research proposal1

Research Proposal

Summary

.

M

y

research

topic

is

“Prenatal

Mercury

Effect

on

Children

as

Loss

in

“

IQ

”

,

my

research

question

is:

Do

children

whose

mothers

have

prenatal

exposure

to

mercury

suffer

a

neurological function damage. Our target population is children whose mothers have prenatal

exposure to mercury, the intervention is the blood and hair mercury level (ug/L) , and we use

the IQ points to measure the neurological status. The appropriate study design is prospective

cohort

design;

we

use

multiple-level

dose-response

regression

with

repeated

measure

to

analysis

the

dataset.

My

analysis

provides

the

ability

to

estimate

the

potential

confounding

and benefits from reduction in mercury exposure by using the Trasande approach

Manli Yan

1

2013-3-28

Introduction

Parental

exposure

to

mercury

through

maternal

consumption

of

fish

has

been

associated

with

reduced

performance

on

tests

of

neurologic

function

in

children,

including

tests

of

cognitive development, attention and behavior, and motor skills. A comprehensive review of

the mercury literature conducted by the NRC(National Research Council) Committee on the

Toxicological

effects

of

Methyl

mercury

concluded

that

based

on

the

evidence

available,

“neurodevelopment

deficits are the most sensitive ,well-documented effects

”

of exposure to

mercury.[1]

If

we

do

a

search

in

our

medical

database,

using

keywords

such

as

“methyl

mercury”

“

prenatal

exposure”

,

“

child

development

”

,

“neurobehavioral”

,

revealed

that

there

are

12

recent investigation have been published so far on this subject, in the sense that a sample of

children was tested on one occasion and

associations between test results and a measure of

prenatal exposure to MeHg we examined, unfortunately a number of cross-sectional studies

have

similarly

inconsistent

results

two

major

longitudinal

investigation

report

contradictory

findings[2].therefore we should try a new study design and a more accurate statistical method,

by using the cohort study we can be very certain of temporal sequence, the determination of

outcome status unlikely to bias, we have better chance to get consistent conclusion. Applying

multiple-level

dose-response regression to

our case will get

very

accurately predicted value

which explains for how much mercury contained in blood will cause the IQ decrease by 1,and

most importantly, we will have a dependable evidence

“

Whether Prenatal Mercury has Effect

on Children as Loss in

“

IQ

”

”

Object and Research question.

Although there were many investigations in this area, so far few of them can give a very

convincible

result

due

to

the

limitation

of

the

study

design

and

inappropriate

statistical

method, none of them has clarified and quantified the specific association between prenatal

mercury

and

children

neurologic

status.

My

research

topic

“Pre

natal

Mercury

Effect

on

Children as

Loss in

“

IQ

””

will investigate and study this issue in at a particular aspect and

angle.

Population: the children whose mothers have exposure history to mercury.

Exposure: the blood and hair mercury level of mothers.

Comparison group: not Applicable

Outcome: IQ scales.

Background and Literature Review

Mercury

is

ubiquitous

environmental

toxicant.

It

exists

in

three

forms,

each

of

which

possesses

different

bioavailability

and

toxicity:

the

metallic

element,

inorganic

salts,

and

organic compounds (methyl mercury, ethyl mercury, phenyl mercury) [3]

Although

volcanoes

and

other

natural

sources

release

some

elemental

mercury

to

the

environment, anthropogenic emissions from coal-fired electric power generation facilities are

the major cause of mercury that are released into the earth

’

s atmosphere.

Manli Yan

2

2013-3-28

Elemental mercury is easily aerosolized because of its low boiling point, and once airborne

it

can

travel

long

distances

to

eventually

deposit

into

soil

and

water.

In

this

kind

of

circumstance, the metallic mercury is transformed into methyl mercury, however this form of

mercury could be existed in marine food chain for very long time, therefore consumption of

contaminated fish is the major route of human exposure to methyl mercury.

The Toxicity of methyl mercury to brain was fist spotted as an evidence at Minimata Bay,

Japan in 1950,the river fish men living on was found highly contaminated with Hg from the

nearby power plant.20 years later there was a report from Iraq about the Hg poisoning where

seed

grain

contaminate

with

MeHg

was

used

to

make

homemade

bread.

The

neurodevelopment

effects

seen

in

children

exposed

inutero

in

both

place

show

a

severe

sensory and central nervous system impairments.

As

I

mentioned

in

introduction,

there

were

several

investigation,

9

are

cross-sectional,

another 3 comprised a cross-sectional

element of a larger longitudinal study, concerning on

the adverse effects of prenatal exposure. The following tables show the result:

Table 1

.Cross-sectional studies[4]

Reference/Country

Response

n

Age at

Exposure

Exposure

Test

rate (%)

testing

measure

level

type

Faroe Islands

64

182

2 week

CB,MH

20.4ug/L

NE

Canada

95

234

12-30

MH

6ug/g

NE,DS

months

Canada

72

212

38

MH

0.5ng/mg

PT

months

Peru

NS

131

NS

MH

7.05ppm

NE,DM

New Zealand

NS

237

6-7

MH

>6mg/kg

PT,AT

years

3-6mg/kg

<3mg/kg

Madeira

99

149

6-7

MH

9.64ug/g

PT,NT

years

Brazil

84

351

7-12

MH

11.6ug/g

PT

Years

Greenland

NS

43

7-12

MH

15.5ug/g

PT,NT

years

French Guiana

88

97

9months

MH

12.7 ug/g

NE,PT

-12 year

AT:

attainment

test;

CB,

cord

blood;

DM:

developmental

milestone;

DS:

developmental

scales;

MH:

maternal

hair;

NE:

neurologic

examination;

PT,

psychological

tests

NS:

not

stated

Manli Yan

3

2013-3-28

Table 2.

