The Bayesian Microbial Subtyping Attribution Model: Robustness to Prior Information and a Proposition

Abstract

Attributing foodborne illnesses to food sources is essential to conceive, prioritize, and assess the impact of public health policy measures. The Bayesian microbial subtyping attribution model by Hald et al. is one of the most advanced approaches to attribute sporadic cases; it namely allows taking into account the level of exposure to the sources and the differences between bacterial types and between sources. This step forward requires introducing type and source-dependent parameters, and generates overparameterization, which was addressed in Hald's paper by setting some parameters to constant values. We question the impact of the choices made for the parameterization (parameters set and values used) on model robustness and propose an alternative parameterization for the Hald model. We illustrate this analysis with the 2005 French data set of non-typhi Salmonella. Mullner's modified Hald model and a simple deterministic model were used to compare the results and assess the accuracy of the estimates. Setting the parameters for bacterial types specific to a unique source instead of the most frequent one and using data-based values instead of arbitrary values enhanced the convergence and adequacy of the estimates and led to attribution estimates consistent with the other models' results. The type and source parameters estimates were also coherent with Mullner's model estimates. The model appeared to be highly sensitive to parameterization. The proposed solution based on specific types and data-based values improved the robustness of estimates and enabled the use of this highly valuable tool successfully with the French data set.

Source attribution of human campylobacteriosis at the point of exposure by combining comparative exposure assessment and subtype comparison based on comparative genomic fingerprinting

“…The prior for a j was assigned an Exponential (0.02, 0.02), also suggested by Mullner [7]. Following previous work [30], the value for each q i “anchor” was set to fixed value for each subtype that was source-specific, meaning those subtypes i that were present from human isolates and found in only one source. In those cases the q i value was: …”

Risk Factors for Human Salmonellosis Originating from Pigs, Cattle, Broiler Chickens and Egg Laying Hens: A Combined Case-Control and Source Attribution Analysis

“…Salmonella source attribution is being performed in several countries to ascertain the main food-producing animal reservoirs towards which control efforts should be directed and to assess the impact of such interventions [15]–[20]. Classical case-control studies can only trace back the source of human infections up to the exposure (e.g.…”

Scientific Opinion on the evaluation of molecular typing methods for major food‐borne microbiological hazards and their use for attribution modelling, outbreak investigation and scanning surveillance: Part 1 (evaluation of methods and applications)

“…One limitation is that these subtype-dependent and source-dependent parameters estimated to account for these relative impacts are arbitrary and can best be described as multiplication factors. This helps the model to arrive at the most likely solution given the observed data (Hald et al, 2004;David et al, 2012). The specific parameters are consequently difficult to interpret and consistency between models using different datasets is not always seen.…”