Send to:

Choose Destination
See comment in PubMed Commons below
J Clin Epidemiol. 2006 Apr;59(4):387-92.

Summary measures were a useful alternative for analyzing therapeutic clinical trial data.

Author information

  • 1Department of Clinical Epidemiology and Biostatistics, McMaster University, 1200 Main Street W., Hamilton, Ontario L8N 3Z5, Canada.



To compare use of both a pooled index and area under the curve (AUC) to the standard analysis of a randomized controlled trial of antibiotics for patients with Alzheimer's disease.


Using data from a randomized, placebo-controlled trial of antibiotics for patients with Alzheimer's disease, a pooled index of six outcome measures was constructed. Each change score was standardized by dividing by the standard deviation, and the six standardized change scores were averaged. The Standardized Alzheimer's Disease Assessment Scale cognitive subscale and the pooled index were plotted against time and the AUC was calculated.


The AUC analysis of the pooled index showed significant treatment effect over the 12-month period. In contrast, none of the individual measures showed an effect at more than one time point. The AUC analyses of the Standardized Alzheimer's Disease Assessment Scale cognitive subscale, the primary outcome of the trial, showed no significant difference over the entire 12-month period, although the original individual time-point analysis showed a difference at 6 months alone.


Pooled indices and AUC analyses may provide important insight into therapeutic effect of agents tested in randomized clinical trials of patients with Alzheimer's disease.

[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk