First Aid - Epidemiology and Biostatistics Flashcards
Cross-sectional study
Measures what?
Cross-sectional study
Disease prevalence.
Can show risk factor association with disease, but
does not establish causality.
Cross-sectional study
Design of the Study?
Cross-sectional study
Frequency of disease and frequency of riskrelated
factors are assessed in the present.
Asks, “What is happening?”
Case-control study
Design of the Study?
Case-control study
Compares a group of people with disease to a
group without disease.
Looks to see if odds of prior exposure or risk
factor differ by disease state.
Asks, “What happened?”
Case-control study
Measures what?
Case-control study
Odds ratio (OR).
Patients with COPD had higher odds of a
smoking history than those without COPD.
Cohort study
Measures What?
Cohort study
Relative risk (RR).
Smokers had a higher risk of developing COPD
than nonsmokers.
Cohort = relative risk.
Cohort study
Measures What?
Cohort study
Relative risk (RR).
Smokers had a higher risk of developing COPD
than nonsmokers.
Cohort = relative risk.
Crossover study
Design of the Study?
Crossover study
Compares the effect of a series of ≥2 treatments
on a participant.
Order in which participants receive treatments
is randomized. Washout period occurs
between each treatment.
Crossover study
Advantage of the Study?
Crossover study
Allows participants to serve as their own
controls.
Twin concordance study
Measures What?
Twin concordance study
Measures heritability and influence of environmental factors (“nature vs nurture”).
Twin concordance study
Design of the Study?
Twin concordance study
Compares the frequency with which both
monozygotic twins vs both dizygotic twins
develop the same disease.
Adoption study
Measures What?
Adoption study
Measures heritability and influence of
environmental factors.
Adoption study
Design of Study?
Adoption study
Compares siblings raised by biological vs
adoptive parents
Clinical Trials
Experimental study involving humans. Compares therapeutic benefits of ≥2 treatments, or of
treatment and placebo. Study quality improves when study is randomized, controlled, and _________
(ie, neither patient nor doctor knows whether the patient is in the treatment or control
group). Triple-blind refers to the additional blinding of the researchers analyzing the data.
Four phases (“Does the drug SWIM?”).
Clinical Trials
Experimental study involving humans. Compares therapeutic benefits of ≥2 treatments, or of
treatment and placebo. Study quality improves when study is randomized, controlled, and doubleblinded
(ie, neither patient nor doctor knows whether the patient is in the treatment or control
group). Triple-blind refers to the additional blinding of the researchers analyzing the data.
Four phases (“Does the drug SWIM?”).
Clinical Trials
Experimental study involving humans. Compares therapeutic benefits of ≥2 treatments, or of treatment and placebo. Study quality improves when study is randomized, controlled, and doubleblinded. \_\_\_\_\_\_\_\_ refers to the additional blinding of the researchers analyzing the data. Four phases (“Does the drug SWIM?”).
Clinical Trials
Experimental study involving humans. Compares therapeutic benefits of ≥2 treatments, or of treatment and placebo. Study quality improves when study is randomized, controlled, and doubleblinded. **Triple-blind** refers to the additional blinding of the researchers analyzing the data. Four phases (“Does the drug SWIM?”).
Clinical Trials
Four phases (“Does the drug ____?”).
Clinical Trials
Four phases (“Does the drug SWIM?”).
Safe, Works, Improves, postMarketing safe
Clinical Trials - Phase I
Typical study sample?
Clinical Trials - Phase I
Small number of either healthy volunteers or
patients with disease of interest
Clinical Trials - Phase I
PURPOSE?
(SWIM)
Clinical Trials - Phase I
“Is it Safe?” Assesses safety, toxicity,
pharmacokinetics, and pharmacodynamics.
Clinical Trials - Phase II
PURPOSE
(SWIM)
Clinical Trials - Phase II
“Does it Work?” Assesses treatment efficacy,
optimal dosing, and adverse effects
Clinical Trials - Phase II
Typical study sample?
Clinical Trials - Phase II
Moderate number of patients with disease of
interest.
Clinical Trials - Phase III
Typical study sample?
Clinical Trials - Phase III
Large number of patients randomly assigned
either to the treatment under investigation or
to the standard of care (or placebo).
Clinical Trials - Phase III
PURPOSE?
(SWIM)
Clinical Trials - Phase III
“Is it as good or better?” Compares the new
treatment to the current standard of care (any
Improvement?).
Clinical Trials - Phase IV
PURPOSE?
Clinical Trials - Phase IV
“Can it stay?” Detects rare or long-term adverse
effects (eg, black box warnings). Can result in
treatment being withdrawn from Market.
Clinical Trials - Phase IV
Typical Study Sample?
Clinical Trials - Phase IV
Postmarketing surveillance of patients after
treatment is approved.
Evaluation of Diagnostic tests
Sensitivity and specificity are ____ properties
of a test
Evaluation of Diagnostic tests
Sensitivity and specificity are fixed properties
of a test
Evaluation of Diagnostic tests
PPV and NPV vary depending on
disease prevalence in population being tested.
