Recognizing Structural Nonidentifiability: When Experiments Do Not Provide Information About Important Parameters and Misleading Models Can Still Have Great Fit

Schmidt, Philip J.; Emelko, Monica B.; Thompson, Mary E.

doi:10.1111/risa.13386

Cited by 13 publications

(14 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For these groups, neither r nor m were independently identifiable, as indicated by long narrow ridges in each parameter's posterior sample (Figs. 2c–e; Schmidt, Emelko, & Thompson, ). That is, neither parameter could be estimated with confidence on its own, and it appeared that changes in one parameter could compensate for changes in the other (because the products r × m had somewhat well‐defined posterior distributions; see Figs.…”

Section: Resultsmentioning

confidence: 99%

Outbreak‐BasedGiardiaDose–Response Model Using Bayesian Hierarchical Markov Chain Monte Carlo Analysis

Burch

2019

Risk Analysis

View full text Add to dashboard Cite

Giardia is a zoonotic gastrointestinal parasite responsible for a substantial global public health burden, and quantitative microbial risk assessment (QMRA) is often used to forecast and manage this burden. QMRA requires dose-response models to extrapolate available dose-response data, but the existing model for Giardia ignores valuable dose-response information, particularly data from several well-documented waterborne outbreaks of giardiasis. The current study updates Giardia dose-response modeling by synthesizing all available data from outbreaks and experimental studies using a Bayesian random effects dose-response model. For outbreaks, mean doses (D) and the degree of spatial and temporal aggregation among cysts were estimated using exposure assessment implemented via two-dimensional Monte Carlo simulation, while potential overreporting of outbreak cases was handled using published overreporting factors and censored binomial regression. Parameter estimation was by Markov chain Monte Carlo simulation and indicated that a typical exponential doseresponse parameter for Giardia is r = 1.6 × 10 −2 [3.7 × 10 −3 , 6.2 × 10 −2 ] (posterior median [95% credible interval]), while a typical morbidity ratio is m = 3.8 × 10 −1 [2.3 × 10 −1 , 5.5 × 10 −1 ]. Corresponding (logistic-scale) variance components were σ r = 5.2 × 10 −1 [1.1 × 10 −1 , 9.6 × 10 −1 ] and σ m = 9.3 × 10 −1 [7.0 × 10 −2 , 2.8 × 10 0 ], indicating substantial variation in the Giardia dose-response relationship. Compared to the existing Giardia dose-response model, the current study provides more representative estimation of uncertainty in r and novel quantification of its natural variability. Several options for incorporating variability in r (and m) into QMRA predictions are discussed, including incorporation via Monte Carlo simulation as well as evaluation of the current study's model using the approximate beta-Poisson.

show abstract

Section: Resultsmentioning

confidence: 99%

Outbreak‐BasedGiardiaDose–Response Model Using Bayesian Hierarchical Markov Chain Monte Carlo Analysis

Burch

2019

Risk Analysis

View full text Add to dashboard Cite

show abstract

“…They alluded to this aggregation with the inclusion of an electron micrograph image of a cluster, and attempted to estimate the degree of aggregation from the corresponding dose-response data [25]. However, it has been shown that such inference is not possible due to model non-identifiability limitations associated with the impossibility of concurrently evaluating dose response and estimating an unmeasured aggregation parameter [29,56].…”

Section: Discussionmentioning

confidence: 99%

“…It is important to note that these limitations apply to any dose-response experiment in which (1) disaggregation of administered pathogens was not confirmed or (2) aggregation was known to exist but not well quantified. For this reason, it has recently been suggested that ethics approval of future dose-response experiments should be contingent upon use of disaggregated pathogens (or pathogens aggregated to a known extent) [56].…”

Section: Discussionmentioning

confidence: 99%

Evaluation of the 50% Infectious Dose of Human Norovirus Cin-2 in Gnotobiotic Pigs: A Comparison of Classical and Contemporary Methods for Endpoint Estimation

