Fragility Index

from Wikipedia, the free encyclopedia

The Fragility Index is a statistical measure that v. a. is used in the medical literature to assess study results. It describes the minimum number of patients whose status would have to change from a non-event to an event in order to convert a statistically significant result of the study into a non-significant result. The smaller the Fragility Index, the more fragile the results of the study.

Derivation

The result of scientific studies is designated as statistically significant if the calculated probability of error for a type 1 error (assumption of a connection that does not really exist) is below the arbitrarily chosen but generally accepted limit of 5%. Study results that meet these significance criteria can be published more easily and in higher-ranking journals ( publication bias ).

The sole consideration of the level of significance for the assessment of study results is increasingly seen as problematic (see also Problems with the interpretation of statistical significance ). The fragility index describes how much the ascription of statistical significance is subject to random influences. It indicates the number of patients or test subjects whose results would have to change in order to change the statistical evaluation of the study. The Fragility Index only refers to studies that examine dichotomous results (deceased / surviving, cured / not cured).

meaning

A median fragility index of 2 was calculated in a study of 56 intensive care medical studies that had a significant result in terms of mortality; in 40% of the studies it was 0 or 1. That is, it would have been in at least half of the studies examined it was sufficient that 2 patients had a different outcome and the result of the study would no longer have been considered significant. In 12.5% ​​of the studies examined, the fragility index was smaller than the number of patients for whom no follow-up data were available (lost to follow-up ). So it is quite possible that more complete data collection would have undone the significant result.

This problem of fragility is v. a. discussed in intensive care literature. That will u. a. attributed to the fact that the patient numbers in these studies are naturally small. But this problem of the fragility of study results has also been described in other areas of medicine.

literature

  1. ↑ a b Walsh M et al .: The statistical significance of randomized controlled trial results is frequently fragile: a case for a Fragility Index . In: Journal of Clinical Epidemiology . tape 67 , no. 6 , June 2014, p. 622–628 , doi : 10.1016 / j.jclinepi.2013.10.019 (English, jclinepi.com [accessed on May 10, 2018]).
  2. ^ Ridgeon EE et al .: The Fragility Index in Multicenter Randomized Controlled Critical Care Trials . In: Critical Care Medicine . tape 44 , no. 7 , July 2016, p. 1278-84 , doi : 10.1097 / CCM.0000000000001670 , PMID 26963326 (English).
  3. ^ Owen Boyd: The Fragility of Evidence in Critical Care Research: Is It Just a House of Cards? In: Critical Care Medicine . tape 44 , no. 7 , July 2016, p. 1423-4 , doi : 10.1097 / CCM.0000000000001756 (English).