1 / 52

Explicit, Or Not, Every Decision is Based on a Model

Explicit, Or Not, Every Decision is Based on a Model . Don Burke Dean, Graduate School of Public Health University of Pittsburgh Institute for Systems Science and Health 22 May 2011 . Outline. Systems thinking, modeling, and dynamics Toy (simple) epidemic models

ozzie
Download Presentation

Explicit, Or Not, Every Decision is Based on a Model

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Explicit, Or Not, Every Decision is Based on a Model Don Burke Dean, Graduate School of Public Health University of Pittsburgh Institute for Systems Science and Health 22 May 2011

  2. Outline Systems thinking, modeling, and dynamics Toy (simple) epidemic models Serious H1N1 pandemic decision support models Health behavior models In defense of modeling Conclusions

  3. Jay Forrester (1918- )MIT Professor of Computer Science and ManagementFounder of the field of System Dynamics • "All decisions are made on the basis of models. Most models are in our heads. Mental models are not true and accurate images of our surroundings, but are only sets of assumptions and observations gained from experiences ... Computer simulation models can compensate for weaknesses in mental models" (Forrester, 1994).

  4. System Dynamics • “System dynamics is the necessary foundation underlying effective thinking about systems”- Jay Forrester, 1999

  5. Example: Classical Predator-Prey Oscillations in Ecological Systems (Lotka–Volterra) Animals Time

  6. Example: Linked Oscillators in Mechanical Systems

  7. Example: Linked Oscillators in Mechanical Systems

  8. BusinessWeek / Best Leaders

  9. Prediction can be very difficult . . . especially if you are trying to predict the future Neils Bohr Physicist

  10. Predictive Approaches: A modest revision of oFRescher’s categories Don Burke INFORMAL = (1) INFORMED GUESS, NO EXPLICIT MODEL ( = “LAZY” ) FORMAL RUDIMENTARY = (2) EMPIRICAL PROJECTION FROM A STATISTICAL MODEL ( =“BORING”) FORMAL SCIENTIFIC = (3) PROJECTION FROM A DYNAMIC DETERMINISTIC MODEL ( = “ARROGANT” ) (4) PROJECTION FROM A DYNAMIC STOCHASTIC SIMULATION MODEL ( = “TOO DAMNED COMPLICATED”)

  11. Some Systems ● Solar System ● Eco System ● Health Care System ● Financial System

  12. 10^8 m Earth 10^6 m Country 10^4 m City 10^2 m Village 10^0 m Human 10^-2 Tonsil 10^-4 Lymph follicle 10^-6 Cell 10^-8 DNA 10^-10 Nucleotide 10^-12 X ray 10^-14 Atomic nucleus “Systems Public Health” Systems Biology

  13. Association of Schools of Public Health “Systems Thinking” Competency

  14. Systems Thinking Competencies: Upon graduation a student with an MPH should be able to… 1. Identify characteristics of a system. 2. Identify unintended consequences produced by changes made to a public health system. 3. Provide examples of feedback loops and “stocks and flows” within a public health system.4. Explain how systems (e.g. individuals, social networks, organizations, and communities) may be viewed as systems within systems in the analysis of public health problems. 5. Explain how systems models can be tested and validated. 6. Explain how the contexts of gender, race, poverty, history, migration, and culture are important in the design of interventions within public health systems. 7. Illustrate how changes in public health systems (including input, processes, and output) can be measured. 8. Analyze inter-relationships among systems that influence the quality of life of people in their communities. 9. Analyze the effects of political, social and economic policies on public health systems at the local, state, national and international levels. 10. Analyze the impact of global trends and interdependencies on public health related problems and systems.

