DIMACS/MBI US - African BioMathematics Initiative: Workshop on Economic Epidemiology

August 3 - 5, 2009
Makerere University in Kampala, Uganda

Matt Bonds, Harvard University (ASI program co-organizer), mbonds at hsph.harvard.edu
Nina Fefferman, Rutgers University (Workshop program co-chair), fefferman at AESOP.Rutgers.edu
Alison Galvani, Yale University (ASI program co-organizer), alison.galvani at yale. edu
Wayne Getz, UC Berkeley, getz at nature.berkeley.edu
Abba Gumel, U. of Manitoba (ASI program organizer), gumelab at @cc.umanitoba.ca
Ramanan Laxminarayan, Resources for the Future (Workshop program co-chair), ramanan at rff.org
Simon Levin, Princeton University, slevin at princeton.edu
Jan Medlock, Clemson University (ASI program co-organizer), medlock at clemson.edu
Joseph Mugisha, Makerere University, jytmugisha at math.mak.ac.ug
Fred Roberts, DIMACS, (Workshop program co-chair), froberts at dimacs.rutgers.edu
Dave Smith, University of Florida, davesmith at ufl.edu
Presented under the auspices of the DIMACS/MBI US-African BioMathematics Initiative.

This workshop and an associated Advanced Study Institute (ASI) and are jointly organized with the Mathematics of Information Technology and Complex Systems (MITACS).

This workshop and ASI are jointly sponsored by:


Nelly Biondi, SACEMA

Title: Introduction to Health Economics

How can we meet the ever increasing demand for health care given our limited resources? All too often, Health Economics is seen as a synonym for savings and financial pressure. However, what it really aims to do is to help prioritise health care interventions so that maximum health benefits are provided to the population within available resource constraints. To address this issue, Health economists perform economic evaluation of health care interventions to help decision-makers decide on how best to allocate limited resources. This lecture offers a general definition on what Health Economics is and what economic evaluation seeks to achieve. It also provides a method on how to perform an economic evaluation including both cost-side and benefit-side before explaining the different types of economic evaluations which exist.

Matthew H. Bonds, Harvard School of Public Health, Donald C. Keenan, Pejman Rohani, Jeffrey D. Sachs

Title: Poverty Trap Formed by Feedback Between Economics and the Ecology of Infectious Diseases

While most of the world has enjoyed exponential economic growth, more than one sixth of the world is roughly as poor today as their ancestors were many generations ago. Widely-accepted general explanations for the persistence of such poverty have been elusive and are in critical need by the international development community. Building on a well-established model of human infectious diseases, we show how formally integrating simple economic and disease ecology models can naturally give rise to poverty traps, where initial economic and epidemiological conditions determine the long- term tra jectory of the health and economic development of a society. This poverty trap may therefore be broken by improving health conditions of the population. More broadly, we demonstrate that simple human ecological models can help explain broad patterns of modern economic organization.

Seydou Doumbia, Malaria Research Traning Center (MRTC), Mali

Title: Putting Non-Parametric Methods in the Service of Public Health

The popularity and widening applications of non-parametric methods, including "learning" approaches such as artificial intelligence, derive from its ability to solve otherwise exceedingly complex real-world problems for which mathematical and statistical descriptions are lacking or unmanageable. Furthermore, most epidemiological systems in Africa lack complete datasets, their possible mathematical and or statistical treatability notwithstanding. Therefore, these non-parametric methods could expand the epidemiology prediction and risk assessment toolbox by "learning" from experience and pattern extraction. Regrettably, this revolutionary field?particularly artificial intelligence?has not yet disseminated among epidemiologists and public health practitioners. The authors discuss that although the non-parametric toolbox may be complex from a theoretical perspective, its application, operationalization, and result interpretation are rather straight forward and its utility vast. This discussion focuses on risk analysis, spatial mapping (e.g. prevalence), and temporal predictions (e.g. incidence) with methods from econometrics (e.g. exponential smoothing methods), engineering (e.g. fuzzy logic), and artificial intelligence (e.g. kernel methods).