Evaluation of Diagnostic tests
PPV and NPV vary depending on
disease _______ in population being tested.
Evaluation of Diagnostic tests
Positive Predictive Value is the number of ____ Positives Devided by the Number of ____ Positives and the Number of ____ Positives
Evaluation of Diagnostic tests
Positive Predictive Value is the number of True Positives Devided by the Number of True Positives and the Number of False Positives
(PPV=TP/(TP+FP)
Evaluation of Diagnostic tests
Positive Predictive Value is the number of ____ Positives Devided by the Number of ____ Positives and the Number of ____ Positives
Evaluation of Diagnostic tests
Positive Predictive Value is the number of True Positives Devided by the Number of True Positives and the Number of False Positives
(PPV=TP/(TP+FP)
Evaluation of Diagnostic tests
Negative Predictive Value - Definition
Evaluation of Diagnostic tests
Probability that a person with a negative test
result actually does not have the disease.
Evaluation of Diagnostic tests
Positive Predictive Value - Definition
Evaluation of Diagnostic tests
Probability that a person with a positive test
result actually has the disease.
Evaluation of Diagnostic tests
Sensitivity (True Positive Rate) is actually number of True Positives devided by the number of True ____ and the Number of ____ Negatives.
Evaluation of Diagnostic tests
Sensitivity (True Positive Rate) is actually number of True Positives devided by the number of True positives and the Number of False Negatives.
Sensitivity = TP / (TP + FN)
Evaluation of Diagnostic tests
Specificity (True Negative Rate) is actually number of True Negatives devided by the number of True ____ and the Number of ____ Positives.
Evaluation of Diagnostic tests
Specificity (True Negative Rate) is actually number of True Negatives devided by the number of True Negatives and the Number of False Positives.
Specificity = TN / (TN + FP)
Evaluation of Diagnostic tests
High Specificity indicates a Low ____ _____ Rate!
Evaluation of Diagnostic tests
High Specificity indicates a Low False Positive Rate!
(Specificity = True Negative Rate)
Evaluation of Diagnostic tests
High Sensitivity indicates a Low ____ _____ Rate!
Evaluation of Diagnostic tests
High Sensitivity indicates a Low False Negative Rate!
(Sensitivity = True Positive Rate)
Evaluation of Diagnostic tests
___ varies inversely with prevalence or pretest
probability
Evaluation of Diagnostic tests
NPV varies inversely with prevalence or pretest
probability
Evaluation of Diagnostic tests
Prevelance = (TP+FN)/(TP+FP+TN+FN)
Evaluation of Diagnostic tests
Prevelance = (TP+FN)/(TP+FP+TN+FN)
Evaluation of Diagnostic tests
___ varies directly with pretest probability
(baseline risk, such as prevalence of disease):
high pretest probability → high ___
Evaluation of Diagnostic tests
PPV varies directly with pretest probability
(baseline risk, such as prevalence of disease):
high pretest probability → high PPV
Evaluation of Diagnostic tests
A?
Evaluation of Diagnostic tests
A = 100% Sensitivity cutoff value
Evaluation of Diagnostic tests
C?
Evaluation of Diagnostic tests
C = 100% specificity cutoff value
Evaluation of Diagnostic tests
B?
Evaluation of Diagnostic tests
B = practical compromise cutoff between Specificity and Sensitivity
Evaluation of Diagnostic tests
Lowering the cutoff value: B→A
Will cause FP↑ and FN↓
What will happen to the Sensitivity?
Evaluation of Diagnostic tests
Sensitivity Rises!
Sensitivity↑ = TP / (TP + FN↓)
Evaluation of Diagnostic tests
Lowering the cutoff value: B→A
Will cause FP↑ and FN↓
What will happen to the Specificty?
Evaluation of Diagnostic tests
Specificty Falls!
Specificty↓ =TN / (TN + FP↑)
Evaluation of Diagnostic tests
Lowering the cutoff value: B→A
Will cause FP↑ and FN↓
What will happen to the NPV?
Evaluation of Diagnostic tests
NPV Rises!
NPV↑ = TN / (TN + FN↓)
Evaluation of Diagnostic tests
Lowering the cutoff value: B→A
Will cause FP↑ and FN↓
What will happen to the PPV?
Evaluation of Diagnostic tests
PPV Falls!
PPV↓ = TP / (TP + FP↑)
Evaluation of Diagnostic tests
Raising the cutoff value: B→C
Will cause FP↓ and FN↑
What will happen to the Specificity?
Evaluation of Diagnostic tests
Specificity Rises!
Specificity = TN / (TN + FP↓)
Evaluation of Diagnostic tests
Raising the cutoff value: B→C
Will cause FP↓ and FN↑
What will happen to the NPV?
Evaluation of Diagnostic tests
NPV Falls!
