14.01.2015 Views

Evidence-Based Medicine

Evidence-Based Medicine

Evidence-Based Medicine

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

etween – 1 and +1. When r is +1, there is a prefect positive relationship (when<br />

one variable increases, so does the other, and the proportionate difference<br />

remains constant). When r is –1 there is a perfect negative relationship (when<br />

one variable increases the other decreases, or vice versa, and the proportionate<br />

difference remains constant). This, however, does not rule out a relationship —<br />

it just excludes a linear relationship.<br />

Cost-Benefit Analysis<br />

Converts effects into the same monetary terms as the costs and compares<br />

them.<br />

Cost-Effectiveness Analysis<br />

Coverts effects into health terms and describes the costs for some additional<br />

health gain (e.g. cost per additional MI prevented).<br />

Cost-Utility analysis<br />

Converts effects into personal preferences (or utilities) and describes how<br />

much it costs for some additional quality gain (e.g. cost per additional quality<br />

life-year, or QALY).<br />

Crossover randomised trial<br />

A trial in which participants receive one treatment and have outcomes measured,<br />

and then receive an alternative treatment and have outcomes measured again.<br />

The order of treatments is randomly assigned. Sometimes a period of no<br />

treatment is used before the trial starts and in between the treatments (washout<br />

periods) to minimise interference between the treatments (carry over effects).<br />

Interpretation of the results from crossover randomised controlled trials (RCTs)<br />

can be complex.<br />

Crossover studies have the risk that the intervention may have an effect after<br />

it has been withdrawn, either because the washout period is not long enough<br />

or because of path dependency. A test for evidence of statistically significant<br />

heterogeneity is not sufficient to exclude clinically important heterogeneity. An<br />

effect may be important enough to affect the outcome but not large enough to<br />

be significant.<br />

Crossver Study Design<br />

The administration of two or more experimental therapies one after the other in<br />

a specified or random order to the dame group of patients.<br />

-32-

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

Saved successfully!

Ooh no, something went wrong!