Elamin H. Elbasha, Merck Research, Erik J. Dasbach, and Ralph P. Insinga

Title: Assessing the public health and economic impact of HPV vaccination strategies

A prophylactic quadrivalent (types 6/11/16/18) vaccine against genital human papillomavirus (HPV) was first approved for use in various countries in 2006. This paper presents a nonlinear, deterministic, age-structured, mathematical model of the transmission dynamics of HPV 6, 11, 16, and 18 and associated disease occurrence in a population stratified by gender and sexual activity group. The model can assess both the epidemiologic consequences and cost effectiveness of alternative vaccination strategies. Several gender- and or age-based vaccination strategies are compared based on analyses conducted for the United States as a case study. Inputs for the model were obtained from public data sources, published literature, and analyses of clinical trial data. The results of computer simulations suggest that a prophylactic quadrivalent HPV vaccine can: (i) substantially reduce the incidence of cervical HPV disease and genital warts, (ii) increase survival among women, (iii) improve quality of life for both male and female persons, (iv) be cost-effective when administered to female persons age 12-26 years, and (v) be potentially cost-effective when implemented as a strategy that combines vaccination of both female and male persons age 12-26 years. The robustness of these conclusions with respect to changes in epidemiologic parameters, economic parameters, properties of the vaccine, and levels of vaccination coverage are tested.

Nina Fefferman, Rutgers University

Title: The Impact of Household Capital Models on Targeted Epidemiological Control Strategies for Diseases with Age-Based Etiologies

Many epidemiological models investigating outbreak control strategies have determined that targeting 'populations of concern' for prevention/treatment can both help limit overall numbers of adverse outcomes, and also limit spread to the broader population (in which fewer adverse outcomes are expected). For example, we frequently discuss how children and the elderly should be targeted for flu vaccination/antiviral treatments. However, these models have assumed that resource limitation occurs only at the community level, rather than the individual household. I will discuss a model in which infection status impacts the ability of the household to recruit economic resources for the support of general health and nutrition of its members and demonstrate how this constraint can compromise the population-level epidemiological goals of a child-oriented intervention strategy.

Patricia Geli, Resources for the Future

Title: Cost v. benefit of strategies to control antibiotic resistance

How to best control antibiotic resistance is a controversial matter. This fact is well illustrated by the vast contrast of components included in interventions around the world dealing with penicillin resistance in Streptococcus pneumoniae, which in many countries kills more young children than any other microorganism. Screening to find and isolate colonized, but healthy, individuals with resistant bacteria is one example of intervention which is in stark contrast to other strategies aiming at increasing the hygiene level in a population or vaccination. In this seminar I will talk about different aspects related to different types of interventions: Why have the analyses related to the control of antibiotic resistance been such a controversial thing, sometimes even with conflicting results, and what can we do to increase our understanding for policies that offer the greatest epidemiological impact at reasonable cost?

Ramanan Laxminarayan, Resources for the Future

Title: Incentives for surveillance and reporting of disease outbreaks

Our ability to minimize the risk of emerging diseases depends critically on effective surveillance and early detection. Current mechanisms to encourage better reporting of disease have generally ignored the economic dilemma and strategic behavior of countries with emergent outbreaks. In this lecture we assess the effectiveness of both sanctions and the local capacity for outbreak containment on surveillance and reporting of outbreaks.

Madhav Marathe, Virginia Tech

Title: Studying the Relationships Among Individual Behavior, Social Networks, Public Policies and Epidemic Process

Human behavior, social networks, and the civil infrastructures are closely intertwined. Understanding the coupled evolution is critical for situational awareness and decision support for crises preparedness and response.

For example, human behaviors and day to day activities of individuals create dense social interactions that are characteristic of modern urban societies. These dense social networks provide a perfect fabric for fast, uncontrolled disease propagation. Conversely, people's behavior in response to public policies and their perception of how the crises is unfolding as a result of disease outbreak can dramatically alter the normally stable social interactions. Understanding these epidemiological processes is of immense societal importance. Effective planning and response strategies must take these complicated interactions into account.

Recent quantitative changes in high performance and wireless computing capability have created new opportunities for collecting, integrating, analyzing and accessing information related to such large socio-technical networks. The advances in network and information science that build on this new capability provide entirely new ways for reasoning and controlling epidemics over these networks.

The talk will describe a high performance computing approach to support policy informatics as it pertains to epidemic processes in large realistic social networks. As an illustration of this approach we will describe a recent computational study for analyzing the socio-economic impact of various intervention strategies for containing epidemics.

