- Количество слайдов: 121
Syndemic Thinking in the CDC Futures Initiative Bobby Milstein NCCDPHP Strategy Planning Process December 22, 2003
CDC Futures Initiative Directions and Open Questions What are the major challenges that stand in the way of greater effectiveness For CDC as an organization? For public health work as a societal endeavor? Is the basic problem organizational disarray, which requires rearranging, or disorientation, which demands new approaches to thinking, problem framing, decision making, and organizing itself?
What Does it Mean to Approach Public Health Work from a Syndemic Orientation? Planned as a three-year study of innovations in health planning and evaluation Member network includes 158 individuals 123 organizations 11 countries Centers for Disease Control and Prevention. Spotlight on syndemics. Syndemics Prevention Network, 2001. Accessed December 15, 2002 at
Phases of the Syndemics Project Phase 1: explore trends, dilemmas, and innovations; develop working definitions; identify core concepts and methods Phase 2: articulate the foundations of a syndemic orientation; work with others to use this perspective in transforming public health work at the CDC and beyond
CDC Futures Initiative Observations on Our Present Position Silos (organizational fragmentation) Arrogance Inefficient internal processes (molasses) Focus on processes, not impact Difficulty differentiating CDC work from other agencies Persistent gap between science and practice (intrusion of politics? ) Disconnect between CDC work and public perception (niches of opportunity) Window of opportunity for CDC leadership (on health system reform, informatics, chronic illness, and prevention broadly)
CDC Futures Initiative Possible Stakes in the Ground View public as the primary customer Lead emphasis on prevention (and protection? ) throughout the Health care delivery system (certainly) Public health system (certainly) Society (? ) Loosen fragmentation, but don’t lose specialization Stimulate organizational evolution, not revolution (i. e. , do not disrupt or dismantle existing infrastructure but make it grow in new, more balanced directions) Use, but don’t become captured by the language of customer and products
CDC’s Shifting Orientations Transformational Orientation Medical Orientation Customer Service Orientation 1946 1990 2000 What is Transforming? • People’s health status • Conditions for health (threats) • Health response systems • CDC culture • Scientific methods • The way we think about and organize public health work Elements of a syndemic orientation may help as we navigate these transitions
Why Do We Do Public Health Work? CDC Vision & Mission Healthy people, in a healthy world, through prevention To promote health and quality of life by preventing and controlling disease, injury, and disability Institute of Medicine The purpose of public health is to fulfill society’s interest in assuring the conditions in which people can be healthy How we reconcile these two frames of reference will shape the possibilities for leading health system change
Solving Problems vs. Creating Value "In problem solving we seek to make something we do not like go away. In creating, we seek to make what we truly care about exist… We can get so caught up in reacting to problems that it is easy to forget what we actually want. Organizations must do both–resolve day-to-day problems and generate new results. But if your primary role is to fix problems, individually or collectively, rather than create something new and meaningful, it's hard to maintain a sense of purpose, and. . it's difficult to harness the energy, passion, commitment, and perseverance needed to thrive in challenging times. " -- Peter Senge PM. Creating desired futures in a global society. Reflections 2003; 5(1): 1 -12.
Public Health Goals Are Expanding …and Accumulating “The perfection of means and confusion of goals characterizes our age. ” -- Albert Einstein Prevent disease and injury (~1850 -- present) Promote health and development (1974 -- present) Assure the conditions in which people can be healthy (1988 -- present)
What does it mean to organize science and society around the goal of assuring healthful conditions?
Plan for Today Core ideas leading to a syndemic orientation Health system dynamics Health Systems Workgroup report (Figure 1) Working toward a more balanced health system Living conditions and the role of individual behavior Importance of simulation modeling Progress in systems modeling (if we have time) Steps for putting maps in motion Examples from three on-going modeling projects
Seeing Syndemics “You think you understand two because you understand one. But you must also understand ‘and’. ” -- Sufi Saying The word syndemic signals a special concern for relationships Mutually reinforcing character of health problems Connections between health status and living conditions Synergy/fragmentation within the health system (e. g. , by issues, sectors, organizations, professionals and citizens)
Ideas About Interaction Co-occurring Events Confounding Connecting* Synergism Syndemic Systems * Includes several forms of connection or inter-connection such as synergy, intertwining, intersecting, and overlapping
Placing Health in a Wider Set of Relationships Health Capacity to Act Living Conditions A syndemic orientation is one of a few approaches that includes within it our power to respond
CDC’s Vision for the 21 st Century Healthy People, in a Healthy World-Through Prevention Vision Element Main Task Healthy People • Improve health Healthy World • Enhance living conditions Through Prevention • Strengthen capacity (public work for health)
Working Across Multiple Scales Goals People Places • Prevent disease and injury • People with affliction • Neighborhoods Whole • Promote health and development • Sub-groups with greater burden of affliction • Regions Greater Whole • Assure conditions in which people can be healthy • Society with a recurring problem of inequitable burden • Planet Part
Innovations in Public Health Work Steps in Public Health Problem Solving Trends and Emerging Priorities Define the problem • Eliminate health disparities • Avoid activity limitation • Promote life satisfaction • Increase healthy days Determine the cause • Social determinants of health • Income inequality • Eroding social capital • Unhealthy built environment • Adverse childhood experiences Develop and test interventions • Comprehensive community initiatives • Ecological perspectives • Inter-sector collaboration • Health impact assessments Implement programs and policies • Policy interventions • Community and systems change • Adaptation to local context And scores more….
