Statistics

Question 1

What is another name for Type 1 error and Type 2 error?

Answer 1

Type 1 = alpha

Type 2 = beta

Question 2

Draw out the Bayesian analysis tale that demonstrates the relationship between alpha, beta, and the null hypothesis

Answer 2

Question 3

What statistical tool is used to determine responsiveness?

Answer 3

Receiver Operating Characterisitc (ROC) curve.

Question 4

What statistical tool is used to evlauate the presence of publication bias?

Answer 4

Funnel plot

Question 5

What statistical tool is used to determine the survivorship of a population?

Answer 5

Kaplan-Meier curve

Question 6

What is a forest plot utilized for?

Answer 6

To graphically depict a meta-analysis of the results of randomized controlled trials.

Question 7

What is the Cronbach alpha coefficient used for?

Answer 7

To measure internal consistency and indicates how well individual items in a test or questionnaire are correlated.

Question 8

Describe Type I and Type II errors?

Answer 8

Question 9

What are the measures of central tendency in statistics?

Answer 9

MODE- value that occurs most often. Best for data which is allocated into distinct categories nominal data.

MEDIAN- Value that occurs at the middle of all values. Not affected by extreme values. Goot for all levels of measurement except nominal data.

MEAN- the arithmetic average. Uses all values of data. Highly sensitive to extrem values especially skewed distribution.

Question 10

Describe sensitivity?

Answer 10

probability that test results will be positive in patients with disease.

Question 11

Describe specificity?

Answer 11

probability test result will be negative in patients without disease.

Question 12

How do you calculate postive predictive value?

Answer 12

Question 13

How do you calculate negative predictive value?

Answer 13

Question 14

What is the likelihood ration in statistics?

Answer 14

Likelihood that a given test result would be expected in a patient with the target disorder compared to the likelihood that the same result would be expected in a patient without the target disorder.

POSITIVE LIKELIHOOD RATIO = sensitivity/ (1-specificty)

NEGATIVE LIKELIHOOD RATIO: (1-sensitivity)/specificity

Question 15

What is relative risk?

How do you calculate it?

What type of studies is it usually obtained from?

Answer 15

Question 16

What is definition of an odds ratio?

How is it calculated?

What studies is it usually obtained from?

Answer 16

Question 17

What is the number needed to treat?

How do you calculate it?

Answer 17

number of patients that must be treated in order to achieve on additional favorable outcome.

Numer need to treat = (1/absolute risk reduction)

Question 18

What is the definition of power in statistics and how is it calculated?

Answer 18

An estimate of the probability a study will be able to detect a true effect of the intervention.

A power analysis to determine sample size should be performed prior to initiation of the study.

Equation: Power = 1 - Type-II error

Question 19

What is effect size?

Answer 19

Magnitude of the difference in the means of the control and experimental groups in a study with respect to the pooled stnadard deviation.

Question 20

What is variance in statistics?

Answer 20

estimate of the variability of each individual data point from the mean.

Question 21

What is the definition of the confidence internal?

Answer 21

The interval that will include a specific paramter of interest, if the experiment is repeated.

Usually set at 95% by convention.

Question 22

Define a Tyupe II Error (beta)

Answer 22

Question 23

Define a Type I Error (alpha)

Answer 23

Question 24

What is the minimal clinically important difference in statistics?

Answer 24

The difference in outcome measures that will have clinical relevance.

Difficult to study and measure, very few outcome tools have estabilished and universally accepted MCID.

Helps to reconcile the statistical significance and clinical relevance of study results that use outcome tools.

Question 25

What is the Kaplan-Meier method used to analyze?

Answer 25

used to analyze survivorship data to determine things like implant survivial in joint replacement. Other methods for survivorship analysis: **Life table method-** annual success rate, determined from the failure rate, is cumulated to five a survival rate for each successive year, this can change only once per year. **Product limit method-** same as life table method, but the survival rate is recalculated each time a failure occurs.

Question 26

What is the definition of a receiver operator curve? How is it interpreted?

Answer 26

A graphical representation of the diagnostic ability of different tests. Used to determine responsiveness. X-AXIS- False positive rate Y-AXIS- Ture positive rate Area under the curve (called C-statistic) determines diagnostic ability. The higher the better. \< .5 and the test is considered a useless test.

Question 27

What is the definition and clinical significance of a funnel plot?

Answer 27

DEFINTION: Simple scatter plot of the intervention effect estimates from individual studies against some measure of each study's size or precision and is used to detect publication bias in meta-analyses CLINICAL SIGNIFICANCE: This method is based on the fact that larger studies have smaller variability, whereas small studies, which are more numberous have larger variability. Thus the plot of a sample of studies publication bias will produce a symmetrical, inverted-funnel-shaped scatter, whereas a biased sample will result in a skewed plot.

Question 28

What is statistical inference? What are the different classifications?

Answer 28

Used to test specific hypotheses about associations or differences among groups of subjects/sample data. PARAMETRIC INFERENTIAL STATISTICS- continuous data that is normally distributed NONPARAMETRIC INFERENTIAL STATISTICS- continuous data that is not normally distributed (skewed). categorical data. image is example of non-parametric.

Question 29

What are the different study types for statistical inference? What tests would you use for parametric and non-parametric data for the following studies? When comparing two means? When comparing proprotions? When comparing three or more groups?

Answer 29

Question 30

What are the five steps of evidence based medicine?

Answer 30

Formulate an answerable question Gather the evidence Appraise the evidence Implement the evidence Evaluate the process

Question 31

What are the different levels of evidence? Give an example of a study type for each level of evidence?

Answer 31

Question 32

What are the different Randomized controlled trial desings?

Answer 32

**Parallel-group**- each participant is randomly assigned to a group, and all the participants in the group receive (or do not receive) an intervention. **Crossover**- Administration of two or more therapies, one after the other in a random order. Susceptible to bias if washout period is inadequate. Wahsout period is the time between therapies. **Factorial**- each participant is randoly assigned to a group that recives a particular combination of interventions or non-interventions. **Cluster**- pre-existing groups of participants (e.g. villages, schools) are randomly selected to receive (or not recieve an intervention. **Single blinded study vs double blinded study**- patients are blinded vs patients and providers both blinded.

Question 33

Describe a cohort study?

Answer 33

Question 34

Describe a case-control study?

Answer 34

Question 35

Describe a cross-sectional study?

Answer 35

Question 36

Describe a case series study?

Answer 36

Question 37

Briefly describe the SF-36 survery?

Answer 37

A generic, multi-purpose, short-form health survey consisting of 36 questions. useful for surveys of general and specfic populations, comparing the relative burden of diseases, and differentiating the health beneftis produced by a wide range of different treatmetns. Consists of 8 scaled scores, which are the weighted sums of the questions in their section. Each scale is directly transformed into a 0-100 calse on the assumption that each questions carries equal weight.