Eduardo Massad, Universidade de Sao Paulo, Ronald H. Behrens, Marcelo N. Burattini, and Francisco A. B. Coutinho

Title: Modeling the Risk-Benefit of Chemoprophylaxis for Travelers to Areas with Stable Malaria Transmission

Malaria has emerged as a frequent problem in international travelers. The risk depends on destination, duration and season of travel. We used mathematical models to estimate the risk of non-immune persons to acquire falciparum malaria when traveling to the Amazon region. From the force of infection we calculated the risk of malaria dependent on duration of stay and season of arrival. Our results highlight that the risk for non-immune travelers to acquire malaria in the Amazon region is substantial but varies greatly with seasons and epidemic cycles. For instance, for a traveler who stays in the Amazon for 120 days during the high season, the risk of acquiring malaria was 0.5%. Cost-effectiviness analysis allows us to conclude that for people arriving at summer time, the period they can safely remain without chemoprophylaxis is restricted to one week, whereas for people arriving in the winter, that period is extended to 2 months, with intermediary values for the other seasons. Risk estimates based on mathematical modelling will help the travel medicine provider give better evidence based advice for travelers to malarial countries.

Fred Matovu, University of Makerere

Title: The Economic Impact of HIV/AIDS In Uganda

The wide spread occurrence of HIV and AIDS in Uganda and sub-Saharan Africa (SSA) in general has raised great concern about the potential adverse economic impact both at an overall macroeconomic level as well as the sectoral and household level. The economic impact is manifested in terms of reduction in worker productivity; an increase in expenditure on health care; reduced savings and investment in physical and human capital. Estimates show that at least 2-3% of GDP is lost due to HIV/AIDS illness in SSA ? the World's most affected by the pandemic. HIV/AIDS has extensive effects on enterprises through the impacts on the workforce, investment and markets. As workers become sick, their productivity falls due to time taken off for sick leave and reduced ability to perform while at work. Worker absenteeism increases thereby interrupting production levels .The death or retirement of workers on grounds of medical recommendation results in the loss of experienced labor force and more investment in skills development is required to train new work force. The analysis of the impact of HIV/AIDS on the economy is done both from the micro (household/individual) and the macro (sectoral and economy-wide) perspectives. While there have been tremendous efforts towards combating the HIV/AIDS pandemic, prevalence rates have stagnated at about 7%, while the number of new infections is rising. There is need for concerted efforts to deepen interventions such as voluntary Counseling and testing, provision of ARVs, psycho-social support programmes, prevention of mother-to-child transmission (PMTCT) to mitigate the economic and social impact of the diseases.

Jan Medlock, Clemson University

Title: Optimizing influenza vaccine distribution

The criteria applied to assess public-health policies are fundamental to policy optimization. Using a model parametrized with data from the 1957 and 1918 influenza pandemics, we determined optimal vaccination strategies with regard to five outcome measures: deaths, infections, years of life lost, contingent valuation and economic costs. There is a balance between vaccinating children who transmit most and older individuals at greatest risk of mortality. For abundant vaccine supply, all outcome measures prioritized vaccinating children. When vaccine availability or efficacy is reduced, the balance is shifted toward prioritizing those at greatest risk for some outcome measures. We also find that previous and new recommendations of the CDC are dramatically below optimum for all outcome measures. If distributed optimally, the current number of vaccine doses administered is sufficient to prevent annual epidemics and even some pandemics.

Josh Michaud, Johns Hopkins University

Title: Disease Surveillance as a Global Public Good: An Economic Epidemiology Analysis

Surveillance systems for early detection of emerging diseases are generally weak worldwide. Weak surveillance can lead to late recognition of emerging diseases, which in turn can lead to large and costly outbreaks that might have been prevented under a system with better surveillance. The lack of early detection capability can be considered a reflection of the global public good nature of disease surveillance - it is an activity prone to free-riding and a failure of collective action between countries. Still, it is not clear whether or not the current poor state of surveillance can be considered an economically efficient outcome, or if a more efficient outcome would be greater provision of surveillance for early detection.

Methods: I outline a joint epidemic/game theory model to examine the global public good aspects of disease surveillance. In the model, countries facing the possibility of an emerging disease must decide on a level of investment in their domestic disease surveillance system. Benefits from surveillance are defined as the avoided costs from cases prevented through early detection, derived from an epidemic model. In the game, countries based their decisions on the expected payoffs from surveillance, comparing potential benefits to costs. I the determine the global equilibrium outcome from these individual country decisions, and compare it to an equilibrium that is globally efficient. I examine two cases: one in which all countries are symmetric, and another in which two types of countries exist: developed and developing.

