Bayesian versus frequentist methods for estimating true prevalence of disease and diagnostic test performance

Sanogo, Moussa; Abatih, Emmanuel; Saegerman, Claude

doi:10.1016/j.tvjl.2014.08.002

Cited by 8 publications

(7 citation statements)

References 48 publications

(81 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both traditional (frequentist) and Bayesian methods have been developed to estimate true prevalence. When the expected prevalence is low, assumptions required for frequentist approaches may not be met and unrealistic point estimates and wide confidence intervals result [ 27 , 29 ]. In this work, we use a Bayesian approach described by Joseph et al [ 30 ] and utilized through Epitools (Bayesian estimation of true prevalence from survey testing with one test) [ 31 ].…”

Section: Methodsmentioning

confidence: 99%

“…In this work, we use a Bayesian approach described by Joseph et al [ 30 ] and utilized through Epitools (Bayesian estimation of true prevalence from survey testing with one test) [ 31 ]. The accuracy of Bayesian estimates is dependent on the quality of prior knowledge regarding sensitivity and specificity of the test and expectation of the prevalence [ 29 ]. Prior estimates of the sensitivity and specificity, in all cases, were those obtained from the SERoSP1-SERoSP2 diagnostic combination on the validation dataset.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Suicidal Presentations to Emergency Departments in a Large Australian Public Health Service over 10 Years

Stapelberg

Sveticic

Hughes

et al. 2020

IJERPH

View full text Add to dashboard Cite

This paper presents trends and characteristics for 32,094 suicidal presentations to two Emergency Departments (EDs) in a large health service in Australia across a 10-year period (2009–2018). Prevalence of annual suicidal presentations and for selected groups of consumers (by sex, age groups, and ethnicity) was determined from a machine learning diagnostic algorithm developed for this purpose and a Bayesian estimation approach. A linear increase in the number of suicidal presentations over 10 years was observed, which was 2.8-times higher than the increase noted in all ED presentations and 6.1-times higher than the increase in the population size. Females had higher presentation rates than males, particularly among younger age groups. The highest rates of presentations were by persons aged 15–24. Overseas-born persons had around half the rates of suicidal presentations than Australian-born persons, and Indigenous persons had 2.9-times higher rates than non-Indigenous persons. Of all presenters, 70.6% presented once, but 5.7% had five or more presentations. Seasonal distribution of presentations showed a peak at the end of spring and a decline in winter months. These findings can inform the allocation of health resources and guide the development of suicide prevention strategies for people presenting to hospitals in suicidal crisis.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Suicidal Presentations to Emergency Departments in a Large Australian Public Health Service over 10 Years

Stapelberg

Sveticic

Hughes

et al. 2020

IJERPH

View full text Add to dashboard Cite

show abstract

“…Therefore, prior information should be sought and incorporated in terms of prior distributions [ 46 , 47 ]. Additionally, the large sample size (22,126 animals) made a corrective strategy, which allowed the data “to speak for itself” [ 48 ]. In this sense, Model 1 required fewer parameters than Model 2, which would have required prior knowledge of sensitivities and specificities for both tests.…”

Section: Discussionmentioning

confidence: 99%

Bayesian Estimation of the Prevalence and Test Characteristics (Sensitivity and Specificity) of Two Serological Tests (RB and SAT-EDTA) for the Diagnosis of Bovine Brucellosis in Small and Medium Cattle Holders in Ecuador

