ENVIRONMENTAL RISK ASSESSMENT

Hazard identification
Dose-response assessment
Human exposure assessment
Risk characterization

Risk
Hazard implies capability of substance to cause an adverse
effect
Risk is a measure of the probability that the hazard will occur
under specific exposure conditons.

Some risk are well defined (eg. frequency and severity of

automobile accidents)
In contrast, other hazardous activities such as those
resulting from the use of alcohol and tobacco are more
difficult to document. Their assessment requires complex
epidemiological studies.

Risk assessment
Risk assessment is the scientific approaches to collect
data that are used to relate the response to a dose of a
pollutant. Such dose-response data can be combined to
estimate overall risk assessment.
Risk management is the process of deciding what to do
and how to allocate national resources to protect public
health and the environment.
Risk is being expressed as a percentage or as a decimal
fraction, no units.
Example:
In the U.S in 2001, there were about 3.9 million deaths per year. Of these,
about 541,532 were cancer-related.
The risk of dying from cancer in a lifetime was about 0.14 or 14%.
The annual risk (assuming a 70-year life expectancy) is 0.002 or 0.2%

One of the objectives

of RA is to provide a
starting point in
balancing the
tradeoffs between an
acceptable
incremental risk and
the cost of controlling
risk to that level.
RA is divided into 4
steps:

Terminology:
Exposure and dose are often used interchangeably.
Strictly, exposure refers to either a deliberate or non-

deliberate non-quantifiable situation, whereas dose refers

to the quantity of toxicant administered to, or ingested by

an organism.
Threshold - the point that must be exceeded to begin
producing a given effect or result or to elicit a response

Terminology:
Acute toxicity a measure of the amount of a
substance that is needed to cause some acute

response, such as organ injury, coma, or even

death. The effect can be within a short period or

after a single exposure;

At the prolonged exposure a chronic toxicity is
observed.

Terminology:
mutagen an agent that induces a permanent
change in the genetic material of the cell
exposed to it, which can be transmitted to future
generations
carcinogen an agent that causes a cancer
cancer collective noun for 200 different
diseases, all of which are characterized by
unrestrained cell divisions.
teratogen an agent that induces
abnormalities in an embryo/fetus when
administered to the maternal organism

1. Hazard identification: What are we considering?

(LD50; mutagenesis; carcinogenesis; animal tests;
human studies)

Determine whether or not a particular chemical is

causally link to particular health effect, such as cancer or
birth defects

Experimental design:

There are a variety of parameters which have to be considered

when testing the toxicity of a substance, assuming that it is
pure.
-choice of animal model
-route of administration
- physical state of toxicant (gas, liquid, lipophilicity, etc)
- dose and duration of treatment: acute/sub-acute or chronic
toxicity testing

Toxicity testing in animal

Three types of animal tests:

i.Short-term test, called the Ames mutagenicity assay, subjects

special tester strains of bacteria to the certain chemical.
ii.Intermediate testing involve relatively short term (several
months duration) carcinogenesis bioassays in which specific
organs in mice and rats are subjected to known mutagens to
determine whether tumors develop.
iii.The most costly, complex and long-lasting test chronic
carcinogenesis bioassay.
- Two species of rodents must be tested (mice and rats)
- At least 50 males and 50 females of each species for
each dose
- At least two doses must be administered (plus a nodose control)

Human studies
Epidemiologic is the study of the incidence rate of disease in
real populations.
Preliminary data analysis involves setting up a simple 2 x 2
matrix.

Epidemiologic Data Analysis

Suppose 5 percent of individuals exposed to a chemical

get a tumor, and 2 percent of those not exposed get the
same kind of tumor. Find the relative risk, attributable
risk, and odds ratio.
Relative risk = 2.5
Attributable risk = 0.03

Odds ratio = 2.58

The relative risk and the odds ratio both are above 1.0, so they
suggest a relationship between exposure and tumor risk. For those
who were exposed, the risk of tumor has increased by 0.03 over who
were not exposed.

Suppose 30 out of 500 rats exposed to a potential

carcinogen develop tumors. A control group of 300 rats
not exposed to the carcinogen develops only 10 tumors.
Find the relative risk, attributable risk, and odds
ratio. Do these indicators suggest that there might be a
relationship between exposure and tumor risk?
Relative risk = 1.8
Attributable risk = 0.0267
Odds ratio = 1.82
The relative risk and the odds ratio both are above 1.0, so they
suggest a relationship between exposure and risk. For the rats were
exposed, the risk of cancer has increased by 0.0267 over who were
not exposed. All three measures indicate the relationship between
exposure and tumor risk.

Limitation of Animal studies and Human studies

Animal studies

Human studies

- No species provides an exact

duplicate of human response.
Certain effects that occur in
common lab animals generally
occur in people. The exceptions
are toxicities dependent on
immunogenic mechanisms. Most
sensitization reaction are difficult
to induce in lab animals.

- Large populations are required to

detect a low frequency of
occurrence of a toxicological
effect.

- Inability of a bioassay to detect

small risks. Regulatory try to
restrict human risks due to
exposure to carcinogens to level
of about 10-6, yet animal studies
are only capable of detecting
risks down to 0.01 to 0.1.

- Competing causes of the

observed toxicological response
make it difficult to attribute a
direct cause and effect. (cigarette
smoking, the use of alcohol and
drugs, prior disease states)

- A long or highly variable latency

period may be needed between
the exposure to the toxicant and
a measurable effect.

Dose-response mortality curves for acute toxicity.

LD50 = Lethal Dose for 50% of a tested group

2. Dose-response assessment: How bad it is?

(threshold dose; dose-response curve; chronic daily
intake, reference dose, hazard quotient, hazard index)

To obtain a mathematical relationship between the

amount of a toxicant that a human is exposed to and the

risk that will be an unhealthy response to that dose.
The dose of an exposure averaged over an entire
lifetime (for human it is assumed to be 70 years)

Hypothetical dose-response curves. Dose-response

curves for carcinogens are assumed to have no
threshold; that is, any exposure produces some
chance of causing cancer.

The potency factor is the slope of the doseresponse curve.

Incremental lifetime cancer risk = CDI x PF

Chronic Daily Intake:

(how much is consumed every day)

Average _ daily _ dose _(mg / day)

CDI (mg / kg day)
Body _ weight _(kg)
When a risk assessment is made for exposures that do not last the entire lifetime:

For contaminants in water:

CDI

Concentrat ion (mg / L) _ x _ Intake _ rate( L / day) _ x _ Exposure(days / life )

Body _ weight (kg) _ x _ 70( yrs / life ) _ x _ 365(days / yr )
For contaminants in air:

Concentrat ion (mg / m 3 ) _ x _ Intake _ rate(m 3 / day) _ x _ Exposure(days / life )

CDI
Body _ weight (kg) _ x _ 70( yrs / life ) _ x _ 365(days / yr )

If 70 kg people breath 20 m3/day of air containing 10-3

mg/m3 of carcinogenic VOC throughout their entire 70
year lifetime, find the cancer risk. Given the potency
factor = 0.01 (mg/kg-day)-1.

CDI = 0.000285 mg/kg-day

Risk = CDI x PF = 2.9 x 10-6

The drinking water standard for tetrachloroethylene is 0.005 mg/L.

Suppose a 70 kg person drinks 2 L of water every day for 70
years. Find the cancer risk for this individual. Potency factor of
tetrachloroethylene is 5.1 x 10-2 (mg/kg-day)-1.
If a city with 500,000 people in it also drinks the same amount of
this water, how many extra cancers per year would be expected?
Assume the standard 70 year lifetime.
CDI = Average daily dose (mg/day) / body weight (kg)
= 1.43 x 10-4 mg/kg-day
Incremental lifetime cancer risk = CDI x PF = 7.3 x 10-6

So, over a 70-year period, the upper-bound estimate of the probability that
a person will get cancer from this drinking water is about 7 in 1 million.
If there are 7.3 cancers per million people over a 70-year period, then in
any given year in a population of 500,000, the number of cancers
caused by TCE would be 0.052 cancers/year.

The drinking water standard for tetrachloroethylene is

0.005 mg/L. Using the EPA exposure factors for
residential consumption (Daily intake = 2L for adult,
exposure frequency = 350 days/year, exposure
duration = 30 years, body weight = 70kg for adult),
what lifetime risk would this pose?

Lifetime risk = CDI x Potency factor

= 3.0 x 10-6

Mainstream smoke inhaled by a 70 kg smoker contains

roughly 0.03 mg per cigarette of the class B2
carcinogen, benzo(a)pyrene. From an individual who
smokes 20 cigarettes per day for 40 years, estimate
the lifetime risk of cancer caused by that
benzo(a)pyrene (there are other carcinogens in
cigarettes as well). Given that potency factor
(inhalation route) of benzo(a)pyrene is 6.11 (mg/kgday)-1.

CDI = 0.0049 mg/kg-day

Risk = CDI x PF = 0.03

The reference dose for noncarcinogenic effects:

Noncarcinogens - there is an exposure threshold; that is,

any exposure less than the threshold would be expected to
show no increase in adverse effect.
Lowest-observed-effect level (LOEL) : the lowest dose
administered that results in a response
No-observed-effect level (NOEL) : the highest dose
administered that does not create a response
Reference dose (RfD) or acceptable daily intake (ADI) an
indication of a level of human exposure that is likely to be
without appreciable risk (unit : mg/kg-day averaged over a
lifetime)

RfD is obtained by dividing the NOAEL by an appropriate uncertainty

factor. The uncertainty factor is typically between 10 and 1 000.

Reference
Dose

No-ObservedAdverseEffect Levels

Lowest-ObservedAdverse-Effect
Levels

Comparison between actual exposure and RfD to see

whether the actual dose is supposedly safe.
(mg/kg-day)

For non-carcinogens, the daily dose is averaged only over the

period of exposure.
If hazard quotient < 1.0, there should be no significant risk of
toxicity.
If hazard quotient > 1.0 could represent a potential risk.
When exposure involves more than one chemical:

Hazard index = sum of the hazard quotients

Suppose a 50 kg individual drinks 1 L/day of water

containing 2 mg/L of 1,1,1-trichloroethane, 0.04 mg/L of
tetrachloroethylene, and 0.1 mg/L of 1,1dichloroethylene. Given the oral RfD for 1,1,1trichloroethane = 0.035, tetrachloroethylene = 0.010,
1,1-dichloroethylene = 0.009. What is the hazard index?
HQ (1,1,1-trichloroethane) = 1.14

HQ (tetrachloroethylene) = 0.08
HQ (1,1-dichloroethylene) = 0.22
Hazard index = 1.44

3. Exposure assessment: How much of it?

Determine the size and nature of the population that has been
exposed to the toxicant under consideration and the length of
time and toxicant concentration to which they have been
exposed.

Firstly, the toxicants should be transported from the source to

the point of contact with human (pathways), and secondly, a
probability of toxicants contact with human should be
estimated (amount of contact).

Analyze the exposure pathway

Estimate the concentration of toxicants

at a particular exposure point

Human intake estimates (based on a

lifetime of exposure) duration of
exposure, amount of contaminants into
each exposed persons body, number of
people exposed

Bioconcentration factor (BCF, L/kg) : Amount of

consumption of a product containing toxicants by a human.
For example: BCF provides the key link measuring the
tendency for a substance to accumulate in fish tissue.
Concentration of toxicant in fish (mg/kg) = concentration in
water (mg/L) x bioconcentration factor (L/kg)

Another way is the estimation of half-life (days) of various

contaminants in water, air, and soil (contaminant
degradation rate).
Half life is the time required for the concentration to be
reduced by 50%.
t1/2 = ln 2 / k

C(t) = C0e-kt

Bioconcentration of TCE
Given a daily intake of 54 g of fish for a person, 350 days per year for
30 years eating locally caught fish, estimate the lifetime cancer risk
from fish taken from waters containing a concentration of
trichloroethylene (TCE) equal to 100 ppb (0.1 mg/L). The
bioconcentration factor for TCE is given as 10.6 L/kg. The cancer
potency factor for an oral dose of TCE is 1.1 x 10-2 (mg/kg-day)-1.

