Handout Review Flashcards
The number of occurrences at ONE PARTICULAR TIME
Prevalence
The occurrence, RATE, or frequency of a disease
Incidence
???
Outlier
Represented by “r” (rho)
The closer to 1 (or -1) the stronger the relationship. Closer to 0? A weaker relationship.
Pearson correlation is the most common but sensitive to outliers (can be misleading if non-linear relationship)
Correlation coefficient
Measures 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
Must have incidence information (cohort of clinical trials are conducted over time)
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
Proportion of people with the disease who have a positive test for the disease
Sensitivity
The ability of the test to identify correctly those who have the disease
sensitivity
The proportion of people without disease who have a negative test
specificity
Ability of the test to identify correctly those who don’t have the disease
specificity
In this screening, a less expensive/invasive/uncomfortable test is generally performed first… those that screen positive are recalled for further testing with more expensive/invasive/uncomfortable test
Two-stage (sequential) testing
Loss in net sensitivity, gain in net specificity
Two-stage (sequential) testing
Patient is considered positive if they test pos on either/both tests.
Pt considered negative if they test neg on both.
Simultaneous testing
Net gain in sensitivity, net loss in specificity
Simultaneous testing
ability to apply results obtained from a study population to a broader population
External validity
Also called generalizability
External validity
Within the confines of the study, the results appear to be accurate and the interpretation of the investigators is supported
Internal validity
Most valuable in determining the statistical significance of an effect estimate?
confidence interval
More important than p-value? A better determination of significance?
Confidence interval
Produces a range within which the true value most likely lies…
“We be 95% certain that the true value is within the __ range”
Narrower is better…
Confidence interval
Odds ratios calculated in a case-control study are a good approximation of relative risk in the population when what three conditions are met?
When cases studied are representative, with regard to history of exposure, of all people w/ the disease in the population from which the cases were drawn.
When CONTROLS studied are representative, with regard to history of exposure, of all people w/ the disease in the population from which the CONTROLS were drawn.
When the studied disease doesn’t 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 (NNT)w
Expresses the likelihood of the tx to benefit an individual patient
number needed to treat (NNT)
Is there an absolute value for NNT that defines whether something is effective or not?
No. But NNTs for very effective treatments are usually in the range of 2-4.
Usually a lower number b/c we expect large effects in small numbers of people.
Larger NNTs can be found useful where few pts are affected in large populations (use for prophylactic measures)
When an experimental treatment is detrimental, what term would we employ?
number needed to harm (NNH)
Numbers are similar to NNT, except NNH will have a negative absolute risk reduction
Low probability of a false negative?
Sensitivity
Highly useful when negative because it rules out the disease
Sensitivity
Low probability of a false positive
specificity
Highly useful when it is positive because it tends to rule in the disease
Specificity
Means that subjects are analyzed according to the categories into which they were originally randomized.
Even if subjects withdraw/fail to take prescribed/otherwise adjust their tx, they still belong to their original tx group
Intention to treat analysis
note that this was taken from the book the definition couldn’t be found in the slides
Indicates the chance of a random error
p-value
P-value key metrics?
p = or < .05 (statistically significant)
p < .001 (HIGHLY significant)
When a drug/procedure/intervention works under ideal circumstances
efficacy
When we reject the null hypothesis, but the hypothesis is actually true?
Type 1 error
aka Alpha error
When we fail to reject the null hypothesis, but the null hypothesis is false
Type 2 error
aka Beta error
variables that correlate directly or indirectly with the dependent and independent variables
Confounding variables
AKA hidden variable, lurking variable
An extraneous variable in a statistical model that correlates (positively or negatively) with both the dependent and independent variable
confounding variable
A descriptive study that provides a narrative in professional literature that IDs a single incident and discusses pertinent factors related to that patient
Brings a nove/unusual patient to the attention of colleagues
case report
Descriptive study of individual cases that share a commonality
case series
Case series are used to?
Examine adverse events/effects
Catalog new diseases/outbreaks
Determine feasibility/safety of new tx/intervention
Discuss potential efficacy of new tx
An analytic/observational study
Studies in which pts who already have a specific condition are compared with people who do not have the condition
Case control studies
Researcher looks back to identify factors/exposures that might be associated with illness
Tries to capture the cause and effect relationship by comparing frequency of a risk factor among those who are exposed and not exposed
Case-control studies
thorough, comprehensive, and explicit way of interpreting the medical literature
systematic review
STATISTICAL approach to combine the data derived from several seleted studies
meta-analysis
identify a group of patients who are already taking a particulat tx/have an exposure, follow them forward over time and then compare their outcomes with a similar group that has not been affected by the treatment or exposure being studied
prospective cohort studies
Strongest observational study?
cohort study
Strengths of cohort study?
study multiple effects of a single exposure
identify a temporal relationship b/w exposure/outcome
help confirm casue/effect and magnitude of effect
can measure incidence of dz
can calculate relative risk
Highest validity of observational study design?
