Upload
sharleen-neal
View
220
Download
3
Tags:
Embed Size (px)
Citation preview
How to Read How to Read an Article an Article
about a about a Systematic Systematic
ReviewReview
Maribeth Chitkara, MDMaribeth Chitkara, MDRachel Boykan, MDRachel Boykan, MD
Stony Brook Long Island Stony Brook Long Island Children’s HospitalChildren’s Hospital
Session ObjectivesSession Objectives
Demonstrate understanding of general EBM principles.
Demonstrate understanding of validity measures specific to articles about a systematic review.
Critically appraise an article about a systematic review and describe its clinical applicability.
Clinical ScenarioClinical Scenario
You are on inpatient service and have been You are on inpatient service and have been taking care of a 4 month old with bronchiolitis. taking care of a 4 month old with bronchiolitis. The infant was initially admitted to the ICU for The infant was initially admitted to the ICU for respiratory distress, but was transferred to the respiratory distress, but was transferred to the general pediatric ward on the 4general pediatric ward on the 4thth hospital day hospital day with resolved increased work of breathing but a with resolved increased work of breathing but a persistent O2 requirement. Your attending persistent O2 requirement. Your attending suggests starting the child on systemic suggests starting the child on systemic corticosteroids because she is getting inpatient corticosteroids because she is getting inpatient with the prolonged length of stay. You are with the prolonged length of stay. You are skeptical, and decide to see where the medical skeptical, and decide to see where the medical evidence lies. evidence lies.
Systematic ReviewsSystematic Reviews
A summary of the medical A summary of the medical literature literature
Uses explicit methods to Uses explicit methods to systematically search, systematically search, critically appraise, and critically appraise, and synthesize the world synthesize the world literatureliterature
Systematic ReviewsSystematic Reviews
GoalsGoals::Minimize BiasMinimize Bias
randomized trialsrandomized trialsseeks multilingual published and seeks multilingual published and
unpublished reportsunpublished reportsMinimize random errorMinimize random error
by amassing very large numbersby amassing very large numbers
Systematic ReviewsSystematic Reviews
MayMay include some statistical include some statistical method for combining the method for combining the results of individual studiesresults of individual studies
““Meta-analyses”Meta-analyses”
Appraisal of Systematic Appraisal of Systematic ReviewsReviews
Is the Systematic Review Is the Systematic Review ValidValid??
What are the What are the ResultsResults??
Can I Can I ApplyApply the results to my the results to my patient?patient?
Is the Systematic Review Is the Systematic Review ValidValid? ?
Is this a systematic review of Is this a systematic review of randomized trials?randomized trials?
Were the results consistent Were the results consistent from study to study?from study to study?
Is the Systematic Review Is the Systematic Review ValidValid??
Randomized trials reduce biasRandomized trials reduce bias
Combining all relevant Combining all relevant randomized trials further randomized trials further reduces bias and random errorreduces bias and random error
Provides the highest level of Provides the highest level of evidence currently achievableevidence currently achievable
Is the Systematic Review Is the Systematic Review ValidValid??
Systematic Reviews of Non-Systematic Reviews of Non-Randomized Trials Can:Randomized Trials Can:
Compound the problems of Compound the problems of individually misleading trialsindividually misleading trials
Produce a lower quality of Produce a lower quality of evidenceevidence
Is the Systematic Review Is the Systematic Review ValidValid??
If a systematic review includes If a systematic review includes both both Randomized and Non-Randomized and Non-Randomized Trials, Randomized Trials, avoid itavoid it unless these types of trials are unless these types of trials are separated in the analyses.separated in the analyses.
Is the Systematic Review Is the Systematic Review ValidValid??
Does the study have a Methods Does the study have a Methods SectionSection
Seek a description of Seek a description of howhow the the investigators found all the investigators found all the relevant trialsrelevant trials
If not present If not present STOP STOP
The Methods Section The Methods Section
Seek reassurance that the Seek reassurance that the search went beyond standard search went beyond standard bibliographic databasesbibliographic databases
Should include hand-searching Should include hand-searching journals, conference journals, conference proceedings, theses, databanks proceedings, theses, databanks of pharmaceutical companies, of pharmaceutical companies, and contacting authors of and contacting authors of published articlespublished articles
The Methods SectionThe Methods Section
Negative trials are less likely Negative trials are less likely to be submitted and selected to be submitted and selected for publicationfor publication
A Systematic Review A Systematic Review restricted to restricted to published published trials trials can result in a false-positive can result in a false-positive conclusionconclusion
Restriction to a single Restriction to a single language can bias the language can bias the conclusionsconclusions
Is the Systematic Review Is the Systematic Review ValidValid??