NPV = TN / (TN + FN↑)
Evaluation of Diagnostic tests
Raising the cutoff value: B→C
Will cause FP↓ and FN↑
What will happen to the PPV?
Evaluation of Diagnostic tests
PPV Rises!
PPV = TP / (TP + FP↓)
Evaluation of Diagnostic tests
Raising the cutoff value: B→C
Will cause FP↓ and FN↑
What will happen to the Sensitivity?
Evaluation of Diagnostic tests
Sensitivity Falls!
Sensitivity= TP / (TP + FN↑)
Likelihood ratio
What is it?
Likelihood ratio
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.
Likelihood ratio
What does LR+>10 indicates?
Likelihood ratio
LR+ > 10 indicates a highly SPECIFIC test
Likelihood ratio
What does LR–< 0.1indicates?
Likelihood ratio
LR– < 0.1 indicates a highly SENSITIVE test.
Odds ratio
Typically used in Case-Control studies. Represents the odds of exposure among cases (_/c) vs odds of exposure among controls (_/d).
Odds ratio
Typically used in Case-Control studies. Represents the odds of exposure among cases (a/c) vs odds of exposure among controls (b/d).
Odds ratio
If in a case-control study, 20/30 lung cancer patients and 5/25 healthy individuals report smoking, the OR is _; so the lung cancer patients are _ times more likely to have a history of smoking.
Odds ratio
If in a case-control study, 20/30 lung cancer patients and 5/25 healthy individuals report smoking, the OR is 8; so the lung cancer patients are 8 times more likely to have a history of smoking.
OR = (20/10)/(5/20) = 8
Relative risk
Used in Cohort studies. Risk of developing disease in the
exposed group divided by risk in the unexposed group.
If RR = 1 than ____ ?
Relative risk
RR = 1 → NO association between
exposure and disease.
Relative risk
Used in Cohort studies. Risk of developing disease in the
exposed group divided by risk in the unexposed group.
If RR < 1 than ____ ?
Relative risk
RR < 1 → exposure associated with
↓ disease occurrence.
Relative risk
Used in Cohort studies. Risk of developing disease in the
exposed group divided by risk in the unexposed group.
If RR < 1 than ____ ?
Relative risk
RR < 1 → exposure associated with
↓ disease occurrence.
Relative risk
For rare diseases (low prevalence), __
approximates RR.
Relative risk
For rare diseases (low prevalence), OR
approximates RR.
OR = (a/c)/(b/d)
RR = (a/(a+b)/(c/(c+d)
Relative risk
If 5/10 people exposed to radiation are diagnosed with cancer, and 1/10 people not exposed to radiation are diagnosed with cancer, the RR is _ ; so people exposed to radiation have a _ times greater risk of developing cancer.
Relative risk
If 5/10 people exposed to radiation are diagnosed with cancer, and 1/10 people not exposed to radiation are diagnosed with cancer, the RR is 5; so people exposed to radiation have a 5 times greater risk of developing cancer.
RR = (5/10)/(1/10) = 5
Relative Risk Reduction
The proportion of risk reduction
attributable to the ________ as
compared to a control.
RRR = 1 - RR
Relative Risk Reduction
The proportion of risk reduction
attributable to the intervention as
compared to a control.
RRR = 1 - RR
Relative Risk Reduction
If 2% of patients who receive a flu
shot develop the flu, while 8% of
unvaccinated patients develop the flu,
then RR = _ , and RRR = _ .
RRR = 1 - RR
Relative Risk Reduction
If 2% of patients who receive a flu
shot develop the flu, while 8% of
unvaccinated patients develop the flu,
then RR = 2/8 = 0.25, and RRR = 0.75.
RRR = 1 - RR
Attributable Risk
The difference in risk between
________ and ________ groups.
Attributable Risk
The difference in risk between
exposed and unexposed groups.
Attributable Risk
If risk of lung cancer in smokers is 21% and risk in nonsmokers is 1%, then the attributable risk is ___.
Attributable Risk
If risk of lung cancer in smokers is 21% and risk in nonsmokers is 1%, then the attributable risk is 20%
Absolute Risk Reduction
The ______ in risk (not the proportion) attributable to the intervention as compared to a control.
Absolute Risk Reduction
The difference in risk (not the proportion) attributable to the intervention as compared to a control.
Absolute Risk Reduction
If 8% of people who receive a placebo vaccine develop the flu vs 2% of people who receive a flu vaccine, then ARR = _.
Absolute Risk Reduction
If 8% of people who receive a placebo vaccine develop the flu vs 2% of people who receive a flu vaccine, then ARR = 8%–2% = 6% = 0.06.
Number Needed to Treat
Lower number = _____ treatment.
NNT = 1/ARR
Number Needed to Treat
Lower number = better treatment.
NNT = 1/ARR
Number Needed to Treat
Number of patients who need to be treated for ___.
NNT = 1/ARR
Number Needed to Treat
Number of patients who need to be treated for 1 patient to benefit.
NNT = 1/ARR
Number Needed to Harm
Number of patients who need to be exposed to a risk factor for ___
NNH = 1/AR
Number Needed to Harm
Number of patients who need to be exposed to a risk factor for 1 patient to be harmed.
NNH = 1/AR
Number Needed to Harm
Higher number = ___ exposure.
NNH = 1/AR
Number Needed to Harm
Higher number = Safer exposure.
NNH = 1/AR
Case Fatality Rate
Percentage of deaths occurring among ___.
Case Fatality Rate
Percentage of deaths occurring among those with disease.
CFR% = (Deaths/Cases)x100%
Case Fatality Rate
If 4 patients die among 10 cases of
meningitis, case fatality rate is ___.
Case Fatality Rate
If 4 patients die among 10 cases of
meningitis, case fatality rate is 40%.
CFR% = (Deaths/Cases)x100%
Prevelance Vs. Incidence
What is the Difference?
Prevelance Vs. Incidence
Incidence looks at new cases (incidents) while Prevalence looks at all current cases.
Prevelance Vs. Incidence
Incidence = # of new cases / # of people ___
(per unit of time)
Prevelance Vs. Incidence
Incidence = # of new cases / # of people at risk
(per unit of time)
Prevelance Vs. Incidence
Prevalence = # of ___ cases / Total # of people
(at a point in time)
Prevelance Vs. Incidence
Prevalence = # of existing cases / Total # of people
(at a point in time)
Prevelance Vs. Incidence
Prevalence/ (1-Prevelance) = ___ x Average Duration of the Disease
Prevelance Vs. Incidence
Prevalence/ (1-Prevelance) = Incidence rate x Average Duration of the Disease
Prevelance Vs. Incidence
Prevalence ≈ ____ for short duration disease
(eg, common cold).
Prevelance Vs. Incidence
Prevalence ≈ Incidence for short duration disease
(eg, common cold).
Prevelance Vs. Incidence
___ > incidence for chronic diseases, due to
large # of existing cases (eg, diabetes).
Prevelance Vs. Incidence
Prevalence > incidence for chronic diseases, due to
large # of existing cases (eg, diabetes).
Prevelance Vs. Incidence
Prevalence ∼ ______ probability.
Prevelance Vs. Incidence
Prevalence ∼ pretest probability.
Prevelance Vs. Incidence
↑ ____= ↑ PPV and ↓ NPV.
Prevelance Vs. Incidence
↑ Prevalence = ↑ PPV and ↓ NPV.
Prevelance Vs. Incidence
Survival Rate ↑ → Prevalence ↑/↓
Prevelance Vs. Incidence
Survival Rate↑ → Prevalence↑
Prevelance Vs. Incidence
Mortality ↑ → Prevalence ↑/↓
Prevelance Vs. Incidence
Mortality↑ → Prevalence↓
Prevelance Vs. Incidence
Recovery Time ↑ → Prevalence ↑/↓
Prevelance Vs. Incidence
Recovery Time↑ → Prevalence↓
Prevelance Vs. Incidence
Extensive Vaccine Administration ↑ → Prevalence ↑/↓
Prevelance Vs. Incidence
Extensive Vaccine Administration ↑→ Prevalence↓
Prevelance Vs. Incidence
Extensive Vaccine Administration ↑ → Incidence ↑/↓
Prevelance Vs. Incidence
Extensive Vaccine Administration ↑→ Incidence↓
Prevelance Vs. Incidence
Risk Factor ↓ → Incidence ↑/↓
Prevelance Vs. Incidence
Risk Factor ↓ → Incidence↓
Prevelance Vs. Incidence
Risk Factor ↓ → Prevelance ↑/↓
Prevelance Vs. Incidence
Risk Factor ↓ → Prevelance ↓
Accuracy Vs. Precision
The consistency and reproducibility of a test.
The absence of random variation in a test.
True for?
Accuracy Vs. Precision
Precision (Relability)
Accuracy Vs. Precision
Test Precision (Relability)↑ → Random Error↓/↑
Accuracy Vs. Precision
Test Precision (Relability)↑ → Random Error↓
Accuracy Vs. Precision
Test Precision (Relability)↑ → Standard Deviation↓/↑
Accuracy Vs. Precision
Test Precision (Relability)↑ → Standard Deviation↓
Accuracy Vs. Precision
Test Precision (Relability)↑ → Statistical Power (1 − β)↓/↑
Accuracy Vs. Precision
Test Precision (Relability)↑ → Statistical Power (1 − β)↑
Accuracy Vs. Precision
The closeness of test results to the true values.
The absence of systematic error or bias in a test.
True for?
Accuracy Vs. Precision
Accuracy (Validity)
Accuracy Vs. Precision
Test Accuracy (Validity)↑→ Systemic Error↓/↑
Accuracy Vs. Precision
Test Accuracy (Validity)↑→ Systemic Error↓
Receiving Operating Characteristic Curve
ROC curve demonstrates how well a diagnostic
test can ________ between 2 groups (eg,
disease vs healthy).
