Intern
ati
o
n
a
l
Jo
urn
a
l
o
f
P
u
b
lic Hea
l
th Science (IJ
P
HS)
V
o
l.4
,
No
.2
,
Jun
e
2
015
, pp
. 94
~101
I
S
SN
: 225
2-8
8
0
6
94
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJPHS
He
alth R
i
sk of O
b
esit
y
in Nativ
e
Ame
r
ic
an
A
dol
e
scen
ts
Benja
m
in D. K
n
i
s
l
e
y
1
, Gr
ac
e
Crosb
y
2
,
Linn Carothers
1
, M
.
Margar
et
Barth
2
1
Departm
e
nt
of
Natural
&
Math
em
atic
al
Scien
c
e
s
, Cal
i
forni
a
Bap
tist Univ
ersit
y
,
USA
2
Department of Health
Scien
ces
, Nutrition & Food Scien
ces, Cali
f
o
rnia B
a
ptist University
, USA
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Oct 02, 2014
Rev
i
sed
Feb
21
, 20
15
Accepted
May 22, 2015
Recen
t
repor
ts dem
onstrate the
need
to im
pro
v
e m
e
thods for
ident
i
f
y
ing
obesity
among adolescent minority
popu
lations, especially
amo
ng Native
American adolescents. Our
stud
y examined sever
a
l
anthropometr
i
c measures
to see which one was the best indicat
o
r
of obesity
for Native American
adolescents. We compared our data w
ith
that p
r
ovided b
y
the
Centers fo
r
Disease Contro
l
and Preven
tion (
C
DC), and
the n
a
tion
a
l He
alth
a
nd Nutritio
n
Examination Survey
s (NHANESIII). We
ex
amin
ed which measures differ
e
d
significantly
fro
m the NH
ANE
SIII, and which were most e
ffective fo
r
measuring obesity
in Nativ
e American
ado
l
escents. Our stud
y
populatio
n
represented a cr
oss-sectional, ep
idem
iological p
opulation (n
=18
3
) of Native
American students (ages 14-18) from dive
rse tribal backgrounds
at an urban
residential high
school. We
obtained baseline an
th
ropometric measurements
of height, weigh
t
, waist, tr
iceps
skin
-fold, and c
a
lf skin-fold m
easures from
the Nat
i
ve Am
er
ican s
t
ud
ents
an
d com
p
ared th
is
inform
ation wit
h
refer
e
nc
e
da
ta to a
sse
ss we
ight
c
l
a
ssifica
tio
n
b
y
bod
y
mass indices (B
MI) for our
population
.
Und
e
r the cond
ition
s
of our study
,
we found the calf skin-fold
measure to be th
e best indicator
of
norm
a
l weight in m
a
le Nativ
e
Am
erican
adoles
c
e
nts
,
and
the tric
eps
s
k
in-fold m
eas
ure to be the bes
t
i
ndica
tor of
obesit
y
in
m
a
le
Native
Am
erica
n
adol
escents
. T
h
e assessm
ent o
f
hea
lth
risk
b
y
anthropometr
i
c measures we
obtain
e
d appear
s appropriate for
predicting
obesit
y
and
d
e
velop
i
ng eff
e
c
tive
interv
ent
i
o
n
s for Native
Am
erican
adolescents in general.
Correlations between anth
ropometric measures and
obesity
in our
stud
y
population may
prove of significant interest for
monitoring obesity
prev
ention
in
itiative
s
for
Nat
i
v
e
Am
eric
an
adol
es
cents
.
Keyword:
Adolesce
nts
Ant
h
ropom
etric Indicators
B
ody
M
a
ss
I
n
d
e
x
Native Am
eric
an
Ob
esity Prev
entio
n
Copyright ©
201
5 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Ben
j
am
in
D. Kn
isley,
Depa
rtm
e
nt of
Natural & Mat
h
em
atical Sciences,
Califo
r
n
i
a Baptist Un
iv
ersity,
8
432
Mag
n
o
lia Av
enu
e
, Riv
e
rsid
e, C
A
9
2504
U
S
A
.
Em
a
il: b
e
n
j
am
i
n
d
a
v
i
d.kn
isely@calb
a
p
tist.edu
1.
INTRODUCTION
The c
o
ncept
of obesity as a
dis
ease, with well-defined
com
p
licati
ons, i
s
ap
pr
o
x
i
m
at
el
y o
n
e ce
nt
u
r
y
ol
d.
What
ha
s
m
a
de t
h
e pro
b
l
e
m
al
ar
m
i
ng i
s
t
h
e expo
n
e
n
tial in
crease in
th
e p
r
e
v
ale
n
ce and incide
nce of
obe
si
t
y
ove
r t
h
e past
fe
w
dec
a
des;
t
h
i
s
i
n
c
r
e
a
se has l
e
d t
h
e
Wo
rl
d
Heal
t
h
Or
ga
ni
zat
i
on
(
W
H
O
)
t
o
decl
are i
t
a
p
a
nd
em
ic an
d
worldwid
e
pu
blic h
ealth
crisis [1
].
In
Am
erica alone
,
31.8%
of all
childre
n and adolesce
nts are
now
classifie
d
as
overweight or obese [2].T
h
is
s
u
ggests
that the de
gree
of adolesce
nt obesity has incre
a
sed
i
n
recent deca
des
[3]. Furthe
rmore
,
in
t
h
ree
indepe
ndent
studies data re
fl
ecting m
i
nority populations s
u
ggests
t
h
at
u
nde
rre
p
r
esent
e
d
g
r
o
u
p
s
are am
ong
t
h
ose at
ri
s
k
fo
r
devel
opi
ng
hi
g
h
er
rat
e
s
of
o
v
e
rwei
ght
a
n
d o
b
esi
t
y
wh
en
co
m
p
ared
with
wh
ites [4
],[5
]. Th
at is, u
n
d
e
rrep
r
esen
t
e
d
g
r
ou
ps and
min
o
r
ities p
l
ay an
in
sep
a
rab
l
e p
a
rt
fo
r us t
o
f
u
l
l
y
gras
p t
h
e i
m
pact
of o
b
esi
t
y
i
n
ou
r cu
rre
nt
U.
S. p
o
pul
at
i
o
n.
As
s
h
o
w
n
i
n
Fi
gu
re
4,
w
e
fo
u
n
d
t
h
a
t
m
a
l
e
and
fem
a
le
cases
of
o
b
esi
t
y
m
a
de up
38
% o
f
ou
r
p
opu
latio
n
(70
ou
t
of
183
N
a
ti
v
e
American
adoles
cen
ts).