TCE concentration in fish = 1.06 mg TCE/kg fish

CDI = 3.36 x 10-4 mg/kg-day
Incremental lifetime risk of cancer = CDI x PF = 3.6 x 10-6

Suppose an underground storage tank has been leaking for many

years, contaminating the groundwater and causing a contaminant
concentration directly beneath the site of 0.30 mg/L. The
contamination is flowing at the rate of 0.5 ft per day toward a public
drinking water well 1 mile away (1 mile = 5280 ft). The half-life of the
contaminant is 10 years.
a.Estimate the steady-state pollutant concentration expected at the
well.
b.If the potency factor for the contaminant is 0.02 (mg/kg-day)-1,
estimate the cancer risk if a 70 kg person drank 2 L of this water per
day for 10 years.
Time required to travel to the well = 10560 days
The pollutants is assumed degrade exponentially, so the reaction rate
coefficient k = 1.9 x 10-4 /day
Pollutant concentration in well = 0.040 mg/L
CDI = 1.6 x 10-4 mg/kg-day
Lifetime cancer risk = CDI x PF = 3.2 x 10-6

 4. Risk characterization : So, what is the risk?

involves analysis of gathered data to be used in decision-

making process to set regulatory requirements and control

measures.
It is the final stage in the risk assessment process and involves
the prediction of the frequency and severity of effects in
exposed populations.

All data collected from exposure and toxicity assessment are

reviewed to corroborate qualitative and quantitative
conclusions about risk.

 Since most risk assessments include major uncertainties, it is

important that biological and statistical uncertainties are described in
the risk characterization.
In some complex risk assessments such as for hazardous waste sites,
the risk characterization must consider multiple chemical exposures
and multiple exposure pathways.
Simultaneous exposures to several chemicals, each at a
subthreshold level, can often cause adverse effects by simple
summation of injuries.
Individuals are often exposed to substances by more than one
exposure pathway (e.g. drinking of contaminated water, inhaling
contaminated dust). In such situations, the total exposure will
usually equal the sum of the exposure by all pathways.

Estimate the lifetime average chronic daily intake of benzene from

exposure to a city water supply that contains a benzene concentration
equal to the drinking water standard. The allowable drinking water
concentration (maximum contaminant level, MCL) is 0.005 mg.L-1.
Assume the exposed individual is an adult male (78kg) who consumes
water at the adult rate (2.3 L/day) for 63 years, that he is an avid
swimmer and swims in a local pool (supplied with city water) 3 days a
week for 30 minutes from the age of 30 until he is 75 years old.
(Assume a total lifetime of 75 years.) As an adult, he takes a long (30
minutes) shower every day for 63 years. Assume that the average air
concentration of benzene during shower is 5g.m-3. From the
literature, it is estimated that the dermal uptake from water is 0.0020
m3.m-2.h-1. Surface area available for an adult male is 1.94 m2.Direct
thermal absorption during showering is no more than 1% of the
available benzene because most of the water does not stay in contact
with skin long enough. Amount of air breathes daily for adult male is
0.633 m3/hr and water swallowing rate while swimming is 50 ml/hr.
Estimate the risk from exposure to drinking water containing the MCL
for benzene. (Given potency factor of benzene for oral route = 0.015
(mg/kg-day)-1 and inhalation route = 0.029 (mg/kg-day)-1)

Five routes of exposure:

1. Ingestion
2. Dermal contact while showering
3. Dermal contact while swimming
4. Inhalation of vapor while showering
5. Ingestion while swimming
1. CDI = [(0.005 mg/L)(2.3 L/day) (365 days/year)(63 yrs)]/ [(78 kg)
(75yrs)(365 days/yr)] = 1.24 x 10-4 mg/kg-day

2. Absorbed dose (AD) = [(0.005 mg/L)(1.94 m2)(0.002 m3/m2-h)

(0.5h/event)(1event/day)(365days/yr)(63yrs)(103L/m3)]/[(78kg)(75yrs)
(365days/yr)] = (1.04 x 10-4 mg/kg-day
Given that direct dermal absorption during showering is no more that 1%
(because limited contact time)
Actual dose of dermal contact, AD = (0.01) (1.04 x 10-4 mg/kg-day) =
1.04 x 10-6 mg/kg-day
3. AD = 3.19 x 10-5 mg/kg-day
4. CDI = 1.70 x 10-5 mg/kg-day
5. CDI = 4.11 x 10-7 mg/kg-day

Total exposure, CDIT = 1.74 x 10-4 mg/kg-day

Drinking the water dominated the intake of benzene.

There are different slope factors for both oral and inhalation
routes. Because we do not have a slope factor for dermal
contact, we assumed that it is same as oral ingestion.
Risk = 2.85 x 10-6 mg
This is the total lifetime risk (75 years) for benzene in drinking
water.