Receiving Operating Characteristic Curve
ROC curve demonstrates how well a diagnostic
test can distinguish between 2 groups (eg,
disease vs healthy).
Receiving Operating Characteristic Curve
Plots the ____-positive rate (sensitivity) against the ____-positive rate (1 – specificity).
Receiving Operating Characteristic Curve
Plots the True-positive rate (sensitivity) against the False-positive rate (1 – specificity).
Receiving Operating Characteristic Curve
The better performing test will have a higher ___, with the curve closer to the upper left corner.
Receiving Operating Characteristic Curve
The better performing test will have a higher AUC (area under the curve), with the curve closer to the upper left corner.
Bias and Study Errors - Recruiting Participants
Selection Bias - Nonrandom sampling or treatment allocation of subjects such that study population is not representative of _____ population.
Most commonly a sampling bias.
Bias and Study Errors - Recruiting Participants
Selection bias - Nonrandom sampling or treatment allocation of subjects such that study population is not representative of target population.
Most commonly a sampling bias.
Bias and Study Errors - Recruiting Participants
Selection Bias Example : cases and/
or controls selected from hospitals are less healthy and
have different exposures than general population.
This true for ______ bias
Bias and Study Errors - Recruiting Participants
Selection Bias Example : cases and/
or controls selected from hospitals are less healthy and
have different exposures than general population.
This true for Berkson bias
Bias and Study Errors - Recruiting Participants
Selection Bias Example : Participants lost to follow up have a different prognosis than those who complete the study.
This true for ______ bias
Bias and Study Errors - Recruiting Participants
Selection Bias Example : Participants lost to follow up have a different prognosis than those who complete the study.
This true for Attrition bias
Bias and Study Errors - Recruiting Participants
Selection Bias Reduction is done by : __________ and
Ensuring the choice of the right comparison/reference group
Bias and Study Errors - Recruiting Participants
Selection Bias Reduction is done by : Randomization and
Ensuring the choice of the right comparison/reference group
Bias and Study Errors - Performing study
_____ bias is when Awareness of disorder alters
recall by subjects; common in retrospective studies.
Bias and Study Errors - Performing study
Recall bias is when Awareness of disorder alters
recall by subjects; common in retrospective studies.
Bias and Study Errors - Performing study
_____ bias is when Awareness of disorder alters
recall by subjects; common in retrospective studies. For example Patients with disease recall exposure after learning of similar cases.
Bias and Study Errors - Performing study
Recall bias is when Awareness of disorder alters
recall by subjects; common in retrospective studies. For example Patients with disease recall exposure after learning of similar cases.
Bias and Study Errors - Performing study
Recall bias Reduction is done by ___ time from exposure to follow-up
Bias and Study Errors - Performing study
Recall bias Reduction is done by Decreasing time from exposure to follow-up
Bias and Study Errors - Performing study
________ effect—participants change behavior upon awareness of being observed. This is an exampele for Measurement Bias.
Bias and Study Errors - Performing study
Hawthorne effect—participants change behavior upon awareness of being observed. This is an exampele for Measurement Bias.
Bias and Study Errors - Performing study
________ effect—participants change behavior upon awareness of being observed. This is an exampele for Measurement Bias.
Bias and Study Errors - Performing study
Hawthorne effect—participants change behavior upon awareness of being observed. This is an exampele for Measurement Bias.
Bias and Study Errors - Performing study
Reduction of Measurement bias is done by Using objective, ______, and previously tested methods of data collection that are planned ahead of time and Use of _____group
Bias and Study Errors - Performing study
Reduction of Measurement bias is done by Using objective, standardized, and previously tested methods of data collection that are planned ahead of time and Use of placebo group
Bias and Study Errors - Performing study
Procedure bias is when _______ in different groups are
not treated the same. For Example Patients in treatment group spend more time in highly specialized hospital units
Bias and Study Errors - Performing study
Procedure bias is when Subjects in different groups are
not treated the same. For Example Patients in treatment group spend more time in highly specialized hospital units
Bias and Study Errors - Performing study
Reuction of Procedure bias is done by _______(masking) and use of ______ reduce influence of participants and
researchers on procedures and interpretation of outcomes as neither are aware of group aassignments
Bias and Study Errors - Performing study
Reuction of Procedure bias is done by Blinding (masking) and use of placebo reduce influence of participants and
researchers on procedures and interpretation of outcomes as neither are aware of group aassignments
Bias and Study Errors - Performing study
Reuction of Observer-expectancy
bias is done by _______(masking) and use of ______ reduce influence of participants and
researchers on procedures and interpretation of outcomes as neither are aware of group aassignments
Bias and Study Errors - Performing study
Reuction of Observer-expectancy
bias is done by Blinding (masking) and use of placebo reduce influence of participants and
researchers on procedures and interpretation of outcomes as neither are aware of group aassignments
Bias and Study Errors - Performing study
_____________ bias is when Researcher’s belief in the efficacy of a treatment changes the outcome of that treatment (aka, Pygmalion effect). For Example An observer expecting treatment group to show signs
of recovery is more likely to document _______ outcomes.
