April 18, 2014

August 3-13, 2010
Transmission Heterogeneity

Talk Titles and Abstracts

Odo Diekmann (Mathematical Institute, Utrecht University)

Inference concerning the spread of antibiotic resistant bacteria in ICU (Intensive Care Units)

In order to estimate the relative importance of acquisition routes, one can analyse data in the context of a Markov chain model. In the lecture I will explain how one can combine data about admission, discharge and the results of tests, with a stochastic model for transmission and next perform Maximum Likelihood Estimation. This is based on joint work with Martin Bootsma and Marc Bonten, see for more information about the interdisciplinary project and a preprint version of the paper M.C.J. Bootsma, Bonten, M.J.M., Nijssen, A.C. Fluit, and O. Diekmann: An algorithm for real-time monitoring of bacterial transmission routes in hospital settings.
Am. J. Epid. 166 (2007) 841-851 DOI : 10.1093/aje/kwm149.

Zhilan Feng (Purdue University)

Modeling the evolutionary implications of influenza medication strategies

Medication and treatment are important measures for prevention and control of influenza. However, the benefit of antiviral use can be compromised if drug-resistant strains arise. Consequently, not only the epidemic size may increase with a higher level of treatment but also the viruses may become more resistant to the antiviral drugs. We use a mathematical model to explore the impact of antiviral treatment on the transmission dynamics of influenza. The model includes both drug-sensitive and -resistant strains.
Analytical and numerical results of the model show that the conventional quantity for the control reproduction number is not appropriate to use for gaining insights into the disease dynamics. We derive a new reproduction number by considering multiple generations of infection, and demonstrate that this new reproduction number provides a more reasonable measure for evaluating control programs as well as evolutionary implications of influenza medication strategies.

John Glasser (Centers for Disease Control and Prevention)

Modeling the potential impact of heterogeneity in vaccine coverage due to religious and philosophical exemptions

Using the simplest meta-population model capable of informing vaccination policy, we have demonstrated that heterogeneity in vaccine coverage increases the population-immunity threshold (below which outbreaks will occur on the introduction of an infectious person), especially when sub-populations mix non-randomly. Insofar as children with religious or philosophical exemptions not only live in the same households or neighborhoods, or attend the same schools, but also associate preferentially with other children having like-minded parents, non-medical exemptions reduce our ability to prevent outbreaks of vaccine-preventable diseases disproportionately (i.e., more than randomly distributed and mixed susceptible people do). By virtue of higher reproduction numbers or lower vaccine efficacy, some diseases are closer to the random mixing threshold than others. Among measles, mumps and rubella, for example, mumps is closest, followed in turn by measles and rubella, and immunity to mumps may wane absent boosting. Travelers infected abroad who still harbor the pathogens responsible for these diseases on returning to the US may cause outbreaks where immunity is heterogeneous or mixing non-random. Thus, despite average two-dose coverage of the measles, mumps and rubella vaccine above 90%, the mumps outbreaks on residential college campuses in the rural mid-west during 2006 and 2009, and ongoing within an orthodox Jewish community in the northeast, might have been expected in retrospect.

Ying-Hen Hsieh (China Medical University)

Transmission Heterogeneity of 2009 pH1N1 in Central Taiwan: Some Preliminary Results

A sero-epidemiology study was carried out in one urban city (Taichung city) and one rural county (Nantou) in central Taiwan starting before the 2008 fall influenza season in Taiwan and through the beginning of 2010. Households with schoolchildren were recruited for the study and followed sequentially, with serum samples were taken in several pre-season and post-season sampling periods for haemagglutination inhibition (HI) assay against H1N1 viruses of seasonal vaccine, wild- type, and pH1N1 strains. Survey questionnaires regarding their contact patterns, social demographic characteristics, and influenza-like symptoms during the flu seasons are obtained. Using serologic evidence of infection, i.e., 4-fold increase in HI titers, we investigate the temporal changes in pH1N1 infections among this cohort, as well as transmission heterogeneity due to age, household type/size, location, status of vaccination, etc.

Gauthier Sallet (INRIA & IRD)

Differential susceptibility and infectivity. Application to transmission of HBV in Subsaharian Africa


Beate Sander RN, MBA, MEcDev1,2,3, Chris T. Bauch PhD3,4, David Fisman MD MPH5, Robert A. Fowler MD MSc3,6, Jeffrey C. Kwong MD MSc5,7,8, Andreas Maetzel MD, PhD9, Allison McGeer MD10, Janet Raboud PhD5,10, Damon Scales MD PhD6,11, Marija Zivkovic Gojovic MSc3, and Murray Krahn MD MSc3,10,11

Is a Mass Immunization Program for Pandemic (H1N1) 2009 Good Value for Money? Evidence from the Canadian Experience

1 Department of Health Policy, Management and Evaluation, University of Toronto, Toronto, Canada, 2 Division of Clinical Decision-Making and Health Care Research, University Health Network, Toronto, Canada, 3 Toronto Health Economics and Technology Assessment Collaborative, Toronto, Canada, 4 Department of Mathematics and Statistics, University of Guelph, Guelph, Canada, 5 Dalla Lana School of Public Health, University of Toronto, Toronto, Canada, 6 Department of Medicine and Interdepartmental Division of Critical Care Medicine, Sunnybrook Health Sciences Center, Toronto, Ontario, Canada, 7 Institute for Clinical Evaluative Sciences, Toronto, Canada, 8 Department of Family and Community Medicine, University of Toronto, Toronto, Canada, 9 Amgen (Europe) GmbH, Zug, Switzerland, 10 Division of Infectious Diseases, University Health Network, Toronto, Canada, 10 Faculty of Pharmacy, University of Toronto, Toronto, Canada, 11 Department of Medicine, University of Toronto, Toronto, Canada

Background: In response to pandemic H1N1 influenza 2009 outbreak, many jurisdictions undertook mass immunization programs. The objective of this study was to determine the cost-effectiveness of the mass H1N1 immunization program in Ontario, Canada's most populous province (population 13,000,000).

