Risk Assessment Bruce Case

Risk Assessment Bruce Case

Risk Assessment: Lecture Outline 1. 2. 3. 4. 5. Definitions: Risk Analysis, Risk Assessment (Evaluation) and their components A detailed look at HAZARD EVALUATION Risk Perception, Risk Communication, Risk Management An example of risk assessment: Mesothelioma among Quebec asbestos mining area women. Risk and the precautionary principle

Buzzword Alert! There are a number of technical terms in this lecture Yes, you have to know them! These terms have precise meaning, even though you will often see them MIS-used. Since risk assessment is (or aims to be) a scientific activity we must agree on terminology

Risk Analysis This is the overall term for all of Risk Science It has four elements: - Risk Assessment (Risk Evaluation) - Risk Communication - Risk Perception - Risk Management (Risk Characterization (EPA))

Definition of Risk Assessment Risk Assessment, or risk evaluation, is a scientific/ mathematical discipline which is a substantive, changing and controversial field

Definition of Risk Assessment at the margin of our understanding of the health effects of chemicals and other substances best defined as the determination of pathology caused by human production and activity, with the understanding that "pathology" is a change in some aspect of human anatomical structure or function

Risk Assessment: Two Roads Qualitative - virtually the same thing as “hazard evaluation” step of “Quantitative” Risk Assessment - is the material harmful to humans under any circumstances - Codified by agencies, especially for cancer Quantitative A formal process with four steps Ends with a mathematical estimation of actual risk, usually quantified as deaths per 1,000,000 per year or less.

Risk Management: Putting the elements together

“Hazard Evaluation” is the equivalent of Qualitative Risk Assessment. (in many instances the three further steps are not taken) Examples: EPA, IARC Cancer Monographs

Types of Study Available for Hazard Evaluation BEST: Human Evidence (Epidemiology) Next best: Whole animal studies (toxicology; animals exposed to known dose and allowed to live to times of sacrifice or natural death)

Types of Study Available for Hazard Evaluation Other: In-vitro studies (studies on cells in culture) Structure-function relationship study and similar Identification of active compounds in metabolism

Study for Hazard Evaluation: Human Case reports (example: angiosarcoma of liver) Case series (example: mesothelioma in S. Africa) Descriptive epidemiology (much like geographic study; ecological fallacy is a problem)

Study for Hazard Evaluation: Human Analytical epidemiology: best: cohort studies: following exposed humans through time second best: case-referent studies: comparing “cases” of given disease to MATCHED referents and noting differences in exposure.

Study for Hazard Evaluation: Animal Studies of cells (in vitro studies: example O2-) Acute toxicity studies (how much does it take to kill half of all the animals?) Chronic toxicity studies: Best method but very expensive and time-consuming Proper design (doses, sacrifice times, animal selection) a must.

Study for Hazard Evaluation: Animal: Problems Ethical Concerns (see: papers by Peter Singer and Henry Spira): e.g. Rack L, Spira H. Animal rights and modern toxicology Toxicol Ind Health 1989 Jan;5(1):133-43. Conversely: non-realistic models may be useless; e.g. animal intra-tracheal injection versus inhalation e.g. use of rats (who do not have the same respiratory tract structure as humans: HOGS are best!!!)

Study for Hazard Evaluation: Human: Problems Ethical Concerns Expense LATENCY Practical considerations: for example the use of questionnaires or interviews in a case-referent study and: - sample size, response rate - selection bias and other bias

Hazard Evaluation: Synthesis: IARC Group 1 GROUP 1: AGENT CARCINOGENIC TO HUMANS Assignment to this category is based on a finding of "sufficient" evidence of carcinogenicity in humans. This implies a causal relationship between exposure to a chemical and cancer in epidemiological studies in which ".chance, bias and confounding could be ruled out with reasonable confidence"13.

Hazard Evaluation: Synthesis: IARC 2A GROUP 2A: AGENT PROBABLY CARCINOGENIC TO HUMANS “limited” or inadequate evidence in epidemiological studies for carcinogenicity: the agent falls into this category if there is "sufficient" evidence from experimental animal work. Causal relationship has been shown in two or more species of animals OR in two or more independent studies in one species.

Hazard Evaluation: Synthesis: IARC 2B GROUP 2B: AGENT POSSIBLY CARCINOGENIC TO HUMANS sufficient evidence for carcinogenicity neither in humans nor in experimental animals. a "credible" causal HUMAN relationship is suggested but bias, chance and confounding cannot be ruled out AND sufficient animal evidence OR "inadequate" evidence in humans but "sufficient" animal evidence.

Hazard Evaluation: Synthesis: IARC 3: GROUP 3: AGENT NOT CLASSIFIABLE FOR HUMAN CARCINOGENICITY category for agents which cannot otherwise be classified. Really a “garbage” category scientifically, but corresponds to some extent to the Precautionary Principle GMOs could possibly fit this category

Hazard Evaluation: Synthesis: IARC 4 GROUP 4: AGENT "PROBABLY" NOT CARCINOGENIC TO HUMANS animal studies in at least two species showing that the substance is “not carcinogenic”. If there is a large body of negative animal evidence: the agent will fall into this category even if there is some, but "inadequate", epidemiological evidence

Exposure Assessment 1 How much of a pollutant do people inhale/ ingest ? In what period of time? How many people will be exposed? To what? Or which (e.g. PCB)? From what source(s)? With what interaction(s) (e.g. smoking)

Exposure Assessment 2 Example: Total suspended particulates Sample stacks; sample environment; sample PEOPLE Characterize the particles (carbon? Asbestos?)

Exposure Assessment 3 Model the exposure (using for example wind speed) Determine the source(s) Find out anything “special” about the population

Exposure Assessment

Dose-response relationships "how much is dangerous" ? Animal data and (preferably) human occupational data used: example: BEIR IV The problem of thresholds Extrapolation: most common convention is the use of some multiple of the upper bound of the 95% confidence interval

SLOPE (b) of the lung cancer/ exposure curve: SMR 100 [b times (cumulative exposure)] Slope (extent per unit exposure) of risk Textile: 1.0 Manufacture (mixed): 0.2 Mining: .05 Degree of exposure ----

Different principles for cancer and non-cancer

The threshold issue D I S E Issue: Which line is correct? A S E Exposure ---

The threshold problem: points are at high dose Tremblay 1998 Jones 1956 Smith 1989

Risk Characterization/ Management 1 What Is the Extra Risk to Health? Maximum Individual Lifetime Cancer Risks:

Risk Characterization/ Management 2 What Is the Distribution of Individual Risks? Population Cancer Risks can be calculated from the Distributed Individual Risks This is where we “do the math”

Risk Perception; Risk Communication The balance between “risk” and “outrage” High risk/ low outrage: radon and lung cancer? Low risk/ high outrage: asbestos in schools? The media as an "amplifier" Voluntary vs. Involuntary Risk (smoking vs. hazardous waste siting) Known vs. Unknown Risk (lead pipes; lead in gas)

Risk assessment based on the linear exposure-effect model Žy Žx b Žx Žy R.A. 1.0 background risk Exposure assessment Cumulative exposure