EBM

Question 1

Type I Error

Answer 1

FP; we reject H_o when we shouldn’t (if H_o is not false)

There’s no relationship but you found one.

Smoking does not cause cancer but you found that it did.

Question 2

Type II Error

Answer 2

FN; we fail to reject H_o when we should

There’s a relationship there but you didn’t find one.

Smoking does cause cancer but you don’t find more cancer in smokers.

Question 3

Alpha

Answer 3

Willingness to reject the null hypothesis when we shouldn’t (type I error); willingness to be wrong.

There really is no association but our results show an association just by chance.

Question 4

Beta

Answer 4

willingness to make type II error (failing to reject false H_o)

Question 5

H_o

Answer 5

Null hypothesis (no association)

Question 6

H_A

Answer 6

Alternative hypothesis

Question 7

P_alpha

Answer 7

probability of a false positive study

if Pa=0.05, there is a 5% chance we will report a difference that occured by chance alone

Question 8

P_beta

Answer 8

probability of a false negative

if P_beta= 0.10, then 10% chance we will fail to detect a difference that is real

Question 9

Power

Answer 9

ability to detect or verify a difference that is real

Power = 1 - P_beta (or 1-beta)

Ability to avoid a type II error

If you reject Ho, then by definition you cannot lack power (even if n is small)

Question 10

What to consider when H_o is not rejected

Answer 10

• likely that Ho is true
• suspect type II error (P >.05 but <.10)
• consider that the study was underpowered
• small n, esp if P is close to 0.05

Question 11

What to consider when Ho is rejected

Answer 11

• consider possability of type I error, esp if P is close to 0.05
• if P<.01, be confident it is not a type I error

Question 12

T-test

Answer 12

• compares the difference between 2 means
• divided by the variability in the two samples
• parametric: relies on parameters of the distribution (assumes equal variances)

Question 13

t=

Answer 13

meanA - meanB / (varA + varB)^.5

Question 14

df

Answer 14

df= n_A-1) + (n_B-1)

Question 15

critical value

Answer 15

• using alpha and df to determine
• if t is calculated as les than critical value, we fail to reject Ho
• ie: alpha=0.05, df=8, critical value is 2.30
  
  • if t<2.30, we fail to reject Ho

Question 16

Mann Whitney U test

Answer 16

• non-parametric
• distribution free
• ignores the mean and the median

Question 17

Relative Risk

Answer 17

a/(a+b)

_______________

c/ (c+d)

Question 18

Differential (regarding error)

Answer 18

misclassifies subjects differently depending on exposure and/or outcome - biases result in one direction or the other

Question 19

Nondifferential

Answer 19

tends to bias results toward the null

can be systematic or random error

Question 20

Confounding Variable

Answer 20

Must be associated with independent variable and the dependent varible; cannot be in the causal pathway btwn exposure and disease

Question 21

Which designs would you use to answer the fundamental questions of:

1. prevalence?
2. risk of harm?
3. treatment or prevention?
4. prognosis?
5. screening?

Answer 21

1. prevalence? cross sectional
2. risk of harm? cohort, case-control
3. treatment or prevention? RCT, Cohort, Case-control
4. prognosis? Cohort
5. screening? RCT, Cohort, Case-control

Question 22

Disease Prevalence

Answer 22

• point prevalence: instantaneous time period
• period prevalence: longer time period

Prevalence =

(# of ppl with diease at a given pt in time)/
(total # of ppl in pop)

• often called prevalence rate but time not in denominator

Question 23

Disease Incidence

Answer 23

• how quickly ppl are being diagnosed with the disease
• considers only new cases

cumulatie incidence =

(# of new cases) / (# of ppl at risk of developing the disease over a defined time)

Question 24

Incidence Rate

Answer 24

rate at which new diesease has occurd in the pop at risk per some time unit

Question 25

Relative Risk (RR)

Answer 25

(Incidence_exposed) / (Incidence_unexposed)

risk ratio

Question 26

Cohort Studies

Answer 26

Criteria:

1. They do not have diease in question at time of assmbly
2. Should be observed over a meaningful pd of time in the natural history of the diesease in question.
3. Members of cohort should be observed over the full period of the followup.

• Upon entry into the study, ppl in the cohort are classified according to characteristics (possible risk factors) that might be related to outcome.
• Classified as exposed or not-exposed
• aka: longitudinal, prospective/historical/retrospective, incidence studies

Question 27

Case-Cohort Studies

Answer 27

• all exposed people are included in the study and followed for some outcome of interest.
• however, only a small random sample of unexposed ppl is studied (group of unexposed is “enriched” with those who subsequently suffer the outcomes of interest)–results then adjusted to reflect the sampling fractions used to obtain the sample
• need large, computerized, medical database