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
.3
, Sep
t
em
b
e
r
20
15, pp
. 232
~237
I
S
SN
: 225
2-8
8
0
6
2
32
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
Mat
ern
al Mort
ali
t
y in Ghana:
Im
pact
of
t
h
e Fee-F
r
ee Deli
very
Policy and the Nation
al
Health I
n
surance Sch
e
m
e
Seidu S
o
fo
1
, Emmanuel Thompson
2
1
Departm
e
nt of Health
,
Hum
a
n P
e
rform
ance & Recre
a
tion
,
S
out
heas
t
M
i
s
s
ouri
S
t
at
e
Univers
i
t
y
,
US
A
2
Departm
e
nt
of
M
a
them
ati
c
s
,
S
o
utheas
t
M
i
s
s
ouri S
t
at
e Univ
ers
i
t
y
, US
A
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Aug 2, 2015
Rev
i
sed
Au
g
25
, 20
15
Accepted Aug 30, 2015
Maternal mortality
(MMR) is the second
larg
est cause of female deaths in
Ghana. Yet
,
m
a
n
y
households cannot a
fford th
e cost of skilled deliv
er
y
The
stud
y
ut
ili
zed
th
e Panel D
a
ta M
odel to
exam
ine
the im
pac
t
of t
h
e fee-
fre
e
deliv
er
y
(FEP)
and the Nation
a
l Health
Insuran
ce Policy
(NIP) exemptions
on MMR in Gh
ana.
The Demographic and Health Survey
repor
ts on Ghana
from 2002 to 2009 served as th
e
main data
source. Data wer
e
an
aly
zed using
Panel d
a
ta model with with
in g
r
oup fi
xed
eff
e
cts estimator
.MMR declined
signific
a
ntl
y
ov
e
r
the per
i
od stud
ied. Bo
th FEP a
nd NIP positivel
y im
pa
cted
MMR at a 5%
lev
e
l of
signif
i
can
ce.In add
ition, skilled d
e
liver
y
was
a
significant pr
edictor of MMR. Stakehold
e
rs would do well
to ens
u
re that NIP
is
adequ
a
te
l
y
fu
nded in
order
to
s
u
s
t
ain th
e d
ecl
i
n
e in
M
M
R
.
Keyword:
Gha
n
a
Matern
al m
o
rtality
Millen
n
i
u
m
d
e
v
e
lop
m
en
t g
o
a
l
Panel
dat
a
m
odel
Sk
illed
d
e
liv
ery
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
:
Seidu S
o
fo,
Depa
rt
m
e
nt
of
Heal
t
h
,
H
u
m
a
n Per
f
o
r
m
a
nce & R
ecreat
i
o
n,
Southeast Miss
ouri State
Uni
v
ersity,
1
Uni
v
ersity Pl
aza, Cape
Gi
ra
rdea
u,
MO 63701, USA.
Em
a
il: sso
fo@sem
o
.
ed
u
1.
INTRODUCTION
Desp
ite th
e
d
e
clin
e in
m
a
ter
n
al m
o
rtalit
y ratio
(MMR) glo
b
a
lly, it is li
k
e
ly th
at Gh
an
a (as o
t
h
e
r
African
cou
n
t
ries) will no
t attain
th
e Millen
n
i
u
m
Dev
e
lop
m
en
t Go
al
5 b
y
th
e year
2
015
[1
]. Mat
e
rn
al
m
o
rtali
t
y,
wh
i
c
h
accou
n
t
s for 1
4
%
o
f
all female
d
eath
s
,
is still th
e second
larg
est cau
s
e of
fem
a
le d
eath
s
in
Gh
an
a
[2
].
A
maj
o
r
factor in
redu
cing
MMR is th
e
av
ailab
ility o
f
sk
illed
d
e
liv
ery
[3
]. Fo
r ex
am
p
l
e, [4
]
reporte
d
skilled delive
r
y as the fact
or that m
o
st predicted MMR
in Gh
a
n
a. Howeve
r,
m
a
ny hous
eholds in
su
b-Sah
a
ran Africa can
n
o
t
affo
rd
t
h
e co
st of
sk
illed
d
e
liv
ery, esp
ecially emerg
e
n
c
y
ob
stetric care
[5
],[6
].
To
increase accessib
ility to
sk
illed
d
e
livery
,
Gh
an
a in
trodu
ced a fee-free d
e
liv
ery po
licy (FEP) in
fo
ur
regi
ons
of
t
h
e cou
n
t
r
y
i
n
20
03:
N
o
rt
he
r
n
, U
p
per East
,
Up
per
West
,
and C
e
nt
ral
re
gi
o
n
s [
7
]
.
