Value of biopsy in a cohort of children with high-titer celiac serologies: observation of dynamic policy differences between Europe and North America

Abstract

Background: Healthcare systems implement change at different rates because of differences in incentives, organizational processes, key influencers, and management styles. A comparable set of forces may play out at the national and international levels as demonstrated in significant differences in the diagnostic management of pediatric Celiac Disease (CD) between European and North American practitioners.

Methods: We use retrospective clinical cohorts of 27,868 serum tissue transglutaminase (tTG) immunoglobulin A levels and 7907 upper gastrointestinal endoscopy pathology reports to create a dataset of 793 pathology reports with matching tTG results between July 1 of 2014 and July 1 of 2018. We use this dataset to characterize histopathological findings in the duodenum, stomach and esophagus of patients as a function of serum tTG levels. In addition, we use the dataset to estimate the local and national cost of endoscopies performed in patients with serum tTG levels greater than 10 times the upper limit of normal.

Results: Using evidence from a US tertiary care center, we show that in the cohort of pediatric patients with high pre-test probability of CD as determined by serum tTG levels, biopsy provides no additional diagnostic value for CD, and that it counter-intuitively introduces diagnostic uncertainty in a number of patients. We estimate that using the European diagnostic algorithms could avoid between 4891 and 7738 pediatric endoscopies per year in the US for evaluation of CD.

Conclusions: This study considers the North American and European management guidelines for the diagnosis of pediatric CD and highlights the slow adoption in North America of evidence-based algorithms developed and applied in Europe for triage of endoscopy and biopsy. We suggest that system dynamics influences that help maintain the status quo in North America include a variety of social and economic factors in addition to medical evidence. This work contributes to the growing body of evidence that the dynamics that largely favor maintaining status quo management policies in a variety of systems extend to clinical medicine and potentially influence clinical decisions at the level of individual patients and the population.

Keywords: Celiac disease; Health policy; System dynamics; Testing; Value of information.

Fig. 1

Histopathological classification of duodenal biopsies in 535 patients with positive serum tTG results. Box and whisker plot are generated using Microsoft Excel® and show the individual data (circles), as well as minimum value, first quartile, median, third quartile, maximum value, and outliers. Panel a shows all data, while panel b is limited to tTG results between 20 and 200 Chem’U. Three outliers labeled 1, 2 and 3 in the Indefinite Duodenitis group of Panel a are discussed in the body of the manuscript

Fig. 2

Significant and incidental findings in the esophagus and stomach as a function of serum tTG values

Fig. 3

Causal loop diagram highlighting factors that could affect the decision to biopsy. Green standard font shows factors that generally reinforce the decision to biopsy. Red italic font show factors that would generally shift the balance away from doing a biopsy