Th
e relativ
e freq
u
e
n
c
y
of o
u
r statistics
is
ap
pro
x
im
ately
double that re
p
o
rt
ed b
y
t
h
e
CDC
(18%
) [6].
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Assessing
Hea
lth
Risk
o
f
Ob
esity in
Na
tive American
Ado
l
escen
ts (Ben
j
a
min
D. K
n
isley)
95
In add
itio
n, a rev
i
ew of t
h
e cu
rren
t
literatu
re d
e
m
o
n
s
trates th
at st
u
d
i
es on ado
l
escen
t
obesity a
m
o
ng
min
o
r
ity po
pulatio
n
s
are rel
a
tiv
ely rare.
In
fo
rm
atio
n
is esp
ecially scarce co
n
c
ern
i
ng
o
b
e
sity in
Nativ
e
Am
eri
can adol
escent
s
.
Al
t
h
o
u
g
h
s
o
m
e
research
o
n
o
b
esi
t
y
has bee
n
do
ne am
ong
Nat
i
v
e Am
eri
can adul
t
s
fr
om
t
h
e Pim
a
In
di
ans
o
f
Ari
z
ona
, o
u
r
st
udy
s
o
u
g
h
t
t
o
ad
d
r
ess t
h
e
uni
que
nee
d
s
of
Nat
i
v
e
A
m
eri
can
ad
o
l
escen
t
s u
s
i
n
g
a trib
ally div
e
rse sam
p
le
to
i
m
p
r
ov
e th
e relev
a
n
c
y of o
u
r resu
lts to
th
e Nativ
e Am
erican
adol
esce
nt
po
p
u
l
a
t
i
on
i
n
ge
ne
ral
.
Cu
rren
tly, m
o
st eq
u
a
ti
o
n
s
th
at pred
ict
ob
esity are
d
e
riv
e
d
fro
m
p
r
ed
o
m
in
an
tly wh
ite sam
p
le
populations
[7]. This
does
not
take
i
n
to
account
biol
ogical va
riations bet
w
een rac
e
s, a
n
d ca
n l
ead t
o
inaccurate res
u
lts whe
n
t
h
ese
equations
are
a
pplied
to popul
a
tions from
dive
rse racial
bac
k
grounds.
In t
h
e
wo
rl
d o
f
evi
d
e
n
ce-
bas
e
d m
e
di
ci
ne, i
t
i
s
n
ecessary to acknowledge
r
acial differe
nces a
m
ong
our national population in
order to provi
d
e sufficient
evide
n
ce for treating m
i
norities.
For e
x
am
ple, a recent
st
udy
i
n
2
0
1
2
am
ong
Af
ri
can
Am
eri
can st
udent
s a
g
ed 1
8
-
27
descri
bed “
a
n u
n
ex
pect
ed
fi
n
d
i
n
g reveal
i
ng t
h
at
waist circu
m
feren
ce
was sig
n
ifican
tly
associated with cardiova
scular
reac
tiv
ity wh
en
an
alyzed
; h
o
w
ev
er, th
i
s
rel
a
t
i
ons
hi
p w
a
s no l
o
nge
r si
gni
fi
cant
aft
e
r B
M
I was ent
e
r
e
d i
n
t
o
t
h
e eq
uat
i
on” [
8
]
.
T
hus
, sy
st
em
at
ic erro
r
can res
u
lt in the inaccurate es
tim
a
tion
of t
h
e
relative body fat of African
Americans, and there
f
ore de
fini
tions
o
f
ob
esity in
min
o
rities, if th
es
e inh
e
ren
t
d
i
fferen
ces are igno
red
.
2.
R
E
SEARC
H M
ETHOD
Th
e
o
b
j
ectiv
e
o
f
ou
r st
u
d
y
w
a
s to
i
d
en
tify p
r
ev
alen
t an
th
ro
p
o
m
e
tric
in
d
i
cators
o
f
ob
esit
y in
o
u
r
Nativ
e
Am
erican a
dolesce
n
t
p
opulation
a
n
d to
com
p
are
th
ese resu
lts
with
n
a
tion
a
l
statistics.
Th
e
p
r
im
ary
b
a
selin
e an
throp
o
m
e
tric in
d
i
cato
rs
fo
r ou
r researc
h
with
i
n
ou
r
Nativ
e
American
p
o
pu
l
a
tio
n
co
nsisted
of
th
e
w
a
i
s
t
,
t
r
i
cep
s s
k
i
n
-f
ol
d,
an
d
c
a
l
f
s
k
i
n
-
f
o
l
d
m
easure
s
, a
n
d
w
e
re
gat
h
e
r
ed
us
i
n
g
st
an
dar
d
i
z
e
d
pr
ot
o
c
ol
s.
M
easure
s
su
c
h
as
h
e
igh
t
, w
e
ig
h
t
, an
d BMI co
n
t
ained
no
missin
g
data and
utilized
all
1
8
3
Nativ
e
American
adol
esce
n
t
s (
a
g
e
s 1
4
-
1
8)
. M
i
ss
i
ng
dat
a
w
a
s
re
corde
d
withi
n
our
t
h
ree prim
a
r
y
m
easures
.
Our c
o
llected
data
w
a
s
then com
p
iled
and asses
s
ed
as
a pilot st
udy
t
o
sho
w
w
h
i
c
h i
n
d
i
cat
ors o
f
ob
esi
t
y
w
e
re
creating
d
i
screpan
c
ies with
nat
i
onal
dat
a
c
o
m
p
ari
s
ons
. Al
l
st
at
i
s
t
i
cal
graphs
an
d a
n
al
y
s
es
w
e
re
gene
ra
t
e
d
usi
n
g R
-
3.
1
.
0 st
at
i
s
t
i
cal
sof
t
w
a
re
[9]
.
3.
R
E
SU
LTS AN
D ANA
LY
SIS
There
w
e
re t
h
ree basi
c ste
p
s in
our
data a
n
al
ysis. F
o
r
the
first step,
w
e
sele
cted t
w
o
w
e
igh
t
categories
for
co
m
p
ariso
n
,
n
a
mely
, n
o
r
m
a
l (
B
MI lev
e
ls b
e
t
w
een
18
.
5
-2
4.
