Premium
The binomial symptom index: toward an optimal method for the evaluation of symptom association in gastroesophageal reflux
Author(s) -
BarrigaRivera A.,
Elena M.,
Moya M. J.,
LopezAlonso M.
Publication year - 2013
Publication title -
neurogastroenterology and motility
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.489
H-Index - 105
eISSN - 1365-2982
pISSN - 1350-1925
DOI - 10.1111/nmo.12143
Subject(s) - reflux , medicine , association (psychology) , standard deviation , statistics , mathematics , psychology , disease , psychotherapist
Abstract Background The evaluation of symptom association in gastroesophageal reflux is an open problem. The scientific literature reports important deficiencies and clinicians are claiming a new methodology. This article provides an optimal method for the evaluation of symptom association, the binomial symptom index ( BSI ). Methods A mathematical description of the BSI was presented for the study of association and causality. A total of n = 850 000 patients were simulated using a Monte Carlo model to perform a two‐way sensitivity analysis. The average and the standard deviation of the BSI were evaluated in groups of 5000 patients with the same values of the reflux rate, symptom rate, association ratio, window of association, and monitoring time in order to contrast their influence on the estimator. Key Results The BSI decreased with the number of reflux episodes when there was association, and remained constant and below 40% when there was not. The standard deviation was no higher than 40% and decreased with the reflux or symptom rates, and more sharply with the monitoring time, reaching approximately 0% for 50 days. A window length matching the characteristic reflux‐symptom lag maximized the overall BSI and minimized its dispersion. Twenty‐four hour and 96‐h monitorings allowed detecting association ratios of 50% and 25%, respectively. Conclusions & Inferences The BSI is a simple and reliable index for the evaluation of symptom association that considers all the parameters under analysis. Defining an appropriate cut‐off value, the BSI can provide a measure of probability and strength of association simultaneously.