Study your flashcards anywhere!
Download the official Cram app for free >
 Shuffle
Toggle OnToggle Off
 Alphabetize
Toggle OnToggle Off
 Front First
Toggle OnToggle Off
 Both Sides
Toggle OnToggle Off
 Read
Toggle OnToggle Off
How to study your flashcards.
Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key
Up/Down arrow keys: Flip the card between the front and back.down keyup key
H key: Show hint (3rd side).h key
A key: Read text to speech.a key
49 Cards in this Set
 Front
 Back
Measures of central tendency
Mean Median Mode 
Mean = sum of observations / number of observations
Median = observation in the middle when all observations are ordered from smallest to largest; if even, mean of middle two. Mode = observation that occurs most frequently 

Measures of dispersion
Range Standard deviation Standard error of the mean Percentile 
Range = largest  smallest observation
Standard deviation = variability of data around the mean  68% of sample pop will fall within on standard deviation  95% will fall within two Standard error of the mean = how much variability can be expected when measuring the mean from several different samples Percentile = % of a distribution below a specific value 

Frequency of an event
Incidence Prevalence 
Incidence
 Number of new events that occur in a specific time interval / population at risk at the beginning of the time interval  Represents the likelihood of developing an event in that time interval Prevalence  Number of individuals with a given disease at a given point in time / population at risk at that point in time 

Magnitude of effect

Relative risk
Odds ratio Absolute risk Number needed to treat 

Magnitude of effect
Relative risk 
Relative risk or cohort studies
= Indicence in exposed / incidence in unexposed Cohort study  Group of patients who have variable exposure to a risk factor are followed over time for outcome  Then those with and without the outcome are evaluated for some potentially causative variable  Nurses' Health Study:colon ca:fiber intake 

Magnitude of effect
Odds ratio 
Odds ratio
 = Odds an individual with a specific condition has beene xposed to a risk factor / the odds that a control has been exposed  Used in casecontrol studies  Patients with disease are identified  Compared with matched controls for exposure to a risk factor  Does not permit measurement of proportion exposed to the risk factor then developed disease > relative risk and incidence cannot be calculated  Odds ratio approximates relative risk  People with colon ca:compared with match controls:fiber intake compared 

Magnitude of effect
Absolute risk Attributable risk 
Absolute risk
Attributable risk  additional incidence of disease related to an exposure taking into account the background rate of disease = Incidence in exposed  incidence in unexposed 

Magnitude of effect
Number needed to treat 
Number needed to treat
=Reciprocal of absolute risk reduction Placebocontrolled trial with 100 patients 30 died during the study period  10 receiving active drug = 20% mortality rate  20 receiving placebo = 40% mortality rate 40%  20% = 20% = 0.2 1/0.2 = 5 = NNT 

Magnitude of effect
Absolute risk Population attributable risk 
Population attributable risk
Contribution that an exposure has on incidence of a specific disease = Attributable risk * prevalence of exposure to a risk factor  Particularly important when considering public health measures and allocation of resources to reduce incidence of disease 

Quality of measurements
Reliability Validity 
Reliability  Extent to which repeated measurements of a relatively stable phenomenon fall closely to each other.
Validity  Extent to which an observation reflects the "truth" of the phenomenon being measured. 

Diagnostic test accuracy
Sensitivity Specificity 
Sensitivity
 Number of patients with a positive test / all patients who have the disease.  Few false negatives Specificity  Number of patients with a negative test and do not have the disease / number of patients who do not have the disease  Few false positives 

Positive predictive value
Negative predictive value 
Positive predictive value  Likelihood a patient with a positive test has the disease.
Negative predictive value  Likelihood a patient with a negative test does not have the disease. Depend on prevalence of the disease within the population. 

Likelihood ratio

Likelihood ratio
 Odds of having a disease relative to the prior probability of the disease.  Independent of prevalence. Positive likelihood ratio= Sensitivity / (1specificity)  Negative likelihood ratio=(1sensitivity)/specificity 

Accuracy

Number of true positive and true negatives / total number of observations


Confidence interval

 Gives values within which there is a high probability (95%) that the true population can be found
 Narrows as the number of observations increases (ie, its variance or disperson decreases) 

Errors
Type I Type II 
Type I (Alpha)
 Probability of incorrectly concluding that there is a statistically significant difference  The number about a pvalue  p<0.05 = less than 5% chance tha thte difference course have occured by chance Type II (Beta)  Probability of incorrectly concluding that there was no statistically significant difference.  Reflects insufficient power of the study. 