Bias and Study Errors - Performing study
Observer-expectancy bias is when Researcher’s belief in the efficacy of a treatment changes the outcome of that treatment (aka, Pygmalion effect). For Example An observer expecting treatment group to show signs
of recovery is more likely to document positive outcomes.
Bias and Study Errors - Interpreting results
_______ bias is when Factor related to both exposure and outcome (but not on causal path) distorts effect of exposure on outcome (vs effect modification, in which the exposure leads to different outcomes in subgroups stratified by the factor)
Bias and Study Errors - Interpreting results
Confounding bias is when Factor related to both exposure and outcome (but not on causal path) distorts effect of exposure on outcome (vs effect modification, in which the exposure leads to different outcomes in subgroups stratified by the factor)
Bias and Study Errors - Interpreting results
An uncontrolled study shows an association between
drinking coffee and lung cancer. However, coffee
drinkers also smoke more, which can account for the
association - this is an example for __________ bias.
Bias and Study Errors - Interpreting results
An uncontrolled study shows an association between
drinking coffee and lung cancer. However, coffee
drinkers also smoke more, which can account for the
association - this is an example for Confounding bias.
Bias and Study Errors - Interpreting results
Multiple/repeated studies, Crossover studies (subjects act as their own controls), Matching (patients with
similar characteristics in both treatment and control groups)these are reduction methods for __________ bias.
Bias and Study Errors - Interpreting results
Multiple/repeated studies, Crossover studies (subjects act as their own controls), Matching (patients with
similar characteristics in both treatment and control groups)these are reduction methods for Confounding bias.
Bias and Study Errors - Interpreting results
____-time bias is when Early detection is confused
with elevated survival.
Bias and Study Errors - Interpreting results
Lead-time bias is when Early detection is confused
with elevated survival.
Bias and Study Errors - Interpreting results
Early detection makes it seem like survival has increased,
but the disease’s natural history has not changed - this is an example for ________
Bias and Study Errors - Interpreting results
Early detection makes it seem like survival has increased,
but the disease’s natural history has not changed - this is an example for Lead-time bias
Bias and Study Errors - Interpreting results
Early detection makes it seem like survival has increased,
but the disease’s natural history has not changed - this is an example for ________
Bias and Study Errors - Interpreting results
Early detection makes it seem like survival has increased,
but the disease’s natural history has not changed - this is an example for Lead-time bias
Bias and Study Errors - Interpreting results
Measure “back-end” survival (adjust survival according to the severity of disease at the time of diagnosis) this is an example for Reduction of a Bias - ________
Bias and Study Errors - Interpreting results
Measure “back-end” survival (adjust survival according to the severity of disease at the time of diagnosis) this is an example for Reduction of a Bias - Lead-time bias
Bias and Study Errors - Interpreting results
_________ bias is reduced by a randomized controlled trial assigning subjects to the screening program or to no
screening.
Bias and Study Errors - Interpreting results
- *Length-time bias** is reduced by a randomized controlled trial assigning subjects to the screening program or to **no
screening. **
Bias and Study Errors - Interpreting results
_____________ is when Screening test detects diseases with long latency period, while those with shorter latency period become symptomatic earlier. For Example a slowly progressive cancer is more likely detected by a screening test than a rapidly progressive cancer.
Bias and Study Errors - Interpreting results
Length-time bias is when Screening test detects diseases with long latency period, while those with shorter latency period become symptomatic earlier. For Example a slowly progressive cancer is more likely detected by a screening test than a rapidly progressive cancer.
Measures of central tendency
____ = (sum of values)/(total number of values). Most affected by outliers (extreme values).
Measures of central tendency
Mean = (sum of values)/(total number of values). Most affected by outliers (extreme values).
Measures of central tendency
_____ = middle value of a list of data sorted
from least to greatest. If there is an even number of values, the median will be the average of the middle two values.
Measures of central tendency
Median = middle value of a list of data sorted
from least to greatest. If there is an even number of values, the median will be the average of the middle two values.
Measures of central tendency
_____ = most common value. Least affected by outliers.
Measures of central tendency
Mode = most common value. Least affected by outliers.
Measures of Dispersion
Standard ______ = how much variability exists in a set of values, around the mean of these values.
Measures of Dispersion
Standard Deviation = how much variability exists in a set of values, around the mean of these values.
Measures of Dispersion
Standard ____ = an estimate of how much variability exists in a (theoretical) set of sample means around the true population mean.
Measures of Dispersion
Standard Error = an estimate of how much variability exists in a (theoretical) set of sample means around the true population mean.