9
)
and ob
ese w
e
i
g
h
t
categories (BMI greater
than
o
r
eq
u
a
l to
30
.0
) to
assess wh
ic
h
an
thro
p
o
m
e
t
r
ic in
d
i
cators w
e
re b
e
st suited
fo
r i
n
feren
c
e. As
sh
o
w
n i
n
Fi
gu
re
1
,
w
e
also
created
a
ti
m
e
-series
p
l
o
t
of m
e
an
s
fo
r
b
o
t
h
gen
d
e
rs to
d
i
sp
la
y
th
e v
a
riation
c
o
n
t
ai
ne
d i
n
B
M
I
th
ro
ugh
ou
t ti
me. A
t
th
e
ag
e
of
14
,
a
l
a
rge
n
u
m
b
er of m
a
le
and fem
a
le
ad
olescen
ts
w
e
re
already
classifi
ed
as
o
v
e
r
w
e
i
g
h
t
.
By the age of
18,
m
a
n
y
of these
had m
o
v
e
d in
t
o
the ob
ese
w
e
igh
t
category
.
Ov
erall, th
is fig
u
re
dem
onst
r
at
es t
h
at
cl
im
bi
ng
r
a
t
e
s of B
M
I i
n
bot
h ge
n
d
er
cat
ego
r
i
e
s co
rr
esp
o
n
d
t
o
ri
si
n
g
rat
e
s
p
r
edi
c
t
e
d
by
NH
A
N
ES
II
I d
a
ta
[3]
.
Th
e seco
nd
step
o
f
o
u
r d
a
ta
an
alysis,
as sh
o
w
n
in
Fi
g
u
res
2
.
0
-
2
.
1
,
w
a
s
to id
en
tify
tell-tale
sig
n
s
of
n
o
rm
ali
t
y
in
ou
r d
i
stri
b
u
tions. Using
t
h
e
Sh
ap
i
r
o-W
ilks t
e
st, w
e
assessed
w
a
ist, tricep
s sk
in-fo
l
d
,
an
d
cal
f
ski
n
-f
ol
d m
easures t
o
u
n
c
o
v
e
r
sev
e
ral
de
g
r
ee
s of n
o
rm
al
bel
l
cur
v
e ass
u
m
p
t
i
ons. F
r
om
t
h
i
s
, w
e
i
n
fer
r
e
d
whic
h
m
easure w
o
ul
d
b
e
b
e
st
s
u
i
t
e
d t
o
cl
assi
fy
n
o
r
m
a
l
and
ob
e
s
e
w
e
igh
t
categorie
s. For i
n
stance
, in fi
gure
2.0
we ca
n
obs
er
ve t
h
e a
p
pr
oac
h
o
f
no
r
m
al
wei
ght
cl
ass Nat
i
v
e
American m
a
les approach
i
ng
bel
l
cur
v
e ass
u
m
p
t
i
ons
with
th
e calf
mean
(i.e., th
e sh
ap
e
o
f
t
h
e
d
i
stribu
tio
n in
th
is i
m
ag
e is
m
o
st alik
e to
a stan
d
a
rd
b
e
ll cu
rv
e).
Hence
,
it appe
ars that calf-mean m
easures are best
su
ited fo
r pred
ictin
g n
o
r
m
a
l weig
ht classes fo
r male
Native
Am
eric
an a
dolesce
nts. In cont
rast
, the no
rm
al weigh
t
m
a
le tricep
s-sk
i
n
fo
ld and
waist circu
m
feren
ce
measures
devi
ate
m
o
re from
norm
al-bell curve ass
u
m
p
tions
. In fact, waist
circum
ference
m
easures deviate
m
o
re than tri
ceps-ski
n
fold m
eas
ures and s
u
ggest
wai
s
t circum
feren
ce to
b
e
th
e
p
oorest ind
i
cato
r
for
pre
d
i
c
t
i
ng
n
o
r
m
al
wei
ght
cl
asses f
o
r m
a
le Nat
i
v
e Am
eri
can a
dol
esce
nt
s. Li
ke
wi
se,
wi
t
h
fi
g
u
r
e
2.
1 we
obs
erve obese
weight classes for m
a
le
Native Am
ericans and
discove
r
once
again that waist circum
ference is
th
e p
oorest –
in
term
s o
f
its s
h
ap
e
–
in
b
e
ing
m
a
tch
e
d
with
a stan
d
a
rd
b
e
ll cu
rv
e.
Howev
e
r, no
tice th
e switch
bet
w
ee
n cal
f a
nd t
r
i
ceps
ski
n
-f
ol
d m
easures
. Fo
r m
a
l
e
obe
se weigh
t
categ
ories it app
e
ars th
at th
e tricep
s-sk
in
fo
l
d
m
easu
r
e is th
e
b
e
st i
n
d
i
cato
r
fo
r
pred
ictin
g ob
esity
am
o
n
g
our
p
opu
latio
n du
e to
its
relativ
e lik
en
ess
with
a stan
d
a
rd
b
e
ll cu
rv
e. Later
on in
th
is stu
d
y
(see Tab
l
e 1
)
we will o
b
s
erv
e
th
e d
e
fin
itiv
e pro
o
f th
at statisti
cally
dem
onst
r
at
es
whi
c
h i
n
di
cat
o
r
i
s
m
o
st
appropriate for acc
urately predict
i
ng
o
b
ese a
nd
no
rm
al
wei
ght
cl
asses
for m
a
le Nati
v
e
Am
erican
s. Un
til th
en, th
ese fi
g
u
re
s
giv
e
app
r
op
riat
e h
i
n
t
s an
d
su
gg
estion
s
th
at o
u
r
po
p
u
l
a
t
i
ons a
ppea
r
t
o
ap
p
r
oac
h
no
rm
al cur
v
e as
s
u
m
p
tions m
u
ch m
o
re effec
tively a
m
ong certain
an
thropo
m
e
tric
in
d
i
cators.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-88
06
I
J
PH
S Vo
l. 4
,
N
o
. 2
,
Jun
e
201
5
:
9
4
–
101
96
Fi
gu
res
3
.
0
-
3
.
2 sh
o
w
t
h
e t
h
i
r
d
st
ep i
n
ou
r
dat
a
anal
y
s
i
s
, whi
c
h w
a
s
t
o
obt
ai
n a
nd a
n
a
l
y
ze si
de-
b
y
-
si
de n
o
rm
al
w
e
i
g
h
t
an
d ob
e
s
i
t
y
densi
t
y
d
u
al
-
g
en
de
r
pl
ot
s b
a
sed
on a
n
t
h
ro
p
o
m
e
t
r
i
c
m
easures
o
f
w
a
i
s
t
,
tricep
s
skin-fold, and
calf s
k
in-fold.