Longitudinal studies

Age at

Resp

testing

onse

rate

(%)

Faroe Islands

12 months

57

7 years

90

14 years

86

Seychelles Island(pilot)

5-109 weeks

98

66 months

NS

108 months

NS

Seychelles Island(main)

6.5 months

95

19 months

95

29 months

94

66 months

91

108 months

83

n

Exposure

measure

Exposure

level

Test type

583

917

882

789

217

87

740

738

736

711

643

CB,MH

MH

MH

MH

MH

MH

MH

MH

MH

MH

MH

121 nmol/L

114 nmol/L

NS

6.6 ppm

7.1 ppm

<3->9 ppm

5.9 ppm

5.9 ppm

5.9 ppm

6.8 ppm

6.9 ppm

DM

NE,PT,NT

PT,NT

DS,NE

PT

PT

VR,NE,DS

DM,DS

DS

PT

PT,TR

As

we

can

see

from

above

tables,

the

outcome

of

cross- sectional

study

is

similarly

produced

inconsistent

result,

and

two

major

longitudinal

investigation

report

contradictable

findings. Therefore, we can have some brief view of the literature highlight, which is, so far,

we

still

have

lot

uncertainty

in

this

field,

basing

the

current

study

design,

the

most

reason

which cause the gap in our study is the variation of the methods used to assess the two basic

element of the association which is exposure and outcome. The major problem in exposure is

there

is

no

much

clarity

in

the

mercury

concentration;

different

studies

may

use

different

subjects, which include: child hair, maternal hair, cord blood, and maternal blood, the strength

of

relationship

of

each

of

these

elements

is

uncertain,

for

the

outcome,

most

were

predominantly

psychological

tests,

but

some

of

tests

may

be

not

suitable

for

our

case,

for

example,

the

Trailmaking

test[]

used

in

the

Faroes

Island

was

originally

designed

for

the

patient suspected suffer to brain damage, this damage could result from head injury or other

insult or from a degenerative disease, furthermore, when the tester dealing

with the patient,

they more like to treat them as an individual and try to make clinical judgment on the basis of

a number of source of information, that rise the risk the value will be lost when they apply

this

attitude

to

a

large

groups

of

people.

On

the

account

of

all

these

reason,

if

we

keep

applying cross- sectional design in this area, it will increase the uncertainty rather than reduce

it.

Manli Yan

4

2013-3-28

Method

The target population is the pregnant women from 6 areas, which are Nunavik(Inuit ), New

Zealand, French Guiana,

Florida,

Texas, Nevada, the people in these area are all subject to the

exposure

of

mercury

since

the

highly

consumption

of

seafood

or

nearby

coal-fired

power

plant. There were also similar investigation carried our in those areas, thus it will not be hard

to find appropriate subjects, we can check the dataset of the local hospital, find the pervious

record distribution of mercury level of pregnant women who are exposure to mercury such as

fish men, workers from power plant or citizen living near from power plant or sea, make sure

we have most reasonable location to pick our samples. Once we decide the hospitals we could

start

sampling,

during

the

sampling,

we

should

be

noted

that

our

subject

should

not

have

potential risk of confounding our future response, for example, the pregnant women who have

neurologic function problem, or other related issue should not be included in our procedure.

Exposure assessment

We measure from the subjects is Cord blood (ug/L), Maternal blood (ug/L), Maternal hair

first

trimester

(ug/g),

Maternal

hair

second

trimester

(ug/g),

Maternal

hair

third

trimester

(ug/g),for the Maternal hair ,we measure all three period, and take mean of these three when

we analyzing the data, use the mean as single element, in our future analysis we will treat all

these

three

elements

as

a

factor

to

see

whether

there

are

confounding

between

these

measurements.

Outcome measurement

When diagnosing children, basing the formal experience, only applying one test may cause

potential bias, because so far, it is still uncertain that which part of nurological function the

mercury

will

affect,

so

it

is

wise

to

apply

multiple

tests

for

each

child

to

avoid

the

measurement bias.

We use

children’

s intelligence quotient (IQ) as response, to guarantee the accuracy of our

outcome; we use four different tests, the result of them are all scale, they are following:

Wechsler Intelligence Scales for Children, Revised (WISC-R), a test for children between

the

ages

of

4

and

16

inclusive

that

can

be

completed

without

reading

or

writing,

mainly for

diagnose attention-deficit hyperactivity disorder (ADHD) and learning disabilities

California

Verbal

Learning

Test

(CVLT),

well-known

verbal

learning

tests

in

neurophysiology

.

Because

it

contains

so

many

different

components;

the

CVLT

is

fairly

popular as a neuropsychological test of many aspects of verbal learning and memory

Bender-Gestalt

Test

(BGT),

is

a

psychological

test

first

developed

by

child

neuropsychiatrist

Lauretta

Bender.

The

test

is

used

to

evaluate

maturity

to

screen for developmental disorders, or to assess neurological function or brain damage.

McCarthy

Scales

of

Children's

Abilities

(MSCA)

An

individually

administered

test

that

assesses

the

cognitive

development

and

motor

skills

of

children

aged

2-8;

meanwhile,

we

consider a pilot study in our case, which is repeat measure, we follow up those children since

they are four-year old, each year, we examine their neurologic status by using all four tests,

the length of follow up is four years, we have four repeats.

Potential confounding and biases

.

In

our

study,

the

main

biases

is

selection

bias,

since

our

subjects

come

from

a

certain

circumstances, such as health worker effect, this bias can arise in cohort studies, and the loss

Manli Yan

5

2013-3-28