Statistical distribution
Normal Distribution is a Gaussian Disturbtion where
Mean = ___ = ___.
( also called bell-shaped)
Statistical distribution
- *Normal** Distribution is a Gaussian Disturbtion where
- *Mean = Median = Mode.**
( also called bell-shaped)
Non-Normal distributions
______ Distibution Suggests two different populations (eg, metabolic polymorphism such as fast vs
slow acetylators; age at onset of Hodgkin
lymphoma; suicide rate by age).
Non-Normal distributions
Bimodal Distibution Suggests two different populations (eg, metabolic polymorphism such as fast vs
slow acetylators; age at onset of Hodgkin
lymphoma; suicide rate by age).
Non-Normal distributions
______ skew distribution is where:
Typically, mean > median > mode.
Asymmetry with longer tail on right.
Non-Normal distributions
Positive skew distribution is where:
Typically, mean > median > mode.
Asymmetry with longer tail on right.
Non-Normal distributions
______ skew distribution is where:
Typically, mean < median < mode.
Asymmetry with longer tail on left
Non-Normal distributions
Negative skew distribution is where:
Typically, mean < median < mode.
Asymmetry with longer tail on left
Statistical hypotheses
_____ is a Hypothesis of no difference or relationship (eg, there is no association between the disease and the
risk factor in the population).
Statistical hypotheses
Null (H0) is a Hypothesis of no difference or relationship (eg, there is no association between the disease and the
risk factor in the population).
Statistical hypotheses
_________ is Hypothesis of some difference or relationship (eg, there is some association between the disease and the risk factor in the population).
Statistical hypotheses
Alternative Hypothesis (H1) is Hypothesis of some difference or relationship (eg, there is some association between the disease and the risk factor in the population).
Outcomes of statistical hypothesis testing
______ result can State that there is an effect or difference when one exists (null hypothesis rejected in favor of alternative hypothesis).
Outcomes of statistical hypothesis testing
Correct result can State that there is an effect or difference when one exists (null hypothesis rejected in favor of alternative hypothesis).
Outcomes of statistical hypothesis testing
______ result can State that there is an effect or difference when one exists (null hypothesis rejected in favor of alternative hypothesis).
Outcomes of statistical hypothesis testing
Correct result can State that there is an effect or difference when one exists (null hypothesis rejected in favor of alternative hypothesis).
Incorrect result
Type _ error (_) - Stating that there is an effect or difference when none exists (null hypothesis incorrectly
rejected in favor of alternative hypothesis).
Incorrect result
Type I error (α) - Stating that there is an effect or difference when none exists (null hypothesis incorrectly
rejected in favor of alternative hypothesis).
Incorrect result
_ is the probability of making a type I error. p is
judged against a preset _ level of significance
(usually set as 0.05). If p < 0.05 for a study outcome,
the probability of obtaining that result purely
by chance is < 5%.
Incorrect result
α is the probability of making a type I error. p is
judged against a preset α level of significance
(usually set as 0.05). If p < 0.05 for a study outcome,
the probability of obtaining that result purely
by chance is < 5%.
Incorrect result
Type _ Error - Also called false-positive error.
α = you accused an innocent man.(You can never “prove” the alternate hypothesis, but you can reject the null hypothesis as being very unlikely)
Incorrect result
- *Type I Error** - Also called false-positive error.
- *α** = you accused an innocent man.(You can never “prove” the alternate hypothesis, but you can reject the null hypothesis as being very unlikely)
Incorrect result
Type _ error (_) - Stating that there is not an effect or difference when one exists (null hypothesis is not rejected when it is in fact false).
Incorrect result
Type II error (β) - Stating that there is not an effect or difference when one exists (null hypothesis is not rejected when it is in fact false).
Incorrect result
_ is the probability of making a type II error. _ is related to statistical power (1 – _), which is the probability of rejecting the null hypothesis when it is false.
Incorrect result
β is the probability of making a type II error. β is related to statistical power (1 – β), which is the probability of rejecting the null hypothesis when it is false.
Incorrect result
Type II Error is Also called ____-negative error.
β = you blindly let the guilty man go free. If you ↑ sample size, you ↑ power. There is power in numbers.
Incorrect result
Type II Error is Also called False-negative error.
β = you blindly let the guilty man go free. If you ↑ sample size, you ↑ power. There is power in numbers.
Statistical Power
Equals to (1 – β), to ↑ Power and ↓ β:
1) Sample size → ↓/↑
2) Expected Effect size → ↓/↑
3) Precision of Measurement → ↓/↑
Statistical Power
Equals to (1 – β), to ↑ Power and ↓ β:
- *1) Sample size ↑
2) Expected Effect size ↑
3) Precision of Measurement ↑**
Confidence interval
CI is a Range of values within which the true _____
of the population is expected to fall, with a
_______ probability.
Confidence interval
CI is a Range of values within which the true Mean
of the population is expected to fall, with a
Specified probability.
