Strategies for tackling Taenia solium taeniosis/cysticercosis: A systematic review and comparison of transmission models, including an assessment of the wider Taeniidae family transmission models

Abstract

Background The cestode Taenia solium causes the neglected (zoonotic) tropical disease cysticercosis, a leading cause of preventable epilepsy in endemic low and middle-income countries. Transmission models can inform current scaling-up of control efforts by helping to identify, validate and optimise control and elimination strategies as proposed by the World Health Organization (WHO). Methodology/Principal findings A systematic literature search was conducted using the PRISMA approach to identify and compare existing T . solium transmission models, and related Taeniidae infection transmission models. In total, 28 modelling papers were identified, of which four modelled T . solium exclusively. Different modelling approaches for T . solium included deterministic, Reed-Frost, individual-based, decision-tree, and conceptual frameworks. Simulated interventions across models agreed on the importance of coverage for impactful effectiveness to be achieved. Other Taeniidae infection transmission models comprised force-of-infection (FoI), population-based (mainly Echinococcus granulosus ) and individual-based (mainly E . multilocularis ) modelling approaches. Spatial structure has also been incorporated ( E . multilocularis and Taenia ovis ) in recognition of spatial aggregation of parasite eggs in the environment and movement of wild animal host populations. Conclusions/Significance Gaps identified from examining the wider Taeniidae family models highlighted the potential role of FoI modelling to inform model parameterisation, as well as the need for spatial modelling and suitable structuring of interventions as key areas for future T . solium model development. We conclude that working with field partners to address data gaps and conducting cross-model validation with baseline and longitudinal data will be critical to building consensus-led and epidemiological setting-appropriate intervention strategies to help fulfil the WHO targets.

Force-of-infection of Taenia solium porcine cysticercosis: a modelling analysis to assess global incidence and prevalence trends

“…Chagas disease, trachoma) to investigate spatial heterogeneity and temporal incidence trends (secular or elicited by interventions) 55 – 57 . The FoI estimates in this study show variation among settings (between and within countries), reinforcing the importance of understanding local epidemiological dynamics for the parameterisation of mathematical transmission models 58 and the implementation of tailored, setting-specific intervention strategies 59 . A preliminary characterisation of different endemicity settings is also postulated, identifying that PCC FoI estimate ranges based on observed data are 0.0009–0.077 month −1 for hypoendemic settings (0–9.99% all-age (sero)prevalence); 0.002–0.276 month −1 for mesoendemic settings (10–24.99% all-age (sero)prevalence), and 0.097–0.529 month −1 for hyperendemic settings (≥ 25% all-age (sero)prevalence).…”

Modelling for Taenia solium control strategies beyond 2020

“…Recent research has characterized and compared the structures of these models, and their parameters and capabilities to model particular interventions. 13 Building on this research, formal comparisons of models are needed to understand the extent of the biological and epidemiological uncertainties associated with the life cycle of T. solium and its transmission properties in order to identify key unknown factors that would benefit from the collection of new data (further information available in the data repository) 14 and ultimately to build consensus on the most suitable intervention options. 15 Some ways to fill critical data gaps and hence improve our ability to capture baseline epidemiology in models in the absence of interventions include: better characterization of local transmission dynamics, such as force-of-infection estimates (individual rate of infection acquisition in susceptible hosts), which can be obtained from human and pig age-stratified prevalence data; risk factor analysis to improve the defining of parameters, for example, of contact rates with stages infective to humans and pigs; biological factors of adult tapeworm, for example, life span and reproductive output; and environmental factors, which will likely be highly dependent on setting, for example, egg viability studies and dispersal mechanisms to better understand the distribution of environmental contamination.…”

“…19 Formal model comparison research will require initial identification of similarities and differences in key structural and parametric assumptions, which has begun to be addressed by researchers. 13 After this step, the parameters shared between the models need to be harmonized, such as the ratio of human-to-pig population sizes, to test their influence on model outcomes, for example, stable, viable T. solium endemic prevalence's. Through this process, the residual differences between the models will point to important epidemiological, biological and setting-specific uncertainties that will inform the research agenda.…”

“…However, technical definitions of both intensified control and hyperendemicity need clarification for progress towards these goals over the next decade to be effectively evaluated. A key modelling activity should involve predicting, in a variety of epidemiological settings, the effect of various interventions in the currently available set • yes, if realistic control targets are proposed • modelling can inform the design and evaluation of pilot and large-scale control programmes with current and complementary intervention strategies 8,13 • cystiSim and EPICYST, computational transmission models, are applicable; 17,18 cystiSim is already in use in Zambia and Latin America, and an additional transmission model has also been developed for Latin America 23,24 What is required to achieve the target?…”

Understanding transmission and control of the pork tapeworm with CystiAgent: a spatially explicit agent-based model

“…While each of these attempts have moved the T. solium modeling agenda forward, limitations in structure and data quality [19] have prevented these models from providing the detailed insights needed to inform future control strategies. These existing models, like many traditional infectious disease models, rely on assumptions of spatial homogeneity, closed-populations, and parameter values that are averaged across large populations.…”