Cohort study
Cross sectional
An analytic, observational study
examines relationship between otucomes and other variables of interest as they exist in a defined population at one partcular time
Can a cross-sectional determine prevalence?
Yes, but it cannot show causality nor a temporal relationship
Cross sectional strengths/
assess multiple outcomes and exposures simultaneously
can be completed quickly
data generated can lead to further studies
can generate prevalence
Cross sectional weaknesses?
no time reference (snapshot in time)
only useful for common conditions
cannot calculate incidence
Results are dependent on study population
Main purpose of randomization is to?
prevent any potential biases on the part of the investigators
Though randomization strives for comparability, it is not….?
guaranteed
The gold standard?
Double blinded randomized control trial
Limitations of control studies?
Large trials (may effect statistical power)
Long term follow up (possible losses)
Compliance
Expensive
Possible ethical questions
Quasi experimental study aka?
non-randomized control trial
Limitations of non-randomized control trials?
Study group characteristics may not be balanced at baseline and these differences may confoudn study’s results
(typical confounding variables include age, educational level, motivation, severity of illness, socioeconomic status/income, comorbidities)
Variant of case-control study
Each case becomes their own individual control
Used for transient exposures during a discrete occurence?
Case crossover study
Must have a washout period?
Case crossover study
Intention to treat?
Analyzed according to original randomized assignment
If bias occurs, typically biases towards the null; provides a more conservative estimate
Research dealing with phenomena that that are difficul or imposible to quantify mathematically, such as beliefs, meanings, attributes, and symbols; it may invovle content analysis
qualitative research
Basic level: a descriptive account of the date
i.e., this is what was said but no comments or theories as to why or how
(regarding qualitative research)
manifest level
higher level; a more interpretive analysis that is concerned with the response as well as what may have been implied or inferred
(regarding qualitative research)
Latent level
the score that occurs most frequently
mode
Measure of central tendency
middle point
median
Measure of central tendency
the average
mean
Measure of central tendency
least precise measure of central tendency?
mode
percent of patients with positive test who actually have the disease
positive predictive value
assesses reliability of positive test
positive predictive value
Percentage of patients with a negative test who actually do NOT have the disease
negative predictive value
assesses reliability of a negative test
negative predictive value
If you have a low prevalence of disease, in regards to positive predictive value, you’ll have ___ PPV, your false positive ____, and its a ___ reliable positive test result
lower
increase
less
If you have a low prevalence of disease, in regards to negative predictive value, you’ll have ___ NPV, your false negative ____, and its a ___ reliable negative test result
higher
decreased
more
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
Likelihood ratio (LR)
Low prevalence = Less reliable positive test result; therefore, use ___
Likelihood ratio (LR)
LR addresses ?
How much more likely are we to find that a test is positive among patients with disease compared with those without disease?
A positive LR (LR+) is the ratio of the proportion of diseased people with a positive test result (sensitivity) to the proportion of non-diseased people with a positive result (1-specificity).
How good the test is at “Ruling in” disease!
A negative LR (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)
How good the test is at “Ruling out” disease
Measures the strength of association between an exposure and disease
Odds ratio (OR)
If exposure does not affect (either cause or protect from) disease, the OR is ~ 1
If the exposure is related to the disease, the OR > 1
If the exposure is protective against the disease, the OR < 1
Odds ratio (OR)
strives for comparability of the different treatment groups; however, its not guaranteed
randomization
the best approach in the design of a trial, and the critical element of ______ is the unpredictability of the next assignment
randomization
occurs when relationships that exist for groups are assumed to also be true for individuals
ecological fallacy
type of bias
“When investigators know the identity of case and comparison subjects and which exposures are risks, objectivity is put to a strenuous test”
Researcher bias
from book pg 47
Searching medical records more thoroughly or questioning more diligently about exposure to medical asbestos in cases of fibrotic lung disease or about the consumption of artificial sweeteners among bladder patients…
an example of?
Researcher bias
from book pg 47
Both are used for the development of Clinical Practice Guidelines (CPGs)?
Meta analysis
Systematic review
What is the difference between a “systematic review” and a “meta-analysis”?
A “systematic review” is a thorough, comprehensive, and explicit way of interpreting the medical literature
A “meta-analysis” is a statistical approach to combine the data derived from several selected studies
is the study of the distribution and determinants of health-related states or events in specified populations, and the application of this study to control health problems
epidemiology
the branch of statistics that deals with data relating to living organisms
Using the tools of statistics to help answer pressing research questions in medicine, biology, and public health
biostatistics