Innovations Point to the Emergence of a Syndemic Orientation Public health work is becoming more… Inter-connected (ecological, multi-causal, dynamic, systems-oriented) Concerned more with leverage than control Public (broad-based, partner-oriented, citizen-led, inter-sector, democratic) Concerned with many interests and mutual-accountability Questioning (evaluative, reflexive, practical) Concerned with creating and protecting values like health, security, satisfaction, justice, wealth, and freedom in both means and ends Many other orientations rely on disconnected, singular, and unthinking approaches where means and ends have very different qualities (e. g. , security by means of war)
A Complementary Science of Relationships Efforts to Reduce Population Health Problems Problem, problem solver, response Efforts to Organize a System that Protects the Public’s Health Dynamic interaction among multiple problems, problem solvers, and responses
Core Public Health Functions Under a Syndemic Orientation ASSESSMENT Social Navigation Network Analysis Categorical Syndemic Orientation ASSURANCE System Dynamics POLICY DEVELOPMENT
Core Public Health Functions Under a Syndemic Orientation Techniques • Leadership/institutional development • Power and interest mapping • Broad-based, multi-issue organizing • Action planning • Public work • Flow charting (logic mapping) • Journey mapping • Navigational statistics ASSESSMENT Techniques • Problem naming • Network analysis • Time-trend analysis • Summary measures Social Navigation Network Analysis Categorical Syndemic Orientation ASSURANCE System Dynamics Techniques • Causal diagramming • Storytelling, scenario-based planning • Game-based learning • Simulation experiments • Health impact assessment POLICY DEVELOPMENT
The term epidemic, first used in 1603, signifies a kind of relationship wherein something is put upon the people Epidemiology appeared 270 years later, in the title of J. P. Parkin's book "Epidemiology, or the Remoter Causes of Epidemic Diseases“ Ever since then, the conditions that cause health problems have increasingly become matters of public concern and public work
Public Health Began as Public Work “Public death was first recognized as a matter of civilized concern in the nineteenth century, when some public health workers decided that untimely death was a question between men and society, not between men and God. Infant mortality and endemic disease became matters of social responsibility. Since then, and for that reason, millions of lives have been saved…. The pioneers of public health did not change nature, or men, but adjusted the active relationship of men to certain aspects of nature so that the relationship became one of watchful and healthy respect. -- Gil Elliot G. Twentieth century book of the dead. New York, : C. Scribner, 1972.
Changing (and Accumulating) Ideas in Causal Theory What accounts for poor community health? God’s will Humors, miasma, ether Poor living conditions, immorality (sanitation) 1840 Single disease, single cause (germ theory) 1880 Single disease, multiple causes (heart disease) 1950 Single cause, multiple diseases (tobacco) 1960 Multiple causes, multiple diseases (but no feedback dynamics) (social epidemiology) 1980 Dynamic feedback among afflictions, living conditions, and response capacity (syndemic) 2000
Focused Efforts to Prevent and Control Diseases Have Led to Major Achievements Actual and Expected Death Rates for Coronary Heart Disease, 1950– 1998 Age-adjusted Death Rate per 100, 000 Population 700 Rate if trend continued 600 500 Peak Rate 400 300 200 Actual Rate 100 50 1955 1960 1965 1970 1975 1980 1985 1990 1995 Year Marks JS. The burden of chronic disease and the future of public health. CDC Information Sharing Meeting. Atlanta, GA: National Center for Chronic Disease Prevention and Health Promotion; 2003.
Specialization A Proven Problem Solving Approach Identify disease Determine causes Develop and test interventions Implement programs and policies Repeat steps 1 -4, as necessary!
But “Solutions” Can Also Create New Problems Merton RK. The unanticipated consequences of purposive social action. American Sociological Review 1936; 1936: 894 -904. Forrester JW. Counterintuitive behavior of social systems. Technology Review 1971; 73(3): 53 -68.
Side Effects of Specialization Issue Organizations A B C D E Confusion, inefficiency, organizational disarray Competition for shared resources Attention to “local” causes, near in time and space D B Neglected feedback (+ and -) E A Confounded evaluations Coercive power dynamics C Priority on a single value, implicitly or explicitly devaluing others Neighborhood Limited mandate to address context (living conditions) or infrastructure (public strength) Disappointing track record, especially with regard to inequalities
Diseases of Disarray Hardening of the categories Tension headache between treatment and prevention Hypocommitment to training Cultural incompetence Political phobia Input obsession Wiesner PJ. Four disease of disarray in public health. Annals of Epidemiology. 1993; 3(2): 196 -8. Chambers LW. The new public health: do local public health agencies need a booster (or organizational "fix") to combat the diseases of disarray? Canadian Journal of Public Health 1992; 83(5): 326 -8.
The picture looks different if we think about people’s overall state of health or affliction 14% increase Source: Centers for Disease Control and Prevention. Health-related quality of life: prevalence data. National Center for Chronic Disease Prevention and Health Promotion, 2003. Accessed March 21 at
Misleading Framing Assumptions Stepwise progress will lead to system wide improvement Focus on the events Everything that happens must have a cause That cause must be close in time and space Instantaneous impacts Causality runs one-way Independence Impacts are linear and constant These assumptions overlook non-local forces of change, such as feedback and delay Richmond B, Peterson S, High Performance Systems Inc. An introduction to systems thinking. Hanover NH: High Performance Systems, 1997.
