BIO stats Flashcards

1
Q

Continuos data what are the types and the differences of each and examples/

A

Interval and Ratio

Ratio has a meaningful zero age, ht, wt

Interval no meaningful zero, celsius (equal distance between values)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Discrete data is also referred to as?

What are the different types and examples of them?

A

Nominal: Order is arbitrary (gender, ethnicity)

Ordinal ranked in logical order pain scale 0,10

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

The mean is preferred for what type of data?

A

continuos and normally distributed data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the mean preferred for?

A

preferred for ordinal or continuous data that is skewed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

The mode is preferred for what data?

A

nominal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Continuous data tends to follow what?

A

A normal distribution

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

68% fall within how many?

95%?

A

within 1 SD
within 2 SDs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

When is skewed data likely to occur?

A

If sample size is small and or there are outliers

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the best measure of central tendency when you have outliers?

How can the distortion of outliers be reduced?

A

Median is the best judge

Distortion can be reduced with increased sample size

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Right and left skew?

A

low values to the right positive right skew

left, negative left skew

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

If the alpha is 5% what does the p-value need to be to reject the null hypothesis?

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Confidence interval =?

A

CI=1- alpha

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

When are the results statistically significant?

Is there a difference with ratio data? (RR, OR, HR)

A

If the confidence interval doesnt include zero

Statistically significant if values CI does not contain 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is a type 1 error?

A

Rejecting the null when the null is true

saying there is a difference when there is none

p value correlated to probability of type 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Type 2?

A

Accepting the null when the null is false (b)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is power?

A

the probability that the test will reject the null hypothesis correctly

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

How is power calculated?

A

1-B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

why is absolute risk reduction more useful?

A

because it includes the reduction in risk and the incidence rate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

ARR of 12% in metoprolol vs placebo trial what does it mean?

A

means 12% out of every 100 pts benefited from tx

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

How do you round NNT?

NNH?

When do you use absolute value?

A

52.1 round to 53

NNH: 41.9 round down to 41

When calculating NNH use ARR absolute vlue

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What kind of studies use odd ratio instead of relative risk?

How do you calculate OR?

A

Case-controlled studies

OR= AD/BC

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

When are hazard ratios used?

A

survival analysis (analysis of death or disease progression)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

HR and OR interpretation?

A

If OR or HR = 1 the event rate is the same

> 1 event rate is higher in treatment group

<1 event rate is lower in tx group

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Notes on composite endpoints

A

must be similar in magnitude and have similar meaningful importance to the pt

Use the composite endpoint value instead of adding them all up

25
Q

What test should be used for continuous data?

A

normal distribution (parametric test)

Nonnomral (nonparametric)

26
Q

When is a T-test used?

when is a student t-test usd?

A

with continuous and normally distributed data

Student t-test

2 independent variables, medication and placebo

27
Q

When is an analysis of variance used? (ANOVA)

A

or F-test is used to when continuous data with 3 or more sample groups

28
Q

What test is used when nominal or ordinal data is used?

A

Chi-squared test

29
Q

What test is used for correlation?

What is it used for?

A

Spearmans ranked order correlation (RHO)

ordinal or ranked data

30
Q

What test is used for continuous data correlation?

A

Pearsons Correlation Coeffieicent

r which indicates the strength and direction of correlation (-1 to +1)

31
Q

When are regressions used?

What are the 3 types and what are they each used for?

A

described the relationship between a dependent variable and one or more independent variables

commonly used in observational studies

Linear: continuous data

Logical for categorical

Cox regression for categorical in survival analyisis

32
Q

If the test result is positive what is the likelihood of having the disease?

If the test is negative what is the likelihood of not having the disease?

A
33
Q

If a sensitivity is 100% what does that mean?

A

test will be positive in all patients with the condition

34
Q

100% specificy?

A

100% of negative patients will not have the disease

35
Q

Specificity is the percentage of?

A

True negative results

36
Q

Sensitivity and Specificity Formulas

A

Sensitivity: A/(A+C) x 100

Specificity: D/(B+D) x 100

37
Q

What two ways can data from clinical trials be analyzed?

A

Intention to treat and per protocol analysis

38
Q

NonInferiority and Equivalence trials

A

Equivalence new drug is as good as

Non-inferiority: new drug is not much worse

39
Q

When are forest plots used?

A

Composite endpoints in one study

or

meta-analysis

40
Q

Forest plots

A
  • Boxes show the effect estimate, in meta-analysis the bigger the box the more effect from the study is seen
  • Diamonds: represent pooled results from multiple studies
  • Horizonal line is the length of the confidence interval
  • Vertical line: Line of no effect, left illustrates significant benefit, to the right shows significant harm
41
Q

Case control study

A
  • Pts with disease (cases) to those without (control),
  • Retrospective
  • OR,
42
Q

Cohort Study

A

Compares outcomes of a group of pts exposed and not exposed

Groups are followed prospectively

43
Q

Cross-sectional survey

A
  • Estimates the relationship between variables and outcomes (prevalence) at one particular time (cross section) in a defined population
44
Q

Case report and case series

A
  • Describes an adverse reaction or a unique condition that appears in a single pt (case report) or a few pts (series)
  • No real conclusions can be drawn
45
Q

RCTs

A
  • Compared an experimental tx group to a control (placebo or existing tx) to determine which is better, subjects with the design characteristics (inclusion criteria)
46
Q

Benefits and Limitations of Cross over RCT

A
  • Pts serve as there own control this minimizes effects of confounders
  • A washout period between tx is needed to minimize influence of the first drug during second tx
47
Q

Factorial design

A

Randomized to more than the usual to

  • Evaluates multiple interventions in a single experiment
  • With every arm added the more subjects you need
48
Q

Meta-Analysis

A
  • Smaller studies can be pool instead of performing a large one
  • Studies may not be uniform, validity can be compromised if lower quality studies are weighted equally to higher studies
49
Q

Systemic Review article

A
  • Summary of clinical literature that targets something specific
  • Inexpensive
50
Q

Methods for pharmacoeconomic analysis? 4

A
  • Cost effectiveness analysis
  • Cost-minimization analysis
  • cost utility analysis
  • cost benefit analysis
51
Q

What do pharmacoeconomic studies serve to do

A

Guide optimal healthcare resource allocation

52
Q

What model is used to evaluate outcomes?

A

ECHO

  • Economic
  • Clinical
  • Humanistic Outcomes
53
Q

Incremental cost effectiveness ration calculation

A

C2-C1/ E2-E1

54
Q

When is cost-minimization analysis used?

A

when two or more interventions have shown equivalence in outcomes and the cost of the interventions are compared

55
Q

Cost Benefit Analysis

A

comparing benefits and costs of an intervention in terms of monetary units

56
Q

Advantage of Cost Effectiveness Analysis

A

outcomes are easier to quantify most common analysis

Input dollars output clinical (LDL level)

Disadvantage: unable to directly compare different types of outcomes (Diabetes program vs. Asthma program: cant do that)

57
Q

What is Cost Utility Analysis?

A

specialized form of CEA that include a quality of life component

using quality adjusted life years (QALY) and disability adjusted life years (DALYs)

About quality not quanitity

takes into account morbidity and mortality

58
Q
A