The
pol
i
c
y
p
r
ov
id
ed
free
d
e
liv
ery serv
ices at all p
u
b
lic an
d
p
r
i
v
at
e h
e
alth
facilities.
Th
e fee ex
em
p
tio
n
was ex
tend
ed
t
o
t
h
e rest
of t
h
e
cou
n
t
r
y
i
n
2
0
0
5
.
The e
x
em
pt
i
on c
ove
re
d al
l
norm
a
l
del
i
veri
es, m
a
nage
m
e
nt
of al
l
assi
st
ed
del
i
v
eri
e
s,
a
n
d
m
a
nagem
e
nt
o
f
m
e
di
cal
and
s
u
r
g
i
cal
com
p
l
i
cat
i
ons a
r
i
s
i
n
g
out
o
f
del
i
v
eri
e
s [
7
]
.
Th
e
fee ex
em
p
tio
n
p
o
licy resu
lted
i
n
in
crease in
f
acility d
e
liv
ery cov
e
rag
e
[8
]-[10
] th
e in
crease
was
greatest am
ong the poorest. On a
positive note, the
quality
of care did not deteriorate due to inc
r
ease
d
u
tilizatio
n
witho
u
t
a correspon
d
i
n
g
in
crease
in
resou
r
ces
[11
]
. Seco
nd
, th
ere was a si
g
n
i
fican
t drop
in ou
t-o
f
-
poc
ket
paym
en
ts for caesa
r
ea
n section a
n
d norm
al deliver
y after i
n
troduction
of the
policy, especially am
ong
t
h
e po
o
r
est
[1
2]
. Thi
r
d, del
i
very
-related
MMR in two regions dec
r
eased, e
v
en t
h
ough thes
e were not
statistical
ly si
g
n
i
fican
t [8
].
Fin
a
lly, th
e po
licy resu
lted
in
a d
eclin
e i
n
d
i
rect m
a
tern
al d
eat
h
s
bu
t
n
o
t
o
n
i
ndi
rect
m
a
t
e
rn
al
deat
hs
[
13]
.
A m
a
j
o
r ch
allen
g
e
for th
e fee-free d
e
liv
ery p
o
licy wa
s th
at d
o
c
tors were n
o
t
well prepared
to
d
eal
with
th
e
o
b
stetric co
nd
itio
ns co
mm
o
n
l
y s
een
in
d
i
strict h
o
sp
itals [1
0]. In
ad
d
ition
,
so
m
e
reg
i
o
n
s had
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Ma
terna
l
Mo
rt
a
lity in
Gha
na: Impa
ct
o
f
th
e
Fee-Free
Delivery Po
licy
an
d
th
e Na
tiona
l .... (S
eidu
So
fo
)
23
3
ex
h
a
u
s
ted
fun
d
s b
y
late 200
5
[5
]. Th
e lack
of fun
d
s
re
su
lted
in a declin
e i
n
facility d
e
liveries b
e
t
w
een
2
006
and
2
0
07
[
12]
.
To a
d
dress
t
h
i
s
set
b
ac
k, t
h
e
go
ve
rnm
e
nt
i
n
2
0
0
8
e
x
em
pt
ed
pre
gna
nt
wo
m
e
n fr
om
pay
i
ng
t
h
e
National
Health In
su
ranc
e Schem
e
(NH
I
S) re
gistra
tion an
d
p
r
em
iu
m f
ees. Th
e
2
008
po
licy pr
ov
id
ed
pre
g
nant
w
o
m
e
n
wi
t
h
si
x a
n
t
e
nat
a
l
an
d t
w
o
p
o
st
nat
a
l
vi
si
t
s
, chi
l
dcare a
n
d
ot
he
r
pri
m
ary
heal
t
h
care
b
e
nefi
t
s
[8
]. Th
ere is ev
id
en
ce to
su
pp
ort th
e po
sitiv
e im
p
act o
f
h
ealth
in
suran
c
e o
n
m
a
tern
al h
ealth
[14
]
. Howev
e
r,
litt
le is k
nown
ab
ou
t th
e im
p
act o
f
bo
th th
e
fee-free
d
e
liv
ery an
d h
ealth insu
ran
ce
fee exe
m
p
tio
n
p
o
lici
e
s on
MMR n
a
tio
nwid
e.
To
d
a
te, stud
i
e
s on
th
e
FEP and
NIP policies fo
cu
sed o
n
h
ealth
care u
tilizatio
n
and
eco
n
o
m
ic
outc
o
m
e
s, and were c
o
nducted in a fe
w se
lected regi
ons
of t
h
e c
o
unt
ry
.
Furt
herm
ore,
t
h
e few st
udi
e
s
t
h
at
exam
ined the effect of th
ese
pol
i
c
i
e
s on
M
M
R
report
e
d m
i
xed res
u
l
t
s
[8]
,
[
1
3]
. The
r
ef
ore
,
t
h
e ai
m
of t
h
e
current st
udy
was to e
x
am
in
e the im
pact of the fee-free
de
livery and th
e
NIP e
x
em
ptions on MMR in
Gha
n
a.
Th
e sp
ecific
ob
j
ectives
were
to
ex
am
in
e: (a) if th
ere was si
g
n
i
fican
t
d
eclin
e in
MMR
post p
o
licies an
d
(b) th
e
sig
n
i
fican
t
p
r
ed
icto
rs of MM
R after im
p
l
e
m
en
tatio
n
of th
e
p
o
licies.
2.
R
E
SEARC
H M
ETHOD
The
Dem
ogra
phi
c
an
d
Heal
t
h
S
u
rvey
dat
a
fo
r
Gha
n
a
f
r
o
m
200
2 t
o
2
0
0
9
[
1
5]
ser
v
e
d
as t
h
e m
a
i
n
data
sources. The varia
b
les use
d
from
the surveys
in
cl
u
d
ed
: MMR (resp
on
se
v
a
riab
le),
Do
ct
o
r
Po
pu
latio
n
Ratio
(DR), Nu
rse Pop
u
l
ation
Ratio
(NR
)
,
An
ten
a
tal Co
verag
e
(ANC),
Po
stnatal Co
v
e
rag
e
(PNC
), Sk
illed
Delivery (SD), and Fam
ily P
l
anning
Acceptors (FPA). M
M
R va
lues for the peri
ods
before and afte
r the
in
trodu
ctio
n of th
e FEP and
NIP
were co
m
p
ared to
d
e
term
in
e th
e im
p
act o
f
th
e
p
o
licies.
Data we
re a
n
aly
zed usi
n
g
descri
ptive a
n
d in
fere
n
tial statistics. Av
erag
es, stand
a
rd d
e
v
i
atio
n
s
,
coefficients of Ske
w
nes
s
a
n
d Kurtos
is, an
d
co
nfid
en
ce i
n
terv
als
were com
p
u
t
ed
fo
r t
h
e in
tro
d
u
c
tion
of FEP
(200
2-200
5), im
p
act o
f
FEP (2
006
-2
008
),
an
d im
p
act
o
f
NIP (2
009
). Th
e stud
y
u
tilized
t
h
e w
ith
in
group
esti
m
a
to
r to
analyze th
e i
m
p
act o
f
two
h
ealth
in
su
ran
ce
policies in
Gh
an
a. Th
ese po
licies in
clud
ed FEP and
NIP. Th
e FE esti
m
a
to
r is ap
prop
riate fo
r
p
a
n
e
l d
a
ta—
d
a
ta co
llected
on
th
e sam
e
su
b
j
ects o
v
e
r ti
m
e
. Pan
e
l
dat
a
are co
rrel
a
t
e
d wi
t
h
i
n
s
u
bject
s
but
i
n
de
pen
d
e
n
t
bet
w
e
e
n subjects. T
h
is is a charact
eristic th
at th
e
d
a
ta in
the curre
nt study exhibit,
he
nce the
ch
o
i
ce o
f
with
i
n
g
r
ou
p fix
e
d
effect
s esti
m
a
to
r. To
in
trod
u
c
e the with
in
gr
o
u
p
fi
xe
d e
f
f
ect
s est
i
m
a
t
o
r,
we
fi
rst
pr
ovi
d
e
a b
r
i
e
f
desc
ri
pt
i
o
n
o
f
pa
nel
dat
a
m
odel
.
2.
1.
Pa
nel Da
ta
Mo
del
A
Pa
nel
dat
a
m
odel
wi
t
h
o
n
e-
way
er
ro
r
c
o
m
p
o
n
e
n
t
s
di
st
ur
b
a
nces
i
s
gi
ve
n by
y
x
β
μ
ϑ
i
1
,
…
,
N
;
t
1
,
…
,
T
(1)
The
i
a
n
d t
de
not
e
t
h
e
c
r
os
s-
sect
i
o
n
an
d
t
i
m
e-seri
es
di
m
e
nsi
o
ns
res
p
ect
i
v
el
y
.
is
K1
an
d
x
is th
e
it
th
obser
vat
i
o
n
o
n
p
r
ed
icto
r variab
le.
μ
i
s
t
h
e u
n
o
b
se
rve
d
het
e
ro
ge
nei
t
y
acro
ss i
n
di
vi
d
u
al
s t
h
at
i
s
fi
xed
o
v
e
r
ti
m
e
.
ϑ
d
e
no
tes
th
e st
o
c
h
a
stic
term
wh
ich is ind
e
p
e
nd
en
t
,
id
en
tically d
i
stribu
ted (i.i.d
.)
with
m
ean
0 an
d
varia
n
ce
σ
.
2.
2.
Within
Gro
u
p
Fixed
Effec
t
s
Estima
tor
To i
m
pl
em
ent t
h
e wi
t
h
i
n
gr
ou
p est
i
m
at
or, we c
onsi
d
er
(1
) a
nd t
h
en
assum
e
t
h
at
Ε
X
μ
0
,
Ε
X
ϑ
0
, and
Ε
ϑ
ϑ
Σ
. Stack
i
ng
all th
e
ob
servatio
n
for
t1
,
…
,
T
as
p
r
esen
ted in
[16
]
we
ob
tain
y
T1
x
TM
β
M
1
μ
11
1
T1
ϑ
T1
(2
)
Let
V
TT
I
1
1
1
1
I
U
and
U
1
1
1
1
T
1
1
No
te t
h
e
fo
llowing
:
i.
U
1
1
ii.
V
1
0
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
. 3
,
Sep
t
emb
e
r
201
5
:
2
32
–
23
7
23
4
iii.
U
y
1
1
1
1
y
1
y
iv
.
V
y
I
U
y
y
1
y
y
Fro
m
th
e ab
ove, we ob
tain
ed
a tran
sfo
r
m
e
d
fo
rm
o
f
(2
):
V
y
V
X
β
μ
V
1
V
ϑ
(3
)
The
reg
r
essi
on
m
odel
st
acked
by
obs
er
vat
i
o
n
becom
e
s:
y
⋮
y
X
⋮
X
β
ϑ
⋮
ϑ
an
d th
e
with
i
n
g
r
ou
p fix
e
d
effects esti
m
a
to
r is ob
tain
ed
as
fo
llo
ws:
β
V
X
V
X
V
X
V
y
Ho
we
ver
,
V
i
s
i
d
em
pot
ent
,
he
nc
e
β
an
d t
h
e c
o
r
r
e
sp
on
di
n
g
vari
an
ce (Baltaji
(2005)) becom
e
β
X
V
X
X
V
y
and
va
r
β
σ
X
V
X
We di
rect
read
ers
t
o
[
1
6]
-[
18]
f
o
r det
a
i
l
s
o
n
panel
dat
a
m
o
d
e
l
s
.
3.
R
E
SU
LTS AN
D ANA
LY
SIS
3.
1.
Trend of MMR
Tabl
e 1 p
r
o
v
i
d
es t
h
e descri
pt
i
v
e st
at
i
s
t
i
c
s for M
M
R
by peri
od
. The ave
r
a
g
e M
M
R
rose from
220.
2
7
in
200
2-
200
5
t
o
202
.5
6 in
2
0
0
6
-2
008
and
l
a
ter
dr
opp
ed
t
o
171
in
20
09
. A
l
so, it is v
e
ry in
d
i
cativ
e fro
m
th
e
Tab
l
e
1
th
at, th
e
ratio
in th
e th
ree d
i
fferent p
e
ri
o
d
s
we
re no
t sign
ificantly sk
ewed
since the c
o
efficients
of
Ske
w
nes
s
we
r
e
nei
t
h
er sm
al
ler t
h
a
n
-
1
n
o
r
bi
g
g
er t
h
an +
1
. Si
m
i
l
a
rl
y
,
no
ne o
f
t
h
e
val
u
es of t
h
e k
u
r
t
o
si
s fo
r
th
e ratio in
t
h
e th
ree
p
e
riod
s
were less th
an
-3
or
b
i
gg
er t
h
an
+3
.
Hen
ce t
h
e ratio
for th
e th
ree
p
e
riod
s
were
n
o
t
sign
ifican
tl
y h
eav
y tailed
.
Th
e
95% confidence interval
s
for the avera
g
e MMR in
t
h
e po
pul
at
i
o
n i
n
20
02
-
2
005
,
20
06-
200
8, an
d 200
9
w
e
r
e
(1
81
.34
,
2
1
9
.
80
), (18
3
.1
0, 2
2
2
.
02
),
an
d
142
.7
6, 1
9
9
.
2
4
)
r
e
sp
ectively
as
sho
w
n i
n
Ta
bl
e1.
Tab
l
e 1
.