Ramesh

Parreño

Schmidt

et al. 2020

Viruses

Self Cite

View full text Add to dashboard Cite

Human noroviruses (HuNoVs) are the leading causative agents of epidemic and sporadic acute gastroenteritis that affect people of all ages worldwide. However, very few dose–response studies have been carried out to determine the median infectious dose of HuNoVs. In this study, we evaluated the median infectious dose (ID50) and diarrhea dose (DD50) of the GII.4/2003 variant of HuNoV (Cin-2) in the gnotobiotic pig model of HuNoV infection and disease. Using various mathematical approaches (Reed–Muench, Dragstedt–Behrens, Spearman–Karber, exponential, approximate beta-Poisson dose–response models, and area under the curve methods), we estimated the ID50 and DD50 to be between 2400–3400 RNA copies, and 21,000–38,000 RNA copies, respectively. Contemporary dose–response models offer greater flexibility and accuracy in estimating ID50. In contrast to classical methods of endpoint estimation, dose–response modelling allows seamless analyses of data that may include inconsistent dilution factors between doses or numbers of subjects per dose group, or small numbers of subjects. Although this investigation is consistent with state-of-the-art ID50 determinations and offers an advancement in clinical data analysis, it is important to underscore that such analyses remain confounded by pathogen aggregation. Regardless, challenging virus strain ID50 determination is crucial for identifying the true infectiousness of HuNoVs and for the accurate evaluation of protective efficacies in pre-clinical studies of therapeutics, vaccines and other prophylactics using this reliable animal model.

show abstract

“…A Monte Carlo based exposure dose is usually applied in the QMRA of water supplies ( Schijven et al., 2011 ; Whitaker et al., 2005 ) rather than a deterministic (point-value) dose assumption, which is also widely used. A Monte Carlo Markov Chain (MCMC) analysis, based on the Gamma prior distribution ( Schmidt et al., 2013 , 2019 ) was applied in this study to determine the 5%, 50 and 95% percentiles of the target counts of E. coli in both contaminated reclaimed water of vegetables and polluted seawater at the Forcatella beaches. Specifically, we considered the host’s ingestion (or consumption) of raw vegetables, such as salads, tomatoes, cucumbers, and fruits (e.g.…”

Section: Methodsmentioning

confidence: 99%

Application of QMRA to MAR operations for safe agricultural water reuses in coastal areas

Masciopinto¹,

Vurro²,

Lorusso³

et al. 2020

Water Research X

View full text Add to dashboard Cite

A pathogenic Escherichia coli ( E.coli ) O157:H7 and O26:H11 dose-response model was set up for a quantitative microbial risk assessment (QMRA) of the waterborne diseases associated with managed aquifer recharge (MAR) practices in semiarid regions. The MAR facility at Forcatella (Southern Italy) was selected for the QMRA application. The target counts of pathogens incidentally exposed to hosts by eating contaminated raw crops or while bathing at beaches of the coastal area were determined by applying the Monte Carlo Markov Chain (MCMC) Bayesian method to the water sampling results. The MCMC provided the most probable pathogen count reaching the target and allowed for the minimization of the number of water samplings, and hence, reducing the associated costs. The sampling stations along the coast were positioned based on the results of a groundwater flow and pathogen transport model, which highlighted the preferential flow pathways of the transported E. coli in the fractured coastal aquifer. QMRA indicated tolerable (<10 −6 DALY) health risks for bathing at beaches and irrigation with wastewater, with 0.4 infectious diseases per year (11.4% probability of occurrence) associated with the reuse of reclaimed water via soil irrigation even though exceeding the E. coli regulation limit of 10 CFU/100 mL by five times. The results show negligible health risk and insignificant impacts on the coastal water quality due to pathogenic E. coli in the wastewater used for MAR. However, droughts and reclaimed water quality can be considered the main issues of MAR practices in semiarid regions suggesting additional reclaimed water treatments and further stress-tests via QMRAs by considering more persistent pathogens than E. coli .

show abstract

Recognizing Structural Nonidentifiability: When Experiments Do Not Provide Information About Important Parameters and Misleading Models Can Still Have Great Fit

Cited by 13 publications

References 26 publications

Outbreak‐BasedGiardiaDose–Response Model Using Bayesian Hierarchical Markov Chain Monte Carlo Analysis

Outbreak‐BasedGiardiaDose–Response Model Using Bayesian Hierarchical Markov Chain Monte Carlo Analysis

Evaluation of the 50% Infectious Dose of Human Norovirus Cin-2 in Gnotobiotic Pigs: A Comparison of Classical and Contemporary Methods for Endpoint Estimation

Application of QMRA to MAR operations for safe agricultural water reuses in coastal areas

Contact Info

Product

Resources

About