Association between covariates and disease occurrence in the presence of diagnostic error

Lewis, Fraser; Sánchez-Vázquez, Manuel J.; Torgerson, Paul R.

doi:10.1017/s0950268811001932

Cited by 9 publications

(5 citation statements)

References 29 publications

(41 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One potential criticism of risk factor analyses based on simple classifications is that they do not incorporate diagnostic test sensitivity and specificity when classifying the farms as case or control [ 27 , 46 ]. In this study, incorporating the relevant diagnostic test characteristics did not result in any of the control farms being reclassified as case farms, indicating that imperfect sensitivity of liver condemnation was not an issue for our dataset.…”

Section: Discussionmentioning

confidence: 99%

Farm-level risk factors for Fasciola hepatica infection in Danish dairy cattle as evaluated by two diagnostic methods

et al. 2017

View full text Add to dashboard Cite

BackgroundThe prevalence of bovine fasciolosis in Denmark is increasing but appropriate guidelines for control are currently lacking. In order to help develop a control strategy for liver fluke, a risk factor study of farm management factors was conducted and the utility of bulk tank milk (BTM ELISA) as a tool for diagnosis in Danish dairy cattle farms was assessed.MethodsThis case-control study aimed to identify farm-level risk factors for fasciolosis in Danish dairy farms (> 50 animals slaughtered in 2013) using two diagnostic methods: recordings of liver condemnation at slaughter, and farm-level Fasciola hepatica antibody levels in BTM. A case farm was defined as having a minimum of 3 incidents of liver condemnation due to liver fluke at slaughter (in any age group) during 2013, and control farms were located within 10 km of at least one case farm and had no history of liver condemnation due to liver fluke during 2011–2013. The selected farmers were interviewed over telephone about grazing and control practices, and BTM from these farms was collected and analysed by ELISA in 2014. The final complete dataset consisting of 131 case and 63 control farms was analysed using logistic regression.ResultsHeifers grazing on wet pastures, dry cows grazing on wet pastures, herd size, breed and concurrent beef cattle production were identified as risk factors associated with being classified as a case farm. With the categorised BTM ELISA result as the response variable, heifers grazing on wet pastures, dry cows grazing on wet pastures, and purchase of cows were identified as risk factors. Within the case and control groups, 74.8 and 12.7% of farms were positive for fasciolosis on BTM ELISA, respectively. The differences are likely to be related to the detection limit of the farm-level prevalence by the BTM ELISA test, time span between slaughter data and BTM, and the relatively low sensitivity of liver inspection at slaughter.ConclusionsControl of bovine fasciolosis in Denmark should target heifers and dry cows through grazing management and appropriate anthelmintic treatment, and BTM ELISA can be a useful diagnostic tool for fasciolosis in Danish dairy farms.Electronic supplementary materialThe online version of this article (10.1186/s13071-017-2504-y) contains supplementary material, which is available to authorized users.

show abstract

Section: Discussionmentioning

confidence: 99%

Farm-level risk factors for Fasciola hepatica infection in Danish dairy cattle as evaluated by two diagnostic methods

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Thus a number of dogs in the previous study would have been misclassified and would have affected the results of the regression analysis. Techniques are now becoming available to incorporate the latent but unknown infection status in regression analysis [22] and these should be used where possible to avoid reaching inappropriate conclusions about the possible significance of covariates in epidemiological studies.…”

Section: Discussionmentioning

confidence: 99%

Latent-Class Methods to Evaluate Diagnostics Tests for Echinococcus Infections in Dogs

et al. 2013

View full text Add to dashboard Cite

BackgroundThe diagnosis of canine echinococcosis can be a challenge in surveillance studies because there is no perfect gold standard that can be used routinely. However, unknown test specificities and sensitivities can be overcome using latent-class analysis with appropriate data.MethodologyWe utilised a set of faecal and purge samples used previously to explore the epidemiology of canine echinococcosis on the Tibetan plateau. Previously only the purge results were reported and analysed in a largely deterministic way. In the present study, additional diagnostic tests of copro-PCR and copro-antigen ELISA were undertaken on the faecal samples. This enabled a Bayesian analysis in a latent-class model to examine the diagnostic performance of a genus specific copro-antigen ELISA, species-specific copro-PCR and arecoline purgation. Potential covariates including co-infection with Taenia, age and sex of the dog were also explored. The dependence structure of these diagnostic tests could also be analysed.Principle findingsThe most parsimonious result, indicated by deviance-information criteria, suggested that co-infection with Taenia spp. was a significant covariate with the Echinococcus infection. The copro-PCRs had estimated sensitivities of 89% and 84% respectively for the diagnoses of Echinococcus multilocularis and E. granulosus. The specificities for the copro-PCR were estimated at 93 and 83% respectively. Copro-antigen ELISA had sensitivities of 55 and 57% for the diagnosis of E. multilocularis and E. granulosus and specificities of 71 and 69% respectively. Arecoline purgation with an assumed specificity of 100% had estimated sensitivities of 76% and 85% respectively.SignificanceThis study also shows that incorporating diagnostic uncertainty, in other words assuming no perfect gold standard, and including potential covariates like sex or Taenia co-infection into the epidemiological analysis may give different results than if the diagnosis of infection status is assumed to be deterministic and this approach should therefore be used whenever possible.

show abstract

“…An alternative is to use the Expectation Maximization (EM) algorithm [26], which is also a maximum likelihood approach but ideally suited to problems comprising latent class variables, which is exactly what we have here as the true disease status of each observation is only latently observed. Some details of how to implement EM estimation in the context of imperfect diagnostic tests are given in [27] and [20]. The third option is to use a Bayesian approach.…”

Section: Preliminariesmentioning

confidence: 99%

A tutorial in estimating the prevalence of disease in humans and animals in the absence of a gold standard diagnostic

Lewis

Torgerson

2012

Emerg Themes Epidemiol

View full text Add to dashboard Cite

Epidemiological methods for estimating disease prevalence in humans and other animals in the absence of a gold standard diagnostic test are well established. Despite this, reporting apparent prevalence is still standard practice in public health studies and disease control programmes, even though apparent prevalence may differ greatly from the true prevalence of disease. Methods for estimating true prevalence are summarized and reviewed. A computing appendix is also provided which contains a brief guide in how to easily implement some of the methods presented using freely available software.

show abstract

Association between covariates and disease occurrence in the presence of diagnostic error

Cited by 9 publications

References 29 publications

Farm-level risk factors for Fasciola hepatica infection in Danish dairy cattle as evaluated by two diagnostic methods

Farm-level risk factors for Fasciola hepatica infection in Danish dairy cattle as evaluated by two diagnostic methods

Latent-Class Methods to Evaluate Diagnostics Tests for Echinococcus Infections in Dogs

A tutorial in estimating the prevalence of disease in humans and animals in the absence of a gold standard diagnostic

Contact Info

Product

Resources

About