Lecture 12 - Statistics And Probabilities

Question 1

the science of epidemiology rests on ______

Answer 1

statistics

Question 2

ALL of public health relies on _______ to provide and interpret data

Answer 2

statistics

Question 3

the term “statistics” refers to both…….

Answer 3

-the numbers that describe the health of the populations
-The science that helps to interpret those numbers

Question 4

the science of statistics is a set of ____ and ____ used to ____ ____ in order to ____ _____

Answer 4

the science of statistics is a set of CONCEPTS and METHODS used to ANALYZE DATA in order to EXTRACT INFORMATION

Question 5

______ results from epidemiological studies are common

Answer 5

contradictory

Question 6

What are 2 sources of error?

Answer 6

-Biasing (unintentionally guiding the results in one direction)

-confounding variables (very difficult to control all variables)

Question 7

Give an example that demonstrates that people sometimes demand certainty about their public health even when science cannot prove it

Answer 7

1997 - mammography screening - NCI said that each woman should make the decision individually with their dr to be screened or not based on medical and family history
-public response was heated

in the end, NCI recommended that women in their 40s be screened —pressure from politicians who were eager to get credit for supporting women’s health led to a misguided sense of SCIENTIFIC CERTAINTY

Question 8

politics vs science
When there is uncertainty in epidemiology…. _____ may prevail. Why?

Answer 8

politics
politicians pressure scientists to change their opinion (ie: the breast cancer screening)

Question 9

What is the issue with politicians convincing scientists to change their mind on things?

Answer 9

the public finds it frustrating to hear contradictory information when they’re genuinely concerned about protecting their health

Question 10

Science always involves _____

Answer 10

uncertainty

Question 11

How do scientists quantify uncertainty?

Answer 11

by measuring probabilities (requires statistics)

Question 12

think of “probability” as….

Answer 12

the likelihood that an event of this type will occur based on what has happened in the past

Question 13

______ said that the _____ is what usually happens. BUT statisticians know that the ______ happens more than most people think

Answer 13

ARISTOTLE said that the PROBABLE is what usually happens. BUT statisticians know that the IMPROBABLE happens more than most people think

Question 14

What is the P value?

Answer 14

the likelihood that the difference between control and experiment is due to variability.

or the degree of probability or improbability of a certain result in an experiment

Question 15

the p value expresses the probability that…..

Answer 15

the observed result could have occurred by chance alone

Question 16

TYPICALLY, a p value of ____ or less is acceptable

Answer 16

0.05

Question 17

what does a p value of 0.05 mean?

Answer 17

if an experiment were repeated 100 times, the same answer would result 95 of those times, while 5 times would yield a different answer

Question 18

Give a reason that an acceptably low p value could lead to an erroneous conclusion

Answer 18

-bias
-confounding variables

Question 19

bias and confounding variables are ____ errors

Answer 19

systemic (innate)

Question 20

Give an example of a specific case in which the p value was statistically significant (0.001) but the results were erroneus

Answer 20

the study that found a link between pancreatic cancer and coffee drinking.

the conclusion is wrong NOT due to random chance, but because the cancer was caused by smoking rather than coffee drinking

Question 21

What is the reason that large numbers of cancer treatments could be in clinical use that are not effective?

Answer 21

p value of 0.05 indicates that 5 out of every 100 ineffective treatments would appear to be effective, but may actually represent errors caused purely by chance

Question 22

The fact that the probable is not always what happens leads to the law of _______

Answer 22

law of small probabilities

Question 23

What does the Law of Small Probabilities state?

Answer 23

the most improbable things are bound to happen occasionally

Question 24

Due to the law of small probabilities, a few people with apparently fatal diseases will inexplicably recover and may be convinced that their recovery was the result of……

Answer 24

something that they did, giving rise to a new quack therapy. Because their recovery was merely a random deviation, other patients will not get the same benefit

Question 25

Another consequence of the law of small probabilities is the phenomenon of....

Answer 25

cancer clusters

Question 26

Explain how cancer clusters are a consequence of the law of small probabilities

Answer 26

every now and then a community will discover that it is the site of some kind of cancer and everyone will be alarmed -not the result of an environmental agent, but is the result of chance variation

Question 27

If a cancer cluster is very large, however.....

Answer 27

it is likely NOT random variation

Question 28

a large number of cases is said to confer _____ on a study

Answer 28

power

Question 29

power increases the probability of.....

Answer 29

finding an effect (if there is one)

Question 30

___ and ____ can be present in even the largest of studies

Answer 30

bias and confounding variables

Question 31

Studies with low power are likely to produce......

Answer 31

false negative results

Question 32

false positive results occur when...

Answer 32

the study finds an effect that is not real (when random variation appears to be a true effect)

Question 33

In the mission of public health to prevent disease and disability, _________ and ______ plays an important role

Answer 33

secondary prevention (early detection) and treatment plays an important role

Question 34

When the causes of a disease are not well known, little is known about _______ (ex: breast cancer)

Answer 34

primary prevention

Question 35

When little is known about primary prevention of a disease, what is the next best public health measure?

Answer 35

to screen the population at risk so as to detect the disease early

Question 36

Give an example of 2 populations involved in screening to detect disease early

Answer 36

identifying individuals with HIV and counseling them to prevent spreading to others newborns are routinely screened for certain congenital diseases

Question 37

Screening tests may be highly _______ resulting in....

Answer 37

highly sensitive, resulting few false negatives

Question 38

Most lab tests should be highly accurate, but most are likely to yield ____ or ____ to some degree

Answer 38

false positives or false negatives

Question 39

for most public health screening programs, _______ are desirable to avoid missing an individual with a serious disease who could be helped by some intervention

Answer 39

sensitive tests

Question 40

Sensitive, inexpensive tests are chosen to encourage testing of as many at-risk individuals as possible are often not very______

Answer 40

specific

Question 41

When a positive result is found using a non specific test, what happens next?

