October 2007
REVISED 3, 2007new revisions in green Old revisions from 9-10-07 in red Syllabus: Risk Analysis for Science and Technology Policy Fall 2007 PA 8790 3 credits 12:45 p.m.-2 p.m. Tuesdays and Thursdays HHH 20 Instructor: Prof. Jennifer Kuzma Contact: 612-625-6337, or [e-mail] Office Hours: Tuesdays and Thursday, 2:30-3:30 p.m. (it will be announced in class if there is a cancellation); or by appointment SCOPE This class will focus on the interplay between risk analysis, decision making, and risk policy for societal issues involving human, environmental, and ecological health and well-being. The role of S&T in risk analysis, either as the subjects or the tools, will be explored. A mix of readings, class discussions, problem solving and group exercises will be used to facilitate understanding of technical risk assessment methods; risk management processes, issues and methods; the role and treatment of uncertainty; factors in decision making; risk-based rule making; public values about risk; risk communication; and risk perception. Scientific, technical, social, political, and ethical issues for select case studies will be discussed. The class will help students develop skills for formulating questions for risk policy research, structuring risk analyses and problems, critiquing analytical and political choices involving risk, and building conceptual models of systems involving risk and decision-making. We will be spending about half the course on the risk policy dimensions (i.e. where risk and S&T intersect with society) and the other half understanding the methodologies and techniques used in risk analysis so that we can be better analysts, consumers, or policy makers in the face of risk information. Although we will do some basic modeling and calculations, sophisticated mathematical abilities are not required. Rudimentary knowledge of statistics and ability to work with algebraic equations will be helpful. GOALS understand risk analysis methodologies and increase ability to interpret risk assessments, whether qualitative or quantitative be able to develop models for risk assessment and use data in these models for estimating risk develop skills in structuring risk policy options and decision analysis frameworks be familiar with techniques used in risk management, such as cost-benefit analysis understand the role of science and technology in risk analysis, as both the subject of assessment and the provider of information and tools. appreciate the challenges of conducting risk assessments and incorporating the results of technical risk assessment into policy and decision-making appreciate the connections between risk assessment and social, economic, political, and ethical frameworks be able to trace a risk assessment through the U.S. federal rule-making process understand the importance of social factors, stakeholder engagement, and local and specialized knowledge in risk policy and decision-making. become acquainted with issues in communicating about risks
1
2007
ASSIGNMENTS AND GRADING Readings Each student is expected to complete the readings prior to the start of each session starting with session 2. WebVista will be used to post the readings for the class. For instructions and help with WebVista see http://webct.umn.edu/students/. Also, a single hard copy of each reading will be available in a box outside of my office (HHH 254), for students to check out and make one personal copy if they prefer to access readings this way. Readings are listed below for each session. There will be about 50-75 pages of reading per class session. Software--Computers We will be using the Decision Tools software from Pallisade. Copies will be provided for each student in the class. Each comes with a one-year license to use. Each student will be able to load his or her personal copy on a laptop or desktop computer. If you do not own a computer, see me and well figure out a way for you to access the software in class (reserving a laptop from compstaff) or the HHH computer lab. Required text: Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005) Copies may be purchased from the bookstore. Optional text Making Hard Decisions. R.T. Clemen, and T. Reilly. Duxbury Thomson Learning (2001) We will be reading some sections and doing some modeling exercises from this text. These will be posted on Web Vista or provided as handouts. You may also purchase a copy. A few copies should be available from the bookstore. Overall structure of the course For Tuesday sessions, I will lecture on the reading material (45 minutes) but throughout, questions, comments, thoughts and ideas to provoke discussion will be encouraged. In other words, lectures will be active and engaged, so your reading of the material is vital. In smaller discussion groups (2-3 people), you will discuss 1) your answers to the reading question sets (due that day) and/or 2) your or your peers write-ups from the problem sets (due the prior Friday, COB). On selected Tuesdays (Oct. 16th and Dec 4th), you will be presenting your literature critiques (see below for more information). Your group will periodically be asked to present the results of your discussion back to the full class. For Thursday sessions, there will be a brief discussion of the reading material to clarify concepts and answer any questions you might have (15 minutes). This will be less formal and less structured than Tuesday lectures. The rest of the time (1 hour) will be devoted to in-class risk problem solving exercises and/or modeling. Students will work in groups of 2-3 people to complete the problems or models. Discussion and problem-solving/modeling groups will be assigned the first week of class during session 2, Sept. 6th. Your first assigned group will work together for the first half of the semester
2
2007
(until Oct. 19th). Mid-semester, we will reassign groups to work together until the end. That way, you get a bit of continuity and a bit of change. Question sets (TuesdaysSeptember 11th, 18th, 25th, Oct 2nd, 9th, 23rd, 30th, Nov 6th,13th, 20th) Students will be assigned question sets based on the readings to complete for the Tuesday sessions. Answers to the question sets are due at the beginning of class on Tuesday (in hard copy please), but you will hold on to them for your discussion. There will be a very low tolerance for those who try to complete the question sets in class during the lecture or discussion (although you might add notes on the side margins to your question sets as the discussion progresses). The questions are designed to ensure that you are carefully reading the course material and thinking deeply about it. However, the answers should be no more than two pages (at least 1.5 line spacing and 10 pt. fonthandwritten answers are fine too). Note that the assignments should be turned in regardless of whether you miss the class, otherwise, you will get a 0 for that question set. These question sets will be done individually, but discussed in your small groups. 30% of grade (3% x 10) Peer reviewed literature critiques and short presentations (Tuesday-October 16th,Dec. 4th) Students will be asked to find an article of interest to them from a current issue of the journal Risk Analysis (within last few yearslets say post 2000). Summarize it succinctly, critique the methodology of the analysis, and suggest ways to improve the analysis and/or future ways to study the risk. The assignment should be no more than 3 typed pages in length (at least 1.5 line spacing, 10pt font), and you will be asked to present your paper for about 5-10 minutes to the whole class by succinctly summarizing the article you read and your critique of it. The first will be a straight-forward written critique of the article (Oct. 16th). For the second one (Dec. 4th), you will critique the paper in writing, but also derive your own model to account for at least one deficiency in the article or to incorporate a policy lever that could be used to address or mitigate the risk issue analyzed in the paper. Use of the Decision Tools software is encouraged, but not required. These two assignments are designed to help build your critical thinking skills using key concepts from the course, and to familiarize you with peer-reviewed literature in risk analysis. This assignment will be done individually, but presented to the whole group. The two critiques will take the place of the reading question sets in two Tuesday sessions (i.e. there will be no reading question sets when these critiques are due). The assignment should be turned in in hard copy please. 20% of grade (10% x 2) Problem sets (ThursdaysSept 13th, 20th, 27th, Oct 11th, 18th, 25th, Nov 1st, 8th, 15th, 29th) Each Thursday session will have approximately 50 pages of reading. Please do the reading before classeven though we wont cover it in great detail, it is essential to conducting your inclass exercises. In class on Thursdays, there will be a brief introduction to the reading material (15 min). Then, students will work with their small group to complete a problem/exercise (1 hour). Problem sets will be diversesome will involve government decision making process, others will involve modeling using the Decision Tools software, others will involve generating probability distributions from data using the software. In addition to solving the problem in class, each group will write up the problem in a 1-2 page document (1.5 line spacing, 10pt font). There should be some time to help complete this task in class (you might finish it in class or need some additional time outside of class). Problem write-ups will generally be due by Friday COB 5 p.m. and submitted via email to the instructor by that time. However, there might be cases where the in-class exercises take too long and an extra day or two will be given (Tuesday at latest). Come prepared to discuss your work in class on Tuesday during the discussion. 30% of grade (3%x10)
3
2007
Discussion/Attendance We will leave time for discussion in the Tuesday class periods particularly. Students are expected to participate. Participation also involves working diligently to complete the exercises in the Thursday class sessions. Everyone starts with a 4.0 (A) for each class. If you do not participate, either by not listening or not speaking or both (please contact me if you are shy about speaking), this score will drop. If you are absent from class for reasons other than personal emergencies, your discussion grade for that class session will be 0. Work, field trips for other classes, seminars, trips, etc. are not emergencies. If a student has more than two non-emergency absences in the semester, see me for additional assignments. 10% of grade Final Synthesis Exam The final exam will be given during the final exam time assigned for the course (hopefully early in the period of Dec 14-20th). It will involve several essay questions based on the risk policy dimensions of the course material, at least one modeling problem, and a few other risk analysis questions and problems. Students will take the exam individually, yet may draw upon and refer back to the readings and completed assignments (open book format) 10% of grade Overall Grade Grades will be given on a 4.0-0.0 scale corresponding to the letter grades A-F. A =3.8-4.0 A(-) = 3.4-3.7 B+ =3.1-3.3 B =2.7-3.0 B=2.3-2.6 C+ =2.0-2.2 POLICIES For anyone who has a disability which may require some modification of seating, testing, or other class requirements, appropriate arrangements may be made. Please see me after class or during my office hours. Disability Services is located at 16 Johnston Hall. If you are unsure about your performance feedback (aka grades) or anything else about the class, do feel free to talk to me about it. I am generally available through email and usually respond quickly unless Im traveling. However, whenever possible, please visit me during my office hours instead of requesting a separate meeting. I will make exceptions if you have a pressing issue or cannot consistently make that time slot. Thanks. I encourage you to learn from each other as much as you learn from me (or more in some cases!). You all come with knowledge, perspectives, and experiences that relate to the course material. With risk issues, sometimes discussion can generate controversy and debate. This is what makes the course interesting. Disagreement is welcome and fine. However, you should respect your classmates opinions and refrain from disparaging remarks. Civil
4
2007
tones are required. Listen to the opposing viewpoints and try to acknowledge the merits of them. TOPICS Session 1--September 4th (no assignments due, but please read the readings after class, and before session 2) Introduction of each other and review of course syllabus Introduction to risk analysis in the context of public policy a) Glickman, T.S. and M. Gough, eds. Readings in Risk. Part 1. Basic Concepts. Pp. 1-50 Morgan,, Fischoff, Watson, Hope, Derby and Keeny articles. Resources for the Future (1990). b) International Council on Risk Governance. White Paper on Risk Governance: Towards an Integrative Approach. Annex B and C: pp. 139-156. Session 2 Sept. 6thThursdayno assignments due, but go back and read session 1 readings prior to class Designation of groups Load Decision Tools software a) Chapter One: Risk Analysis pp. 1-40. Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text b) Chapter 1, Introduction. Kammen and Hassenzahl. Should We Risk It? Exploring Environmental, Health, and Technological Problem Solving. Chapter 1, pp. 3-30. c) Handouts in-class exercise: Conceptual exercise of risk analysis relationships to its components and policy analysis (time permitting, no-write up due) Session 3 Sept. 11th (Reading Question Set due, group discussion) Health Risk Assessment Ecological Risk Assessment a) Paustenbach, D. J., Retrospective on U.S. Health Risk Assessment: How Others Can Benefit Risk 6: 283 (1995). http://www.piercelaw.edu/risk/vol6/fall/pausten.htm 35 pp. b) Chapter Ten: Ecological Risk Analysis pp. 311-327. Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text
5
2007
Session 4Sept 13thThursdayin class modeling exercise Influence diagrams Decision Trees How do they fit into risk analysis? Intro to risk management and decision analysis a) Chapter 1-3, pp. 1-33 (skip pp. 33-43) , p. 43-83, Making Hard Decisions. R.T. Clemen, and T. Reilly. Duxbury Thomson Learning (2001) b) Handouts for exercise provided in class: Learning influence diagrams and decision trees--In class warm-up modeling exercise p. 83-100, Making Hard Decisions. Risk Assignment--Construction of a decision tree or influence diagram ---group choice of problem 1.7-1.8 (gypsy moth influence) OR 3.10 (medical treatment decision tree) from Making Hard Decisions Identify where the health or ecological risk questions are present in your diagram and where more data is needed in your write-up, include BUT other factors as well in the diagrams (social, political, economic, etc.). Objectiveslearning software tools and thinking broadly about risk problems. Distinguishing technical risk assessment points from management or policy decisions. Session 5Sept 18th (Reading question set due) Microbial Risk Assessment Pest or Non-Indigenous Species Risk Assessment Hazard Identification a) Campbell, F. The Science of Risk Assessment for Phytosanitary Regulation and the Impact of Changing Trade Regulations Bioscience 51: 148-153 (2001) b) Ruesnik, J. et al. Reducing the Risks of Nonindigenous Species Introductions BioScience, 51: 465-477 (1995). c) Brown, M.; Stringer, M. Microbiological Risk Assessment in Food Processing. Woodhead Publishing. Online version available at: http://www.knovel.com/knovel2/Toc.jsp?BookID=683&VerticalID=0 Read Chapters 2 and 3 pp. 5-64. Session 6Sept 20thThursdayin class modelingevent trees (like decision trees) Solving decisions trees Event trees for microbial contamination or pest-entry Risk Management/Policy problem structuring a) Chapter 4, Making Choices, pp. 111-145. Chater 5, Sensitivity Analysis, pp. 175-188. Making Hard Decisions. R.T. Clemen, and T. Reilly. Duxbury Thomson Learning (2001) b) Handouts for exercise provided in class: Solving decision trees--In class warm-up modeling exercise p. 146-159, Making Hard Decisions. Warm-up #2, sensitivity analysis p. 193-201
6
2007 Risk AssignmentMake a very simple event tree for zebra mussel entry into the U.S or
contamination of E.coli-0157:H7 in McDonalds hamburger (case descriptions provided). Identify where the health or ecological risk questions are present in your diagram. Identify where the policy choices are. Identify data gaps that would be useful for making better decisions (S&T part of problem). Think about possible sensitive variables. Time permitting, put in some basic values (make them up) into the event tree and run a sensitivity analysis. Objective--Use of event trees in risk analysis and comparisons to decision trees in business. Thinking about how trees are helpful for structuring complex risk characterizations for microbial or pest RA. Session 7 Sept 25th (Reading question set due) Uncertainty Probability Distributions a) Chapter TwoFunctions, Models, and Uncertainties, pp. 41-96 Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text b) Morgan, and Henrion. Uncertainty : A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis Chapters 4. pp. 47-72 (1990). Session 8Sept 27thThursday--in class modeling More on probability, Bayes theorem Probability (event) trees for human disease risk False negatives and positives a) Chapter 7, Probability Basics, pp. 249-280. Making Hard Decisions. R.T. Clemen, and T. Reilly. Duxbury Thomson Learning (2001) b) Handouts for exercise provided in class: Probability trees. Making Hard Decisions. Risk AssignmentCreate probability tree like figure 7.21 for the AIDS case study (p. 289). Complete questions 1-6. Objective--Use of probability trees (event trees) for medical risk and test performances (false + and -). Understanding of Bayes theorem. Session 9Oct. 2ndReading Question Set Due Exposure Assessment Dosimetry a) Chapters Four and FiveExposure Assessment pp. 113-171, Dosimetry pp. 173-194 Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text
7
2007
Session 10Oct. 4thThursdayNO CLASS Session 11Oct. 9th-- Reading Question Set Due Toxicology Epidemiology Benzene case study a) Chapters Six and SevenEpidemiology pp.197-235, Toxicology pp. 237-269 skim technical discussions Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text b) Glickman, T.S. and M. Gough, eds. Readings in Risk. Health Risk Assessment Part 4 pp 156-178, Resources for the Future (1990)
Session 12Oct. 11thThursday-- in class modeling Creating probability distributions Solving Health Risk Problems a) Chapter 8, 9, & part of 10. Subjective Probability, pp. 295-328, and Theoretical Probability Models pp. 352-373; Using Data pp. 398-405. Making Hard Decisions. R.T. Clemen, and T. Reilly. Duxbury Thomson Learning (2001). b) Handouts for exercise provided in class: Creating probability distributions in Risk Viewwarm-up#1--p.328-336, warm up #2, 373-379. Warm-up #3, 405-411 Making Hard Decisions Risk Assignmentcreate the probability distributions for problem 7-4 in Kammen and Hassenzahl in @ Risk. Should We Risk It? Exploring Environmental, Health, and Technological Problem Solving. Chapter 7, pp. 241-247. Your write-up should discuss why probability and quantitative uncertainty analysis is (or isnt) important in risk and decision analysis and the pros and cons of its use. ObjectiveUnderstanding different types of distributions and skills of creating them in software. Session 13Oct. 16thpaper presentations and subsequent small group peer review Presentation of your paper and critique Discussion/peer review of critiques in small groups Mid-semester evaluations Session 14Oct. 18th-Thursdaymodeling discussionno problem set due Guest lecture in System Dynamics modeling--and risk analysis connections Scott Johnson, BP
8
2007
Do these readings before class, although they will not be specifically discussed until Oct. 23rdor 25th Combining probability distributions Risk Characterization Monte Carlo Analysis a) Chapter EightRisk characterization pp. 271-289. Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text b) Chapter 11. Monte Carlo Simulation pp. 459-466,486-487. Making Hard Decisions. R.T. Clemen, and T. Reilly. Duxbury Thomson Learning (2001).