How did the investigators assess How did the investigators assess the validity of the individual the validity of the individual studies?studies?Randomized?Randomized?Follow-Up sufficient?Follow-Up sufficient?Intention to treat analysis?Intention to treat analysis?Blinded?Blinded?Were the groups similar at the start?Were the groups similar at the start?Were the groups treated equally, Were the groups treated equally,
apart from the experimental tx?apart from the experimental tx?
What are the What are the ResultsResults? ? Forest PlotsForest Plots
Each study is a lineEach study is a line Black square is Black square is
point estimatepoint estimate Size of square is Size of square is
weightweight Line is 95% Line is 95%
confidence intervalconfidence interval Pooled analysis is Pooled analysis is
combined effect combined effect and CIand CI
What are the What are the ResultsResults? ? Funnel Plots and Publication BiasFunnel Plots and Publication Bias
Vertical AxisVertical Axis Measure of study Measure of study
precisionprecision Higher up: more precise, Higher up: more precise,
smaller CIsmaller CI
Horizontal AxisHorizontal Axis Measures treatment Measures treatment
effecteffect
Center LineCenter Line Combined treatment Combined treatment
effecteffect
No publication bias: No publication bias: Inverted funnel shapeInverted funnel shape
Can I Can I ApplyApply the results to my the results to my patient?patient?
Is our patient so different from those in Is our patient so different from those in the study that the results cannot apply?the study that the results cannot apply?
Is the treatment feasible in our setting?Is the treatment feasible in our setting?What are our patient’s potential benefits What are our patient’s potential benefits
and harms from the therapy?and harms from the therapy?What are our patient’s values and What are our patient’s values and
preferences for both the outcome we preferences for both the outcome we desire and the potential side-effects of desire and the potential side-effects of treatment?treatment?
Can I Can I ApplyApply the results to my the results to my patient?patient?
A Systematic Review provides an A Systematic Review provides an overall, average effect of therapy which overall, average effect of therapy which may be derived from a heterogeneous may be derived from a heterogeneous population.population.
SR’s may provide precise information on SR’s may provide precise information on sub-groups that may help individualize sub-groups that may help individualize evidence to our own patients.evidence to our own patients.
BUT...BUT...
Can I Can I ApplyApply the results to my the results to my patient?patient?
A qualitative difference in treatment efficacy A qualitative difference in treatment efficacy among subgroups is likely only when ALL the among subgroups is likely only when ALL the following questions can be answered “YES”:following questions can be answered “YES”:
Does it really make biological and clinical sense?Does it really make biological and clinical sense?
Is the qualitative difference both clinically and Is the qualitative difference both clinically and statistically significant?statistically significant?
Was it hypothesized before the study began (rather Was it hypothesized before the study began (rather than a product of dredging the data) and has it than a product of dredging the data) and has it been confirmed in other independent studies?been confirmed in other independent studies?
Was it one of just a few subgroup analyses carried Was it one of just a few subgroup analyses carried out in this study?out in this study?
Practice CasesPractice Cases
Review the case scenarios in your handoutReview the case scenarios in your handout
Articles for each clinical question have been Articles for each clinical question have been providedprovided
Use worksheet to help guide your appraisal Use worksheet to help guide your appraisal for each article for each article
Practice Case Practice Case ReferencesReferences
Case 1: Case 1: Garrison MM. et al. Systemic Corticosteroids in Infant Bronchiolitis: A Meta-analysis. Pediatrics, 2000; 105:e44.
Case 2: Case 2: Wooditch AC, Aronoff SC. Effect of Initial Corticosteroid Therapy on Coronary Artery Aneurysm Formation in Kawasaki Disease: A Meta-analysis of 862 Children. Pediatrics, 2005;116:989.
Case 3: Case 3: Figuls MR, Giné-Garriga M, Granados Rugeles C, Perrotta C. Chest physiotherapy for acute bronchiolitis in paediatric patients between 0 and 24 months old. Cochrane Database of Systematic Reviews 2012.
EBM EBM References/ResourcesReferences/Resources
Friedland DJ, Shilpak MG, Subak LL, Bent Friedland DJ, Shilpak MG, Subak LL, Bent SW, Mendelson T. Evaluating Integrative SW, Mendelson T. Evaluating Integrative Literature. Literature. Evidence-based medicine: a Evidence-based medicine: a framework for clinical practiceframework for clinical practice, Appleton , Appleton and Lange, 1998: 222-227. and Lange, 1998: 222-227.
Straus SE, Glasziou P, Richardson WS, Haynes RB. Therapy: Reports of Systematic Reviews. Evidence-based medicine: how to practice and teach it, 4th ed, Churchill Livingston Elsevier, 2011: 98-108.