The m
easures
w
e
re
co
m
p
are
d
with t
h
e nat
i
onal m
easures
from
NHANESI
I
I
.
Afterw
ards,
w
e
ob
tain
ed z-sco
r
es to d
e
term
in
e wh
ic
h
m
easu
r
e
m
en
ts fro
m
ou
r Nativ
e
Am
erican
adol
esce
n
t
s
al
i
gne
d
cl
o
s
est
w
i
t
h
NH
A
N
ESI
I
I
dat
a
.
R
ecal
l
t
h
e
pu
rp
ose
fo
r
fi
g
u
res
2.
0
-
2
.
1 w
h
e
r
eby
t
h
e
y
gave
candi
d hints a
b
out the underly
ing distri
butions for each a
n
thropom
etric
measur
e in
relation to a standard bell
curve. Li
kewi
se, Figures
3.0-3.2
provi
d
e
suggestions
ab
out
each unde
rlying dist
ribution within
each
an
thropo
m
e
tric
m
easu
r
e for
male an
d
fem
a
le Nativ
e
American
s in
relatio
n
to
normal curve ass
u
m
p
tions.
Ho
we
ver
,
i
n
st
ead o
f
bei
n
g
p
r
i
m
aril
y
concerne
d wi
t
h
re
latio
n
s
h
i
p
s
to
a relativ
e lik
en
ess with
stand
a
rd
b
e
ll
cur
v
es
we ad
d
e
d an
othe
r feat
ure that
visuall
y
com
p
ares an
d g
r
ap
hs the
m
ean deri
ved
fr
om
NHAN
E
S
II
I in
each
figure. For e
x
am
ple, notice the
verti
cal blue a
n
d
green li
nes in figure
3.0, the
s
e lines re
prese
n
t the
weigh
t
ed
m
e
a
n
ob
tain
ed
from o
u
r
calcu
latio
n
s
with
NHANESIII
4
fo
r
male an
d
fe
male Nativ
e American
s
resp
ectiv
ely.
W
i
t
h
th
is in
fo
rmatio
n
we p
r
ov
id
ed
graph
s
to
p
r
ov
id
e so
me v
i
su
al in
fo
rm
atio
n
th
at sug
g
ests h
o
w
close or fa
r off our
weighte
d
m
eans were
to m
eans
found i
n
each
of
our ant
h
ropom
etric indicators
.
For
i
n
st
ance,
wi
t
h
fi
g
u
re
3.
0
n
o
t
i
ce t
h
e
pr
o
x
i
m
it
y
of
ou
r
bl
ue
vert
i
cal
m
ean (wei
ght
e
d
m
ean e
qual
s
8
3
.
3
3 a
n
d
po
ol
ed S
.
D
.
o
f
3.
62
fo
r M
a
l
e
s) t
o
t
h
e m
e
an an
d S.
D. for Native Am
erican m
a
les wh
en
m
easu
r
in
g
waist
cir
c
u
m
f
e
r
e
n
c
e (
7
3
.
30
an
d S.D
.
of
1
7
.72)
. Th
e m
ean
s
by them
selves m
a
y sugg
est
a dis
p
arity between
NH
A
N
ES
II
I a
nd
o
u
r
Nat
i
v
e
Am
eri
can po
pul
at
i
o
ns,
h
o
w
ev
er
,
w
ith
out a clo
s
er
look
at th
e un
der
l
yin
g
di
st
ri
b
u
t
i
on
we
are m
i
ssi
ng ha
l
f
of t
h
e pi
ct
u
r
e. He
nce, t
h
e
d
e
nsi
t
y
m
a
ps sh
ow
n i
n
fi
g
u
res
3.
0-
3.
2 (e
.g
.,
n
o
t
i
ce
th
at f
i
gu
r
e
s 2.0 and
2.1 ar
e i
n
clu
d
e
d
as
p
a
r
t
o
f
f
i
gu
r
e
s 3.0-3
.
2
in
ou
r m
a
l
e
categ
or
ies)
il
lu
str
a
te how
cl
o
s
e
o
r
rem
o
tely different each distri
bution is
from
a standa
rd
bell curve. T
h
us, a
r
m
e
d with these dual-ge
nde
r
plot
s
w
e
can
easily d
i
stin
gu
ish
wh
ich
an
thr
opometr
i
c
m
easu
r
e is
m
u
ch
m
o
r
e
lik
ely to
be ak
in
t
o
a no
r
m
al
NH
A
N
ES
II
I d
i
stributio
n. T
h
eref
ore
,
we
ca
n see t
h
at wai
s
t circum
ference once agai
n fails to a
ppe
ar as
no
rm
al
ly
di
st
ri
but
e
d
o
r
near
(
i
n t
e
rm
s of t
h
e
m
ean) t
o
NH
ANE
SI
II a
ssu
m
p
ti
ons a
nd
p
a
ves t
h
e
way
f
o
r
us t
o
search
fo
r ot
he
r ant
h
ro
p
o
m
e
t
r
i
c
i
ndi
cat
ors t
h
at
coul
d
be
mu
ch
m
o
re effectiv
e p
r
ed
icto
rs for weigh
t
class in
ou
r Nat
i
v
e Am
eri
can p
o
p
u
l
a
t
i
o
n
(see
fi
gu
res 3.
1-
3.
2)
.
Up
o
n
f
u
rt
her a
n
al
y
z
i
ng t
h
e d
e
nsi
t
y
pl
ot
s,
w
e
fo
u
nd t
h
at
t
h
e wai
s
t
m
ean
appea
r
s t
o
be
f
a
r f
r
om
t
h
e
sha
p
e of a
normal bell- curve; howe
ver
,
t
h
e t
r
i
ceps s
k
i
n
-f
ol
d a
n
d cal
f
sk
i
n
-f
ol
d
m
eans appea
r
e
d
t
o
ap
pr
oa
c
h
th
e fam
i
liar n
o
rm
al
d
i
strib
u
tio
n. Th
e
waist mean
was wi
dely skewed, and served as
th
e p
o
o
r
est ind
i
cato
r for
p
r
ed
ictin
g
no
rmal o
r
ob
ese
weigh
t
fo
r m
a
le Nativ
e Am
erican
ad
o
l
escen
t
s. Th
is im
p
lies th
at cu
rren
t BMI
st
anda
rd
s m
a
y not
be ap
p
r
o
p
r
i
a
t
e
for asses
s
i
ng n
o
r
m
a
l wei
ght
an
d o
b
esi
t
y
i
n
Nat
i
v
e A
m
eri
can adol
es
cent
s
.
Furt
herm
ore,
B
M
I cut
o
f
f
s a
m
ong
ot
her
m
i
no
ri
t
y
po
p
u
l
a
t
i
ons
ha
ve
bee
n
s
h
ow
n t
o
c
r
eat
e sy
st
em
at
ic er
ro
r
wh
en
assessing
g
e
statio
n
a
l d
i
ab
etes
m
i
l
litu
s
(GDM) [1
0
]
. In
fact
,
in
on
e 20
14
st
u
d
y
h
e
aded
b
y
M. Hedderson
h
e
r team
state
d
: “th
e
re is
on
go
ing
d
e
b
a
te su
rroun
d
i
n
g
t
h
e
d
e
fin
ition
o
f
o
v
e
rweigh
t
an
d
ob
esity
i
n
Asian
po
p
u
l
a
t
i
ons:
t
h
e
Worl
d
Heal
t
h
Or
ga
ni
zat
i
o
n
p
r
o
p
o
sed
a B
M
I cut
o
f
f
of
23
.0
k
g
/
m
2
for
o
v
er
wei
g
ht
a
m
ong
Asian
s
i
n
2
000
, co
m
p
ared
with
a cu
to
ff o
f
25
.0
kg
/
m
2
for
no
n
-
A
s
i
a
n p
o
pul
at
i
o
ns [
1
0]
.” T
h
at
i
s
, t
h
e
dem
ogra
phi
c
d
i
st
ri
but
i
o
n
of
o
b
esi
t
y
(hi
ghe
st
am
ong
Af
ri
can
Am
eri
cans and
l
o
west
am
ong
Asi
a
ns)
d
o
e
s
n
o
t
m
i
rror t
h
e
dem
o
g
r
a
phi
c di
st
ri
but
i
o
n o
f
G
D
M
(l
owest
am
on
g A
fri
ca
n A
m
eri
cans and
hi
g
h
est
am
ong
Asi
a
ns)
[10
]
. Thu
s
, racial d
i
sp
arities ex
ist
wh
en
assessin
g
risk
o
f
GDM u
s
ing
stan
d
a
rd
BMI p
r
o
t
o
c
o
l
s
th
at d
o
no
t
account
for
ra
cial diffe
rence
s
. More
ove
r,
without t
h
e
B
M
I m
easure cl
inicians a
r
e
unable
to m
a
ke sound
p
r
ed
ictio
n
s
abo
u
t
wh
o is at risk
fo
r
GDM with
ou
t ta
kin
g
racial/eth
nic d
i
sp
arities in
to
con
s
id
eratio
n.
In
ad
d
ition
,
in
a
2
012
stud
y con
d
u
c
ted
on
66 African
Am
erican
, 72
Hisp
an
ic, and
47
wh
ite
m
e
n
an
d
wo
m
e
n
,
aged
≥
4
5
J
o
a
n
C
a
rr
ol
l
and
her t
eam
di
scove
re
d t
h
at
M
i
ddl
e-a
g
e
d
an
d
ol
der
Af
ri
can
-Am
e
ri
can m
e
n an
d
wo
m
e
n
h
a
d
lower VAT (v
isceral ad
ipo
s
e tissu
e)
d
e
sp
ite similar BMI an
d
waist circu
m
f
e
ren
ce m
easu
r
e
m
en
ts
[1
1]
. T
h
at
i
s
,
al
t
e
red rel
a
t
i
o
nshi
ps
bet
w
ee
n ant
h
r
o
pom
etri
c m
easures and
V
A
T m
a
y
have i
m
pl
i
cat
ions
f
o
r
defi
ni
n
g
m
e
t
a
bol
i
c
ri
sk
i
n
di
f
f
e
rent
p
o
p
u
l
a
t
i
o
ns
base
d
on
ra
ce [1
1]
. T
h
e
r
ef
ore
,
di
ffe
rent
wai
s
t
ci
rcum
ference
or B
M
I c
u
t
o
ff
poi
nt
s m
a
y
be necessa
ry
t
o
adeq
uat
e
l
y
refl
e
c
t
VAT ri
s
k
i
n
di
ffe
rent
ra
ci
al
/
e
t
hni
c gr
o
ups
[1
1]
.
Thu
s
,
d
i
f
f
e
r
e
nces in
ou
r
N
a
tiv
e A
m
er
ican
po
pu
latio
n
app
e
ar
to
appr
o
a
ch
th
is sam
e
tr
en
d an
d
an
thr
opometr
i
c
measures s
u
c
h
as waist ci
rcum
ference fail to m
eet st
andar
d
s fo
r pre
d
i
c
t
i
n
g
o
b
ese o
r
ov
erwei
ght
cat
e
g
ori
e
s
.
Conse
q
uently, if such m
easures continue unabated the
n
we
will inevitably face skew
e
d
results that will
create
great
er sc
hi
sm
s whe
n
i
n
t
e
r
p
r
e
t
i
ng o
b
ese o
r
ove
rwei
ght
d
a
t
a
am
ong m
i
no
ri
t
y
po
pul
at
i
ons
. R
e
sul
t
a
n
t
l
y
, t
h
e
calf skin-fold
mean whe
n
an
alyzed for
our
population appeared to
be
a m
u
ch bet
t
e
r i
ndi
cat
or f
o
r pre
d
i
c
t
i
n
g
norm
al weight categories for m
a
le
Native Am
erican adolescents. More
ove
r, t
h
e t
r
i
c
eps s
k
i
n
-f
ol
d
m
ean
ap
p
e
ared
to
be a
b
e
tter ind
i
cato
r
fo
r assessm
en
t in
pred
ictin
g ob
ese
weigh
t
categories fo
r m
a
le
Nativ
e
Am
erican adolescents.