Session 15Oct 23rd (Reading question set duemoved the problem set here to Th Oct. 25th and then completely deleted the other onefinal result, you are less one problem set for the semester Rule making Regulation
a) Kerwin, C.M. Rulemaking: How government agencies write law and make policy CQ Press, 3rd edition, Chapters 2, The Process of rulemaking (2003) pp 39-85. b) Chapter ThreeRegulation p. 97-113. Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text Session 16Oct. 25thThursdayin class modeling Risk-based decision making Risk management a) Chapter ElevenRisk Management: Values and Decisions. Pp. 329-355. Introduction to Risk Analysis: A Systematic Approach to Science-Based Decision Making Byrd, D.M, and Cothern, C.R. Government Institutes Press: Lanham, MD (2005)Required Text b) Glickman, T.S. and M. Gough, eds. Readings in Risk. Regulatory Issues Part 3 pp 103138, Resources for the Future (1990). c) Kammen and Hassenzahl. Chapter 9. Decision Making. pp. 304-318. Should We Risk It? Exploring Environmental, Health, and Technological Problem Solving. Handouts for exercise provided in class:
9
2007
d) Handouts for exercise provided in class: Doing Monte Carlo in @Riskwarm-up p. 466-481 Risk Assignmentsolve all parts of problem 7-4 in Kammen and Hassenzahl in @Risk using your probability distributions from the prior weeks exercise. Should We Risk It? Exploring Environmental, Health, and Technological Problem Solving. Chapter 7, pp. 241-247. Your write-up should reflect the questions asked 7-4a-d, and 7-B and 7-C. In your write-up, discuss the policy issues about sensitive subpopulations and connections to uncertainty modeling. Also reflect on how Monte Carlo simulation may be useful to a decision or policy maker. Reflect on decision making and rule making in the context of the USDA and heptachlor in beef. ObjectiveUnderstanding effect of uncertainty modeling on risk management, decision making, and policy choices (including ethics and values in those choices) Session 17Oct 30th (Reading question set due) Cost-benefit analysis Arsenic case study a) Kopp, Krupnick, and Toman. Cost-Benefit Analysis and Regulatory Reform: An Assessment of the Science and the Art, Resources for the Future Discussion Paper 97-19 January 1997. http://www.rff.org/Documents/RFF-DP-97-19.pdf, pp. ii-60 b) Kaiser, J. A Second Look at Arsenic Finds Higher Risk Science 293: 2189 (2001) c) Smith et al. Arsenic Epidemiology and Drinking Water Standards Science 296:21452146 (2002). d) Brown and Ross. Arsenic, Drinking Water, and Health: A Position Paper of the American Council on Science and Health Regulatory Toxicology and Pharmacology 36: 162-174 (2002) e) Tchounwou, et al. Arsenic toxicity, mutagenesis, and carcinogenesis a health risk assessment and management approach Molecular and Cellular Biochemistry 255:47-55 (2004). Session 18Nov. 1stThursday-in class modeling Arsenic RM, RIA Cost-Benefit Analysis methods a) EPA. Arsenic in drinking water. EPA 8...