Biostats and Epidemiology

Question 1

Sensitivity

Answer 1

Proportion of all people with a disease who test positive for the disease; the probability that a test detects disease when disease is present

Value approaching 100% is good for ruling OUT disease and indicates a low false-negative rate; high sensitivity is good for screening in diseases with low prevalence

Sensitivity = TP / (TP + FN)

Question 2

Specificity

Answer 2

Proportion of all people without diseas who test negative; probability that a test indicates non-disease when disease is absent

Value approaching 100% is desirable for ruling IN disease and indicates a low false-positive rate

Tests with high specificity used for confirmation after a positive screening test

Specificity = TN / (TN + FP)

Question 3

Positive predictive value

Answer 3

Proportion of positive test results that are true positives; probability that a person actually has the disease given a positive test result

PPV = TP / (TP + FP)

Varies directly with prevalence; high pre-test probability = high PPV

Question 4

Negative predictive value

Answer 4

Proportion of negative test results that are true negative; probability that person actually is disease free given a negative test result

NPV = TN / (FN + TN)

NPV varies inversely with prevalence; high pre-test probability = low NPV

Question 5

Primary Disease Prevention

Answer 5

Aimed at preventing disease occurrence

Examples include: Vaccination, condom distribution

Question 6

Secondary Disease Prevention

Answer 6

Aimed at early detection in patients with disease in an effort to reduce disease-related morbidity and mortality

Ex: Pap smear

Question 7

Tertiary Disease Prevention

Answer 7

Deals with the reduction of disease burden or disability

Ex: Insulin administration in diabetes

Question 8

Quaternary disease prevention

Answer 8

Any action taken to identify patients at risk of over-medicalization, to protect patients from new medical interventions, and to suggest ethical interventions

I.e. Refusal to administer antibiotics for virus-mediated diseases

Question 9

Type I Error

Answer 9

To mistakenly accept the experimental hypothesis and reject the null hypothesis; i.e. to state that there is an effect or difference when none exists

Question 10

P value

Answer 10

The probability of making a Type I error; i.e. the probability of observing an experimentally derived difference when, in fact, no difference exists

Question 11

alpha

Answer 11

The probability of making a type I error (i.e. stating that there is an effect or difference when none exists)

Pre-set to a certain level of significance, usually 5% (.05)

Question 12

Type II Error

Answer 12

Stating that there is not an effect or difference when, in fact, one exists; i.e. to incorrectly reject the null hypothesis

Question 13

Beta

Answer 13

The probability of making a type II error (i.e. to falsely reject the null hypothesis)

Question 14

Statistical Power

Answer 14

The probability of rejecting the null hypothesis when it is, in fact, false

Calculated by 1 - B

Increased by larger sample size, increased expected effect size, increased precision of measurement

Question 15

Case-control study

Answer 15

Compares a group of people with disease to a group disease to compare prior exposures or risk factors

Develops an Odds Ratio for a condition - i.e. patients with X disease have a higher odds of having a history of Y risk factor than patients without the disease

Question 16

Cohort Study - Retrospective

Answer 16

Compares a group with a given risk factor / exposure to a group without that exposure to see if exposure increases the likelihood of disease

Can be prospective (“who will develop the disease?” or retrospective (“who developed the disease - exposed or non-exposed?”)

Generates relative risk - i.e. “smokers had a higher risk of developing COPD than non-smokers”