Basic Problem Solving Orientations Event Oriented View Decisions Feedback View Side Effects Goals Environment Goals of Others Actions of Others Sterman J. Business dynamics: systems thinking and modeling for a complex world. Boston, MA: Irwin Mc. Graw-Hill, 2000.
“When we attribute behavior to people rather than system structure the focus of management becomes scapegoating and blame rather than the design of organizations in which ordinary people can achieve extraordinary results. ” “The tendency to blame other people instead of the system is so strong that psychologists call it the fundamental attribution error. ” -- John Sterman J. System dynamics modeling: tools for learning in a complex world. California Management Review 2001; 43(4): 8 -25.
Is there some way to get a larger, more dynamic overview of the whole health system without loosing sight of unique disease processes? White F. The overview effect: space exploration and human evolution. 2 nd ed. Reston VA: American Institute of Aeronautics and Astronautics, 1998.
A Comprehensive Public Health Framework For Chronic Disease Prevention and Health Promotion A Vision of the Future Social and Environmental Conditions Favorable to Health Behavioral Patterns that Promote Health HP Policy and Environmental Change Behavior Change Low Population Risk 1 Few Events/ Only Rare Deaths Full Functional Capacity/ Low Risk of Recurrence 2 O Risk Factor Detection and Control Emergency Care/Acute Case Management Intervention Approaches Unfavorable Social and Environmental Conditions Adverse Behavioral Patterns O Rehabilitation/ Long-term Case Management The Present Reality Major Risk Factors First Event/ Diagnosis Good Quality of Life Until Death Disability/Risk of Recurrent Episodes End-of-Life Care Fatal Complications Intervention Goals (based on Healthy People 2010, Heart Disease/Stroke) Increase Quality and Years of Healthy Life Eliminate Disparities Goal 1 Total Population 281 million Goal 2 Goal 3 Target Population - US Acute Events Increased Risk tens of millions per factor hundreds of thousands per event Goal 4 Chronic Afflictions hundreds of thousands per condition
Progression of Systems Thinking & Modeling Issue Identification Events Patterns Structure Variable & Behavior Analysis Causal Loop Mapping Implementing Action Plan Understanding Strategy & Policy Implications Simulation Modeling Time Adapted from: Successful Systems, Inc. Issue Identification Variable & Behavior Analysis Causal Loop Mapping
Tools for Policy Development Events Time Series Models Increasing: Describe trends • Depth of causal theory • Degrees of uncertainty Patterns • Robustness for longerterm projection • Value for developing policy insights Structure Developed by Jack Homer, Homer Consulting Multivariate Stat Models Identify historical trend drivers and correlates Dynamic Models Anticipate future trends, and find policies that maximize chances of a desirable path
Different Modeling Approaches For Different Purposes Logic Models (flowcharts, maps or diagrams) • Articulate steps between program actions and results System Dynamics (causal loop diagrams and simulation models) • Improve understanding about the possible effects of a policy over time Forecasting Models • Make accurate forecasts of key variables • Focus on patterns of • Focus on precision change over time (e. g. , long of point predictions delays, worse before better) and confidence intervals
CDC Futures Initiative Report from the Health Systems Workgroup Figure 1 Health system dynamics General protection Society's Health Response Targeted protection Primary prevention Secondary prevention Tertiary prevention Becoming no longer vulnerable Safer, Healthier Population Becoming Vulnerable Population Becoming Afflicted without Complications Developing Complications Adverse Living Conditions From: Milstein B, Homer J. The dynamics of upstream and downstream: why is so hard for the health system to work upstream, and what can be done about it? CDC Futures Health Systems Workgroup; Atlanta, GA; 2003. Afflicted with Complications Dying from Complications
What Kinds of Work are Needed? more inter-organizationally complex, slower rate of improvement Public Work (organizing, governance, citizenship, mutual accountability) FOR SELF INTEREST organizationally complex, faster rate of improvement Professional Work (customers, products, services) FOR OTHERS IN NEED
Dependence on Living Conditions "Each of us has an array of basic needs that must, by and large, be satisfied continuously. We cannot, for instance, do for very long without fresh water, or waste elimination, or sleep. Accordingly, each of us–individually and collectively–requires a synergistic ‘package' of resources and suitable environmental conditions. A society that can reliably provide this package will thrive and possibly grow larger. But if even one of these needs is not satisfied–if any part of the package is deficient–the entire enterprise is likely to be threatened" Corning PA. Presidential speech: the systems sciences in the year 3000. International Society of the Systems Sciences, 2000. Accessed April 23, 2002 at
Definition: Living Conditions “Living conditions are the everyday environment of people, where they live, play and work. These living conditions are a product of social and economic circumstances and the physical environment – all of which can impact upon health – and are largely outside of the immediate control of the individual. ” -- World Health Organization. Health promotion glossary. World Health Organization, 1998. Accessed July 15 at
Prerequisite Conditions for Health Peace Income Shelter Stable eco-system Education Sustainable resources Food Social justice and equity Endorsed at all five world conferences on health promotion (1986 -2000) World Health Organization. Ottawa charter for health promotion. International Conference on Health Promotion: The Move Towards a New Public Health, November 17 -21, 1986 Ottawa, Ontario, Canada, 1986. Accessed July 12, 2002 at
Human Development Freedoms Health Education UNDP Human Development Index Standard of living Political participation Social engagement Physical security Sen AK. Development as freedom. New York: Anchor books, 1999. United Nations Development Programme. Human development report 2002: deepening democracy in a fragmented world. New York: Oxford University Press; 2002.
