163-Análise de Riscos

7/29/2019 163-Anlise de Riscos

1/42

EH&S, Inc. 1

ENVIRONMENTAL HEALTHRISK ASSESSMENTS

SCIENCE OR VOODOO?

Ron Pearson, M.S., CIHEnvironmental Health & Safety, Inc.

St. Paul MN


2/42

EH&S, Inc. 2

The Nature of Risk

200 people die annually in U.S. fromelectrocution (risk level 10-6 per year)

should I replace the wiring in my oldhouse?


3/42

EH&S, Inc. 3

The Nature of Risk

7000 people die annually in U.S. fromfalls in their homes

but ... most are over age 65, so shouldthe rest of us ignore this?

It's all about CHOICES


4/42

EH&S, Inc. 4

Estimating Risk

Probabilities are fine until it happens tome

Some of the uncertainty is due tochance, some of it isn't


5/42

EH&S, Inc. 5

Estimating Risk

Historical risks are easily understood -e.g. car accidents

What kind of car do you drive? Does it have airbags?

Do you drive fast?


6/42

EH&S, Inc. 6

Comparing Risks

action annual risk uncertainty

all cancers 3 in 1,000 10%

pack-a-day 4 in 1,000 150%smoker

mountain 6 in 10,000 50%climber

car accident 24 in 10,000 10% drinking MCL 6 in 1,000,000 1,000%

of chloroformin water


7/42EH&S, Inc. 7

Comparing Risks

Human nature dictates that we tend toworry more about risks that are severeand abrupt, as opposed to somethingthat has some "probability" of occurringdown the road

Many say that we cant compareunlike risks, but in fact, we do it all thetime


8/42EH&S, Inc. 8

The Costs of Risk Reduction

Location Risk Reduction Cost per(geog.) means life saved

Indonesia Death Immunization $100countries (infection)

U.S./ Cancer Pollution $1,000,000

other prevention


9/42EH&S, Inc. 9

Why do we need Risk

Assessment? "Emerging" risks - e.g. hormonalanalogues

Shifts in perception information overload - the "health studies"results that we are bombarded with daily

ability to measure minute amounts of

substances

many traditionally severe health risks(e.g. smallpox) are gone


10/42EH&S, Inc. 10

Environmental Health RiskAssessment health risk = the likelihood that an

adverse effect will occur to a person (orgroup of persons) in a chemical

exposure situation Usually, a higher exposure causes more

serious effects or makes them more

likely At some low exposure level, the risks

become insignificant


11/42EH&S, Inc. 11

Estimates of Risk

Estimates of risk are needed to assist inmaking decisions

Only in extreme cases will risksestimates alone drive decision making

zero risk compels no action, while a

great risk may compel immediate action IN REALITY, risk estimates lie

somewhere in between


12/42EH&S, Inc. 12

Environmental Health Risk

Assessments use two types ofRisk Estimates:

for carcinogens, the increased

probability of individuals' gettingcancer from a particular exposure

for other toxicants, a comparison of

expected exposure to an exposure thatis assumed to be insignificant


13/42

EH&S, Inc. 13

Environmental Health Risk

Assessments use two types ofRisk Estimates

Why? Because they are most often usedin USEPA risk assessments

In general, effects on systems such asthe reproductive or immune system arenot scrutinized nearly as much ascarcinogenic effects


14/42

EH&S, Inc. 14

What DONT risk assessmentsestimate? total number of people affected relative incidence of an adverse effect in

populations known to be exposed with

those not exposed the ratio of the expected risk with the

exposure to that expected without it

reduced life expectancy associated withthe effect

lost income potential, costs to society


15/42

EH&S, Inc. 15

Uncertainties in RiskAssessments Dose Effect = relationship between the

amount of a chemical exposure and thenature and/or severity of the toxic

effect Data on toxic chemicals usually come

from:

laboratory experiments on animals NOTepidemiology studies of humans

moreover, many are inferences based onbacterial and/or human cells


16/42

EH&S, Inc. 16

Uncertainties in Risk

Assessments Dr. Bruce Ames, (Ames salmonellamicrosomal screening test developer),stated repeatedly that he neverintended for his "tool" to be applied as itis today


17/42

EH&S, Inc. 17

Uncertainties in RiskAssessments both of these sources of data cause

problems because:

an animal or cell is not a human being

most animal toxicity data is short-term

relatively high exposures are used

experimentally, to cause statisticallysignificant effects


18/42

EH&S, Inc. 18


Assessments many species arehomogeneous(purposely, to limit

variability inresponse)

By contrast, humansare diverse in theirresponse tochemicals due to:

genetic make up

age

habits

occupation

health status

diet, etc.


19/42

EH&S, Inc. 19

Uncertainties in RiskAssessments

some argue that extrapolations fromanimals to humans are more reliablethan epidemiology studies, due to:

small study populations (lack of"statistical significance")

confounding variables

lack of exposure data differences between study populations

and the population to be protected


20/42

EH&S, Inc. 20

Uncertainties in RiskAssessments When no effect is seen in lab animals, is

there negligible risk to humans exposedat such a level?

a 1% incidence of any disease would beimpossible to detect in a study of 25animals, but would represent more than

2 million cases if the entire U.S.population were exposed

HOW DO WE ANSWER THISQUESTION???


21/42

EH&S, Inc. 21


Assessments In risk assessment, it is often assumedthat:

for cancer: there is no safe dose, and; at low doses, the relationship between

dose-effect is directly proportional(linear)

for other health effects: there is a safedose


22/42

EH&S, Inc. 22

Public Perception and Public

Demands The "vicious circle": public perception ->

media reporting ->congressional action ->agency regulation in response to publicdemands


23/42

EH&S, Inc. 23

Public Perception and PublicDemands

Are we (in the U.S.) better off now thanbefore the "skyrocketing" industrial useof chemicals?

