[LCMATE] Latent class definition question

Mon Feb 11 13:34:08 EET 2019

Dear Braintrust, 
I have several questions and thoughts I want to share with you on interesting questions I get from clinician people and other researchers that have some reluctance with the LCM approach. Most of their concerns can be answered easily but some questions relative to the correct handling of latent status definition are more difficult to be addressed adequately.
Let's imagine I am focusing on bronchopneumonia in calves (one of my research area). I have 2 tests 1 test is for imaging (I), the second test is auscultation of the calves using a computer-aided lung auscultation algorithm (A). I therefore have 2 tests (I and A) that are possibly dependent (or not). These tests have high inter-rater agreement.
Let's imagine I am doing a large cross-sectional survey in feedlot calves where I include 10 000 animals in 1000 different feedlots (huge study).
The definition of I+/I- and A+/A- are fixed. Then I may define 3 different latent classes that are of interest for clinical purpose and rational use of antimicrobial.
1st objective of the study:
Determination of the accuracy of both tests for detection of clinical bronchopneumonia (ie state of disease where infection is active and therefore need to be addressed using antimicrobial)
2nd objective of the study:
Determination of the accuracy to detect any bronchopneumonia lesion (including active lesions, lesions of previous bronchopneumonia (ie still some lung lesion but no active infection/inflammation)
3rd objective of the study:
Determination of the accuracy to detect non-active lesions (ie latent class status 2 -status1)
In this hypothetical study using the same test positive/negative definition, my posterior would only be driven by my priors which are including the different definition when gathering information from the literature or from the experts.
However, the data in that hypothetical study would not change (ie I have the same ++ -+ +- and - stratification of my calves). If running a very large study with many data all my models obtained from objective 1, 2 and 3 will converge to the same posterior densities (SeI, Spi, SeA, SpA and prevalence).
How can I therefore be sure that the accuracy results I obtain are for latent class definition 1, 2 or 3? Maybe there is a simple answer but I did not find it until now.
I thank you in advance for any information on that as well as on any specific reference on that fascinating topic.

Thanks a lot

Sébastien

-----Message d'origine-----
De : lcmate-bounces at lists.uth.gr <lcmate-bounces at lists.uth.gr> De la part de Polychronis Kostoulas
Envoyé : 5 février 2019 07:20
À : lcmate at lists.uth.gr
Objet : [LCMATE] Invitation to the HOTLINE webinar

Dear Colleagues,
This is an invitation to join a webinar presenting the results of the HOTLINE project (www.thehotlineproject.org<http://www.thehotlineproject.org/>). It's about providing Bayesian tools for the harmonization of disease prevalence estimates that are collected under different sampling and testing settings.
The webinar will be on Feb, 19 from 13:00 (CET) till 14:30 (CET) Please see below a brief description of the schedule and guidelines on how to join.

Those interested are advised to register ( https://polychroniskostoulas.clickmeeting.com/the-hotline-project ) because there may be limited slots available at the time of the event.

Kind regards,
Polychronis

HOTLINE: Harmonisation Of TransmissibLe disease INterpretation in the EU Countries typically collect disease data in a way that is best suited for their needs. Therefore, differences exist in the sampling schemes and the diagnostic methods, which produce non-comparable estimates of the prevalence and spectrum of disease.  The HOTLINE project has the objective to make disease information comparable and interpretable across different sampling and testing settings. Like the hotline used to serve for direct communication between the leaders of the superpowers. To serve this objective we are providing a series of Bayesian tools with step by step explanations as well as material for reporting guidelines on surveillance.

Schedule (Central European Time)
13:00 to 13:05: Introductions - Polychronis Kostoulas
13:05 to 13:20: Software & Training material for Bayesian true prevalence estimation - Polychronis Kostoulas
13:20 to 13:35: Software & Training material for Bayesian true prevalence estimation under risk-based settings- Armando Giovannini
13:35 to 13:45: Presentation of IWA: An Interactive Web Application for Bayesian prevalence estimation - Ana Alba
13:45 to 13:55: What data do we need? Reporting guidelines on surveillance - Arianna Comin
13:55 to 14:30: General Questions - Thank you and Event Closure

Choose which way you want to join the event:

1. Join via Website: https://polychroniskostoulas.clickmeeting.com/the-hotline-project

on any browser with Flash

2. Join via Phone:
Ankara
+90 (850) 455-1249
Praha
+420 (2) 3409-3808
Warsaw
+48 (22) 209-2520

Conference PIN: 399214#

See all numbers: https://utilities.clickmeeting.com/phone-bridge/Listener/814946515

3. Join via Mobile Application:
Room ID: 814-946-515

Convert start time to your timezone: http://www.clickmeeting.com/converter/814946515

Polychronis P. Kostoulas
D.V.M., Ph.D., Assistant Professor
Associate Editor of Preventive Veterinary Medicine
Erasmus+ Departmental Coordinator
www.thehotlineproject.org<http://www.thehotlineproject.org>
www.Pre2prO.com<http://www.pre2pro.com/>
LCMATE Administrator (http://lists.uth.gr/mailman/listinfo/lcmate)
---------------------------------------------------------------------------------------------
Laboratory of Epidemiology, Biostatistics and Animal Health Economics Faculty of Veterinary Medicine University of Thessaly Karditsa, 224 Trikalon st.
GREECE. P.C. 43100
Telephone: +30 2441066022
Mobile: +30 6977881857

_______________________________________________
LCMATE mailing list
LCMATE at lists.uth.gr
http://lists.uth.gr/mailman/listinfo/lcmate