Cost‐effectiveness of integrating gut microbiota analysis into hospitalisation prediction in cirrhosis

Abstract

Summary Background Admissions in cirrhosis are expensive and often unpredictable based on purely clinical variables. Admissions could be related to complications associated with gut microbial changes, which can improve prognostication. However, the cost‐effectiveness remains unclear. Aims Determine cost‐effectiveness of adding gut microbiota analysis to clinical parameters in prediction and subsequent reduction of admissions in cirrhosis. Methods Using a Markov model of 1000 cirrhosis patients over 90 days, we modelled microbiota testing using 16srRNA ($250/sample), low‐depth ($350/sample) and high‐depth ($650/sample) metagenomics added to standard‐of‐care (SOC) for prevention of admissions using standard outcome costs and rates of development. We generated quality of life years (QALY) and Incremental cost‐effectiveness ratios (ICER) for the base scenarios and performed sensitivity analyses by varying costs for outcomes (transplant, death, admission) and admission rates (40%, range 25%‐60%). Results Using fixed costs of outcomes and outcome rates, microbiota analysis was cost‐saving ($47K‐$97K) at $250 and $350/sample if admissions were reduced by 5% over SOC and >10% with $650/sample. When costs of liver transplant, death and admissions were varied, these cost‐savings remained robust provided there was >2.1% reduction (range 1.3%‐3.2%) for $250/sample, >2.9% (range 1.8%‐4.4%) for $350/sample and >5.4% (range 3.3%‐8.2%) for $650/sample. These cost‐savings remained robust even when the assumed admission rate was varied for all sample cost values. Conclusions Gut microbiota analysis is cost‐effective for predicting and potentially preventing 90‐day admissions in cirrhosis over current SOC. This cost‐saving remained robust even after sensitivity analyses that varied the background admission rates.

Healthy microbiome - moving towards functional interpretation

“…Stool-based methods are particularly promising because they can be collected non-invasively and frequently, accompanied by a short analysis time. Decreasing costs of next-generation sequencing-based stool sample characterization makes microbiome-based characterization and diagnostics a strong competitor in the field (Bajaj et al, 2020).…”

BiGAMi: Bi-Objective Genetic Algorithm Fitness Function for Feature Selection on Microbiome Datasets

“…Microbiome datasets are small when compared to other fields that benefit from artificial intelligence (AI), with often a few dozen to a few thousand data points due to sampling, experimental, and sequencing costs, which could go as high as hundreds of thousands of dollars to even millions [ 11 , 12 ]. However, microbiome datasets have yet another characteristic that can increase this challenge.…”

Clinical Metagenomic Next-Generation Sequencing has the Potential to Reduce Costs in the Management of Microbial Keratitis Patients

“…Therefore, we deem that a metagenomics approach to microbial keratitis can reduce the overall cost of medical treatments, as similarly reported in gastroenterology, hepatology, and pneumology departments. 6,7…”