Biostats

Question 1

What is sensitivity?

Answer 1

(TP/All the people with disease) = TP/(TP+FN)

Question 2

What is specificity?

Answer 2

(TN/All the well people) = TN/(TN+FP)

Question 2

1. What is the Positive Predictive Value?

What is its relation to Prevalence?

What is its relation to Specificity?

Answer 2

1. The proportion of people who test positive who actually have the disease.
   TP/ All the people with a positive test result =
   TP/(TP +FP.)
2. As Prevalence increases the PPV increases and as Prevalence decreases the PPV decreases.
3. The more specific a test is a catching all the people without the disease the higher the PPV.

Question 3

1. What is Negative Predictive Value? 2.

Answer 3

1. The proportion of people who test negative who actually are healthy/dont have the disease.

TN/ All the people with a negative test results

TN/(TN+FN.)

Question 4

This study is a brief objective report of clinical characteristic outcome from a single clinical subject or event.

Answer 4

Case report (One man with TB)

Question 5

This study takes a sample of patients from one point in time to help determine the prevalence.

Answer 5

Cross-sectional study (looks at one point in time)

Question 6

This involves many subjects/patients that already have the disease and this allows us to study the history of the disease and its manifestations.

Answer 6

Case Series Report (Many patients with TB)

Question 7

This study identifies a group of people with the disease and compares them with a comparison group without the disease.

Answer 7

Case-Control study. (Always retrospective) Can go back in time and identify the risk factors and determine these Odds ratio (looks at the increased odds of getting the disease with the exposure vs getting the disease without the exposure.

Question 8

This study shows a population group is idenitified who has been exposed to risk factor is followed over time and compared with a group not exposed to the risk factor group. Outcome is disease incidence in each group.

Answer 8

Cohort Study. Watch the group as they develop the disease. (Prospective study.)

Question 9

“How much more likely is the exposed person going to get the disease compared to the nonexposed. This is?

Answer 9

Relative risk. Incidence rate of exposed / the incidence rate of nonexposed group. Used in Cohort study where you can get incidence.

Question 10

Name the experiemental study: Experiment in which the unit of allocation to receive a preventive or therapeutic regimen is an entire community or political subdivision.

Answer 10

Community Trial.

Question 11

Name the experiemental study: For ethical reasons, no group involved can remain untreated. Al subjects receive intervention but at different times.

Answer 11

Crossover study.

Question 13

Name the bias: Sample not representative

Answer 13

Selection bias. Need to random, independent samples.

Question 13

Name the bias: Gathering information distorts it

Answer 13

Measurement bias. Measuring patients satisfaction with their respective physcians by using leading questions ie “You dont like your doctor do you?” Associated with Hawthrone effect (Behavior is altered because they are being studied.) Need a control goup/placebos group.

Question 14

Name the bias: Researcher’s beliefs affect outcome

Answer 14

Experimenter expectancy bias. Need to do double blind designs.

Question 15

Name the bias: Early detection confused with increased survival

Answer 15

Lead time bias. This is associated with benefits of screening ie early breast cancer detection prevents development of cancer. Need to measure the “back end: survival.

Question 16

Name the bias: Subjects cannot remember accurately

Answer 16

Recall bias. This associated with retrospective studies when you have to go back in time a remember a risk. Need to confirm information with others to avoid this.

Question 17

Name the bias: Unanticipated factors

Answer 17

Confounding obscure results bias. Associated with Hidden factors affecting results. Need to do multiple studies and have a good research design.

Question 19

Absolute Risk

Answer 19

The absolute difference in rates

Experimental group events - Control group events

ARR = EER-CEF

Question 20

Relative Risk

Answer 20

RRR = (EER - CER)/ CER

Question 21

Odds ratio

Answer 21

AD/BC

Question 23

1. Number Needed to Treat

2. Number Needed to Harm

Answer 23

1. Number of patient needed to receive a treatment for one additional patient to benefit

NNT = 1/ARR

2. Number of patients needed to receive a treatment for one additional patient to be harmed

NNH = 1/ARR

Question 24

Failure to reject null hypothesis when it actually is false?

Answer 24

Type 2 II Error

Question 25

Rejecting the null hypothesis when it actually is true?

Answer 25

Type I Error

Question 26

The probability of an event occurring in a group during a specified time period?

Answer 26

AR= Patients with Event in group/ Total Patients in group

Question 27

What does it mean when the confidence interval is 1?

Answer 27

No risk or benefit exist.

Question 28

1. LR +

2. LR

Answer 28

The measure of the odds of having a disease independent of the disease prevalence.

LR + = Sensitivity/ (1-Specificity)

LR - = (1- Sensitivity)/ Specificity

Question 29

The ratio of the probability of developing a disease with ar risk factor present to the probability of developing the disease without the risk factor present?

Answer 29

RR = EER/ CER