Su
mmar
y
Statistics
of
MMR
in
Gh
an
a f
o
r
2
002
-200
5, 2
006
-20
0
8
,
and
2
009
Variable
Per
i
od
2002-
2
005 (
I
n
tr
o.
FE
P )
2006-
2
008 (
I
m
p
act FE
P)
2009 (
I
m
p
act of NI
P)
Aver
age 200.
57
202.
56
171
Standar
d
Deviation
62.
05
54.
39
45.
55
Skewness 0.
59
0.
29
0.
82
Kur
t
osis 0.
43
-
0
.
37
-
0
.
73
95% CI
for
the M
e
an of M
M
R
(
181.
34,
219.
8
0
)
(
183.
10,
222.
0
2
)
(
142.
76,
199.
2
4
)
The
distrib
u
tio
ns
of M
M
R
fo
r the t
h
ree
pe
riods
are
sh
o
w
n
in Fig
u
re
1
.
M
M
R were m
o
re
sp
read
o
u
t
i
n
20
0
2
-
2
00
5,
fol
l
o
we
d by
2
0
0
6
-
2
0
0
8
,
an
d
t
h
en 2
0
09 as
i
ndi
cat
ed
by
t
h
e u
p
p
e
r w
h
i
s
kers
of t
h
e
b
o
xpl
ot
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Ma
terna
l
Mo
rt
a
lity in
Gha
na: Impa
ct
o
f
th
e
Fee-Free
Delivery Po
licy
an
d
th
e Na
tiona
l .... (S
eidu
So
fo
)
23
5
These confirm
that the st
andard
deviation for MMR decre
a
sed m
onotoni
cally from
2002-2005 to 2009 (see
Tabl
e 1
)
. T
h
e
gene
ral
t
r
e
nd i
n
t
h
e
bo
x
p
l
o
t
s
i
ndi
cat
ed t
h
at
M
M
R
decl
i
n
ed
du
ri
n
g
t
h
e
pe
ri
ods
u
n
d
er
ob
se
rvat
i
o
n
(see Figu
re
1
)
. It is i
m
p
o
r
tan
t
to
m
e
n
tio
n
th
atFig
u
re 1
shows po
ten
tial ou
tliers in
th
e MMR d
a
ta set
fo
r the
p
e
r
i
o
d
s
200
2-20
05
and
2
006
-2
008
.
Fig
u
r
e
1
.
D
i
str
i
b
u
tion
of
MM
R in
G
h
an
afor
2
002
-20
0
5
,
200
6-2
008
, and
20
09
3.
2.
P
o
st
Po
lic
y
Eff
e
c
t
s
Tab
l
e
2
d
i
sp
lays th
e estim
ate
d
p
a
ram
e
ters with
th
eir
co
rrespo
n
d
i
ng
stand
a
rd
er
ro
rs,
t-
valu
es an
d p-
v
a
lu
es. Th
e esti
m
a
ted
p
a
ram
e
ters were ob
tain
ed
b
y
u
s
i
n
g
t
h
e with
i
n
fi
x
e
d
effects esti
mato
r.
Data in
Tab
l
e
2
in
d
i
cate th
at
bo
th
FEP and
NIP
were fo
und
to
b
e
statistical
ly sig
n
i
fican
t i
n
p
r
ed
ictin
g MMR wh
en
t
h
e l
e
v
e
l of
si
gni
fi
ca
nce
w
a
s set
at
5%.
For
i
n
st
ance
,
M
M
R
dro
p
p
ed
si
gni
fi
ca
nt
l
y
by
4
5
.
22
(t
=
-2
.4
5
6
,
p-
val
u
e < 5%)
after th
e in
trodu
ctio
n
o
f
FEP. Si
m
i
la
rly, MMR decreased
marked
ly b
y
51
.4
3
(
t
= -
2
.369
, p-
v
a
l
u
e < 5%)
af
ter
th
e in
tro
d
u
c
tion
of
NIP. In
ad
d
ition
,
Tab
l
e
2
sh
ows th
at
SD was a sign
ifican
t p
r
ed
ictor o
f
MMR,
wh
i
l
e DR,
NR,
ANC
, P
N
C, an
d
FPA
we
re
not.
Tab
l
e
2
.
Esti
mated
co
ef
f
i
cen
t
s fo
r
pr
ed
ictor
s
of
MMR in
Gh
an
a fo
r 200
2-2
005
,
20
06-
200
8, an
d 200
9
Coef
f
i
cient
Standard
Er
ror
t-Value
P-Value
DR 0.
001
0.
001
0.
499
.
619
NR 0.
004
0.
021
0.
206
.
838
ANC -
1
.
280
1.
135
-
1
.
127
.
264
PNC -
0
.
