Demonstrating the absence of an infection from a population can be problematic since it would theoretically require the testing of all individuals with a test that had both 100% sensitivity and specificity. However, this is hardly realistic since every diagnostic test retains some inherent measure of uncertainty and disease surveys are typically carried out on a subset of individuals sampled from the target population. This problem can be overcome by estimating the confidence of freedom from a disease, which represents the probability that an infection is truly absent given that a determinate number of individuals from that population are all negative to a specific diagnostic test. Confidence of freedom extends the concept of negative predictive value to the whole population by taking into account sample size, expected prevalence of the disease and population-level specificity and sensitivity of the diagnostic test of choice. This approach can be most useful in the context of biological invasions, when there is a concrete risk of alien hosts introducing new pathogens that may threaten public health or native wildlife but, at the same time, there is no previous epidemiological history about either the host or the disease in the introduction area. To illustrate this method, we will apply confidence of freedom on two case studies involving the potential introduction of a disease by alien hosts. Although in both surveys we did not find any evidence of the target pathogen, an accurate interpretation of negative results was crucial for either public health (case study I) or conservation (case study II) purposes. Case study I – Raccoons (Procyon lotor) and Baylisascaris procyonis: we applied confidence of freedom to preliminary results obtained during a survey aimed at detecting whether introduced raccoons had carried along to northern Italy the zoonotic nematode B. procyonis. Case study II – Grey squirrels (Sciurus carolinensis) and squirrelpoxvirus: we applied confidence of freedom to data obtained through two different diagnostic methods to determine whether alien grey squirrels had introduced to Italy a poxvirus which is known to severely threaten native squirrels in the UK. Through these examples, we will show how assuming different prevalences and choosing different diagnostic tests (i.e. with different specificity and sensitivity) and sample sizes will affect confidence of freedom estimation.

When zero is a result: how to demonstrate the absence of an infection / C. Romeo, N. Ferrari - In: Atti del IV congresso nazionale di ecopatologia della fauna / [a cura di] L. Stancampiano, E. Armaroli, R. Viganò, N. Ferrari. - [s.l] : Società Italiana di Ecopatologia della Fauna, 2017 Dec. - ISBN 9788894297300. - pp. 83-83 (( Intervento presentato al 4. convegno Congresso Nazionale di Ecopatologia della Fauna tenutosi a Domodossola nel 2017.

When zero is a result: how to demonstrate the absence of an infection

C. Romeo
Primo
;
N. Ferrari
Ultimo
2017

Abstract

Demonstrating the absence of an infection from a population can be problematic since it would theoretically require the testing of all individuals with a test that had both 100% sensitivity and specificity. However, this is hardly realistic since every diagnostic test retains some inherent measure of uncertainty and disease surveys are typically carried out on a subset of individuals sampled from the target population. This problem can be overcome by estimating the confidence of freedom from a disease, which represents the probability that an infection is truly absent given that a determinate number of individuals from that population are all negative to a specific diagnostic test. Confidence of freedom extends the concept of negative predictive value to the whole population by taking into account sample size, expected prevalence of the disease and population-level specificity and sensitivity of the diagnostic test of choice. This approach can be most useful in the context of biological invasions, when there is a concrete risk of alien hosts introducing new pathogens that may threaten public health or native wildlife but, at the same time, there is no previous epidemiological history about either the host or the disease in the introduction area. To illustrate this method, we will apply confidence of freedom on two case studies involving the potential introduction of a disease by alien hosts. Although in both surveys we did not find any evidence of the target pathogen, an accurate interpretation of negative results was crucial for either public health (case study I) or conservation (case study II) purposes. Case study I – Raccoons (Procyon lotor) and Baylisascaris procyonis: we applied confidence of freedom to preliminary results obtained during a survey aimed at detecting whether introduced raccoons had carried along to northern Italy the zoonotic nematode B. procyonis. Case study II – Grey squirrels (Sciurus carolinensis) and squirrelpoxvirus: we applied confidence of freedom to data obtained through two different diagnostic methods to determine whether alien grey squirrels had introduced to Italy a poxvirus which is known to severely threaten native squirrels in the UK. Through these examples, we will show how assuming different prevalences and choosing different diagnostic tests (i.e. with different specificity and sensitivity) and sample sizes will affect confidence of freedom estimation.
Settore VET/06 - Parassitologia e Malattie Parassitarie degli Animali
dic-2017
Book Part (author)
File in questo prodotto:
File Dimensione Formato  
Romeo&Ferrari_Abstract_SIEF 17.pdf

accesso riservato

Tipologia: Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Dimensione 152.73 kB
Formato Adobe PDF
152.73 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/541344
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact