Professor David L. Smith, PhD
University of Washington
\(\odot\)
Professor Smith leads a team that has been developing adaptive management for malaria, or Adaptive Malaria Control. The project has been a collaboration with Uganda’s National Malaria Control Division and Department of Health Information and with the Bioko Island Malaria Elimination Program (BIMEP).
Adaptive malaria control, inspired by adaptive management, is a pragmatic methodology for malaria decision support that can quantify and manage uncertainty. To deal with the complexity of malaria, the need for information, the reality of having enormous data gaps, and all the associated uncertainty we have developed a bespoke inferential framework called RAMP (Robust Analytics for Malaria Policy) using conventional and simulation-based analytics. To respond malaria programs’ needs for highly realistic simulation models, we developed a new mathematical framework for nimble model building with accompanying software. This applied research has exposed some limitations of existing models, prompting new approaches and advances in mosquito ecology and malaria theory.
For more information, please see my publications and an overview of the software (see SimBA).
Featured:
Hergott DEB, et al., Impact of six-month COVID-19 travel moratorium on Plasmodium falciparum prevalence on Bioko Island, Equatorial Guinea. (2024). Nat Commun 15, 8285. (View at Nat Comm)
Wu SL, et al., Spatial Dynamics of Malaria Transmission. (2023), PLoS Computational Biology 19(6):e1010684. (View at PLoS CB)
SimBA – open access software for Simulation-Based Analytics for malaria and other mosquito-transmitted pathogens
Adaptive Malaria Control
Adaptive malaria control is a structured, iterative approach to developing malaria intelligence and robust analytics for malaria policy. Malaria intelligence is developed around stable information systems that involve involve data processing and curation, the transformation of data into information through data analysis, and translation of information into malaria intelligence using simulation to support operational decisions, optimize the allocation of resources, improve technical efficiency, and develop evidence-based policies to reduce burden and eliminate malaria. Malaria is managed on monthly and annual cycles, and on multi-year funding and strategic planning cycles. Adaptive malaria control must support these policy cycles through development of protocols and procedures for repeated analysis to ensure policy advice is consistent, that it is of the highest quality, and that it is updated to be responsive to changing needs. Through careful analysis of uncertainty and through consultation with program managers, surveillance systems are reviewed to identify key data gaps, and plans are made to fill those gaps and improve future decisions.
More about Adaptive Malaria Control
Nimble Model Building
A challenge for malaria research and policy has been to build models fit for purpose. In policy, there is a need for models that can serve the needs of a malaria program in a moment as well as models that can be used in repeated analysis over time. The need for building models can exceed the capacity of existing technology. To support malaria policies, some models might need spatial dynamics, mosquito ecology with the potential for weighing integrated vector control, exogenous forcing by weather or other factors, realistic representation of the process of drug taking and chemo-protection, models with realistic human malaria infection and immunity, including disease, infectiousness, and detection by age.
How well has academic research responded to these needs? In a systematic review of mathematical models for mosquito-transmitted pathogens, published in 2013, covering more than 500 published models published before 2010, we found a distinct pattern. The need for models that were complex enough to address policy questions drove development of complex, individual-based simulation models (IBMs, such as OpenMalaria, eMod, MalariaTools). With few exceptions, most of the remaining models were very similar to the Ross-Macdonald model. While these models handle complexity with aplomb, they require large computer systems and specialized skills, so they are impractical for use in some settings.
To address the need for building dynamical systems models that can scale complexity to any level of realism – without the need for individual-based simulation – we developed a mathematical framework for the epidemiology, transmission dynamics, and control of malaria and other mosquito-transmitted pathogens that is modular, flexible, and extensible. The framework makes it easy to build, solve, and apply models with virtually unbounded complexity.
More about Nimble Model Building
Mosquito Ecology
Mosquito ecology – all the factors that determine the distribution and abundance of mosquito species – play a key role in malaria epidemiology, transmission dynamics and control. Mosquito age has drawn attention, but there are many other basic features of mosquito behavior and ecology that are important for malaria transmission by mosquitoes and their responses to control. Many of these remain poorly characterized from the perspective of malaria transmission dynamics.
