Bio Medical Literature

Critical Evaluation

1. Critical Evaluation
of
BiomedicalLiterature
SreenuThalla
Clinical Pharmacologist

2. •Reviewingthe‘BiomedicalLiterature’posesagreatchallengeto
theclinicalprofessionals.
•Evaluatingascientificarticleisacomplextask.
•Knowledgeofthestandardanatomyofanarticleandidiosyncrasy
ofvarioustypesofstudieswillassistthereadertoreviewthe
‘BiomedicalLiterature’efficiently
Introduction

3. Biomedicalresearch
•Itisthebasicresearch,appliedresearchortranslational
researchesconductedtoaidandsupportthebodyofknowledge
inthefieldofClinicalPractice.
Literatureevaluation
•Itistheprocessofreadingandevaluatingarticle,journal,
literature,andscientificstudyinasystematicwaytoreachata
conclusionthatonecaninterpretitselfwhetherthestudyresults
arescientificallyprovedandapplicabletoClinicalPractice.

4. ScientificStudy
•Itisawrittenandpublishedreportthatdescribesoriginalresearch
results.
•Scientificstudiesarewritteninastylethatisexceedinglyclearand
concise.
•Awell-writtenscientificliteratureexplainstheauthor'sinterest.
•Authorshouldsummarizeandgiveanideaonpreviousresearch,
andReadershouldbeabletodistinguishbetweenprevious
researchandtheactualcurrentstudy

5. CriticalAppraisal
•Itistheprocessofsystematicallyexaminingresearchevidenceto
assessitsvalidity,results,andrelevance”.
•CriticalAppraisalisawaytoassessthescientificvalueand
trustworthinessofastudypublishedinaresearcharticle.
•Ithelpspeopleindevelopingnecessaryskillstomaketrue
judgmentsofascientificliterature

6. SelectingtheArticle
•Primarystepinevaluationofaliteratureistoselectanarticle
whichhasagreaterimpactinclinicalpractice.
•Initially,readthetitle,authorsandabstract.
•Thetitleshouldbecomprehensivethatthereadercanefficiently
analyzethearticle’spotentialanditsimportanceincurrentclinical
practice.
•Ifnot,readercanrejectitandmoveontothenextarticle.

7. ReadingtheLiterature
•Reviewingbeginswithreadingandunderstandingtheabstract
orshortsummarythatgivesabriefbackgroundaboutthe
research.
•Initialreadinggivestheconceptofobjectives,methodology,
resultsandproposedsignificanceofthestudy.
•Aproperunderstandingofresearchstudy’snatureismustfora
reader

8. BiomedicalLiteratureincludescriticalappraisalofthe
followingcontents:
➢Title
➢Abstract
➢Introduction
➢Objective
➢MaterialsandMethods
➢StudyDesigns
➢Bias
➢Statistics
➢StudyResultsandAnalysis
➢Discussion
➢Conclusion
➢References

9. Title
•‘Title’describesthebreadthanddepthofthecurrentstudyand
indicatesthemethodologyused.
•Itisthelimitedpossiblewordsthatadequatelydescribecontents
ofthestudy.
•Thetitleofanarticleshouldbebrief,definiteandconciseand
shouldcatchtheattentionofthereadersinterestedinthestudy

10. EvaluationofTitle
•Basedonthetitleitselfreadercannotreviewordiscardthe
study.
•Titleshouldnotcontainabbreviations,proprietarynames,
chemicalformulae,andjargons.
•Thetitleshouldinformtherealsubjectofthearticle.
•Titleshouldnotreflectitscontent.
•Firstimpressionisthebestimpression;thetitleshouldbe
specificandstudiedwell.
•Titleshouldnotindicateauthor’spreferenceforanyspecific
subject

11. Abstract
•Anabbreviatedaccuraterepresentingthecontentsofadocument,
mainlypreparedbyitsauthor(s)forpublicationinit.
•Abstractcanbedefinedasasummaryoftheinformationina
document.
•Asynopsis(notmorethan250words)shouldbementionedbefore
introductioninthearticle.