Seeing Conditions as Freedoms Adverse living conditions are circumstances that inhibit people's freedom to be safe and healthy and develop their full potential They include, at a minimum, any deviation from prerequisite conditions for life and human dignity (e. g. , physical extremes, violence, deprivation, disconnection) Phenomena like hunger, homelessness, joblessness, illiteracy, war, environmental decay, and various forms of injustice, including racism, are all examples of adverse living conditions
Linking Living Conditions to Freedoms Healthy State Freedom From… Selected Examples Physical security Physical extremes • Crash, fire, fall • Heat, cold • Radiation • Hazardous substances • Natural disaster • Infectious diseases Peace Violence • Homicide • Suicide • War • Rape Minimal standard of living Deprivation • Malnutrition • Homelessness • Poverty • Joblessness • Overcrowding • Illiteracy • Inadequate education Social engagement Disconnection • Inequality • Injustice • Dependency • Incarceration • Runaway • Neglect Stable organic processes Impaired metabolism • Heart disease • Cancer • Stroke • Diabetes • Arthritis • Obesity Mental/emotional balance Impaired cognition or emotion • Depression • Anxiety • Attention deficit • Lack of recreation Successful reproduction Impaired reproduction • Infertility • Miscarriage • Birth defects • Infant mortality
Choice and Non-Choice “Choices are always made from among alternatives presented by the social environment, or by circumstances that were themselves not chosen…When we recognize the elements of non -choice in choice, we can escape the contradiction between social causation and individual responsibility and understand the interactiveness of the two. ” Levins R, Lopez C. Toward an ecosocial view of health. International Journal of Health Services 1999; 29(2): 261 -93.
Balancing Two Areas of Emphasis Healthy Public Policy & Public Work DEMOCRATIC SELF-GOVERNANCE World of Transforming… By Strengthening… • Deprivation • Dependency • Violence • Disconnection • Environmental decay • Stress • Insecurity • Etc… • Leaders and institutions • Foresight and precaution • The meaning of work • Mutual accountability • Plurality • Democracy • Freedom • Etc… Medical and Public Health Policy DISEASE AND RISK MANAGEMENT World of Providing… • Education • Screening • Disease management • Pharmaceuticals • Clinical services • Physical and financial access • Etc…
On Protection and Leadership "How do you know, " I asked, "that in twenty years those things that you consider special are still going to be here? " At first they all raised their hands but when they really digested the question every single one of them put their hands down. In the end, there was not a single hand up. No one could answer that question…” -- Nainoa Thompson N. Reflections on voyaging and home. Polynesian Voyaging Society, 2001. Accessed July 18 at
Two Policy Orientations Healthy Public Policy and Public Work Medical and Public Health Policy • Concerned chiefly with expanding people’s freedom to be safe and healthy • Concerned chiefly with preventing and alleviating specific diseases, managing complications, and delaying premature death or disability • Relies heavily on multiple, small-scale, local solutions, with low technology • Relies heavily on specific high-technology solutions, widely applied • Combines analyses into a broad systems view, transcending sector boundaries • Confines analyses to the health sector • Future-oriented (reacting to long-term dynamics) • Present-oriented (reacting to immediate events) • Questions the givens, focuses on plausible outcomes • Accepts the givens, focuses on probable outcomes • Evaluated first through simulation, then through implementation • Evaluated through implementation • Main resources are citizen leadership and broad-based public work (including that of professionals) • Main resources are money, professional expertise, and technology (often excluding citizen leadership) Adapted from: Hancock T. Beyond health care: from public health policy to healthy public policy. Can J Public Health 1985; 76 Suppl 1: 9 -11.
How Do We Craft High Leverage Policies? Could the behavior of this system be modeled mentally, or with conventional epidemoiological methods (e. g. , logistic or multi-level regression)?
Dynamic Models Let Us Search for Policies with the Greatest Leverage Computer technology makes it feasible to put system maps in motion, to learn how health patterns change under different conditions, and to seriously evaluate or rehearse the long-term effects of response options: they provide added foresight Prototype of a health system simulation model Such models open new avenues for planning and formally evaluating prevention policies
Re-Directing the Course of Change Questions from System Modeling and Social Navigation ? e her W Ho w ? ? y Wh Wh o ? 2010 2020
“Let me assure you, we will survive any crisis that involves funding, political support, popularity, or cyclic trends, but we can't survive the internal crisis, if we become provincial, focus totally on the short term, or if we lose our philosophy of social justice. ” -- William Foege WH. Public health: moving from debt to legacy. American Journal of Public Health 1987; 77(10): 1276 -8.
