Biostats 2 Flashcards by Cal Schiller

Internal validity

extent to which a piece of evidence supports a claim about cause and effect, within the context of a particular study

That is, cause precedes the effect and they happen together, with little possibility for confounders – within the population

How well did you know this?

Not at all

Perfectly

External validity

applies well to a population outside of the study

How well did you know this?

Not at all

Perfectly

Least important of mean, median, mode?

Mode

How well did you know this?

Not at all

Perfectly

quantifies the amount of variability, or spread, around the mean of the measurements.

Variance (σ2 )

How well did you know this?

Not at all

Perfectly

a measure of variation of scores about the mean

more commonly used

standard deviation

How well did you know this?

Not at all

Perfectly

The statement that establishes a relationship between variables being assessed

Alternative hypothesis (Ha or H1)

How well did you know this?

Not at all

Perfectly

The statement of no difference or no relationship between the variables

Null hypothesis (Ho)

How well did you know this?

Not at all

Perfectly

A ___ error is made if we reject the null hypothesis when null hypothesis is true.

type I

How well did you know this?

Not at all

Perfectly

A ___ error is made if we fail to reject null hypothesis

Type 2

How well did you know this?

Not at all

Perfectly

More important than p value – a better determination of significance
Any statistic is simply an estimate of the true value of that statistic

Confidence interval (CI)

How well did you know this?

Not at all

Perfectly

95% CI states that we can be 95% certain that the “true” value is within the CI range

Narrower CI is better

How well did you know this?

Not at all

Perfectly

Odds ratio of 1

no assocation

How well did you know this?

Not at all

Perfectly

A screening test is used to separate from a large group of apparently well persons those who have a high probability of having the disease, so that they may be given a diagnostic work up, and if diseased can be treated.

What are the conditions for a screening?

The target disease is an important cause of mortality and morbidity.

A proven and acceptable test exists to detect individuals at an early stage of disease.

There is a treatment available to prevent mortality and morbidity once positives have been identified.

How well did you know this?

Not at all

Perfectly

Sensitivity

ability of test to correctly identify those who have the disease

How well did you know this?

Not at all

Perfectly

Specificity

ability of the test correctly identify those who DO NOT have the disease

How well did you know this?

Not at all

Perfectly

Tends to rule OUT the disease

High Sensitivity means low probability of false positive

Sensitivity

How well did you know this?

Not at all

Perfectly

Screening test’s ability to identify presence of disease

A test with high sensitivity will not miss many patients
who have the disease

A highly useful test when NEGATIVE

Sensitivity

How well did you know this?

Not at all

Perfectly

Tends to rule IN the disease

High Specificity means low probability of false negative

Specificity

How well did you know this?

Not at all

Perfectly

Screening test’s ability to truly identify absence of disease

That is, how likely is a negative test actually reporting the right answer?

A highly useful test when it is POSITIVE

Specificity

How well did you know this?

Not at all

Perfectly

A highly ___ test is most useful to the clinician when it is NEGATIVE

sensitive

How well did you know this?

Not at all

Perfectly

A highly ___ test is most useful to the clinician when it is POSITIVE

specific

How well did you know this?

Not at all

Perfectly

Allows up to calculate the net sensitivity and net specificity of using both tests in sequence. After completing both tests there is a loss in net sensitivity and net gain in specificity.

Sequential (Two-Stage) Testing

How well did you know this?

Not at all

Perfectly

proportion of patients who HAVE the disease and a positive test

POPULATION related

(e.g., HIV prevalence in suburban city in US vs. HIV prevalence in sub-Saharan Africa)

Positive Predictive Value (PPV)

With low prevalence (% of population) of disease:
Lower PPV
False positives increase
Less reliable positive test result

How well did you know this?

Not at all

Perfectly

proportion of patients who DO NOT HAVE the disease, and have a negative test

Negative Predictive Value (NPV)

How well did you know this?

Not at all

Perfectly

the occurrence, rate, or frequency of a disease

Incidence Obtained from cohort studies Must follow a cohort through time

the number of occurrences at one particular time

Prevalence- Obtained from cross-sectional studies | No time line, only a snap shot

Relationship between incidence and prevalence

slide 33

Allows the researcher to explore the relationship between two continuous variables

Regression analysis

A method of predicting change in the dependent variable by changing one or more independent variables What % of variation in the dependent variable can be explained by a change in the independent variable

Regression analysis

Two broad data types

Categorical Continuous

Categorical

Nominal Ordinal

Continuous

Interval Ration

named categories with no implied order

Nominal

sequenced or ranked data

Ordinal E.g., smallest to largest, lightest to heaviest, easiest to most difficult

intervals along the scale are equal to one another (i.e. integers)

Interval Continuous Data

characterized by the presence of absolute zero on the scale Most precise

Ratio Continuous Data

In ____ screening, a less expensive, less invasive, or less uncomfortable test is generally performed first, and those who screen positive are recalled for further testing with a more expensive, more invasive, or more uncomfortable test, which may have greater sensitivity and specificity. It is hoped that bringing back for further testing only those who screen positive will reduce the problem of false positives.

sequential or two-stage

Summarizes the same kind of information sensitivity and specificity and can be used to calculate the probability of disease in a low prevalence setting