12. EvaluationofAbstract
•‘Abstract’shouldoutlineabrief
•Summaryofeachsection;Introduction,objectives,scopeof
investigation,materialsandbriefmethodology,results,and
conclusionwhichindicatesstudyfindings.
•Abstractdoesnotprovidecompleteinformationaboutthestudy
andshouldnotbeusedalonetoevaluatethestudy

13. Introduction
•Itservestwopurposesinthestudy,creatingreadersinterestin
subjectandprovidingthemwithenoughinformationto
understandthestudy.
EvaluationofIntroduction
•Introductionshouldbepresented,withallpossibleclarity,the
natureandscopeoftheprobleminvestigated.
•Itshouldprovidepertinentliteraturetoorientthereader.
•Itshouldexplainthereasonwhythecurrentresearchisneeded?

14. Objective
•‘Objective’ofastudyiswhattheauthoristryingtoachieve.
•Itisaspecific,clearandsuccinctstatementofintendedoutcomes
fromresearch.
•Objectiveshouldbestatedinaclearandconcisemanner.
EvaluationofObjective
➢Establishingnewhealthprograms
➢Implementingnewpolicies
➢Tryingtosettleacontroversy
➢Showingthevalidityofanewtechnique
➢Openingupanewfieldofinquiry.

15. MaterialsandMethods
•Methodologyisoneofthemostimportantsectionsofastudy.
•Itspurposeistodescribethemethodsusedintheexperimentand
materialsbywhichtheexperimentwascarriedout.
•Thedescriptionofthissectionshouldbedetailedenoughtoallow
otherresearcherstoreplicatethework.
EvaluationofStudyMethods
•Methodsused,andtheirdescriptionshouldbeelucidated.
•Whatarethedoseregimens,routeandfrequencyofadministration,
theoveralllengthofthestudytobementioned.
•Whatwasthelengthofwash-outperiod

16. StudyDesigns
•Thefirstpartofthissectionisgenerallyanoverviewofthetypeof
studydesignthatisutilizedindoingresearch.
•Asoundstudydesignsupportsstudyconclusionandresult.
•Studydesignshouldbeclearandprovideenoughdetailssothat
potentialreadercanrepeattheresearch

17. VarioustypesofStudyDesigns
Observationalstudies
•Datacollectionfromoneormoregroupofsubjects
•Observationalstudiesmaybeprospectiveorretrospective
Single blind
•Eithersubjectsorinvestigatorsareunawareoftreatmentallocation
Doubleblind
•Neithersubjectsnorinvestigatorsareawareoftreatmentallocation.
Tripleblind
•Subjectsandinvestigatorsareunawareoftreatmentallocation;
anothergroupinvolvedwithinterpretationofdataisalsounaware
oftreatmentallocation.

18. Parallelstudy
•Allsubjectsreceiveonlyonetreatment
Prospective
•Dataiscollectedforwardintimefromthestartofthestudy.
Retrospective
•Historicaldata(i.e.,datareferringtopastevents)iscollected.
Cohortstudies
•Cohortstudiesconsistofprospectiveobservationofoneormore
groupswithcertaincharacteristics.
Randomizedcontroltrial
•Subjectsarerandomlyallocatedtoeitheraninterventiongroupor
controlgroup.
•Randomizedcontrolledtrialsaredescribedasthe“goldstandard”
inclinicalresearch.

19. EvaluationofStudyDesigns
•Thestudydesignselectedbyaninvestigatorshouldbesoundand
likelytoanswertheresearchquestions.
•Author(s)mustdescribestudypopulationwellenoughsothatthe
readerisabletovisualizethesamplepopulationpreciselyunder
investigation

20. Bias
•Itisasystemicvariationinwhichtreatmentgroupsunderstudy
aretreatedormeasureddifferentlyonaconsistentbasis.
•Biascanmisleadonetogetintoanerroneousoutcome.
•Thereadershouldbeabletofindoutthesourceofbiasandits
influenceonthefinaloutcomeofstudy

21. Types of Bias
Missingclinicaldatabias
•Certainclinicaldatamaybemissingbecausetheywerenormal,
negativeornevermeasured.
Withdrawalbias
•Patientswhowithdrawfromastudymaydifferfromthosewho
remain.
Samplesizebias
•ToosmallSamplesareinsignificant;sampleswhicharetoolargeare
provedtobehelpful.
Instrumentbias
•Defectsinthecalibrationormaintenanceofinstrumentsmaylead
tosystematicdeviationsinresults.

