Overcoming the Scalability Challenges of Epidemic Simulations on Blue Waters

Modeling dynamical systems represents an important application class covering a wide range of disciplines including but not limited to biology, chemistry, finance, national security, and health care. Such applications typically involve large-scale, irregular graph processing, which makes them difficult to scale due to the evolutionary nature of their workload, irregular communication and load imbalance. EpiSimdemics is such an application simulating epidemic diffusion in extremely large and realistic social contact networks. It implements a graph-based system that captures dynamics among co-evolving entities. This paper presents an implementation of EpiSimdemics in Charm++ that enables future research by social, biological and computational scientists at unprecedented data and system scales. We present new methods for application-specific processing of graph data and demonstrate the effectiveness of these methods on a Cray XE6, specifically NCSA's Blue Waters system.

Transmission Potential of Rift Valley Fever virus over the course of the 2010 epidemic in South Africa

A Rift Valley fever (RVF) epidemic affecting animals on domestic livestock farms was reported in South Africa during January-August 2010. The first cases occurred after heavy rainfall, and the virus subsequently spread countrywide. To determine the possible effect of environmental conditions and vaccination on RVF virus transmissibility, we estimated the effective reproduction number (R) for the virus over the course of the epidemic by extending the Wallinga and Teunis algorithm with spatial information. Re reached its highest value in mid-February and fell below unity around mid-March, when vaccination coverage was 7.5%-45.7% and vector-suitable environmental conditions were maintained. The epidemic fade-out likely resulted first from the immunization of animals following natural infection or vaccination. The decline in vector-suitable environmental conditions from April onwards and further vaccination helped maintain R below unity. Increased availability of vaccine use data would enable evaluation of the effect of RVF vaccination campaigns.

The worldwide epidemic of diabetic retinopathy

Diabetic retinopathy (DR), a major microvascular complication of diabetes, has a significant impact on the world′s health systems. Globally, the number of people with DR will grow from 126.6 million in 2010 to 191.0 million by 2030, and we estimate that the number with vision-threatening diabetic retinopathy (VTDR) will increase from 37.3 million to 56.3 million, if prompt action is not taken. Despite growing evidence documenting the effectiveness of routine DR screening and early treatment, DR frequently leads to poor visual functioning and represents the leading cause of blindness in working-age populations. DR has been neglected in health-care research and planning in many low-income countries, where access to trained eye-care professionals and tertiary eye-care services may be inadequate. Demand for, as well as, supply of services may be a problem. Rates of compliance with diabetes medications and annual eye examinations may be low, the reasons for which are multifactorial. Innovative and comprehensive approaches are needed to reduce the risk of vision loss by prompt diagnosis and early treatment of VTDR.

Evidence for an "epidemic" of myopia.

INTRODUCTION:

It has been widely suggested that the prevalence of myopia is growing worldwide, and that the increases observed in East Asia, in particular, are sufficiently severe as to warrant the term "epidemic". Data in favour of a cohort effect in myopia prevalence are reviewed, with attention to significant shortcomings in the quality of available evidence. Additional factors contributing to myopia prevalence, including near work, genetics and socioeconomic status, are detailed.

MATERIALS AND METHODS:

Medline search of articles regarding myopia prevalence, trends and mechanisms.

RESULTS:

Age-related changes in myopia prevalence (increase during childhood, and regression in the fifth and sixth decades) are discussed as an alternative explanation for cross-sectional patterns in myopia prevalence. There have only been a handful of studies that have examined the relative contribution of longitudinal changes in refraction over life and birth cohort differences on age-specific myopia prevalence as measured in cross-sectional studies. Available data suggest that both longitudinal changes and cohort effects may be present, and that their relative contribution may differ in different racial groups.

CONCLUSIONS:

In view of the relatively weak evidence in favour of a large cohort effect for myopia in East Asia, and the even greater lack of evidence for increased prevalence of secondary ocular pathology, there appears to be inadequate support for large-scale interventions to prevent or delay myopia at the present time.

TELL ME Design: Protective Behaviour During an Epidemic

The TELL ME agent based model simulates the connections between health agency communication, personal decisions to adopt protective behaviour during an influenza epidemic, and the effect of those decisions on epidemic progress. The behaviour decisions are modelled with a combination of personal attitude, behaviour adoption by neighbours, and the local recent incidence of influenza. This paper sets out and justifies the model design, including how these decision factors have been operationalised. By exploring the effects of different communication strategies, the model is intended to assist health authorities with their influenza epidemic communication plans. It can both assist users to understand the complex interactions between communication, personal behaviour and epidemic progress, and guide future data collection to improve communication planning.

