Beschreibung:
<jats:sec><jats:title>Objectives</jats:title><jats:p>To highlight the utility of using an effect size analysis to communicate the effectiveness of treatment interventions.</jats:p></jats:sec><jats:sec><jats:title>Design</jats:title><jats:p>Secondary analysis.</jats:p></jats:sec><jats:sec><jats:title>Setting</jats:title><jats:p>Previously published systematic review on cholinesterase inhibitors (Ch<jats:styled-content style="fixed-case">EI</jats:styled-content>s) in <jats:styled-content style="fixed-case">A</jats:styled-content>lzheimer's disease.</jats:p></jats:sec><jats:sec><jats:title>Participants</jats:title><jats:p>Individuals with mild to moderate <jats:styled-content style="fixed-case">A</jats:styled-content>lzheimer's disease.</jats:p></jats:sec><jats:sec><jats:title>Intervention</jats:title><jats:p>Six‐month randomized controlled trials involving a placebo group and a Ch<jats:styled-content style="fixed-case">EI</jats:styled-content> group (donepezil, galantamine, or rivastigmine).</jats:p></jats:sec><jats:sec><jats:title>Measurements</jats:title><jats:p>Cognitive function was assessed according to performance on the cognition subscale of the <jats:styled-content style="fixed-case">A</jats:styled-content>lzheimer's <jats:styled-content style="fixed-case">D</jats:styled-content>isease <jats:styled-content style="fixed-case">A</jats:styled-content>ssessment <jats:styled-content style="fixed-case">S</jats:styled-content>cale (<jats:styled-content style="fixed-case">ADAS</jats:styled-content>‐Cog). Global <jats:styled-content style="fixed-case">F</jats:styled-content>unction was quantified using the <jats:styled-content style="fixed-case">C</jats:styled-content>linician's <jats:styled-content style="fixed-case">I</jats:styled-content>nterview‐<jats:styled-content style="fixed-case">B</jats:styled-content>ased <jats:styled-content style="fixed-case">I</jats:styled-content>mpression of <jats:styled-content style="fixed-case">C</jats:styled-content>hange—<jats:styled-content style="fixed-case">P</jats:styled-content>lus (<jats:styled-content style="fixed-case">CIBIC</jats:styled-content>‐Plus). Harm was defined as withdrawal from a trial because of an adverse event. Several effect size indices were computed based on these domains: the success rate difference (<jats:styled-content style="fixed-case">SRD</jats:styled-content>), the harm rate difference (<jats:styled-content style="fixed-case">HRD</jats:styled-content>), the number needed to treat (<jats:styled-content style="fixed-case">NNT</jats:styled-content>) or harm (<jats:styled-content style="fixed-case">NNH</jats:styled-content>), and the area under the curve (<jats:styled-content style="fixed-case">AUC</jats:styled-content>). Harm:benefit ratios were also computed to compare effect size indices across domains of function.</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>In terms of benefit, the <jats:styled-content style="fixed-case">NNT</jats:styled-content> for cognition ranged from 4 to 14 (corresponding <jats:styled-content style="fixed-case">AUC</jats:styled-content> values: 0.64–0.54), and the <jats:styled-content style="fixed-case">NNT</jats:styled-content> for global function ranged from 6 to 100 (corresponding <jats:styled-content style="fixed-case">AUC</jats:styled-content> 0.59–0.51). In terms of harm, the <jats:styled-content style="fixed-case">NNH</jats:styled-content> ranged from 6 to 20 (corresponding <jats:styled-content style="fixed-case">AUC</jats:styled-content> 0.58–0.53). Only one of the four studies had favorable harm:benefit ratios in both the cognition and global function domains.</jats:p></jats:sec><jats:sec><jats:title>Conclusion</jats:title><jats:p>Effect size indices should be reported in clinical trials because they provide important insight into the clinical meaningfulness of results. Additional benefit is gained by comparing effect size indices across domains of function to reveal harm:benefit ratios.</jats:p></jats:sec>