In
o
u
r m
a
l
e
po
pul
at
i
o
n
35%
of t
h
e s
u
b
j
ect
s
were cl
assi
fi
ed as
obe
se (
3
5
out
of
9
9
m
a
les),
whi
l
e
i
n
ou
r
fem
a
l
e
pop
ul
at
i
on
4
1
%
of
t
h
e s
u
b
j
ect
s
w
e
re cl
assi
fi
ed
a
s
o
b
ese
(3
5
o
u
t
of
8
4
fem
a
l
e
s).
W
e
use
d
o
u
r
t
i
m
e
-
seri
es pl
ot
i
n
F
i
gu
re 3 t
o
i
d
e
n
t
i
f
y
cases of o
v
er
wei
g
ht
an
d
obe
si
t
y
t
r
endi
n
g
an
d cl
im
axi
n
g.
We f
o
u
n
d
t
h
at
by
the age
of
14
bot
h m
a
les and
fem
a
les were already clas
s
i
fied
as overweight with
a
BMI greater t
h
a
n
25.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Assessing He
alth
Risk of
Obes
ity in Native A
m
erican A
d
oles
cents (Benj
a
min D. K
n
isley)
97
Ho
we
ver
,
by
t
h
e a
g
e
of
18
,
si
gni
fi
ca
nt
se
g
m
ent
s
of
bot
h
p
o
p
u
l
a
t
i
o
ns
h
a
d m
ove
d i
n
t
o
t
h
e
o
b
ese cat
ego
r
y
.
Waist, triceps
skin-fol
d, and calf skin-fol
d m
easures
were analyzed for both
ge
nde
r
s by weight class.
Afterward
s
, we
rep
o
rted
t
h
e Sh
ap
iro
-
W
i
l
k
s test
resu
lts
with thei
r re
specti
v
e
p-values
(listed
in
Tab
l
es 1-3).
Figure 1. Time-series
plot-of-means
fo
r m
a
l
e
a
nd fe
ma
le
Na
ti
v
e
Ameri
c
a
n
ad
ol
esce
n
t
s (a
ges
14
-1
8)
Figure
2.
No
r
m
al w
e
igh
t
class densit
y plots to
visualiz
e
normalit
y for male Na
tiv
e
American
adolescen
ts
(ages 14
-1
8)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-88
06
I
J
PH
S Vo
l. 4
,
N
o
. 2
,
Jun
e
201
5
:
9
4
–
101
98
Figure
3.
Ob
e
s
e
w
e
igh
t
class densit
y plots
to
visual
ize
n
o
rmalit
y for male
Nat
i
v
e
American
adolescen
ts
(ages 14
-1
8)
Figure
4.
W
e
i
g
h
t
cl
ass sp
ecifi
c d
u
a
l
-
g
e
n
d
e
r
de
ns
it
y
plots visualizing normalit
y and
mean
pro
x
imit
y to
NHANESI
I
I
data
with w
a
is
t mean for
male and femal
e
Nativ
e Ameri
can adolescen
ts
(ages 14-
18
)
Figure 5. W
e
igh
t
class
sp
ecific dua
l-gender densit
y
plo
t
s visualizing
normalit
y and
mean pro
x
im
it
y
to
NHANESI
I
I
data wi
th
ca
lf
skinfold
mean
for b
o
th male and female
Nativ
e Americ
a
n
adolescen
ts
(ages
14-
18)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Assessing He
alth
Risk of
Obes
ity in Native A
m
erican A
d
oles
cents (Benj
a
min D. K
n
isley)
99
Figure
6.
W
e
igh
t
-class sp
ecific d
u
a
l
-
g
e
n
d
e
r
dens
it
y
plots visualizing no
rmalit
y and
mean
pro
x
imit
y to NHANESI
I
I
data
with tric
e
p
s skin-fold mean
for
male a
nd female Nativ
e Americ
an
a
dolescen
ts (ages 14
-1
8)
Figure 7.
Total population w
e
igh
t
-class
pi
e c
h
art
visualizing w
e
ig
h
t
dynamics
3.1
.
Male Results
Our most important results
for th
e male
c
a
tegories and
this pape
r are
lis
ted below
in Table
1.
For
nor
mal weight, we saw that waist
was a very p
o
o
r
w
e
ight indicator
(p-
v
alue < 0.
001
). Calf s
k
in-fol
d
appeared to b
e
a somewhat bette
r predictor of n
o
rmal weight th
an triceps skin-fold, although
bot
h
measures wer
e
strong
(
p
-va
l
ue 0f
0.07
8
and 0.
06
6,
re
spectively). For
obese
weight, triceps skin-fol
d
appeared to be a better wei
ght indicator
than calf
skin-fold (
p
-
v
alue of
0.69 a
nd
0.57, res
p
ectively).
However, we
found that waist was
st
il
l not a rel
i
able
predictor for
o
b
esity and ov
erweight (p-v
alue <
0.001). Due to the small sa
mple of female subjects
in our stud
y p
o
p
u
l
ation, we
were unable to separate
them reasonably into unique
weight
categories like norm
a
l and obese.
Fi
gu
re 2
.
0
-
2.
1
sho
w
s
t
h
at
t
h
e cal
f
ski
n
-f
ol
d
an
d
t
r
i
ceps
-
skin
fol
d
m
eas
ures a
p
p
e
a
r
to b
e
the m
o
st
co
nsisten
t
fo
r
pred
icting
wei
g
h
t
classes for
male
Nativ
e American
adolescen
ts, a
n
d
tha
t
the
w
a
ist see
m
s
to
b
e
th
e least
co
nsisten
t
w
e
ig
h
t
cl
ass
pre
d
i
c
t
o
r. Fi
gu
res
3.
0-
3.
2 c
o
n
f
i
r
m
s
t
h
ese
cal
cul
a
t
i
ons
b
y
f
u
rt
her
pr
o
v
i
d
i
n
g
vi
su
al
s t
h
at
com
b
ined weighte
d
means
with
their res
p
ective
densit
y
plots. In effect,
w
e
us
ed
our
NHANESI
I
I
data in
o
r
d
e
r to d
i
sco
v
er h
o
w
at
y
p
ical o
u
r an
th
ro
p
o
m
e
tric
Nativ
e Am
erican adolescen
t
factors
w
e
re
wh
en
com
p
ared
to
th
ese n
a
tio
n
a
l
statistics (see T
a
b
l
e
2
)
.
Afterwa
r
ds
, we com
p
ared
our
da
ta with
NHANESIII su
rv
eys and
di
scovere
d
three
m
easures
of
statistical inter
e
st for
norm
al
and obese
wei
ght m
a
le
Native Am
erican Adolesce
nts:
waist, tricep
s sk
i
n
-fol
d
and calf
ski
n
-fold.
