Intern
ati
o
n
a
l
Jo
urn
a
l
o
f
P
u
b
lic Hea
l
th Science (IJ
P
HS)
Vol
.
4,
N
o
.
4
,
D
ecem
b
er 20
1
5
, pp
. 33
2~
33
7
I
S
SN
: 225
2-8
8
0
6
3
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
Magnitude of Ou
t of
P
o
ck
et
Health Expenditures and
Ass
o
ci
at
ed Fact
ors amon
g Ci
vil S
e
rvants
Mekuane
n
et Geremew
1
,
Mo
lla
Gedefa
w
2
, Girm
ay Tsegay
1
, Ge
tac
h
ew
Mullu
Ka
ssa
3
1
School of pub
lic Health
, Colleg
e of M
e
dicine an
d Health Scien
ces, DebreMarkos
University
, Debr
eMarkos,
Ethiop
ia
2
GAMBY College of M
e
dical Sc
iences, B
a
hir D
a
r, Ethiopia
3
Department of Midwifer
y
,
Co
llege
of
Medicine
and Health
Scie
nces, D
e
breMar
kos University
,
DebreMarkos,
Ethiopia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Oct 28, 2015
Rev
i
sed
No
v
26
, 20
15
Accepted Nov 28, 2015
In Ethiopia,
as other dev
e
lopin
g
count
ries
, pub
lic health
car
e is
provided at
nominally
low prices and free to
those
that does
not afford to pay
.
But the
heal
th care
consum
er population
is still
to m
a
ke
considerab
le
am
ount of out-
of-pocket h
ealth expenditu
re
for
various reasons. A cross sectional
quantitativ
e stud
y
from Janu
ar
y
to
May
2013 w
a
s done. Stud
y
populatio
n
was civil servan
ts in DebreMarkos town. A tot
a
l of 467stud
y
participan
ts
were selected b
y
using si
mple random
sa
mpling method. The collected data
were enter
e
d into a computer b
y
using
Epi-Data v
e
rsion 3.1 and analy
s
is was
performed b
y
u
s
ing SPSS versi
on 16 fo
r wind
ows. Possible associations
between ou
t of p
o
cket h
eal
th exp
e
nditur
e
and i
t
s
predic
tors
were
anal
yz
ed
b
y
using both bivariat
e and m
u
ltiv
aria
te an
al
y
s
is.
The m
ean age
of the stud
y
parti
c
ipan
ts were 41
y
e
a
r
s. Maj
o
rities were be
t
w
een 25 and 44
ye
ars of age,
258 (55.2%). Th
e level of education am
ong the stud
y
particip
ants indicated
that most 380 (81.4%) were grad
uates of
higher
education (HE) and majority
were Orthodox Christian which
accoun
te
d 446 (
95.5%) followed
b
y
Muslim
s
13 (2.8%).
To p
u
t it br
iefly
,
th
e stud
y
identified
that
the median
of out of
pocket h
e
alth
care exp
e
nditur
e
account
ed 8.26%
of total househ
old incom
e
.
Health st
atus of
the household (
w
ith or without
chronic
illn
ess),
debt on an
y
of the househo
l
d, house on
con
s
truction owned
b
y
an
y
househ
old member,
educational fee f
o
r at least one
memb
er of the household and pr
edominantly
used heal
th ins
titut
i
on were
th
e associa
t
ed f
a
ctors tha
t
hav
e
signific
a
nt
impact on
household out of
po
ck
et h
ealth
expend
iture.
Keyword:
Out
o
f
poc
ket
m
oney
Health
insurance
Civ
il serv
an
ts
Ethiopia
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
:
Getach
ew M
u
llu
Kassa,
Depa
rtem
ent of Midwifery,
DebreMarkos Uni
v
ersity,
Ethiopia.
Em
a
il: g
ech
m
2
0
05@g
m
ail.co
m
1.
INTRODUCTION
Out
-
o
f
-
p
ocket
heal
t
h
pay
m
ent
s
ar
e expenditures
borne di
rectly by
a patient where i
n
surance
does
n
o
t
cov
e
r th
e fu
ll co
st
of the
h
ealth
go
od
or serv
ice.
Thes
e include c
o
st-shari
ng, sel
f-medication and
othe
r
expe
n
d
i
t
u
re
pa
i
d
di
rect
l
y
by
pri
v
at
e h
o
u
seh
o
l
d
s
[
1
]
.