Power

= 1Beta
 Ability of a study to detect a true difference  The larger the difference, the fewer observations that will be required 

Multivariate analysis
Multiple regression Logistic regression 
Multiple regression  Used when there's a continuous variable (eg, BP)
Logistic regression  Used when the outcome is dichotomous 

Survival analysis
KaplanMeier analysis Cox proportional hazards analysis 
KaplanMeier analysis  Measures the ratio of surviving patients / total number of patients at risk
 Curve depicts probability of survival  If curves are close together or cross, likely insignificant difference Cox proportional hazards analysis  Accounts for many variables that are relevant for predicting a dichotomous outcome  Permits time to be included (unlike logistic regression) 

Power in a negative study

 Power is statistically probability of avoiding a type 2 error
 Probability that a study will not mistakenly accept the null hypothesis and conclude there was no effect when there really was one  Assess the confidence interval to see if clinically important values exist within the range of the likely values represented by the confidence interval 

Graphic relative risk

[A/(A+B)]/[C/(C+D)]


Graphic odds ratio

(A'/D')/(C'/B')


Graphic
Sensitivity Specificity PPV NPV 
Sensitivity = A / (A+C)
Specificity = D / (B+D) PPV = A / (A+B) NPV = D / (C+D) 

Graphic accuracy

(A+D)/(A+B+C+D)


Statistics: Fisher's Exact Test

Statistics: Fisher's Exact Test
Used to test statistically whether there is any relation between two categorical variables where sample sizes are small 

Statistics: ANOVA

Statistics: ANOVA
Observed variance is partitioned into components due to different explanatory variables when the number of groups is > two 

Statistics: Student's ttest

Statistics: Student's ttest
Tells whether the variation between two groups is significant 

Statistics: Sensitivity

Statistics: Sensitivity
The proportion of true positives that are correctly identified by the test. 

Statistics: Specificity

Statistics: Specificity
The proportion of true negatives that are correctly identified by the test. 

Statistics: Positive predictive value

Statistics: Positive predictive value
Proportion of patients with positive test results who are correctly diagnosed. 

Statistics: Negative predictive value

Statistics: Negative predictive value
Proportion of patients with negative test results who are correctly diagnosed. 

Statistics: Logrank test

Statistics: Logrank test
Used to compare the survival experience of two or more groups of individuals Used to test the null hypothesis that there is no difference between the populations in the probably of an event (eg, death) 

Statistics: Chisquare test

Statistics: Chisquare test
 For estimating how closely an observed distribution matches an expected distribution ("goodness of fit")  Estimating whether two random variables are independent 

Statistics: Validity vs reliability

Statistics: Validity vs reliability
Validity  The extent to which we are measuring what we hope to measure Reliability  Measurement process yields consistent scores over repeat measurements 

Statistics: Onesample ttest

Statistics: Onesample ttest
The level of outcome for a group is compared to a known standard Useful for small numbers 

Statistics: Twosample ttest

Statistics: Twosample ttest
Outcome levels of two groups are compared to each other Useful for small numbers 

Statistics: Wilcoxon RankSum Test

Statistics: Wilcoxon RankSum Test
Nonparametric alternative to the twosample ttest which is based soley on the order in which the observations from the two samples fall by sorting the values as one group then ranking the ordinally and summing each groups ordinal scores 

Statistics: Accuracy

Statistics: Accuracy
The degree of conformity of a measured or calculated quantity to its actual value Degree of accuracy 

Statistics: Precision

Statistics: Precision
Reproducibility or repeatability The degree to which repeated measurements show the same or similar results Degree of reproducibility 

Statistics: Incidence

Statistics: Incidence
Annual number of people who have a case of the disease The number of specified new events during a specified period in a specified population 

Statistics: Prevalence

Statistics: Prevalence
The number of people who currently have the condition The number of cases of a disease existing in a given population at a specific period of time or at a particular moment in time 

Statistics: Oneway ANOVA

Statistics: Oneway ANOVA
A way to test the equality of three or more means at one time by using variances 

Statistics: Twoway ANOVA

Statistics: Twoway ANOVA
Extension of oneway ANOVA but with two independent variables (factors) 

Statistics: KruskalWallis test

Statistics: KruskalWallis test
Commonly used when there is one nominal variable and one measurement variable and the measurement variable does not meet the normality assumption of an ANOVA 

Statistics: Phase I trial

Statistics: Phase I trial
 Initial introduction of drug into humans  Determine the actions and side effects associated with increasing drugs 

Statistics: Phase II trial

Statistics: Phase II trial
 Controlled clinical trials to evaluate the drug's effectiveness for a paticular condition  Well controlled, closely monitored, relatively small number of patients 

Statistics: Phase III trial

Statistics: Phase III trial
 Administration of new drug to larger number of patients to determine 1 Safety 2 Efficacy 3 Appropriate dosage 

Statistics: Phase IV trial

Statistics: Phase IV trial
Postmarketing studies to delineate additional info  Risks  Benefits  Optimal use  Different doses  Different schedules  Other populations  Other stages of disease 

Statistics: Pilot

Statistics: Pilot
 Smallscale or trial run in preparation for a major study  Pretesting of a particular research instrument  "Do not take risk. Pilot test first." 