The
n
, w
e
gathe
r
ed
m
eas
ures
of ce
n
t
er
to
ind
i
cate
whic
h
an
thro
p
o
metric
i
ndi
cat
ors
w
e
r
e
b
e
st-su
ited
to
measu
r
ing
ob
esit
y in
m
a
le Nativ
e Am
erica
n
ado
l
escen
ts,
an
d
t
o
id
en
tify
whi
c
h d
e
vi
at
ed
m
o
st
significa
n
tly from
accepted
norm
s (see
Figure
3.0-3.
2).
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-88
06
I
J
PH
S Vo
l. 4
,
N
o
. 2
,
Jun
e
201
5
:
9
4
–
101
10
0
Tab
l
e
1
.
D
i
scov
er
i
n
g no
r
m
ali
t
y f
o
r an
t
h
rop
o
me
tric
m
easu
r
es with Sh
ap
iro-W
i
l
k
s an
alysis
Anthropo
m
e
tr
ics
sorted by w
e
ig
ht class
Shapiro-Wil
k
s
p-value
Nor
m
al
weight cla
ss
W
a
ist m
ean
0.
8286
<
.
001
T
r
iceps skin-
f
old
0.
9313
0.
0664
*
Calf skin-
f
ol
d
0.
9341
0.
0782
*
Obese weight class
W
a
ist m
ean
0.
8257
<
.
001
T
r
iceps skin-
f
old
0.
9741
0.
693*
Calf skin-
f
ol
d
0.
9696
0.
570*
Table
2. Male
Ant
h
ropom
etrics com
p
ared to NHASES
Anthropo
m
e
tr
ics
N
Mean
S.D.
df
T
p-value
W
a
ist m
ean
73
76.
08
27.
88
72
-
2
.
22
<
.
05
T
r
iceps skin-
f
old
70
19.
74
7.
18
69
7.
79
< .
001
Calf skin-
f
ol
d
70
30.
90
5.
59
69
-
9
.
58
< .
001
Tab
l
e 3
.
Male N
HAN
ESI
I
I
data,
ag
es
14
-18
Anthropo
m
e
tr
ics
N
Weighted Mean
Pooled S.D
.
W
a
ist m
ean
821
83.
33
3.
62
T
r
iceps skin-
f
old
811
13.
05
0.
53
Calf skin-
f
ol
d
1373
37.
30
0.
92
Our an
alysis so
ugh
t to
an
swer th
is
qu
estion
:
wh
at
is t
h
e
b
e
st an
th
rop
o
metric in
d
i
cator to
m
easu
r
e
o
b
e
sity with
in ou
r sam
p
le po
pu
latio
n
?
Fo
r m
a
le Nativ
e Am
erican adolescents,
our
analysis was two- fol
d
.
Our fi
rst step
analyzed the
norm
al we
i
ght
c
l
ass. Fi
rst
,
we
had t
o
c
h
o
o
se
bet
w
ee
n t
h
e t
r
i
ceps s
k
i
n
-f
ol
d
m
ean
and t
h
e cal
f s
k
i
n
-
f
ol
d m
ean (p-val
ues o
f
0.
0
6
6
an
d 0.
0
7
8
,
resp
ectiv
ely). Al
th
ou
gh
it ap
p
e
ared
m
o
re lik
ely th
at
t
h
e cal
f
ski
n
-
f
o
l
d
m
ean w
a
s
deri
ved
f
r
om
a n
o
rm
al
po
pul
at
i
on,
o
u
r
z-sc
ores
t
o
l
d
u
s
a
n
ot
he
r st
o
r
y
.
B
e
t
w
een
these m
easures we ha
d a z-va
lue of
-1.03
for the calf s
k
i
n
-
f
ol
d m
ean vers
us a z-
val
u
e
of
0.
60
fo
r t
h
e t
r
i
ceps
ski
n
-f
ol
d m
ean based
up
o
n
N
H
A
N
ES
II
I.
Du
e t
o
t
h
ese resu
l
t
s, we close to
favor the calf s
k
in-fold m
easure as
t
h
e best
assess
or
of
n
o
rm
al
wei
ght
f
o
r m
a
l
e
Nat
i
v
e Am
erican adolesce
nts, and t
h
e triceps skin-fold m
easure
as the
best asse
ssor of
obese
weight
for m
a
le Native
Am
erican adolesce
nts.
4.
CO
NCL
USI
O
N
In
s
u
m
m
ary
,
t
h
i
s
st
udy
w
a
s
de
si
gne
d t
o
ex
pl
o
r
e w
h
i
c
h a
n
t
h
ro
p
o
m
e
t
r
i
c
i
ndi
cat
ors
b
e
st
p
r
e
d
i
c
t
e
d o
b
esi
t
y
in m
a
le Nativ
e Am
erican adolesce
n
t
s. Under the
cond
itio
n
s
of
ou
r
stu
d
y
, w
e
fou
nd th
e calf
sk
in
-
fol
d
measure to b
e
the b
e
st indica
tor of norm
al
w
e
ig
h
t
f
o
r
ou
r
p
o
pul
at
i
o
n, a
nd t
h
e
t
r
i
c
e
p
s ski
n
-f
ol
d
m
ean to b
e
th
e b
e
st i
n
d
i
cato
r
of
o
b
esit
y fo
r
ou
r
p
o
p
u
latio
n
.
In
add
itio
n
to
t
h
is
p
ilo
t stu
d
y
, mo
re
r
e
sear
c
h
exp
l
or
ing
th
e reliab
ilit
y o
f
NHANESI
I
I
d
a
ta to
assess
o
b
esit
y in
Nati
v
e
Am
erican
ad
o
l
escen
t
s
is
reco
mmend
ed
i
n
ord
e
r
to
co
nfirm
o
u
r d
i
sco
v
ery
.
ACKNOWLE
DGE
M
ENTS
Special thanks to Mr.
Rol
a
nd D
o
epner,
Sher
man I
n
dian High
School; Han
n
ah
Te Stipek,
Washingto
n
State University; and Dr. Wayne Flet
cher, C
a
lifornia B
a
ptist
University; for their
collaboration and sup
p
o
r
t.
REFERE
NC
ES
[1]
W
o
rld Health
Organiz
a
tion
,
“
D
iet,
n
u
trit
ion and
the pr
ev
en
tion
of c
h
roni
c dis
eas
es
.