Thunderstorm Related Asthma - the Epidemic of 24/25 June 1994

Background A large epidemic of asthma occurred following a thunderstorm in southern and central England on 24/25 June 1994. A collaborative study group was formed. Objectives To describe the epidemic and the meteorological, aerobiological and other environmental characteristics associated with it. Methods Collation of data from the Meteorological Office, the Pollen Research Unit, the Department of the Environment's Automatic Urban Network, from health surveillance by the Department of Health and the National Poisons Unit, from clinical experience in general practice and hospitals, and from an immunological study of some of the affected cases from north east London. Results The thunderstorm was a Mesoscale Convective System, an unusual and large form of storm with several centres and severe wind gusts. It occurred shortly after the peak grass pollen concentration in the London area. A sudden and extensive epidemic occurred within about an hour affecting possibly several thousand patients. Emergency services were stretched but the epidemic did not last long. Cases had high serum levels of IgE antibody to mixed grass pollen. Conclusion This study supports the view that patients with specific IgE to grass pollen are at risk of thunderstorm-related asthma. The details of the causal pathway from storm to asthma attack are not clear. Case-control and time series studies are being carried out.

Trends in lung cancer among young European women: the rising epidemic in France and Spain.

Lung cancer mortality in young women in the European Union (EU) has steadily increased until the mid 1990 s and has levelled off thereafter, but trends have been heterogeneous in various countries. We analyzed therefore age-standardized trends in lung cancer mortality in young women (20-44) for the 6 major European countries, using joinpoint regression. In the early 1970s the highest lung cancer mortality in young women was in the UK (2.1/100,000). UK rates, however, steadily declined and in 2000-2004 they were the lowest of all 6 major EU countries (1.2/100,000). The second lowest rate in 2000-2002 was in Italy, whose rates remained around 1.1/100,000 between 1970 and 1994, and increased to 1.4 thereafter. In Germany and Poland, lung cancer rates in young women rose from 0.8-1.0/100,000 in the early 1970s to 1.7-1.9 in the mid 1990 s and levelled off during the last decade. Major rises over recent years were observed in France (from 0.8/100,000 in 1985-1989 to 2.2 in 2000-2003) and in Spain (from 0.8 in the 1985-1989 to 1.7 in 2000-2004). Thus, France showed both the highest rate observed over the last 3 decades and the largest rise over the last 2 decades. Since recent trends in the young give relevant information to the likely future trends in middle age, the female lung cancer epidemic is likely to expand in southern Europe from the current rates of 5.0/100,000 in Spain and 7.7 in France to approach 20/100,000 within the next 2-3 decades. Urgent interventions for smoking cessation in women are therefore required.

Use and abuse of mathematical models: an illustration from the 2001 foot and mouth disease epidemic in the United Kingdom

Foot and mouth disease (FMD) is a major threat, not only to countries whose economies rely on agricultural exports, but also to industrialised countries that maintain a healthy domestic livestock industry by eliminating major infectious diseases from their livestock populations. Traditional methods of controlling diseases such as FMD require the rapid detection and slaughter of infected animals, and any susceptible animals with which they may have been in contact, either directly or indirectly. During the 2001 epidemic of FMD in the United Kingdom (UK), this approach was supplemented by a culling policy driven by unvalidated predictive models. The epidemic and its control resulted in the death of approximately ten million animals, public disgust with the magnitude of the slaughter, and political resolve to adopt alternative options, notably including vaccination, to control any future epidemics. The UK experience provides a salutary warning of how models can be abused in the interests of scientific opportunism.

Risk of foot and mouth disease associated with proximity in space and time to infected premises and the implications for control policy during the 2001 epidemic in Cumbria

An analysis was made that calculated the risk of disease for premises in the most heavily affected parts of the county of Cumbria during the foot-and-mouth disease epidemic in the UK in 2001. In over half the cases the occurrence of the disease was not directly attributable to a recently infected premises being located within 1.5 km. Premises more than 1.5 km from recently infected premises faced sufficiently high infection risks that culling within a 1.5 km radius of the infected premises alone could not have prevented the progress of the epidemic. A comparison of the final outcome in two areas of the county, south Penrith and north Cumbria, indicated that focusing on controlling the potential spread of the disease over short distances by culling premises contiguous to infected premises, while the disease continued to spread over longer distances, may have resulted in excessive numbers of premises being culled. Even though the contiguous cull in south Penrith appeared to have resulted in a smaller proportion of premises becoming infected, the overall proportion of premises culled was considerably greater than in north Cumbria, where, because of staff and resource limitations, a smaller proportion of premises contiguous to infected premises was culled