Amsterdam 2008
Upcoming SlideShare
Loading in...5
×
 

Like this? Share it with your network

Share

Amsterdam 2008

on

  • 208 views

 

Statistics

Views

Total Views
208
Views on SlideShare
208
Embed Views
0

Actions

Likes
0
Downloads
0
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Amsterdam 2008 Presentation Transcript

  • 1. How to pass statistics review in Acta Orthopaedica Jonas Ranstam PhD
  • 2. Statistics is about calculating p-values
  • 3. Statistics is about calculating p-valuesStatistics is about rational interpretation of incomplete and imperfect data
  • 4. The responsibilities of a statistical reviewer“To make sure that the authors spell out for thereader the limitations imposed upon theconclusions by the design of the study, thecollection of data, and the analyses performed.”Shor S. The responsibilities of a statistical reviewer. Chest 1972;61:486-487.
  • 5. The responsibilities of a statistical reviewer“To make sure that the authors spell out for thereader the limitations imposed upon theconclusions by the design of the study, thecollection of data, and the analyses performed.”And to help authors improve their manuscriptsby suggesting possible solutions to theirproblems.Shor S. The responsibilities of a statistical reviewer. Chest 1972;61:486-487.
  • 6. Hint for a successful review
  • 7. Hint for a successful review Comply with the Uniform Requirements forManuscripts Submitted to Biomedical Journals
  • 8. Statistical Methods“Describe statistical methods with enough detail toenable a knowledgeable reader with access to theoriginal data to verify the reported results.”
  • 9. Statistical Methods“Describe statistical methods with enough detail toenable a knowledgeable reader with access to theoriginal data to verify the reported results.”Required for analytical methods (hypothesis tests andconfidence intervals), do not describe how data arepresented when this is self-evident.
  • 10. Results“When possible, quantify findings and present themwith appropriate indicators of measurement error oruncertainty (such as confidence intervals).”
  • 11. Results“When possible, quantify findings and present themwith appropriate indicators of measurement error oruncertainty (such as confidence intervals).”Uncertainty indicators (p-values and confidence inter-vals) are necessary for generalization of results beyondexamined patients, not for case reports or “qualitative”studies.
  • 12. Results“Avoid relying solely on statistical hypothesis testing,such as the use of P values, which fails to conveyimportant information about effect size.”
  • 13. Results“Avoid relying solely on statistical hypothesis testing,such as the use of P values, which fails to conveyimportant information about effect size.”Clinical and statistical significance are two completelydifferent things. Present clinically interpretable effectestimates and their precision (confidence intervals orp-values).
  • 14. P-values vs. confidence intervals P-value Conclusion from confidence intervals Statistically and clinically significant effect p < 0.05 p < 0.05 Statistically, but not necessarily clinically, significant effect n.s. Inconclusive n.s. Neither statistically nor clinically significant effect p < 0.05 Statistically significant reversed effectEffect 0 Clinically significant effect
  • 15. Results“Where scientifically appropriate, analyses of thedata by variables such as age and sex should beincluded.”
  • 16. Results“Where scientifically appropriate, analyses of thedata by variables such as age and sex should beincluded.”Observational studies usually require adjustment forknown and suspected confounding factors.
  • 17. Results“Where scientifically appropriate, analyses of thedata by variables such as age and sex should beincluded.”Observational studies usually require adjustment forknown and suspected confounding factors.Automatic stepwise regression is not an adequatemethod for selecting adjustment factors.
  • 18. Clinical trials“The ICMJE member journals will require, as acondition of consideration for publication in theirjournals, registration in a public trials registry.”“The ICMJE recommends that journals publish the trialregistration number at the end of the Abstract.”
  • 19. Clinical trials“The ICMJE member journals will require, as acondition of consideration for publication in theirjournals, registration in a public trials registry.”“The ICMJE recommends that journals publish the trialregistration number at the end of the Abstract.”Many trials are registered with Clinicaltrials.gov,see www.clinicaltrials.gov.Note: the registers of national medical products agencies and the Eudract are not public.
  • 20. Clinical trials“When reporting experiments on human subjects,authors should indicate whether the proceduresfollowed were in accordance with the ethicalstandards of the responsible committee on humanexperimentation (institutional and national) and withthe Helsinki Declaration of 1975, as revised in 2000(5).”
  • 21. Clinical trials“When reporting experiments on human subjects,authors should indicate whether the proceduresfollowed were in accordance with the ethicalstandards of the responsible committee on humanexperimentation (institutional and national) and withthe Helsinki Declaration of 1975, as revised in 2000(5).”Example: The trial was approved by the ethicscommittee of ... and performed in compliance with theHelsinki Declaration of 1975, as revised in 2000.
  • 22. Clinical trials“For reports of randomized controlled trials authorsshould refer to the CONSORT statement.”
  • 23. Clinical trials“For reports of randomized controlled trials authorsshould refer to the CONSORT statement.”The CONSORT Statement is an evidence-based,minimum set of recommendations for reporting RCTs.http://www.consort-statement.orgInclude the CONSORT checklist and flowchart whensubmitting a manuscript.Include to Acta, and many other journals, also a copyof the study protocol.
  • 24. Clinical trialsInternational regulatory guidelinesTopic E9 - Statistical Principles for Clinical TrialsEMEA Points to consider - baseline covariates - missing data - multiplicity issues - etc.These guidelines can all be found on the internet.
  • 25. General statistical suggestionsMany methods require independent observationsA patients multiple observations are dependent andshould not be analysed using methods requiringindependence, e.g. chi-squared test, Students t-test,Mann-Whitney U-test, ANOVA.Unless the statistical methods can deal with dependentdata, analyse patients, not knees, hips, shoulders, etc.
  • 26. General statistical suggestionsAvoid vague and ambiguous expressions- significant clinically or statistically?- no difference statistically insignificant?- statistical difference statistically significant?- matched selected or just comparable?- correlation relation, regression?- etc.
  • 27. Thank you for your attention!