal from market.
Study sample: Postmarketing surveillance trial of patients after approval.
Below is a 2x2 table obtained from data gathered during a diagnostic test evaluation study.
What does each individual letter represent?
What about each horizontal and vertical sum?
A = True positive
B = False positive
C = False negative
D = True negative.
A+B = Total positive tests
C+D = Total negative tests
A+C = Total amount of people with disease
B+D = Total amount of people without disease
What does sensitivity measure?
Proportion of all people with disease who test positive, or the probability that a test detects disease when disease is present.
AKA the true positive rate.
How do you calculate sensitivity?
Sensitivity = TP / (TP + FN)
Where TP = true positive. FN = false negative.
On a typical 2x2 plot, this is A / (A+C)
OR
Sensitivity = 1 - false negative rate
What is a highly sensitive test useful for?
SN-N-OUT - A highly SeNsitive test, when Negative, rules OUT disease.
A highly sensitive test indicates a low false negative rate.
Used for screening in diseases with low prevalence.
What does specificity measure?
Specificity is the proportion of all people without disease who test negative, or the probability that a test indicates non-disease when a disease is absent.
AKA true negative rate.
How do you calculate specificity?
Specificity = TN / (TN + FP)
In a 2x2 plot of test versus disease, this = D / (D+B)
OR
Specificity = 1 - false-positive rate
When is a highly specific test useful?
SP-P-IN = highly SPecific test, when Positive, rules IN disease
Value approaching 100% is desirable for ruling in disease and indicates a low false-positive rate. Highly specific tests are used for confirmation after an initial positive screening test.
Do specificity and sensitivity vary with disease prevalence? If so, how?
NO. Specificity and sensitivity are fixed properties of a test.
What does positive predictive value measure?
The proportion of positive test results that are true positive.
Or
The probability that a person actually has the disease given a positive test result.
How do you calculate positive predictive value?
PPV = TP / (TP + FP)
On a 2x2 plot, PPV = A / (A+B)
What does negative predictive value measure?
Proportion of negative test results that are true negative.
Or
Probability that person is actually disease free given a negative test result.
How do you calculate negative predictive value?
NPV = TN / (FN + TN)
On a 2x2 plot NPV = D / (D+C)
Do PPV and NPV vary with disease prevalence? If so, how?
Yes.
When prevalence increases (higher pretest probability), PPV increases
When prevalence increases, NPV decreases (inversely related to prevalence)
What is incidence?
How do you calculate it?
The number of new cases per unit time.
Incidence rate = (# of new cases in a specified time period) / (Population at risk during same time period.)
Incidence looks at new cases (incidents).
What is prevalence?
How do you calculate it?
Prevalence is the amount of disease in a population at risk.
Prevalence = (# of existinc cases) / (population at risk)
and
Prevalence is approximately = incidence rate x average disease duration
Prevalence > incidence for chronic diseases
Prevalence is approximately = incidence for short disease duration (colds, etc)
Prevalence looks at all current cases.
What is an Odds ratio? How is it calculated?
When is it used?
Used in case-control studies.
Odds that the group with the disease (cases) was exposed to a risk factor (a/c) divided by the odds that the group without the disease (controls) was exposed (b/d)
OR = (a/c) / (b/d)
or
OR = AD/BC
What is relative risk?
How is it calculated, and when is it used?
Used in cohort studies. Risk of developing disease in the exposed group divided by risk in unexposed group.
If prevalence is low, RR approximates OR.
RR = (a/a+b) / (c/c+d)
What is relative risk reduction?
How is it calculated?
The proportion of risk reduction attributable to the intervention as compared to a control.
RRR - 1 - RR
What is attributable risk?
How is it calculated?
The difference in risk between exposed and unexposed groups, or the proportion of disease occurrences that are attributable to the exposure.
AR = (a/a+b) - (c/c+d)
What is absolute risk reduction?
How is it calculated?
ARR = differenc ein risk (not the proportion) attributable to the intervention as compared to a control.
(e.g., if 8% of people who receive a placebo vaccine develop flu versus 2% of people who receive a flu vaccine, then ARR = 8% - 2% = 6% or .06)
What is number needed to treat?
How is it calculated?
Number of patients who need to be treated for one patient to benefit.
NNT = 1/ARR
ARR = absolute risk reduction
What is number needed to harm?
How is it calculated?
Number of patients who need to be exposed to a risk factor for one patient to be harmed.
Number needed to harm = 1/AR
AR = attributable risk
What is precision?
The consistency and reproducibility of a test (reliability)
The absence of random variation in a test.
Random error - reduces precision in a test.
Higher precision reduces a test’s standard deviation.
What is accuracy?
The trueness of test measurements (validity).
The absence of systematic error or bias in a test.
Systematic error reduces the accuracy of a test.
