Epidemiology and Biostats Pt. 3

Question 1

Epi triad

Answer 1

agent
host
environment

sometimes vector is shown in the center of the triangle with lines radiating out to each of the three parts of the triad

Question 2

factors that affect the agent

Answer 2

type, presence or absence, pathogenicity, dose, chemical and physical causes of injury and disease

Question 3

factors that affect the host

Answer 3

exposure, susceptibility, age, sex, genetic composition, nutritional and immunologic status, anatomic structure, presence of disease or medications, psychological makeup, behaviors

Question 4

factors that affect the environment

Answer 4

geology and climate, biologic factors that transmit agent (vectors or fomites), socioeconomic factors (crowding, sanitation, etc)

Question 5

2 types of herd immunity

Answer 5

innate and acquired

Question 6

2 main mechanisms of disease transmission

Answer 6

1) direct (direct contact, droplet spread)
2) indirect (airborne - from droplets suspended in the air or dust and blown around, vehicle-born (food, water, blood, fomites), vector-borne(mosquitoes, fleas, ticks))

Question 7

examples of diseases spread by droplet spread

Answer 7

pertussis, meningococcal infection

Question 8

modifiable risk factors

Answer 8

tobacco use
alcohol abuse
unhealthy diet
physical inactivity

Question 9

non-modifiable risk factors

Answer 9

age
sex
heredity

Question 10

intermediate risk factors

Answer 10

increased BP, BG, overweight/obesity, etc.

Question 11

aims of levels of prevention

Answer 11

primordial: establish and maintain conditions that minimize hazards to health
Primary: reduce incidence of disease
Secondary: reduce prevalence of dz/shorten its duration
Tertiary: reduce number/impact of complications

Question 12

Koch’s postulates

Answer 12

1) organism must be in every case of dz
2) organism must be isolated from diseased organism and grown in pure culture
3) cultured organism should cause dz when introduced into a healthy organism
4) organisms must be reisolated from the inoculated, diseased experimental host and identified as being identical to the original specific causative agent

nowadays, not all diseases fit perfectly into this model (ie. prion dz)

Question 13

Hill’s causal criteria

Answer 13

There should be:
-strength (there is a strong association between cause and effect)
-consistency (repeatable results)
-specificity (cause leads to specific effect and not multiple ones)
-temporality (cause should precede effect)
-biological gradient (presence of unidirectional dose-response curve)
-plausibility
-coherence (studies must not contradict each other)
-experiment (experimental evidence is provided)
-analogy (same effect shown in animal studies prior to human studies)

Question 14

necessary vs. sufficient causes

Answer 14

necessary - a factor must happen for disease to occur (a critical ingredient)

sufficient - that factor alone is sufficient in causing disease (other factors are not needed), but it doesn’t necessarily have to come from that factor

Question 15

What are directed acyclic graphs

Answer 15

causal path diagrams that consist of nodes (variables) and arrows that indicate causal effect

Question 16

Effect Modification

Answer 16

association between exposure and disease is different for different levels of a third variable
-the effect of the exposure on the disease is modified by a third variable

Question 17

to be a confounder, a variable must be:

Answer 17

-associated with the outcome (ie. is a risk factor for the disease)
-associated with the exposure, but is not a result of the exposure (ie. is not on the casual pathway from exposure to outcome)

Question 18

variables can be classified into 1 of 4 types depending on the type of scale used to characterize their values

Answer 18

1) Nominal: values are categories (ie. country of residence)

2) Ordinal: values can be ranked but not necessarily evenly spaced (ie. stages of cancer)

3) Interval: variable is measured on a scale with equally spaced units, but without a true zero (ie. date of birth), a continuous variable

4) Ratio: interval variable with a true zero point (ie. height in cm), a continuous variable

Question 19

A frequency distribution displays:

Answer 19

values a variable can take and the number of persons or records with each value

Question 20

categorical variables can be summarized as ratios, proportions, and rates

Answer 20

T

Question 21

Central location aka central tendency

Answer 21

the clustering of a distrubution at a particular value (ie. the top of the bell curve)

Question 22

Range

Answer 22

the difference between its largest (maximum) value and smallest (minimum) value

Question 23

Spread

Answer 23

the distribution out from a central value; how wide the bell curve is

Question 24

how to calculate standard deviation

Answer 24

1) calculate the mean
2) subtract the mean from each observation, then square the difference
3) sum the squared differences
4) divide the sum of squared differences by (n-1) where n = the total number of values. This is the variance.
5) take the square root of the variance. This is the standard deviation.

Question 25

Steps for general hypothesis testing

Answer 25

1) specify the null and alternative hypotheses
2) specify the type 1 error level
3) calculate a test statistic under the assumption that the null hypothesis is true. (The test statistic is a random variable number reflecting how well the data conforms to the null hypothesis)
4) compare the test statistic to what is expected if the null hypothesis were true (compute the p-value)

Question 26

How to decide what statistical test to use

Answer 26

1) decide what scale of measurement your data is (nominal, ordinal, interval, etc.)
2) consider the type of analysis required (comparison of independent or paired groups?)

Question 27

parametric vs. non-parametric tests

Answer 27

Parametric tests assume a normal distribution in the underlying population

non-parametric tests rely on no or few assumptions about the population distribution.
-Has less statistical power with the same sample size compared to parametric tests

Question 28

correlation coefficient

Answer 28

represents the strength of a linear relationship between two continuous variables. Ranges from -1 to 1. Does NOT prove causation.

0 = variables are uncorrelated
positive = variables positively correlated
negative = variables negatively correlated

Question 29

Modeling

Answer 29

the representation of physical processes, designed to increase appreciation and understanding of them

Types:
density model
prevalence model
stochastic model
empirical simulation modeling
Monte Carlo simulation modeling

Question 30

Density model

Answer 30

considers the absolute number of infectious agents in each host

Question 31

Prevalence model

Answer 31

considers the presence/absence of infection in various host cohorts

Question 32

Deterministic model

Answer 32

Fixed input parameters do not allow for random variation

Question 33

stochastic model

Answer 33

model that allows for random variation

Question 34

empirical simulation modeling

Answer 34

simulation of performance of dz in relation to changing parameters

Question 35

Monte Carlo simulation modeling

Answer 35

repeated simulations using random numbers to decide whether or not an event takes place