Since 1940 - life expectancy hasincreased nearly 15 years

Since 1970 - infant mortality has

decreased by 1/2 Since 1970 - heart disease has dropped

by nearly 1/3


24/42

EH&S, Inc. 24

Public Perception and Public

Demands Cancer deaths have increased...or have

they?

Many feel the this is due simply to: smoking (increases lung, pharyngeal,

pancreatic and bladder cancer incidence)

sun exposure (malignant melanoma hasincreased eight-fold)

the increase in life expectancy (you have todie of something...)


25/42

EH&S, Inc. 25

Where has this all brought us?

Many times, the science ofepidemiology simply confirms theobvious - rarely has an epidemiologicalstudy drawn attention to an agent thatwas not already recognized by an astuteobserver in the field (e.g. Fen-Phen)


26/42

EH&S, Inc. 26

Where has this all brought us?

We fear carcinogens in our drinkingwater ... but what about Milwaukee'spublic water supply and an outbreak ofcryptosporidium? Would we better offtaking some of our money from theformer and spending it on the latter?

Asbestos: we won't even get into it ...


27/42

EH&S, Inc. 27

Regulatory Reform

What agencies conduct health riskassessments?

OSHA EPA

FDA

USDA


28/42

EH&S, Inc. 28

Regulatory Reform

1983 - the NAS published the "RedBook" - "Risk Assessment in the FederalGovernment: Managing the Process"

defined four steps of risk assessment,but more importantly, discussed how toseparate the "science" from the "policy"

R l R f


29/42

EH&S, Inc. 29

Regulatory Reform

1987: EPA published "Unfinished

Business" - ranked items such aspesticides in food and radon as higherhealth risks than items such as

groundwater contaminants orhazardous waste sites BUT...failed torank airborne lead as a high risk - again,it was driven bycarcinogens....essentially we have noscientific methods for comparing cancerwith non-cancer risks

R l R f


30/42

EH&S, Inc. 30

Regulatory Reform

1991 - Federal Focus, Inc. called for an

executive order (16 CRR 171),essentially prohibiting the use of overlyconservative assumptions

Risk Assessment has been representedby many as a "value free" process, whenin fact it is full of judgements

Risk Assessment and Risk Managementare, and should remain, separateprocesses

R l R f


31/42

EH&S, Inc. 31

Regulatory Reform Risk Assessment produces very precise

numbers of questionable accuracy


32/42

EH&S, Inc. 32

Regulatory Reform

Federal Trend: legislators pursuing riskassessment as means of telling us whatthe "real" risks are, so we can spend ourmoney accordingly - represents anothereasy answer for attacking what ails us -after all, what politician in their right

mind would outright oppose legislationthat is supposedly "good for theenvironment"?

R l t R f


33/42

EH&S, Inc. 33

Regulatory Reform State Trends: decreasing funding for

public health/environmental healthprograms but increasing environmentalregulatory spending - in 1994 we spent$4.09 per capita on the former and$18.87 per capita on the latter

In the Republican party's "Contractwith America" a bill called the "JobCreation and Wage Enhancement Act"bolstered risk assessment and costbenefit analyses requirements


34/42

EH&S, Inc. 34

Cost-Benefit Analysis

A good example of cost-benefit analysisand the fallacies that can be put forth:

OSHA's proposed IAQ rule estimatedthat a facility manager would spend anaverage of 15 minutes documenting

each complaint

"Ad t " i th S i


35/42

EH&S, Inc. 35

"Advancements" in the Scienceof Environmental Health Risk

Assessment ASTM RBCA - Risk Based Closure

Assessment methodology many states have jumped on the

bandwagon for this approach toscreening UST sites, especially as statefunds have become more scarce

ASTM RBCA Risk Based Closure


36/42

EH&S, Inc. 36

ASTM RBCA - Risk Based ClosureAssessment methodology

uses a tiered approach

Tier I: "lookup tables"

Tiers 2 - 4: incorporate more site specificvalues for: ground water

soil types

specific information on receptors

Still, much of the conclusions dependon mathematically modeled results -"garbage in - garbage out" still applies


37/42

EH&S, Inc. 37

Conclusion: Where do we gofrom here?

Most environmental problems areextremely complicated technically


38/42

EH&S, Inc. 38

Conclusion: Where do we gofrom here? We live in the age of entitlement: we

want the government to provide us arisk-free society, and we want it now!

We MUST decide how much minusculereductions in risk we are willing to payfor

We MUST question our legislatorsAND regulators motives and actions


39/42

EH&S, Inc. 39

Conclusion: Where do we gofrom here? We live in the age of technology, and

science can solve all of our ills: modernscience has it's limitations, particularlywhen it comes to the analysis of livingsystems - it may never suffice inaccurately predicting health effects or

their potential from low level exposures We MUST decide how much

uncertainty we are willing to tolerate


40/42

EH&S, Inc. 40

Conclusion: Where do we gofrom here? We live in the age of the sound byte:

most of the public gets the lion's shareof this information from the media

We MUST improve communication ofthese issues dramatically


41/42

EH&S, Inc. 41

Conclusion: Where do we gofrom here?

We live in the age of cancer paranoia:most of the EPA's regulatory efforts

focus on cancer We MUST shift the emphasis equally to

non-cancer endpoints


42/42

"Security is mostly a superstition. It

does not exist in nature, nor do thechildren of men as a whole experienceit. Avoidance of danger is no safer inthe long run than outright exposure.Life is either a daring adventure, ornothing."... Helen Keller

DON'T WORRY - BE HAPPY ...Bobby McFerrin

Documents

163-Análise de Riscos