In developing models and theory, we are interested in the problem of relevant details for mosquito populations and their responses to vector control. On the one hand, we have developed a highly detailed and realistic individual-based adult mosquito simulation model called MBITES. The model is extremely complex – deliberately so – to beg a question about how much detail is useful? MBITES has drawn our attention to resource availability as an important factor in adult mosquito ecology. We are developing models of intermediate complexity capture some of the essential features, including behavioral state, micro-simulation models on complex resource landscapes. We are exploring models like this, where mosquitoes in some behavioral state (e.g. blood feeding, egg laying) search for and use resources. We can also build models describing how mosquito populations are regulated, with realistic responses to mean crowding in dynamic aquatic habitats. Adult mosquito populations, habitat dynamics, and mosquito population dynamics create an important component of malaria transmission systems and their responses to vector control.
More about Mosquito Ecology
Malaria Theory
Malaria can be understood as a large set of managed, complex adaptive systems. Lewis Hackett has famously sad that,
…malaria is so moulded and altered by local conditions that it becomes a thousand different diseases and epidemiological puzzles. Like chess, it is played with a few pieces, but is capable of an infinite variety of situations.
This local peculiarity presents enormous challenges for managing malaria because it implies that each system could be unique in the way it responds to malaria control.
More about Malaria Theory
\(\odot\)
David L Smith (he/him/his) is a Professor in the Department of Health Metrics Sciences (HMS) in the School of Medicine (SOM), University of Washington (UW). He is a member of the Instute for Health Metrics and Evaluation (IHME), and he is affiliated with the Center for Quantitative Science (CQS) through the training program in Quantitative Ecology and Resource Management (QERM).
Professor Smith has published extensively on malaria, including malaria elimination and eradication; malaria spatial dynamics and control; malaria epidemiology; and evolution of resistance to anti-malarial drugs. He has also published on spatio-temporal dynamics and control of rabies; evolution of antibiotic resistance in hospital-acquired bacterial pathogens; and the ecology, transmission dynamics and control of other mosquito-transmitted pathogens and infectious diseases.
Professor Smith was a member of the Malaria Atlas Project (MAP), RAPIDD (Research and Policy for Infectious Disease Dynamics), the Malaria Elimination Group (MEG), and a working group on mathematical modeling as part of malERA (the malaria Eradication Research Agenda). He headed the Malaria Modeling Consortium (MMC).
Up-to-date publication lists are maintained at ORCID, Google Scholar, and Research Gate.
Professor Smith studied Ecology and Evolutionary Biology (EEB) at Princeton University (1993-1998), with Professor Simon A. Levin. Before that, was a graduate student and undergraduate student in mathematics at Brigham Young University (BYU). Professor Smith currently lives in Seattle, Washington.
Jobs
Before the University of Washington, Professor Smith held academic positions at universities, government, and non-governmental organizations including:
1999-2003 – Assistant Professor, Department of Epidemiology & Preventive Medicine, University of Maryland, Baltimore, Maryland, USA
2003-2007 – Research Scientist, Fogarty International Center for International Health, Bethesda, Maryland, USA
2007-2009 – Associate Professor, Department of Zoology, University of Florida, Gainesville, Florida, USA
2009-2011 – Professor, Department of Biology, University of Florida, Gainesville, Florida, USA
2007-2011 – Assistant Director for Disease Ecology, Emerging Pathogens Institute, University of Florida, Gainesville, Florida, USA
2007-2011 – Center for Disease Dynamics, Economics and Policy (CDDEP), Resources for the Future, Washington, DC
2011-2014 – Professor, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
2014-2015 – Senior Fellow, Sanaria Institute for Global Health and Tropical Medicine (SIGHTM), Rockville, Maryland, USA
2014-2016 – Visiting Professor, Department of Zoology, University of Oxford, Oxford, England, UK
Team
University of Washington
Current
Former
Sean L Wu
John M Henry
Daniel Citron
Uganda
Catherine Maiteki-Sebuguzi
Doreen Mbabazi Ssebuliba, Lecturer, Department of Mathematics and Statistics, Kyambogo University, Kampala, Uganda
Juliet Nakakawa Nsumba, Lecturer, Department of Mathematics, School of Physical Sciences, College of Natural Science, Makerere University, Kampala, Uganda
Meddy Rutayisire
Tom Eganyu
BIMEP
Guillermo Garcia
Carlos Guerra
David Galick
Archive
The links to some of the computer code from old studies are stale. Over the next few months, I plan to create archival versions of old studies.