Progress in Dynamic Modeling The Dynamics of Upstream and Downstream Why is it so hard for the health system to work upstream, and what can be done about it? (Milstein & Homer, with the CDC Futures Health System Workgroup) The System-wide Drivers of Diabetes What are the system-wide drivers of type 2 diabetes incidence and progression, including other chronic illnesses, risk factors, and multiple types of prevention and protection programs? (CDC Diabetes System Modeling Project) The Problem of Outside Assistance What types of outside assistance are most effective in reducing the overall burden of affliction (unhealthy days) in communities with multiple afflictions? (Homer & Milstein 2002, 2004; web game)
For Additional Information http: //www. cdc. gov/syndemics
Background on Dynamic Modeling Projects Work in Progress
Benefits of Game-Based Learning “Artful scenario spinning…ensures not that you are always right about the future but--better--that you are almost never wrong. " -- Stewart Brand Formal means of evaluating options Compressed time Actions can be stopped or reversed Experimental control of all conditions Complete, undistorted, immediate results Rehearse worse-before-better scenarios Early warning of unintended effects Opportunity to assemble stronger support
Progress in Dynamic Modeling Problem Focus Outside assistance in communities with multiple afflictions Stage of Development • Most exploratory • Designed to explore interactions between afflictions, living conditions, and public strength • More empirically supported Dynamics of upstream and downstream health work Dynamics of diabetes incidence and progression • Designed to understand an observed phenomenon, the 97% -- 3% split in health care expenditures • Most empirically supported
Steps for Putting Maps in Motion Identify a persistent problem that exists, in part, due to dynamic complexity (i. e. , forces of feedback, delay, non-linearity, etc…) Develop a preliminary dynamic hypothesis (i. e. , what causal forces are at work? ) Convert that hypothesis into a formal computer model (i. e. , by writing a system of differential equations; and calibrating it based on available data; areas of uncertainty are noted and become the focus for sensitivity analysis) Use the computer model to conduct controlled simulation studies, with the goal of learning how the system behaves and how to govern its evolution over time Iteratively repeat the process, creating better hypotheses, better models, better policy insight, and more effective action
Why Is it So Hard to Work Upstream? A Preliminary Dynamic Hypothesis Upstream Prevention and Protection -----------------Total 3% Downstream Care and Management ----------------Total 97% Brown R, Elixhauser A, Corea J, Luce B, Sheingod S. National expenditures for health promotion and disease prevention activities in the United States. Washington, DC: Battelle; Medical Technology Assessment and Policy Research Center; 1991. Report No. : BHARC-013/91 -019.
Health System Dynamics
Health System Dynamics
Health System Dynamics
Health System Dynamics
Health System Dynamics
Health System Dynamics
Health System Dynamics
Health System Dynamics Downstream lock-in: Delay in upstream effort guarantees continued growth in affliction prevalence and emphasis on treatment, which further delays upstream effort, as does mounting social disparity.
Prototype of a Dynamic Health System Simulation Work in Progress
Building a Dynamic Hypothesis B General Protection Targeted protection effect on vulnerability reduction General protection effect on vulnerability onset B Vulnerability reduction General Population Vulnerability onset Vulnerable Population Affliction incidence - Public health response Targeted Protection Complicated B afflicted percent of popn Treatment Afflicted without Complications Affliction progression - Afflicted with Complications Tertiary prevention effect on complications Death from Complications - B Note: for this initial model, the system being modeled includes only a subset of the dynamics that were identified in the conceptual map. Secondary prevention effect on progression Secondary Prevention B Primary Prevention Primary prevention effect on incidence
Writing Differential Equations Active Equations (01) Afflicted = Afflicted without Complications + Afflicted with Complications (02) Afflicted with Complications = INTEG( Affliction progression - Death from Complications , 0) (03) Afflicted without Complications = INTEG( Affliction incidence - Affliction progression , 0) (04) Affliction incidence = Vulnerable Popn * Affliction incidence rate baseline * Primary prevention effect on incidence (05) Affliction incidence rate baseline = 0. 05 (06) Affliction progression = Afflicted without Complications * Affliction progression rate baseline * Secondary prevention effect on progression (07) Affliction progression rate baseline = 0. 1 (08) Complicated afflicted percent of popn = 100 * Afflicted with Complications / Total popn (09) Complicated afflicted percent required to elicit maximum PH response = 20 (10) Complicated percent of afflicted = 100 * ZIDZ ( Afflicted with Complications , Afflicted ) (11) Complications death rate baseline = 0. 1
