Biostats

Question 1

_______ bias is the bias that is associated with researchers choosing who is included in the study. This can lead to no external or internal validity

Answer 1

Selection

Question 2

To increase external validity, you will want to do what in selecting your participants?

Answer 2

Get a sample population that is representative of the target population; where they live, age, race, socioeconomics

Question 3

______ bias is a type of selection bias in which some inidividuals in a community may not be selected: selecting via addresses; some people do not have a permanent address

Answer 3

Sampling

Question 4

_______ bias can be corrected with incentives for filling out surveys

Answer 4

non response bias

Question 5

The _______ _______ occurs when a study recruits the exposed group from a general occupational cohort and then the unexposed group from the general population

Answer 5

healthy worker effect: most of the workers are healthy and able to go to work

Smoking group (occupational) will be of good health where the people who are not working could have various health levels

Question 6

_______ bias is when individuals in two groups are associated with the characteristic of the exposure

Two groups are chosen from the hospital with and without lung cancer and compare how many people smoke in each group in hopes of finding how much smoking can cause lung cancer….but smoking could cause other disease too in the group of controls (Heart disease, diabetes, etc) and some of those patients could be admitted to the hospital because of those diseases that are also associated iwth a higher risk from smoking

Answer 6

Berkson’s bias…. can overcome by having a control group with an outcome (say a broken bone) that does not correlate with the exposure (smoking)

Question 7

_______ can skew the results by underestimating the results of the exposure on the overall outcome

Answer 7

Loss to follow up: you now have less people who have positive disease at the end bc they may have died, moved, got bored, etc

Question 8

_______ bias is common in case control studies and can lead to an over or underestimation of the exposure because there is a difference in the accuracy and completeness of the data

Answer 8

Recall bias

Question 9

Individuals with skin cancer might be more likely to report tanning bed use and may over estimate whereas people without skin cancer may under-report because the visits can blur and become forgotten…. what bias is this?

Answer 9

Recall bias

Question 10

how can you decrease recall bias?

Answer 10

Written records

Study exposures in recent pass

Question 11

_______ bias is when researchers use inadequate methods to collect data: imprecise, self reported, etc

Answer 11

Measurement bias

Question 12

Hoe can you avoid measurement bias?

Answer 12

Carefully research procedures

Regular calibration

Question 13

________ _______ is when people change their behavior because they know that they are being studied; increased adherence, etc

Answer 13

Hawthorne effect

Question 14

How can you avoid the Hawthorne effect?

Answer 14

secrete observation

Question 15

_______ bias is when groups are treated differently;

Cavities in people who floss and who do not; dentists might do a more thorough exam on people who do not floss because they will expect cavities

Answer 15

Procedure bias

Question 16

_______ _______ bias occurs when the people conducting the experiment expect an outcome and therefore round up, etc

Answer 16

Observer expectancy bias

Question 17

How can you overcome an observer expectancy bias?

Answer 17

Double blinding

Question 18

_______ bias can occur when there is an independent factor that is associated with the exposure and the outcome

Answer 18

Confounding

Obesity and colorectal cancer—- red meat and processed meat; obese poeple may consume more

Question 19

How can you eliminate confounders?

Answer 19

Multiple repeated studies

Crossover studies: participants are their own controls

Randomization

Matching participants

Question 20

_______ _______ are when participates are stratified with a specific factor of an outcome

Obesity and colorectal cancer in MALES OR FEMALES; the sex becomes the ______ _______

Answer 20

Effect modifier

Question 21

_______ ________ bias occurs when the disease is diagnosed early and seems to survive longer (just because they were diagnosed early)

Answer 21

Lead time bias

Early detection can be confused with increased survival

Question 22

How can you overcome lead time bias?

Answer 22

Back end survival

Question 23

______ ______ bias where slowly progressive diseases are more prone to be detected compared to aggressive diseases

Answer 23

Length time bias