079
0.
887
-
0
.
089
.
929
SD -
3
.
430
0.
688
-
4
.
984
.
000**
*
FPA 0.
219
0.
725
0.
302
0.
764
FE
P (Post Policy
E
ffect)
-
45.
215
18.
409
-
2
.
456
.
017**
NI
P(
Post Policy
E
ffect)
-
51.
425
21.
708
-
2
.
369
.
021**
**
* P-
Val
u
e
1
%
, **
P-
Val
u
e
5%,
an
d * P-
V
a
l
u
e
5
%
4.
D
I
SC
USSION AN
D CONCLU
S
ION
The fi
ndi
ng
of
t
h
e prese
n
t
st
udy
t
h
at
t
h
e F
E
P res
u
l
t
e
d i
n
a si
gni
fi
ca
nt
d
ecl
i
n
e i
n
M
M
R
i
s
consi
s
t
e
nt
with
th
at of [9
]
ev
en
t
h
oug
h
t
h
e latter was
no
t statisti
cal
ly
si
gni
fi
ca
nt
.S
pe
cifically, [9] re
ported t
h
at delivery-
r
e
lated
MMR
in
two
r
e
g
i
o
n
s
o
f
th
e cou
n
t
ry d
ecr
eased
after
th
e i
n
tr
oductio
n
o
f
FEP.
Cr
o
s
s
et al.
[13
]
also
reporte
d
a
decl
ine in
direct
m
a
t
e
rnal
deat
hs
post
FEP
.
Data in
th
e p
r
esen
t stu
d
y
ind
i
cated
th
at th
e NI
P
ha
d a significant im
pact
on MMR—M
MR declined
after t
h
e in
t
r
od
u
c
tion of t
h
e po
licy. Th
is
su
ppo
rts th
e assertion
th
at
health
in
su
ran
c
e cov
e
rag
e
positiv
ely
im
pacts m
a
ternal
health
[14].
An
ot
he
r fi
n
d
i
n
g o
f
t
h
e cu
rre
nt
st
udy
was t
h
at
su
per
v
i
s
ed
del
i
v
ery
was
a si
gni
fi
ca
nt
p
r
edi
c
t
o
r o
f
MMR after th
e in
trodu
ctio
n
of bo
th
t
h
e
FEP an
d
NIP.
Th
is find
ing
is
con
s
iste
n
t
with
prev
iou
s
stud
ies
[4
],[1
9
]
t
h
at
re
po
rt
ed t
h
e r
o
l
e
of
ski
l
l
e
d del
i
v
e
r
y
i
n
red
u
ci
n
g
M
M
R
.
As
[
20]
no
t
e
d, st
ak
eh
ol
d
e
rs i
n
t
h
e
Gha
n
ai
a
n
h
ealth
sector
need
to
m
itig
at
e th
e so
cial facto
r
s th
at
p
r
even
t m
a
n
y
ex
p
e
ctin
g
m
o
th
ers
fro
m
seek
in
g
sk
illed
d
e
liv
ery.
Th
e st
u
d
y
u
tilized
Pan
e
l Data to
inv
e
stig
ate th
eim
p
act o
f
th
e fee-free
d
e
liv
ery and
t
h
e
n
a
tio
nal
h
ealth
in
sur
a
nce p
o
licy ex
em
p
t
io
n
s
o
n
m
a
ter
n
al m
o
r
t
alit
y r
a
tio
in
Gh
an
a.Th
e in
t
r
oductio
n
s
o
f
b
o
t
h
th
e FEP
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
. 3
,
Sep
t
emb
e
r
201
5
:
2
32
–
23
7
23
6
an
d
NIP resu
lt
ed
in
si
g
n
i
fican
t
d
eclin
e in
MMR.Th
e
pr
esen
t stud
y u
tilized
d
a
ta at th
e n
a
tio
n
a
l lev
e
l;fu
t
ure
researc
h
s
h
oul
d investigate t
h
e effects
of
F
E
P an
d
NI
P
on
M
M
R
i
n
t
h
e
re
gi
o
n
s.
REFERE
NC
ES
[1]
United Nations,
“Millennium
developm
ent go
al
s
:
2013 r
e
port
”
,
New York, Unit
ed Nat
i
ons, 2013
.
[2]
Asa
m
oa
h,
B.
O.,
Moussa
,
K.
M
.
,
Sta
f
ström,
M.,
Musi
nguz
i,
G.,
“Distribution
of c
a
u
se
s of ma
te
rnal mortality
among different
socio-demograp
hic groups
in Gh
ana: A descriptive stud
y
”
,
BMC Public
Hea
lth
, v
o
l/issue: 11(159)
,
2011, doi: 10
.11
86/1471-2458-11-159.