Writing Differential Equations (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) Death from Complications = Afflicted with Complications * Complications death rate baseline * Tertiary prevention effect on complications General Popn = INTEG( Net increase in genl popn + Vulnerability reduction - Vulnerability onset , Total popn initial * ( 100 - Vulnerable percent initial ) / 100) General protection effect from max PHR = 0. 5 General protection effect on vulnerability onset = 1 - ( 1 - General protection effect from max PHR ) * Public health response / 100 Net increase in genl popn = Death from Complications * ( 1 - Vulnerable percent of nonafflicted / 100) Net increase in vulnerable popn = Death from Complications * Vulnerable percent of nonafflicted / 100 Nonafflicted = General Popn + Vulnerable Popn Primary prevention effect from max PHR = 0. 5 Primary prevention effect on incidence = 1 - ( 1 - Primary prevention effect from max PHR ) * Public health response / 100 Public health response = DELAY 1 I ( 100 * MIN ( 1, Complicated afflicted percent of popn / Complicated afflicted percent required to elicit maximum PH response ) , Time for public health to respond to affliction prevalence , 0)
Writing Differential Equations (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) Secondary prevention effect from max PHR = 0. 5 Secondary prevention effect on progression = 1 - ( 1 - Secondary prevention effect from max PHR ) * Public health response / 100 Targeted protection effect from max PHR = 2 Targeted protection effect on vulnerability reduction = 1 + ( Targeted protection effect from max PHR - 1) * Public health response / 100 Tertiary prevention effect from max PHR = 0. 5 Tertiary prevention effect on complications = 1 - ( 1 - Tertiary prevention effect from max PHR ) * Public health response / 100 Time for public health to respond to affliction prevalence = 2 Total popn = Nonafflicted + Afflicted Total popn initial = 100000 Vulnerability onset = General Popn * Vulnerability onset rate baseline * General protection effect on vulnerability onset Vulnerability onset rate baseline = 0. 05
Writing Differential Equations (33) (34) (35) (36) (37) Vulnerability reduction = Vulnerable Popn * Vulnerability reduction rate baseline * Targeted protection effect on vulnerability reduction Vulnerability reduction rate baseline = 0. 07 Vulnerable percent initial = 10 Vulnerable percent of nonafflicted = 100 * Vulnerable Popn / Nonafflicted Vulnerable Popn = INTEG( Net increase in vulnerable popn + Vulnerability onset - Affliction incidence - Vulnerability reduction , Total popn initial * Vulnerable percent initial / 100)
Developing Assumptions For Response Scenarios Parameter Assumption Population Characteristics • Total population initially 100, 000 • Percent afflicted initially 0% • Percent vulnerable initially 10%
Developing Assumptions For Response Scenarios Parameter Assumption Baseline Epidemiological Characteristics • Vulnerability onset rate (% per year among general pop) 5% • Vulnerability reduction rate (% per year among vulnerable) 7% • Affliction incidence rate (% per year among vulnerable) 5% • Affliction progression rate (% per year among afflicted without complications) 10% • Complications death rate (% per year among afflicted with complications) 10%
Developing Assumptions For Response Scenarios Parameter Assumption Health System Characteristics • Complicated affliction prevalence required to elicit maximum health system response (lower prevalence elicits proportionally smaller response) • Time for organizing a health system response to complicated affliction prevalence 20% 2 years
Making Decisions About How to Respond Parameter Assumption Effect of Health System Responses • Tertiary prevention effect on deaths from complications ? • Secondary prevention effect on affliction progression ? • Primary prevention effect on affliction incidence ? • Targeted protection effect on vulnerability reduction ? • General protection effect on vulnerability onset ?
Developing a Scenario-based Research Design Effect of Health System Response on… Response Scenario 40 -Year Simulation Results Deaths Affliction Progress Affliction Incidence Vulnerable Reduction Vulnerable Onset 1 1 1 Prev 3 0. 5 1 1 Prev 2+3 0. 5 1 1 1 Prev 1+2+3 0. 5 1 1 Prev 1+2+3 Prot 2 0. 5 2 1 Prev 1+2+3 + Prot 1+2 0. 5 2 0. 5 No Response Percent Afflicted w/ Complication (T 0 = 0%) Percent Vulnerable (T 0 = 10%) PH Response (T 0 = 0%)
Putting the System in Motion
Interpreting Behavior Over Time
Interpreting Behavior Over Time Effect of Health System Response on… Response Scenario 40 -Year Simulation Results Deaths Affliction Progress Affliction Incidence Vulnerable Reduction Vulnerable Onset Percent Afflicted w/ Complication (T 0 = 0%) 1 1 11% Prev 3 0. 5 1 1 14% Prev 2+3 0. 5 1 12% Prev 1+2+3 0. 5 1 1 11% Prev 1+2+3 Prot 2 0. 5 2 1 10% Prev 1+2+3 + Prot 1+2 0. 5 2 0. 5 9% No Response Percent Vulnerable (T 0 = 10%) PH Response (T 0 = 0%)
Interpreting Behavior Over Time
Interpreting Behavior Over Time Effect of Health System Response on… Response Scenario 40 -Year Simulation Results Deaths Affliction Progress Affliction Incidence Vulnerable Reduction Vulnerable Onset Percent Afflicted w/ Complication (T 0 = 0%) 1 1 11% 25% Prev 3 0. 5 1 1 14% 24% Prev 2+3 0. 5 1 12% 24% Prev 1+2+3 0. 5 1 1 11% 26% Prev 1+2+3 Prot 2 0. 5 2 1 10% 22% Prev 1+2+3 + Prot 1+2 0. 5 2 0. 5 9% 19% No Response Percent Vulnerable (T 0 = 10%) PH Response (T 0 = 0%)
Interpreting Behavior Over Time
Interpreting Behavior Over Time Effect of Health System Response on… Response Scenario 40 -Year Simulation Results Deaths Affliction Progress Affliction Incidence Vulnerable Reduction Vulnerable Onset Percent Afflicted w/ Complication (T 0 = 0%) 1 1 11% 25% 0% Prev 3 0. 5 1 1 14% 24% 69% Prev 2+3 0. 5 1 12% 24% 60% Prev 1+2+3 0. 5 1 1 11% 26% 53% Prev 1+2+3 Prot 2 0. 5 2 1 10% 22% 50% Prev 1+2+3 + Prot 1+2 0. 5 2 0. 5 9% 19% 47% No Response Percent Vulnerable (T 0 = 10%) PH Response (T 0 = 0%)