Likelihood ratio (LR)

provides indication of the test’s discriminatory power Predictive values are lower with a low prevalence can be defined for the entire range of test result values

Likelihood ratio (LR)

is the proportion of diseased people with a negative test result (1-sensitivity) divided by the proportion of non-diseased people with negative test results (specificity)

negative LR (LR-) How good the test is at “Ruling out” disease The smaller the better (Desirable: 0.2 or less)

Loss in net specificity, and gain in net specificity

Sequential (Two-stage) Testing

Net gain in sensitivity, and net loss in specificity. Patient is considered positive if they test positive on any test or both. Patient is considered negative if they test negative on all the tests performed

Simultaneous Testing

one of the most common ways to examine relationships between two or more categorical variables

Chi-square

tests the null hypothesis that the variable are independent of each other, that there is no relationship between the two variable

chi-square of independence

does not give any information about the strength of the relationship.

Chi-square statistic

Computed the same way as the chi-square test for independence, but instead tests the hypothesis that the distribution of some variable is the same in all populations

Chi-square test for equality of proportion

Is used to test they hypothesis that the distribution of a categorical variable within a population followed a specific pattern of proportion

Chi-square test of goodness of fit

Fisher’s exact test

A non-parametric test similar, similar to the chi-square tests, but can be used with small or sparsely distributed data sets

a type of chi-square test used when the data comes from paired samples.

McNemar’s Test for Matched Pairs

Measures the strength of association between an exposure and disease the effect of one intervention v another = (AD)/(BC)

Odds ratio (OR)

If exposure does not affect (either cause or protect from) disease, the OR is ____ If the exposure is ____ to the disease, the OR > 1 If the exposure is ___ against the disease, the OR < 1

~ 1 related protective

to be able to estimate the probability of an outcome associated with a dichotomous response for a single or multiple variables

Logistic Regression

a single outcome (or set of outcomes) from an experiment

Event

The proportion of subjects in a study group in whom the event is observed. Usually seen as a %.

Rate

A measure of how often a particular event (such as response to a drug, adverse event or death) occurs within the scientific control group of an experiment

Control Event Rate (CER) %

A measure of how often a particular event (such as response to a drug, adverse event or death) occurs within the experimental group of an experiment

Experimental Event Rate (EER) %

Basic risk statements express the likelihood that a particular event will occur within a particular population Identifies what in our environment can lead to beneficial or adverse medical outcomes

Relative risk SAME AS RISK RATIO

measures the magnitude of an association between an exposed and non-exposed (control) group. calculated using cumulative incidence data to measure the probability of developing disease

Relative risk (same as Risk Ratio) Must have incidence information to calculate Cohort or clinical trials are conducted over time

The percentage difference in outcome between control (C) and experimental (E) groups

Relative risk reduction RRR= (CER-EER)/CER

Not a good way to compare outcomes Does not report the baseline risk of outcome Measures such as percent reduction in mortality, is selected because it gives a more optimistic view of the effectiveness of a preventive measure. Makes insignificant findings appear significant

Relative risk reduction

The actual reduction in events in the treated group (EER) The arithmetic difference in outcomes between treatment and control groups The “true difference” between the experimental and control intervention

Absolute risk reduction ARR = CER - EER

Odds ratio are an ___ estimate of risk,

indirect not a direct measure of risk

In a case-control study, only the odds ratio can be calculated as a measure of association ,whereas in a ___ study, either the relative risk or odds ratio can be calculated.

cohort

Odds ratios calculated in a case-control study are a good approximation of relative risk in the population when the following conditions are met:

When cases studied are representative, with regard to history of exposure, of all people with the disease in the population from which the cases were drawn. When the controls studied are representative, with regard to history of exposure, of all people without the disease in the population from which the controls were drawn. When the disease being studied does not occur frequently.

The number of patients who need to receive the new intervention instead of the standard alternative in order for one additional patient to benefit

Number needed to treat

Expresses the likelihood of the treatment to benefit an individual patient There is NO absolute value for NNT that defines whether something is effective or not. NNTs for treatments are usually low because we expect large effects in small numbers of people

Number needed to treat

NNTs for very effective treatments are usually in the range of 2 to 4 Rule of thumb: NNT 10 or less for therapy NNT 20 or less for prevention

Larger NNTs can be found useful where few patients are affected in large populations Use for prophylactic measures Example: Aspirin prevents one death at five weeks after a myocardial infarction, NNT of 40

When an experimental treatment is detrimental, the term number needed to harm (NNH) is often used. The equations and approach are similar to those described above, except that NNH will have a negative absolute risk reduction

Number needed to harm

Generally is used to analyze continuous data Compares the means and standard deviations of two populations Data be must be normally distributed Computes a p-value to test the null hypothesis

T-test

Assesses whether a difference between two groups’ averages is unlikely to have occurred because of random chance in sample selection. A difference is more likely to be meaningful and “real” if- The difference between the averages is large. The sample size is large. Responses are consistently close to the average values and not widely spread out (the standard deviation is low).

T-test

Independent variable

the variable you're interested in

Prevalence

cross-sectional prevalance = disease burden