[3]
Cam
pbell, O
.
M.
R., Grah
am
, W
.
J., “
S
trateg
ies f
o
r
reducing
m
a
te
rnal m
o
rtal
it
y:
Getting
on with w
h
at works”,
The
L
ancet,
vol. 368
, pp.1284–1299
,
2006.
[4]
Thompson, E.,
Sofo, S., “Maternal morta
lity
in Ghana: An econometric an
aly
s
is”,
Journal
of Statistical an
d
Econometric M
e
thods
, vol/issue:
3(2), pp
. 115-12
4, 2014
.
[5]
W
itter, S.
, Arhinful, D. K., Kusi, A., Z
a
kari
a-A
koto,
S., “
T
he e
xperien
ce of Ghana in im
plem
e
n
ting a user fre
e
exemption po
licy
to prov
ide free deliver
y
car
e”,
Reproductive Health Ma
tters,
vol/issue: 15(30)
, p
p
. 61-71
, 2007
.
[6]
Yates
,
R
.
,
“
U
nivers
al h
e
a
lth
car
e
and th
e r
e
m
oval
of us
er f
ees
”
,
The
Lanc
et,
vol. 37
3, pp
. 2078-208
1, 2009
.
[7]
Ministr
y
of Health, “Guidelines f
o
r implementing
the
exem
ption p
o
licy
on
matern
al delive
ri
es
”,
Ac
cra:
M
i
nis
t
r
y
of
Health
, 2004
.
[8]
Dzakpasu, S., S
o
remekun, S., Manu A.,
ten Asbroek, G., Tawiah
, C., et al.,
“Impact of free deliver
y
care on health
facility
deliver
y
and insurance
co
vera
ge in Ghan
a’s BrongAhafo
Region”,
PL
oS ONE,
vol/issue:
7(11), pp. e4943
0,
2012. doi:10.137
1/journal.pone.0
049430.
[9]
Bosu, W. K., Bell, J. S
., Armar-
Kl
emesu, M., A
n
song-Tornui, J., “Effect of
deliv
er
y
car
e user f
e
e exemption policy
on institution
a
l m
a
ternal de
aths in the Centra
l a
nd Volta region
s of Ghana”,
Ghana Medical Journal,
vol/issue:
41(3), pp
. 118–1
24, 2007
.
[10]
Penfold, S., Har
r
ison, E., Bell,
J
., Fitzmaurice,
A., “Evalu
ation
of the de
l
i
ver
y
fee ex
em
ption p
o
lic
y in Ghan
a:
Population estimates of ch
a
nges in deliver
y
s
e
rvice utilization in
two regions”,
Gh
ana Medica
l Jo
urnal,
vol/issue:
41(3), pp
. 100–1
09, 2007
.
[11]
Ansong-Tornui,
J., Armar-Klemesu, M., Arhin
f
ul, D., Pe
nfold
,
S.,
Hussein,
J., “Hospital
based maternity
car
e
in
Ghana - Findings of a confidential enquir
y
in
to maternal deaths
”,
Ghana Medical Journal,
vol/issue: 41(3), pp.
125–132, 2007
.
[12]
Asante, F. A., C
h
ikwama, C., D
a
niels, A., Armar-Kle
mesu, M.,
“Evaluating the
ec
onomic outco
mes of the policy
of fee
exem
ptio
n for m
a
terna
l
deliv
er
y c
a
re
in
Ghana”
,
Ghana Medica
l Journ
a
l,
vol/issue: 41
(3), pp. 110–11
7,
2007.
[13]
Cross, S., Bell,
J. S., Graham,
W. J., “What
y
ou count
is what
y
o
u
targ
et: Th
e imp
lications o
f
maternal death
classification for
tracking progr
ess towards reduci
ng matern
al mortality
in
dev
e
loping countries”,
Bull
et
in of the
World Health
Organization,
vo
l/issue: 88(2), pp.
147-153, 2010
.
[14]
W
a
ng, W
.,
Tem
s
ah, G.
, Mal
lick
,
L
.
,
“
H
ealth
insurance
cov
e
rage and its impact
on ma
terna
l
he
a
lth ut
ili
zat
ion i
n
low- and middle-income coun
tries”, DHS
Analy
tical Studies No
.
45, ICF Inte
ran
t
ional, Rockvi
ll,
MD, USA, 2014.
[15]
Ghana Health Service, “The h
ealth sector in Gh
an
a:
Facts and figu
res 2010”, Accra, Gh
ana: Ghana
Health Serv
ice,
2010.