  15. Wordle “Systems Thinking Competencies / ASPH” by don burke

  16. Toy (simple) models of system dynamics in public health

  17. Growing Artificial Societies Joshua M. Epstein Robert Axtell Brookings Institution 1996 Generative Social Science Joshua M. Epstein Princeton Univ Press 2007 “If you can’t grow it, you don’t understandit.” Josh

  18. Smallpox modeling: Building Individual-based social structures and contact networks

  19. Computer screen at start of model run: one infected individual [N.B. “night-time” = all individuals at home, not at work or school morgue hosp

  20. Next Step: Scale-up of social networks & change from cartoon to “real” social structures

  21. Models of Infectious Disease Agent Studies NIH/NIGMS National Center of Excellence Pitt ---- PSC ---- CMU Imperial---Hopkins---Brookings

  22. H1N1 pandemic decision support using large scale agent based simulations

  23. Model of a USA pandemic Ferguson NM, Cummings DA, Fraser C, Cajka JC, Cooley PC, Burke DS. Strategies for mitigating an influenza epidemic Nature July 27, 2006; 442: 448-52

  24. FRED: Framework for Reconstruction of Epidemic Dynamics Simulation Information Management System (FRED SIMS) FRED Core Synthetic Population Request DB Request Queue FRED Web Page Pathogen Parameters FRED Core FRED Web Service FRED Simulation Engine Natural History, Viral Evolution FRED Interface PSC Intervention Policies Health Belief Model Social Network Influences Vaccination Antivirals School Closure Preventive Behaviors FRED Client Results DB FRANCIS Analysis and Visualization Tools Behavior Change Model GAIA

  25. Pennsylvania Visualization

  26. Typical MIDAS - ASPR /BARDA decision support team meeting at Pitt • Don Burke, PI • Ron Vorhees, Allegheny County Epidemiologist • Rick Zimmerman, Community Health Physician • John Grefenstette, Computer Scientist • Cho-Cho Lin, Economist • Sandra Quinn, Behavioral Scientist • Jim Stark, Epidemiology Graduate student • Shanta Zimmer, Infectious Disease Physician • Shawn Brown, Computational Scientist • Roni Rosenfeld, Computer Scientist • Maggie Potter, Lawyer & Public Health Practice • Bruce Lee, Internal Med Physician & Operations Research Bruce or Shawn on phone in DC

  27. Washington DC Metro Visualization

  28. Behavior Change Theories • Health Belief Model • Trans-Theoretical Model • Social Cognitive Theory • Theory of Planned Behavior • Social Ecological Model

  29. I can calculate the motions of heavenly bodies, but not the madness of people. Isaac Newton, 1721

  30. Decision to Stay Home Cases Cases Child Stays Home

  31. In defense of modeling Some tips on how to win over the skeptics

  32. 1. What is a model?

  33. Rene Magritte, 1929

  34. Cecin’est pas uneepidemie

  35. 2. Can you trust a model?

  36. “All models are wrong, some are useful” George E. P. Box Robustness in the Strategy of Scientific Model Building 1979

  37. MODELERS UNITE: STOP SAYING THIS ! “All models are wrong, some are useful” aphorisms

  38. 3. Aren’t models totally dependent on the quality of data?

  39. “Garbage in, garbage out” George Fuechsel IBM technician/instructor in New York ca 1956

  40. MODELERS STOP APOLOGIZING ! “Garbage in, garbage out” • This is just a true for passive non-computational • mental models as it is for computationally • explicit models • AND it is arguably better to be explicit • about your garbage than blissfully ignore it

  41. OK Where do we go now?

  42. EPISTEMOLOGY The branch of philosophy concerned with the nature and scope ( and limitations) of knowledge. It addresses the questions: • What is knowledge? • How is knowledge acquired? • How do we know what we know?

  43. A Bit About The University of Pittsburgh • Public Health Dynamics Laboratory • Approach • Organization • Support • Output

  44. The Pitt Public Health Dynamics Lab Systems Thinking Data mining Time series decomposition Social networks GIS Game theory Remote sensing Machine learning & A.I. High-performance computing Info Sci Epid Data Biostat Patterns & Parameters Math Clara Comp Sci Equation Based Models IndustEngin Agent & Network Models Enviro Engin Behav Modeling of Interventions Law Modeling of Pragmatics Econ Policy Cost Benefit Analyses Philos Benter Foundation Decision Support / Clients Modeling Steps Software Academic Disciplines

More Related