JLE

Cahiers Santé Médecine Thérapeutique

MENU

Alternative analyses of randomised trials: average gains in life expectancy or event-free time as patient-centred measures of therapeutic benefit Volume 31, issue 1, Janvier-Février 2022

Figures


  • Figure 1.

  • Figure 2.

  • Figure 3.

  • Figure 4.

  • Figure 5.

  • Figure 6.
Authors
1 Département de cardiologie, centre hospitalier universitaire Henri Mondor, Créteil, France
2 Inserm, CIC 1412, centre hospitalier universitaire de Brest, Brest, France
3 Tulane University Heart and Vascular Institute, 1430 Tulane Avenue, SL-48, New Orleans, LA, 70112, USA.
4 Clinical Pharmacology and Pharmacy, Department of Public Health, University of Southern Denmark, Odense, Denmark
5 Department of Clinical Biochemistry and Pharmacology Odense University Hospital, Denmark
Tirés à part : PV Ennezat

The impact of randomised controlled trials is highly dependent on the presentation of results.

Objective

Traditionally, the results of randomised controlled trials are presented as relative risk reductions or hazard ratios. These presentations of results are the most favourable to the acceptance and marketing of the drug. In addition to the classic absolute risk reduction, which estimates the number of patients that need to be treated to prevent or most often delay an event (NNT for number needed to treat), an alternative and informative way of presenting the results of randomised controlled trials is to estimate the gain in mean survival or event-free time. Study selection: recent randomised controlled cardiovascular trials that evaluated ezetimibe (IMPROVE-IT), évolocumab (FOURIER) and alirocumab (ODYSSEY OUTCOMES).

Results

Event-free time gains for the primary endpoints ranged from 10 to 35 days over the duration of the studies; mean NNTs ranged from 50 to 67 for delaying any of the composite endpoint events.

Conclusion

publication of survival gains as a complementary measure of treatment effect is probably more useful than relative risk reduction for both patient and physician.