Online Program Home
My Program

Abstract Details

Activity Number: 539 - SPEED: Bayesian Methods and Applications in the Life and Social Sciences
Type: Contributed
Date/Time: Wednesday, August 1, 2018 : 11:35 AM to 12:20 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #332693
Title: A Bayesian Meta-Analysis to Adjust Diagnostics Tests for Trend Analysis of Clostridium Difficile Infection in the Emerging Infections Program, U. S
Author(s): Yi Mu*
Companies: Centers for Disease Control and Prevention
Keywords: Bayesain; meta-analysis; Clostridium difficile infection; trend; diagnostics test
Abstract:

Nucleic acid amplification test (NAAT) are increasingly used to diagnose Clostridium difficile infection (CDI) in the Emerging Infections Program (EIP). Because NAAT is highly sensitive compared to EIA, adjustment of cases diagnosed by EIA is needed to monitor incidence trends. We conducted a literature review and obtain the individual excess positive rate by NAAT compared to EIA. We then developed a Bayesian approach to conduct the random-effects meta-analysis to estimate the pool excess positive rates by NAAT. The advantage of Bayesian approach is that the random effects variance is acknowledged in the pooled effect, unlike a classical analysis where the pooled estimate uses just a point estimate of the variance. From 2011-2015, the proportion CDI cases diagnosed by NAAT increased from 52% to 88%, due to highly sensitivity of NAAT , crude CDI incidence increased from 2011 to 2015, after applying the estimated excess positive rates by NAAT to EIA diagnosis cases, the adjusted CDI incidence declined over the years. Our results demonstrated the importance of adjusting CDI incidence based on testing types for understanding trends in CDI incidence in EIP.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program