22. Statistics
•Knowledgeof‘Statistics’canhelpanindividualtoevaluate
whetherthestatisticaltestsusedinastudyareappropriateor
not.
•Differenttypesofdata(orvariables)areencounteredinstatistics.
•Errorsinstatisticalanalysisofdataleadtoinvalid
result/conclusion

23. TYPES OFDATA TWO
COMPARISI
ON
GROUPS
MORE THAN TWO
COMPARISION
GROUPS
UNPAIRED
DATA
PAIRED
DATA
UNPAIRED
DATA
Nominal Chisquare McNemar Chisquare
Ordinal Mann-WitneyU
test
Wilcoxon Kruskal-Wallis
Parametric Studentt-testPaired student
test
Analysis-of
variance
(ANOVA)

24. EvaluationofStatistics
•Readersshoulddeterminewhetherappropriatestatisticalmethods
wereusedfordataanalysis.
•Useofinappropriatemethodswillresultsinmisleadingconclusion.
•Methodsectionofanyscientificliteratureshouldincludea
summarydescriptionofthestatisticalteststhatwereusedto
evaluatedata.
•Qualitativeandquantitativedataareexamineddifferently.

25. StudyResultsandAnalysis
•‘Results’shouldbedescribedandpresentedinfigures,tablesand
charts,astheyaretheheartofthescientificliterature.
•Figures,tablesandchartswillassistthereaderindecidingwhether
itisworthtoreadtherestofthearticleortodiscardit.
•Aproperlyconductedstudyshouldpresentdataonsubjects
involvedinthestudy.
•Allthedatacollectedinthemethodsectionshouldalsobe
presented

26. EvaluationofStudyResultsandAnalysis
•Readershouldhaveaproperunderstandingofstudyandshould
evaluateclinicalandstatisticalreliabilityofthestudy
•Sometimesauthorstrytopresentresultsinaconfusingway,which
mostlikelyreflectshaphazardousdatacollectionandlackof
clearlydefinedstudyobjectives.
•Arethenegativeresultsbeenquoted?Incaseofanynegative
resultsthoseshouldbequotedandthelimitationshavetobe
specified.

27. Discussion
•‘DiscussionSection’ofastudyprovidesanopportunityforthe
authortointerpretresultsandexplaintheirclinicalimportanceby
relatingorcomparingwithpreviousworkorpractice.
Conclusion
•‘Conclusion’istheauthor’sgeneratedinferences,opinionsand
hypothesesaboutresults.
•Thissectionshouldcontainviewsthattheauthordrawsfromdata
obtainedbythestudy

29. EvaluationofDiscussionandConclusion
•Istheconclusionover-generalized?
•The‘Conclusion’mustbeclearandunderstandabletothereader.
•Conclusionmustbeconsistentwithstudyobjectivesandjustified
byresults.
•Conclusionshouldnotbeamatterofdispute.
•Howdoestheresearchfitintothecontextofitsconclusion?
•Itshouldgivetheanswerofthestudyobjectiveforwhichclaim
wasmadepriortostudy.
•Readersmustunderstandtherelationshipbetweenthedataand
theconclusions.

30. References
•Whilewritingarticle,authorsalwaysrefertosomeinformation
fromothersources.
•Allthesesourcesarelistedin‘ReferenceSection’,sometimes
referredtoas‘Bibliography’.

31. EvaluationofReferences
•Arethereferencesgiven?Whetherappropriateandadequate
referencesareusedinthestudy?
•Arethereferencesquotedappropriatelyintheresearcharticle?
•Arethereferencesgivenrecentandimportant?
•Whatisthesizeof‘ReferenceSection’?
•Howthereferencesareusedforsupport,rebuttaletc.?
•Dothereferencesmatchcitationsinthetext?
•Authorsshouldavoidcitingtheirownresearcheffortsand
publication.