Answer 41

more specific tests are conducted to confirm the initial finding

Question 42

if a SENSITIVE mammogram finds a suspicious spot in a woman's breast, the test is usually followed by....

Answer 42

a biopsy to determine whether or not the spot is cancerous

Question 43

true positives in the cases of rare diseases are _____.

Answer 43

rare

Question 44

in the case of a rare disease, the rate of false positives may be _____ than the number of true positives when sensitivity tests are used

Answer 44

HIGHER

Question 45

give a specific case in which false positives were a big issue

Answer 45

when louisiana mandated premarital screening for HIV

Question 46

Describe what a sensitivity test is

Answer 46

used INITIALLY to screen people for disease. Results in many false positives.

Question 47

What is the goal of sensitivity tests?

Answer 47

to make sure that everyone that actually has the disease is identified. (ensure that there are NO false negatives)

Question 48

What are specific tests?

Answer 48

they are the follow up test after testing positive using a sensitivity test. more expensive, complicated, and can be invasive. If you were a FALSE POSITIVE, this test will most likely show that you dont have the disease

Question 49

what is a rate?

Answer 49

percentage of occurrence in a population

Question 50

epidemiology makes extensive use of rates to....

Answer 50

study disease distribution and determinants

Question 51

Raw data needs to be converted to ____ to be valuable

Answer 51

rates

Question 52

birth rate (like the death rate) is defined as.... while the fertility rate is defined as....

Answer 52

the number of live births per 1000 people fertility rate = number of live births per 1000 women AGES 15-44 Both rates start with the same raw number (number of live births) but use a DIFFERENT POPULATION FOR REFERENCE

Question 53

What are 2 rates commonly used as indicators of a community's health?

Answer 53

infant mortality rate maternal mortality rate

Question 54

what is the infant mortality rate?

Answer 54

the number of infants that dies before their first birthday in a year, divided by the number of live births in that year

Question 55

what is the maternal mortality rate?

Answer 55

the number of deaths among women associated with pregnancy divided by the number of live births in that year

Question 56

For some purposes, the numbers can be made more useful by converting crude rates to adjusted rates. give an example

Answer 56

crude mortality rate in florida is much higher than the crude mortality rate in alaska ----no cause for alarm: the average age of a floridian is higher than the alaskan after adjusting the mortality rate for age differences, the mortality rates are similar

Question 57

rates can be adjusted based on ___, ____, ____, and -___

Answer 57

age, gender, race, and ethnicity

Question 58

Rates can be calculated on a _____ specific basis

Answer 58

group

Question 59

who has a higher mortality rate - males or females??

Answer 59

males

Question 60

What is YPLL

Answer 60

years of potential life lost

Question 61

YPLL gives greater weight to the deaths of.....

Answer 61

young people

Question 62

What are leading causes of YPLL?

Answer 62

Accidental injuries, infectious disease, and suicide because they are likely to kill young people

Question 63

The formal process of risk assessment.....

Answer 63

identifies events and exposures that may be harmful to humans and estimates the probabilities of occurrence as well as the extent of the harm they may cause

Question 64

assessing risk is often done on the basis of......

Answer 64

historical data

Question 65

assessing risk is often done on the basis of historical data. give an example

Answer 65

can predict that the # of motor vehicle crashes next year will be similar to the number this year

Question 66

To estimate the probability of a mishap in new technology, various possible chains of events are considered, and the.....

Answer 66

risk for something going wrong is estimated in each step

Question 67

risks of the individual steps are then added or multiplied to obtain a risk for the whole. give a specific example of when this was done

Answer 67

when nuclear powerplants were first introduced - helped engineers identify what kind of safety devices should be incorporated to reduce the probability of failure

Question 68

According to an analysis published in 2000, experts said that the most risky activities and technologies were:

Answer 68

-motor vehicles -smoking -alcoholic beverages -handguns -surgery

Question 69

however, when the general public was asked about their perception of risk, what was at the top of their list?

Answer 69

nucelar power

Question 70

people are more likely to accept a risk from an activity when....

Answer 70

it is perceived to be beneficial

Question 71

in terms of the public, what are the 3 ways in which they assess risk?

Answer 71

-familiar risks are more acceptable than unfamiliar risks -risks that people perceive they have control over are more acceptable than those that are uncontrollable -a risk with potentially catastrophic consequences is unacceptable, even if it is highly unlikely to occur

Question 72

Risk perception researchers classify risks on 2 scales:

Answer 72

dreadness knowability

Question 73

in terms of automobiles,

Answer 73

driving is one of the riskiest activities, yet it does not arouse anxiety because it is neither dreaded nor unknown. people also perceive that they have control when they're driving - benefit is obvious

Question 74

Why is a nuclear reactor perceived as a great risk by the public while driving is not?

Answer 74

highly dreaded, lack of control, benefit is unclear

Question 75

When did the public's perception about nuclear power gain credibility?

Answer 75

after the 1979 accident at the 3 mile island nuclear reactor in pennsylvania

Question 76

What is the difficulty with a cost-benefit analysis when it comes to public health?

Answer 76

determining what monetary value to place on a life saved

Question 77

What is a cost benefit analysis? What is a cost effectiveness analysis?

Answer 77

cost benefit analysis = weighs the estimated cost of implementing a policy against the estimated benefit, usually in monetary terms cost effectiveness analysis = compares the efficiency of different methods of attaining the same objective

Question 78

Should cost-benefit analysis and cost-effectiveness analysis serve as the sole determinant of a health care decision?

Answer 78

no - but it can improve decision making