Results: In a symmetric game, model results indicate the rational equilibrium for individual countries is to only invest fully in surveillance when they can be assured that most other countries have good surveillance as well, despite the fact that full surveillance in all countries is the most efficient global outcome. In the asymmetric game a similar result holds, but under certain conditions developed countries achieve a higher payoff through subsidizing surveillance in developing countries.

Conclusions: Surveillance for emerging disease outbreaks is a coordination game between countries - all countries are better off when early disease detection systems exist everywhere, but countries making individual, uncoordinated decisions will likely choose not to invest in early detection. This outcome is reflective of the classic collective action problem associated with many pubic goods, and is consistent with the observation that disease surveillance is generally weak worldwide. A more efficient global outcome is likely to be achieved through a robust institution that effectively coordinates international subsidies for surveillance from developed to developing countries.

Paul Mugambi, Makerere University

Title: Trends and Challenges of Infectious Diseases in Africa

Sub-Saharan Africa is a host to more than three quarters of the world's disease infected population and the majority of them have no access to basic health care. Disease burden in Africa is fuelled by many factors such as ignorance about some diseases, lack of access to prevention, inadequate treatment, large populations, seasonal labor migrations, unemployment, and conflict and wars. The impact of the disease burden is reflected in all sectors of economic development. This paper highlights the economic impact of the disease burden to African countries.

Joseph Mugisha, Makerere University

Title: Some Perspectives of the Dynamics of Leishmaniasis Infection in Poor Nations of the World

The economic burden of caring for leishmaniasis illness in poor communities of developing countries, together with the subsequent coping mechanisms contribute to a cycle of poverty and undermines the development in endemic communities. In this paper, we highlight the features related to the dynamics of this disease using a mathematical model, and discuss the socio-economic impact of the disease burden.

Tim Reluga, Penn State

Title: Accounting for Individual and Community Interests in the Public-Health Management of Infectious Diseases

In his history of the Peloponnesian war, Thucydides provides one of the earliest accounts of the devastation that infectious diseases can cause cities and communities. Despite 2000 years of advancement, infectious diseases continue to plague nations around the world. While vaccines and modern medicine have greatly reduced disease burdens throughout the world, pressures from growing human populations and microbial evolution are eroding our advances. Today, public health problems are as much social as biological. In this talk, I'll describe some contemporary challenges we face in managing infectious disease, and how mathematical biology can help us understand these challenges. Using dynamical systems, Markov processes, and game theory, we can formulate and solve a rich variety of problems with practical applications related to vaccines, disease prevention and treatment, and public health in general.

Fred Roberts, DIMACS, Rutgers University

Title: Sometimes it Pays to be Greedy: Greedy Algorithms in Economic Epidemiology

Many problems in Economic Epidemiology can be formulated as optimization problems. The simplest approach to solving such a problem is often a greedy algorithm, one that always chooses the best available (cheapest, highest rated, ...) alternative at each step. We review problems for which the greedy algorithm in fact gives an optimal solution, others for which it can be guaranteed to be close, and others where it is not very good. Examples will be chosen from: assigning workers to health care tasks; choosing medical supplies to maximize value and minimize cost; locating a health care facility so as to minimize the travel times of users. Other examples will include strategies for vaccination each time period given a limited supply; and optimal strategies for sequencing medical tests in order to minimize costs and maximize success.

Abdul-Aziz Yakubu, Howard University

Title: Fish Diseases: The Impact of Periodic and Constant Harvesting Policies on TAC-regulated Fisheries Systems

Diseases are spreading in the ocean and scientists are not sure if these are simply isolated occurrences or a sign of a global problem. According to a long term study by Cornell University and the National Center for Ecological Analysis and Synthesis, disease is increasing among most kinds of marine organisms. Some of the factors contributing to the increasing sickness among the world's marine life include global warming, habitat destruction and human overfishing. In this talk, we use mathematical models to study the impact of constant and periodic fishing pressures on the persistence of total allowable catch (TAC)-regulated fisheries systems. We use Pacific Halibut and Georges Bank Cod stock data to show that under mild environmental conditions and current fishing exploitation levels, it is likely that the Pacific Halibut will persist, while the cod is endangered. However, in the presence of increased weather extremes, both species are endangered. We will also introduce a bioeconomic model for examining the net present value and dynamics of harvest under the optimal periodic and constant proportion policies for the exploited fisheries systems.

Previous: Program
Workshop Index
DIMACS Homepage
Contacting the Center
Document last modified on July 20, 2009.