Diabetes System Modeling Project Work in Progress
Forecast of Diabetes Prevalence Key Constants • Incidence rates (%/yr) • Death rates (%/yr) • Diagnosed fractions (Based on year 2000 data, per demographic segment)
Focusing on the More Modifiable Drivers
Re-Directing the Course of Change ? e her W Ho w ? ? y Wh Wh o ? 2010 2020
Diabetes System Modeling Project Observers and Other Constituents NCCDPHP Senior Staff Initial Model Conceptualization Team Adult and Community Health Smoking and Health Cardiovascular Health Division of Diabetes Translation Public Health Practice Nutrition and Physical Activity Adolescent and School Health Program Branch Epidemiology Branch CDC-Wide Champions Office of Director Dynamic Modeling Experts
Diabetes Population Flows
Diabetes Population Flows Non-Diabetics
Diabetes Population Flows Non-Diabetics
Diabetes Population Flows Non-Diabetics Tertiary Prevention: Disease Management for Stage 2 Diabetics
Diabetes Population Flows Non-Diabetics Secondary Prevention: Screening for Diabetes & Disease Mgmt for Stage 1 Diabetics
Diabetes Population Flows Non-Diabetics Primary Prevention: Screening for Prediabetes & Nutrition/Activity for High Risk Individuals Diabetics
Diabetes Population Flows Non-Diabetics Targeted Protection: Risk Factor Elimination for High Risk Individuals Diabetics
Diabetes Population Flows Non-Diabetics General Protection: Changing Risk Conditions of the General and High Risk Populations Diabetics
Where is the Greatest Leverage? Comparing Program Strategies, Alone and in Combination Diabetes education for the public Diabetes education for providers Weight reduction programs for the obese Resources for diabetes disease management Resources for glycemic screening Facilitating greater food choice Facilitating greater availability of affordable health care Facilitating greater social and cultural support for the underserved Facilitating improvements in built environment and public safety Facilitating more work opportunities Others….
Conducting Policy Experiments
Population Breakdown Based on National Statistics Total Population High risk 22% Diabetics 6% General population 72% Within the high risk group, we calculate 25 -45% have IGT and 33 -60% are Prediabetic (IGT or IFG)* * Lower numbers based on Benjamin et al. 2003, higher numbers based on NIDDK estimates. IGT: Impaired Glucose Tolerance, 2 hr. non-fasting, 140 -199 mg/dl; IFG: Impaired Fasting Glucose, 110 -125 mg/dl
Data Sources for Initial Calibration High Risk Population, Incidence, Prevalence, Deaths “National Diabetes Statistics”: http: //diabetes. niddk. nih. gov/dm/pubs/statistics/index. htm “Prevalence of Selected Chronic Conditions: United States, 1990 -1992”: www. cdc. gov/nchs/data/series/sr_10/sr 10_194. pdf “Healthy People 2000 Review, 1997”: www. cdc. gov/nchs/data/hp 2000/hp 2 k 97. pdf “Deaths: Preliminary Data for 2000”: www. cdc. gov/nchs/data/nvsr 49/nvsr 49_12. pdf “Estimated number of adults with prediabetes in the U. S. in 2000: Opportunities for prevention”, Benjamin SM et al (DDT/CDC), Diabetes Care 26: 645 -9, 2003. “A Dynamic Markov Model for Forecasting Diabetes Prevalence in the United States through 2050”, Honeycut AA et al. (DDT/CDC), Health Care Mgmt Sci 6: 155 -164, 2003. Complications and Benefits of Control “Model of Complications of NIDDM--1. Model Construction and Assumptions”, Eastman RC et al, Diabetes Care 20: 725 -734, 1997. “Model of Complications of NIDDM--2. Analysis of the Health Benefits and Cost-Effectiveness of Treating NIDDM with the Goal of Normoglycemia”, Eastman RC et al. , Diabetes Care 20: 735744, 1997. “The Prevention or Delay of Type 2 Diabetes”, position statement from ADA and NIDDK, Diabetes Care 25: 742 -749, 2002 “Effect of Improved Glycemic Control on Health Care Costs and Utilization”, EH Wagner et al. , JAMA 285: 182 -189, 2001 “Health Economic Benefits and Quality of Life During Improved Glycemic Control in Patients with Type 2 Diabetes Mellitus: A Randomized, Controlled Double-Blind Trial”, Testa MA and Simonson DC, JAMA, 280: 1490 -6, 1998 One benefit of the modeling process can be knowledge integration
Developing a Scenario-based Research Design Effect of Health System Response on… Response Scenario Controled Fraction of S 2 Controled Fraction of S 1 Diabetes Diagnosis Rate Onset Rate for Dx Pre. D Onset Rate for Un. Dx Pre. D Diagnosis Rate Pre. D Onset Rate Rehab of Dx Pre. D Rate Rehab of High Risk Un. Dx Pre. D Rate Become High Risk Rate Steady State . 2 . 06 . 025 . 04 . 065 . 06 . 03 . 02 Prev 3 . 5 " " " " " Prev 2+3 . 5 . 12 " " " " Prev 1+2+3 . 5 . 12 . 020 . 040 . 08 . 55 " " " Prev 2+3 Prot 2 . 5 . 12 " " . 10 . 05 " Prev 1+2+3 + Prot 1+2 . 5 . 12 . 020 . 040 . 08 . 55 . 10 . 05 . 01
Model Tests Start in a Steady-State The model is initialized in a steady state or “dynamic equilibrium” No change in total population: Births = Total deaths Every population stock has inflows exactly cancelled by outflows In the base (“flatline”) run, all input time series are unchanging Any changes in the time series inputs (pink) will disturb the steady state and cause the population system to move toward a new equilibrium By putting the model in an initial equilibrium we are able to examine the effects of changes in the time series inputs in isolation from general population trends (which will be added in a future version)
Interpreting Behavior Over Time