et al. 2021

Self Cite

View full text Add to dashboard Cite

In Ecuador, a national program for bovine brucellosis control has been in implementation since 2008. Given the costs, small- and medium-sized livestock holders are not completely committed to it. The objective of this study was to determine true prevalence (TP) of bovine brucellosis in small- and medium-sized herd populations, as well as the diagnostic sensitivity and specificity of the Rose Bengal (RB) test and the sero-agglutination test (SAT)-EDTA using a Bayesian approach. Between 2011 and 2016, 2733 cattle herds were visited, and 22,592 animal blood samples were taken in nineteen provinces on mainland Ecuador. Bayes-p and deviance information criterion (DIC) statistics were used to select models. Additionally, risk-factor analysis was used for herds according to their brucellosis test status. True prevalence (TP) in herds was estimated by pool testing. National seroprevalence of farms was 7.9% (95% CI: 6.79–9.03), and TP was 12.2% (95% CI: 7.8–17.9). Apparent prevalence (AP) in animals was 2.2% (95% CI: 1.82–2.67), and TP was 1.6% (95% CrI: 1.0–2.4). Similarly, the sensitivity of the RB was estimated at 64.6% (95% CrI: 42.6–85.3) and specificity at 98.9% (95% CrI: 98.6–99.0); for the SAT-EDTA test, sensitivity was 62.3% (95% CrI: 40.0–84.8) and 98.9% (95% CrI: 98.6–99.1) for specificity. Results of the two tests were highly correlated in infected and uninfected animals. Likewise, high spatial variation was observed, with the Coastal Region being the zone with the highest TP at 2.5%. (95% CrI: 1.3–3.8%) in individual animals and 28.2% (95% CI: 15.7–39.8) in herds. Risk factors include herd size, type of production (milk, beef, and mixed), abortions recorded, and vaccination. The results of this study serve to guide authorities to make decisions based on parallel testing at the beginning of a bovine brucellosis program for small livestock holders to increase sensitivity level of the screening tests in Ecuador.

show abstract

“…This status is given by a 'gold standard' test. In the absence of a 'gold standard' test, a Bayesian approach is helpful to estimate test Se, Sp and prevalence, as has been done for several diseases (19,20,27,28,29,30). This is also the case for estimation of the BVDV prevalence (29).…”

Section: Discussionmentioning

confidence: 99%

Bayesian assessment of two competitive enzyme-linked immunosorbent assays for the detection of bovine viral diarrhoea virus antibodies in bovine sera

Saegerman¹,

Quinet²,

Czaplicki³

et al. 2018

Rev. Sci. Tech. OIE

View full text Add to dashboard Cite

Infections due to bovine viral diarrhoea virus (BVDV) are endemic in most cattleproducing countries throughout the world and bovine viral diarrhoea is considered a transboundary disease. The key elements of a BVDV control programme are vaccination, biosecurity, elimination of persistently infected (PI) animals and surveillance. The aim of this study was to assess the sensitivity (Se) and the specificity (Sp) of two commercial competitive enzyme-linked immunosorbent assays (ELISAs) based on selected immune-dominant BVDV proteins: the nonstructural protein NS3 (p80) and the recombinant envelope glycoprotein E0 (Erns). Both tests were used on individual serum samples from randomly sampled young bovines in southern Belgium in order to detect specific BVDV antibodies. The Se and Sp were assessed using a Bayesian approach and were estimated, respectively, at 97.2% (with 95% credibility interval [Cr I]: 95.1-99.8) and 98.7% (95% Cr I: 96.6-99.9) for the first test and 95.8% (95% Cr I: 91.1-99.7) and 96.1% (95% Cr I: 95.1-97.7) for the second test. The results obtained with the two tests were not significantly different. In addition, using both ELISAs, the current BVDV exposure among young bovines in southern Belgium was estimated at 23.3% (95% Cr I: 20.6-26.2). Combining virological testing of all newborns to detect PI animals with regular serological testing of young stock using ELISAs is recommended in the surveillance of BVDV.

show abstract

Bayesian versus frequentist methods for estimating true prevalence of disease and diagnostic test performance

Cited by 8 publications

References 48 publications

Suicidal Presentations to Emergency Departments in a Large Australian Public Health Service over 10 Years

Suicidal Presentations to Emergency Departments in a Large Australian Public Health Service over 10 Years

Bayesian Estimation of the Prevalence and Test Characteristics (Sensitivity and Specificity) of Two Serological Tests (RB and SAT-EDTA) for the Diagnosis of Bovine Brucellosis in Small and Medium Cattle Holders in Ecuador

Bayesian assessment of two competitive enzyme-linked immunosorbent assays for the detection of bovine viral diarrhoea virus antibodies in bovine sera

Contact Info

Product

Resources

About