10.08.2013 Views

Report in English with a Dutch summary (KCE reports 45A)

Report in English with a Dutch summary (KCE reports 45A)

Report in English with a Dutch summary (KCE reports 45A)

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

<strong>KCE</strong> <strong>reports</strong> vol.45 Screen<strong>in</strong>g for Colorectal Cancer: Appendices 195<br />

This non-uniform consensus does not represent a major<br />

disagreement, rather it recognizes that given imperfect<br />

<strong>in</strong>formation, <strong>in</strong>stitutions may adopt different approaches. A<br />

Category 2B designation should signal to the user that more<br />

than one approach can be <strong>in</strong>ferred from the exist<strong>in</strong>g data.<br />

Category 3: <strong>in</strong>clud<strong>in</strong>g the recommendation has engendered a<br />

major disagreement among the panel members. The level of<br />

evidence is not pert<strong>in</strong>ent <strong>in</strong> this category, because experts can<br />

disagree about the significance of high level trials. Several<br />

circumstances can cause major disagreements. For example, if<br />

substantial data exist about two <strong>in</strong>terventions but they have<br />

never been directly compared <strong>in</strong> a randomized trial, adherents<br />

to one set of data may not accept the <strong>in</strong>terpretation of the other<br />

side's results. Another situation result<strong>in</strong>g <strong>in</strong> a Category 3<br />

designation is when experts disagree about how trial data can be<br />

generalized. A Category 3 designation alerts users to a major<br />

<strong>in</strong>terpretation issue <strong>in</strong> the data and directs them to the<br />

manuscript for an explanation of the controversy.<br />

National Health & Medical Research Council (NHMRC Australia)<br />

Levels of evidence<br />

Strength of recommendations<br />

I - Evidence obta<strong>in</strong>ed from a systematic review of all relevant<br />

randomised controlled trials<br />

II - Evidence obta<strong>in</strong>ed from at least one properly designed<br />

randomised controlled trial<br />

III-I - Evidence obta<strong>in</strong>ed from well-designed pseudo randomised<br />

controlled trials (alternate allocation or some other method)<br />

III-2 - Evidence obta<strong>in</strong>ed from comparative studies (<strong>in</strong>clud<strong>in</strong>g<br />

systematic reviews of such studies) <strong>with</strong> concurrent controls and<br />

allocation not randomised, cohort studies, casecontrol studies,<br />

or <strong>in</strong>terrupted time series <strong>with</strong> a control group<br />

III-3 - Evidence obta<strong>in</strong>ed from comparative studies <strong>with</strong><br />

historical control, two or more s<strong>in</strong>gle arm studies, or<br />

<strong>in</strong>terrupted time series <strong>with</strong>out a parallel control group<br />

IV - Evidence obta<strong>in</strong>ed from case series, either post-test or pretest/post-test.<br />

The strength of recommendations are determ<strong>in</strong>ed by a expert advisory panel<br />

tak<strong>in</strong>g <strong>in</strong>to account the level of evidence, quality of studies, size of effect and<br />

cl<strong>in</strong>ical importance for all the <strong>in</strong>cluded studies, and ranges from Strongly<br />

recommended to Strongly not recommended . These levels of<br />

recommendation are modified from The Canadian Task Force on the Periodic<br />

Health:<br />

Strongly recommended : cl<strong>in</strong>ically significant level I <strong>in</strong> favour<br />

of cl<strong>in</strong>ical question strongly recommended <strong>in</strong> favour.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!