Interpreting Behavior Over Time
DRAFT: Diabetes System Sector Sketch (December 4 -5, 2003) Civic Participation Included in version 1 Forces Outside the Community • Social cohesion • Responsibility for others • Macroeconomy, employment • Food supply • Advertising, media • National health care • Racism • Transportation policies • Voluntary health orgs • Professional assns • University programs • National coalitions Not included in version 1 Health Care & Public Health Agency Capacity Personal Capacity Local Living Conditions • Understanding • Motivation • Social support • Literacy • Physio-cognitive function • Life stages • Availability of good/bad food • Availability of phys activity • Comm norms, culture (e. g. , responses to racism, acculturation) • Safety • Income • Transportation • Housing • Education • Provider supply • Provider understanding, competence • Provider location • System integration • Cost of care • Insurance coverage Health Care Utilization Metabolic Stressors • Nutrition • Physical activity • Stress Population Flows • Baseline Flows • Ability to use care (match of patients and providers, language, culture) • Openness to/fear of screening • Self-management, monitoring • Percent of patients screened • Percent of people with diabetes under control
How Does Outside Assistance Affect Communities Facing Multiple Afflictions? Work in Progress
What Are the Dynamics of Outside Assistance in Communities Facing Multiple Afflictions? A Preliminary Dynamic Hypothesis Outside assistance to alleviate and prevent affliction Affliction cross-impacts R 1 B 1 a Affliction prevalence & burden Effort to alleviate and prevent affliction R 3 c Effort to build public strength R 2 a At-risk fraction Outside assistance to build public strength B 1 c Public work fraction R 2 c Social disparity R 2 b Public strength R 3 a R 3 b United efforts Adverse living conditions Key Rectangle: Stock/state variable Blue arrow: same-direction link Green arrow: opposite-direction link Circled “B”: balancing causal loop Circled “R”: reinforcing causal loop B 1 b Divided efforts Effort to improve living conditions Outside assistance to improve living conditions R 3 d Magnitude of ameliorative efforts
About the Feedback Loops Syndemic: Each affliction increases vulnerability to other afflictions, thereby amplifying the effect of increases or decreases in the prevalence of individual afflictions. Community Response: Community residents make efforts to fight affliction and adverse living conditions in response to their prevalence, and to build greater public strength when it is perceived as low. Outside assistance may bolster such efforts. Social Disparity and Public Strength: These efforts, especially those to fight adverse living conditions, are greater in magnitude when citizens are strong and unified through democratic public institutions. But public strength is hindered by social disparity, which, in turn, is made worse by the very afflictions and adverse living conditions the efforts are trying to fight. Public Strength and Public Work: Public strength is also affected by the efforts themselves. When problems spread in a community with strong democratic institutions, a united response (public work) reinforces the community’s strength. Conversely, when problems spread in a community with weaker democratic institutions, a divided response (consisting of only professional work) reinforces the community’s weakness. Outside assistance given to a weaker community for problem fighting may amplify the divided response and undermine the community’s internal response capability. Outside assistance to build public strength may prepare the residents to make a more united response.
Syndemics Simulation Game http: //broadcast. forio. com/sims/syndemic 2003/
Development of a Syndemic Four Scenarios for Affliction Burden Avg unhealthy days person per month Basic scenario: Poor living conditions, weak community, intertwined afflictions Greater community strength Weaker cross-impacts among afflictions Better living conditions
Evaluating Policy Scenarios Focus assistance on… Fighting affliction Different proportions Improving adverse living conditions Different combinations Building public strength Different sequences
Alternative Investment Strategies Public Health Programming Social Programming Democratic Organizing
Comparing Affliction Burden under Basic Setting and Four Different Assistance Schemes Conditions assistance only “LC 111” Affliction assistance only “AF 111” Strength assistance only “CS 111” Optimal assistance scheme “CS 1 AF 11” Avg affliction burden T 4 -T 20: 8. 1 8. 5 8. 8 8. 3
Policy Hypotheses Invest Early in Building Strength The first priority of philanthropies and government in addressing communities that are weak and struggling against multiple afflictions should be to assist in building public strength (enabling a greater degree of citizen-led public work), perhaps even before substantial assistance is provided for direct fighting of prevalent diseases.
Policy Hypotheses Beware the Side Effects of Outside Assistance Related to Living Conditions Outside assistance aimed directly at improving living conditions may often be insufficiently cost -effective, due to time lags and unintended side effects, to warrant making such assistance a high priority in the absence of widespread citizen participation
Structural Reasons for Policy Resistance Problem-fighting programs may have perverse effects on public strength when the community residents are weak and divided to begin with