• Embed Doc
  • Copy Link
  • Readcast
  • Collections
  • CommentGo Back
Download
 
.com
Study Designs
n
 
How Research Is Classified
n
 
Terminology
n
 
Important epidemiologicconcepts
Descriptive Statistics
n
 
Measures of centraltendency
n
 
Measures of spread
n
 
Measures of frequency of events
n
 
Measures of Association
n
 
Terms used to describe thequality of measurements
n
 
Measures of diagnostic testaccuracy
n
 
Expressions used whenmaking inferences aboutdata
n
 
Multivariable RegressionMethods
References
MedPage Tools
Guide to Biostatistics
Here is a compilation o important epidemiologic concepts andcommon biostatistical terms used in medical research. You canuse it as a reerence guide when reading articles published on
MedPage Today 
or download it to keep near the reading standwhere you keep your print journals. For more detailed inor-mation on these topics, use the reerence list at the end o thispresentation.
Study Designs in Clinical Research
CohortstudyCross-sectionalstudyCase controlstudyYesNoNoExposure andOutcome atthe same timeDidresearcherassignexposures?Is there aComparisongroup?ObservationalstudyExposure
OutcomeExposure
OutcomeDirection of the study?DescriptiveStudyAnalyticalStudyNoExperimentalstudyYesYes
Non-Randomisedcontrolled TrialRandomisedcontrolled
IsallocationRandom?
How research is classifed
1
 
 
.com
Study Designs
n
 
How Research Is Classified
n
 
Terminology
n
 
Important epidemiologicconcepts
Descriptive Statistics
n
 
Measures of centraltendency
n
 
Measures of spread
n
 
Measures of frequency of events
n
 
Measures of Association
n
 
Terms used to describe thequality of measurements
n
 
Measures of diagnostic testaccuracy
n
 
Expressions used whenmaking inferences aboutdata
n
 
Multivariable RegressionMethods
References
Terminology
Clinical Trial
Experimental study in which the exposure status(e.g. assigned to active drug versus placebo) is determined bythe investigator.
Randomized Controlled Trial
A special type o clinical trialin which assignment to an exposure is determined purely bychance.
Cohort Study
Observational study in which subjects with anexposure o interest (e.g. hypertension) and subjects withoutthe exposure are identied and then ollowed orward in time todetermine outcomes (e.g. stroke).
Case-Control Study
Observational study that rst identies agroup o subjects with a certain disease and a control groupwithout the disease, and then looks to back in time (e.g. chartreview) to nd exposure to risk actors or the disease. This typeo study is well suited or rare diseases.
Cross-Sectional Study
Observational study that is done to ex-amine presence or absence o a disease or presence or absenceo an exposure at a particular time. Since exposure and outcomeare ascertained at the same time, it is oten unclear i the expo-sure preceded the outcome.
Case Report or Case Series
Descriptive study that reports ona single or a series o patients with a certain disease. This typeo study usually generates a hypothesis but cannot test a hy-pothesis because it does not include an appropriate comparisongroup.
Important Epidemiologic Concepts
Bias
Any systematic error in the design or conduct o a studythat results in a mistaken estimate o an exposure’s eect on risk o disease.
Selection Bias
Bias introduced by the way in which participantsare chosen or a study. For example, in a case-control study usingdierent criteria to select cases (e.g. sick, hospitalized population)versus controls (young, healthy outpatients) other than the pres-ence o disease can lead the investigator to a alse conclusionabout an exposure.
Conounding
 This occurs when an investigator alsely concludesthat a particular exposure is causally related to a disease withoutadjusting or other actors that are known risk actors or thedisease and are associated with the exposure.
 
.com
Study Designs
n
 
How Research Is Classified
n
 
Terminology
n
 
Important epidemiologicconcepts
Descriptive Statistics
n
 
Measures of centraltendency
n
 
Measures of spread
n
 
Measures of frequency of events
n
 
Measures of Association
n
 
Terms used to describe thequality of measurements
n
 
Measures of diagnostic testaccuracy
n
 
Expressions used whenmaking inferences aboutdata
n
 
Multivariable RegressionMethods
References
Descriptive Statistics
Measures o Central Tendency
Mean
equals the sum o observations divided by the number o observations.
Median
equals the observation in the center when all observa-tions are ordered rom smallest to largest; when there is an evennumber o observations the median is dened as the average o the middle two values.
Mode
equals the most requently occurring value among allobservations.
Measures o Spread
Spread (or variability) describes the manner in which data arescattered around a specic value (such as the mean). The mostcommonly used measures o spread are:
Range
is the dierence between the largest observation andthe smallest.
Standard Deviation
measures the spread o data around themean. One standard deviation includes 68% o the values in asample population and two standard deviations include 95% o the values.
Standard Error o the Mean
describes the amount o variabilityin the measurement o the population mean rom several dier-ent samples. This is in contrast to the standard deviation whichmeasures the variability o individual observations in a sample.
Percentile
equals the percentage o a distribution that is belowa specic value. As an example, a child is in the 80th percentileor height i only 20% o children o the same age are taller thanhe is.
Interquartile Range
reers to the upper and lower bound-ary dening the middle 50 percent o observations. The upperboundary is the 75th percentile and the lower boundary is the25th percentile.
Measures o Frequency o Events
Incidence
The number o new events (e.g. death or a particulardisease) that occur during a specied period o time in a popula-tion at risk or developing the events.
Incidence Rate
A term related to incidence that reports thenumber o new events that occur over the sum o time indi-viduals in the population were at risk or having the event (e.g.
of 00

Leave a Comment

You must be to leave a comment.
Submit
Characters: ...
You must be to leave a comment.
Submit
Characters: ...