Confidence interval
CI for sample mean = x¯ ± Z(SE)
The 95% CI (corresponding to α = .05) is often
used. As sample size increases, CI _______.
Confidence interval
CI for sample mean = x¯ ± Z(SE)
The 95% CI (corresponding to α = .05) is often
used. As sample size increases, CI Narrows.
Confidence interval
CI for sample mean = x¯ ± Z(SE)
For the 95% CI, Z = ____.
For the 99% CI, Z = ____.
(SE = SD/√n)
Confidence interval
CI for sample mean = x¯ ± Z(SE)
For the 95% CI, Z = 1.96.
For the 99% CI, Z = 2.58.
(SE = SD/√n)
Confidence interval
If the 95% CI for a mean difference between 2
variables includes _, then there is no significant
difference and H0 is ____.
Confidence interval
If the 95% CI for a mean difference between 2
variables includes 0, then there is no significant
difference and H0 is not rejected.
Confidence interval
If the 95% CI for odds ratio or relative risk
includes _, H0 is _____.
Confidence interval
If the 95% CI for odds ratio or relative risk
includes 1, H0 is not rejected.
Confidence interval
If the CIs between 2 groups do not overlap → statistically significant difference _____.
Confidence interval
If the CIs between 2 groups do not overlap → statistically significant difference exists.
Confidence interval
If the CIs between 2 groups overlap → usually significant difference ______.
Confidence interval
If the CIs between 2 groups overlap → usually significant difference doesn’t exists.
Meta-Analysis
A method of statistical analysis that pools summary data (eg, means, RRs) from multiple studies for a more precise estimate of the ____ of an effect. Also estimates ________ of effect sizes between studies.
Meta-Analysis
A method of statistical analysis that pools summary data (eg, means, RRs) from multiple studies for a more precise estimate of the size of an effect. Also estimates heterogeneity of effect sizes between studies.
Meta-Analysis
Improves power, strength of evidence, and ___________of study findings. Limited by quality of
individual studies and ____ in study selection.
Meta-Analysis
Improves power, strength of evidence, and generalizability of study findings. Limited by quality of
individual studies and bias in study selection.
Common Statistical Tests
_____ - Checks differences between means of 2 groups. (Tea is meant for 2). Example: comparing the mean blood pressure between men and women.
Common Statistical Tests
t-test - Checks differences between means of 2 groups. (Tea is meant for 2). Example: comparing the mean blood pressure between men and women.
Common Statistical Tests
_____ - Checks differences between means of 3 or more groups. (3 words: ANalysis Of VAriance.)
Example: comparing the mean blood pressure
between members of 3 different ethnic groups.
Common Statistical Tests
ANOVA - Checks differences between means of 3 or more groups. (3 words: ANalysis Of VAriance.)
Example: comparing the mean blood pressure
between members of 3 different ethnic groups.
Common Statistical Tests
__________ - Checks differences between 2 or more
percentages or proportions of categorical outcomes (not mean values). Pronounce Chi-tegorical. Example: comparing the percentage of members of 3 different ethnic groups who have essential hypertension.
Common Statistical Tests
Chi-square (χ²) Checks differences between 2 or more
percentages or proportions of categorical outcomes (not mean values). (Pronounce Chi-tegorical). Example: comparing the percentage of members of 3 different ethnic groups who have essential hypertension.
Common Statistical Tests
____________- Checks differences between 2 percentages or proportions of categorical, nominal outcomes. Use instead of chi-square test with small
populations. Example: comparing the percentage of 20 menand 20 women with hypertension.
Common Statistical Tests
- *Fisher’s exact test** - Checks differences between 2 percentages or proportions of categorical, nominal outcomes. Use instead of chi-square test with small
populations. Example: comparing the percentage of 20 menand 20 women with hypertension.
Pearson correlation Coefficient
r is always between __ and __.
Pearson correlation Coefficient
r is always between −1 and +1.
Pearson correlation Coefficient
The closer the absolute value of r is to _, the stronger the linear correlation between the 2 variables.
Pearson correlation Coefficient
The closer the absolute value of r is to 1, the stronger the linear correlation between the 2 variables.
Pearson correlation Coefficient
_______ is how much the measured values differ from the average value of r in a data set.
Pearson correlation Coefficient
Variance is how much the measured values differ from the average value of r in a data set.
Pearson correlation Coefficient
______ r value → _____correlation (as one variable ↑, the other variable ↑).
Pearson correlation Coefficient
Positive r value → positive correlation (as one variable ↑, the other variable ↑).
Pearson correlation Coefficient
______ r value → _____correlation (as one variable ↓, the other variable ↑).
Pearson correlation Coefficient
Negative r value → negative correlation (as one variable ↓, the other variable ↑).
Pearson correlation Coefficient
Coefficient of determination = __ (amount of variance in one variable that can be explained by variance in another variable).
Pearson correlation Coefficient
Coefficient of determination = r2 (amount of variance in one variable that can be explained by variance in another variable).