R
e
p
o
rt
o
f
a
W
HO
S
t
u
d
y
Group, Genev
a
,”
WHO T
e
chnic
a
l
R
e
p
o
r
t
Series,
no
.
797
-TRS
797,
1990.
[2]
Ekonoya
n, G
.
,
“A
history of o
b
esity, or how
w
h
at w
a
s good became ugly an
d then bad”,
Adv Ch
ron
i
c Kidney
Dis
, vol/issue: 1
3
(4), pp. 421-4
27, 2006.
[3
]
F
r
y
a
r CD.,
Gu
Q.,
Ogden C
L
.,
“An
t
hrop
omet
ric ref
e
rence data
for
c
h
ildren
an
d
adults:
United States,
2007
20
10.
Nationa
l C
e
n
t
er
for H
eal
th St
at
istics”
,
Vita
l
He
alth
S
t
at
.,
vol/issue: 11(
252)
, 20
12.
[4]
Brancati FL.,
K
a
o WH
., Folsom A
R
.,
Watson RL., Szkl
o M., “Incident t
ype 2 dia
b
etes
mellitus A
fric
a
n
A
m
e
r
ican and
w
h
ite adults:
th
e A
t
herosclerosis Risk in Communities Study”
,
JAMA
, vol. 283, pp. 2253-225
9,
2000.
[5]
H
e
dley A
A
.
, Ogden CL., John
son CL., et al.,
“P
revalence
of overw
eight
an
d obesi
ty amon
g U
S
children,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Assessing
Hea
lth
Risk
o
f
Ob
esity in
Na
tive American
Ado
l
escen
ts (Ben
j
a
min
D. K
n
isley)
10
1
adolescen
ts, and adults, 1999-2
002”,
JAMA
, vo
l. 291, pp. 2847-2850, 2004.
[6]
Ogden CL., Car
r
oll MD., Kit BK., Flegal
KM., “Prevalence of childhood and a
dult obesity
in the United States
,
2011-2012”,
Jou
r
nal of
the Amer
ican M
e
dica
l
Association
, vol/issue: 311(8)
, pp
. 8
06-814, 2014
.
[7]
Mc
Do
w
e
l
l
M.
,
F
r
y
a
r
CD.,
Ogde
n CL
.,
F
l
e
g
a
l
KM.,
“
A
n
t
hro
p
ometric
reference d
a
ta
for
c
h
ildren and
ad
ults:
United
States, 20
03-2006. Nation
a
l C
e
n
t
er
for Health
Statistics,”
V
i
tal He
al
th
Stat.,
vol/issue:
10(14
5), 2008
.
[8]
Clark V
R
., G
r
e
e
n
b
erb
B.
, H
a
r
r
is
TS
.,
Ca
rson
BL.,
“
B
o
d
y
Mass Index
and w
a
ist
circumf
e
rence predictor
s
of
cardio
v
as
cula
r
ri
s
k
in
Afric
a
n
A
m
ericans
”
,
Ethnic Dis.,
vol/issue: 22(2), pp. 1627, 2012.
[9]
R
Core T
eam,
“
R
: A
lan
guag
e
and
en
viron
m
en
t
for statist
i
cal c
o
mputin
g,”
R F
o
u
nda
tion
fo
r
S
t
a
tistical
Co
mpu
ting
, 2014.
[10]
H
e
dderson M.,
Ehrlich S., Srid
har S.,
D
a
rbinian J., M
oore S., Ferrara A., “
R
acial/Et
hnic
D
i
sparities in t
h
e
Prevalence of Gestationa
l Diabetes by BMI”,
Di
a
b
e
tes Care
, vo
l. 35, pp. 1492-8, 2012.
[11]
Zilanawala A., Davis-Kean P.,
Nazroo
J.,
Sacker A., Simonton S.,
Kelly Y
., “
R
ace/eth
n
ic
dis
p
arities in
early
childh
ood BMI, obesity and o
v
e
rw
eight in the
U
n
ited K
i
ngdo
m and U
n
ited States”,
In
terna
tio
na
l J
o
urn
a
l
o
f
Ob
e
s
ity
, vol. 39, pp. 520-529, 2014.
BIOGRAP
HI
ES OF
AUTH
ORS
Benjamin Knisley
is
a senio
r
completing his
B.S. in
Appl
ied St
atisti
ca
l
Anal
y
s
is
at
California Baptist University
. I
n
2014, he rece
ived the award
for excellen
ce in statistics
res
earch
from
th
e dep
a
rtm
e
nt o
f
natur
a
l
and m
a
t
h
em
atic
al s
c
ien
c
es
. He p
l
ans
to
begin his
M.S. in
Applied
Mathematics at
Californi
a B
a
ptist University
in
September, 2015
.
Grace Cros
b
y
is
a s
e
n
i
or
com
p
leting
her B
.
S
.
i
n
Heal
thca
re A
d
m
i
nis
t
ration
at
Cali
forni
a
Baptist Universi
t
y
. In 2014, she receiv
e
d th
e Scholarship of E
x
cel
lenc
e award
from
the
Association of S
c
hools of Allied
Health Prof
essions (ASAHP). She plans
to beg
i
n
her MPH
with an em
phasis in Food, Nutrition
and Hea
lth at Cal
i
forn
ia Bapt
ist Univ
ersit
y
in
September, 2015
.
Margaret Barth
,
PhD, MPH, MB
A, is Program
Director of Nutr
ition and Food Sciences
an
d
Professor of Nu
trition
and Food Sciences at Ca
lifornia Bap
tist
University
. Prev
iously
, she
was Associate
Professor of Nutrition
and
Food Scien
ce
at
th
e Universi
t
y
of
Kentuck
y
,
Lexington
. Dr.
Barth received
her PhD in
Food Science and
Hum
a
n Nutritio
n from
the
University
of
Illinois.
Linn Caroth
ers,
PhD, is Program Director of Ap
p
lied Statistics an
d Phy
s
ics and Professor of
M
a
them
ati
c
s
at
Californi
a Bapt
i
s
t Univers
i
t
y
. Dr
. Caroth
ers
rece
i
v
ed his
P
h
D in Educat
ional
Psy
c
holog
y
and
Statistics from the Un
iversity
of
Southern Califor
nia.
Evaluation Warning : The document was created with Spire.PDF for Python.