Methods: A cost-utility analysis comparing the H1N1 mass immunization program in Ontario to no intervention was performed from the health care payer perspective. The economic evaluation is informed by a simulation model of a pandemic H1N1 2009 outbreak in a city in Ontario. Health outcomes measured included number of cases, number of deaths and QALYs. Probabilities for health care resource use (office visits, emergency department (ED) visits, hospitalizations) and deaths were based on Ontario pandemic H1N1 surveillance data and administrative data. Costs included immunization program cost and health care cost for treating H1N1 cases and were drawn from Ontario administrative data sources. Primary outcomes were quality adjusted life-years (QALYs), costs in 2009 Canadian dollars, and cost per QALY gained (incremental cost-effectiveness ratio [ICER]).

Results: We estimated that 4.1 million cases of symptomatic influenza would have occurred (31.5% symptomatic attack rate) in the absence of an immunization program. Our model predicted that 22% of symptomatic cases, 22% of office and ED visits, 23% of hospitalizations, and of 25% of death were prevented by the program. While the program was costly (C$180,400,000), it was also highly cost effective at C$9,400/QALY gained. Projections were most sensitive to the timing of the immunization program and less sensitive to immunization program cost and QALYs. In all deterministic sensitivity analyses the ICER remained well below WHO thresholds for cost-effectiveness. Finally, probabilistic sensitivity analysis showed the H1N1 mass immunization program in Ontario to be cost-effective in all simulations (100%) at a willingness-to-pay of $25,000 per QALY.

This analysis suggests that a mass immunization program as carried out in Ontario in response to H1N1 2009 was not only effective in preventing influenza cases and health care resource use but was also highly cost-effective despite the substantial program cost.

Yiming Shao, M.D. & Ph. D. (National Center for AIDS/STD Control and Prevention & China Center for Disease Control and Prevention)

Progress, Challenge and New Strategy of AIDS Prevention and Control in China

The HIV epidemics in China started among IDUs in border region in late 1980s, followed by the 2nd wave in the paid plasma donors in central China in the mid 1990s, and the 3rd wave of sexual transmission at the turning of the century. Both hetero-sexual and homosexual transmission have taken more than half of the total reported HIV infections since 2007. Chinese government has launched an impressive AIDS prevention campaign, with strong support from international communities, and foundations such as Global Fund and the Gates Foundation. Various intervention packages, including condom promotion, methadone substitutions and needle exchange programs are conducted. Even though a lot of progress has been made, the current AIDS control strategies are still not effective enough to significantly block the epidemic spreading from high risk groups to the general population. Most all the best practices for AIDS prevention are developed by other countries, which may not fit to the local environments and some of the current programs are lacking long term sustainability. There is urgent need to develop new strategies through research and international collaboration addressing the scientific and policy challenges facing China's AIDS control efforts.

The presentation will discuss some of the challenges and provide examples of the type of researches needed to overcome those technical obstacles. Those research areas include new techniques of measuring the epidemic and its trends, a strategy to use HIV drug surveillance data to support public health approach ART program, as well as using mathematic model to design comprehensive and sustainable prevention, treatment and care program, targeting both the epidemic and its roots. Shown by the model's simulation the new strategy will lead to effectively control the prevalence and incident of HIV/AIDS in the high epidemic region within 10 years. Finally the presentation will discuss the innovative research strategy moving HIV vaccine design to clinical trials as well as how to strengthen national, regional and international cooperation towards an effective vaccine.

Yanni Xiao (School of Sciences, Xi'an Jiaotong University)

Community-based measures for mitigating the 2009 H1N1 pandemic in China

During the 2009 A/H1N1 influenza pandemic, very strict interventions including Fengxiao were taken in mainland China to prevent imported cases and to slow down the spread to the community. How effective the local control measures are remains to be assessed, and answers to this question may inform decisions whether there should be enforced and/or improved local control measures to mitigate the outbreak. We estimated the mean reproduction number and analyzed the effect of varying interventions on outbreak and attack rate by simulating the spatially stratified compartmental model. We found that early implementation of Fengxiao, although hardly avoid a global outbreak, did and can substantially delay the outbreak peak time. Such a measure, when coupled with other strong quarantine and hygiene precaution measures, can reduce significantly the final size and outbreak peak altogether. This finding strongly suggests that prompt interventions including Fengxiao should be implemented to respond to future waves and emerging epidemics in China.

Joint work with S. Tang, Y. Yang, Y. Zhou, J. Wu and Zima

Dr. Xiaoqiang Zhao (Memorial University of Newfoundland)

A Tuberculosis Model with Seasonality

The statistical data of tuberculosis (TB) cases show seasonal fluctuations in many countries. A TB model incorporating seasonality is developed and the basic reproduction ratio is defined. It is shown that the disease-free equilibrium is globally asymptotically stable and the disease eventually disappears if the basic reproduction ratio is less than one; and there exists at least one positive periodic solution and the disease is uniformly persistent if this ratio is greater than one. Numerical simulations indicate that in the latter case there may be a unique positive periodic solution which is globally asymptotically stable. Parameter values of the model are estimated according to demographic and epidemiological data in China. The simulation results are in good accordance with the seasonal variation of the reported cases of active TB in China. This talk is based on a joint work with Drs. Luju Liu and Yicang Zhou.

Back to top