[16]
Zivot E., “Fixed
effects estimatio
n of pan
e
d d
a
ta”, Pamphlet, Univ
ersity
of
Washin
gton, USA, 2012
.
[17]
Baltag
i
, B. H., “Econometric
analy
s
is of
pa
n
e
l data”, John Wile
y
& S
ons, In
c., New York, 2005.
[18]
Elhorst, J. P., “
S
patial
econometrics
: From cross-sectional data to spatia
l panels”, Springer: B
e
r
lin, New York,
Dordrecht, Lond
on, 2014
.
[19]
Graham, W. J., Bell, J. S., Bullough,
C
.
H. W., “Can skilled
attendan
ce
at
deliver
y
redu
ce matern
al mor
t
ality
in
develop
i
ng coun
tries
?
”,
Safe motherhood strategies: a
review
of the evid
ence,
vo
l. 17, pp. 97-130,
2001.
[20]
Sappor, M., Esena, R. K.
, ‘Utilization of skil
led deliv
er
y
ser
v
ices in
Ghana:
Qualit
y
im
provem
e
nt issues in
m
a
ternal
he
alth
”
,
International Journal of I
nno
va
tive Research an
d Studies
, vo
l/issue: 2(5)
, pp
. 353
-367, 2013
.
BIOGRAP
HI
ES
OF AUTH
ORS
Dr. Sofo is
currently
a full Profes
sor of Ph
y
s
i
cal
Educa
tion Te
ach
er
Edu
c
ation
in the Department
of Health
, Hum
a
n P
e
rform
ance a
nd Recre
a
tion
at
S
outheas
t M
i
s
s
ouri S
t
at
e Unive
r
s
i
t
y
, M
i
s
s
ouri
,
USA.
He holds
a Ph.
D
.
in Human Performance
Studies with a concen
tration in Ph
y
s
ical
Education Ped
a
gog
y
from
The
University
of
Alab
am
a, US
A. H
e
obt
ained
a M
S
.Ed. (P
h
y
s
i
c
a
l
Education Pedag
o
g
y
) from the State University
of
New York College at Bro
c
kport, NY. Prior to
that, he graduated from the University
of Cape
Coas
t, Ghana, with a B.Ed
. in P
h
y
s
i
c
a
l
Education,
and a Diploma in Biolog
y
.
Profe
ssor Sofo has published and presented numerous
peer r
e
vi
ewed p
a
pers
a
t
th
e n
a
t
i
onal
and
intern
ation
a
l
leve
ls
o
n
: ph
y
s
ic
al
edu
cat
ion t
each
er
education, children’s ph
y
s
ical
act
ivit
y,
per
c
eptua
l
-m
otor
development, an
d postcolonial
pedagog
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Ma
terna
l
Mo
rt
a
lity in
Gha
na: Impa
ct
o
f
th
e
Fee-Free
Delivery Po
licy
an
d
th
e Na
tiona
l .... (S
eidu
So
fo
)
23
7
Emmanuel Tho
m
pson is an Assistant Professor of St
atisti
cs/Act
uaria
l
Scien
ce in
the Depar
t
m
e
nt
of M
a
them
ati
c
s
at S
outhe
as
t
M
i
s
s
ouri S
t
ate
Univers
i
t
y
, M
i
s
s
ouri, US
A. He is
a Gradua
te
Statisti
cian m
e
m
b
er of the Am
erican Statist
i
c
a
l
Association and a
l
so an Affiliate
m
e
m
b
er of the
Institute
and Facult
y
of Actu
aries
in the Unit
ed K
i
ngdom
. Em
m
a
nuel holds a Ph.D
. in Statisti
cs
(Actuari
al Sc
ien
ce op
tion)
and
a
MSc. in
Actuar
ial Sc
ien
ce bo
th
from
Universit
y
of Cal
g
ar
y
in
Canada. He also graduated from Univ
ers
i
t
y
of Ca
pe Coas
t in Ghana, with a BS
c. (F
irs
t
Clas
s
)
in
Statistics and h
a
s spent over six
y
e
ars working in
the insur
a
nce
in
dustr
y
in Ghana
and brief
l
y in
Canada. Emmanuel’s resear
ch age
nda cove
rs the are
a
s of risk
theor
y
& a
c
tuar
ial m
odeling
,
applied statistics
,
and app
lie
d
econometrics. Em
manuel has pub
lis
hed and pr
esented diff
erent
peer revi
ewed p
a
pers in Applied
Statistics and A
c
tuar
ial Sci
e
nc
e at the na
tion
a
l a
nd interna
tiona
l
leve
ls.
Evaluation Warning : The document was created with Spire.PDF for Python.