I
n
Et
hi
o
p
i
a
, as
ot
he
r d
e
vel
opi
ng
c
o
u
n
t
r
i
e
s,
p
u
b
l
i
c
heal
t
h
care is prov
id
ed
at no
m
i
n
a
lly
lo
w
p
r
ices and
free to
th
ose
that cannot afford t
o
pa
y
[2
]
.
Bu
t th
e
h
ealth
care
co
nsu
m
er po
pu
latio
n
is still to
m
a
k
e
con
s
id
erab
le am
o
u
n
t o
f
ou
t-o
f
-po
c
k
e
t h
ealt
h
ex
pen
d
iture
(OOPE)
for
vari
ous
reas
o
n
s
[
3
]
.
For instance
,
according to
Nationa
l
Health account of
Ethiopia, th
e magnitude
of out of pocket
h
ealth
exp
e
nd
i
t
u
r
e reach
e
s
ab
ou
t 4
0
% o
f
th
e
to
tal h
ealth
exp
e
nd
iture. Ou
t
-
of-po
c
k
e
t h
ealth
p
a
ym
en
ts
h
a
v
e
sub
s
t
a
nt
i
a
l
neg
a
t
i
v
e si
de ef
fec
t
s. They
m
a
y
lead t
o
i
m
pove
r
i
shm
e
nt
and f
u
rt
her
ha
rds
h
i
p
.
The re
q
u
i
r
em
ent
of
Ou
t
-
of-po
c
k
e
t
h
ealth
p
a
ym
en
ts is p
a
rticu
l
arl
y
h
a
rd
on
t
h
e
p
oor,
wh
o
s
e illn
ess
will eith
er rem
a
in
un
treated
or
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Magnitude of Out of
P
o
cket Health
Ex
penditures
and
As
s
o
ciated F
a
ctor
s Among
.... (M
ekuane
n
et Ge
r
e
mew)
33
3
f
o
r
ce p
a
tien
t
s i
n
to
deep
er
pover
t
y. Th
e poo
r
may n
o
t
seek
med
i
cal car
e an
d, as a r
e
su
lt, r
e
m
a
in
tr
ap
p
e
d in
th
e
v
i
cio
u
s circle o
f
illn
ess and
p
o
v
e
rty [2
]. Ho
wev
e
r, ou
t-
o
f
-po
c
k
e
t h
ealth
ex
p
e
nd
itu
re
remain
s p
o
o
r
ly v
i
sib
l
e,
an
d it is in
n
e
ed
o
f
survey.
M
o
re
ove
r, t
h
e
go
ve
rnm
e
nt
of
Et
hi
o
p
i
a
pl
a
n
s t
o
i
n
t
r
o
duce
soci
al
heal
t
h
i
n
su
ra
nce f
o
r f
o
rm
al
sect
or
e
m
p
l
o
y
ees. The Eth
i
op
ian
g
o
v
e
rn
m
e
n
t
is in
th
e pro
cess
of in
itiatin
g
h
eal
th
in
su
ran
ce sch
e
m
e
s: so
cial
h
ealth
insurance
(SHI) for t
h
e formal sect
or and com
m
uni
t
y
-base
d
heal
t
h
i
n
su
ra
nce (CB
H
I) for citizens in the
inform
al and agric
u
lture sec
t
ors. T
h
e nece
ssary lega
l fra
meworks are
already in
p
l
ace fo
r th
e
p
ilotin
g
of
CBHI sch
e
m
e
s
as well as for i
n
itiatio
n
of th
e SHI
prog
ra
m
.
Th
e SHI ag
en
cy h
a
s alread
y
been
estab
lished an
d
i
s
bei
ng st
af
fe
d wi
t
h
re
qui
re
d pr
o
f
essi
o
n
al
s
.
SHI
was ex
p
ect
ed t
o
be o
p
e
rat
i
onal
f
o
r ci
vi
l
serva
n
t
s
be
gi
n
n
i
n
g
in
Ju
ly 20
12
.
It
will g
r
adu
a
lly ex
p
a
nd
to
co
ver all fo
rm
al secto
r
em
p
l
o
y
ees. Sin
ce
2
011
,
CBHI sch
e
m
e
s
h
a
v
e
been
pi
l
o
t
e
d i
n
1
3
di
st
ri
ct
s i
n
Am
hara,
Or
om
i
a
, SN
NP
,
an
d
Tig
r
ay Reg
i
onal States [
3
]. Th
er
ef
or
e, pr
odu
cing
dat
a
rel
a
t
e
d t
o
t
h
e m
a
gni
t
u
de o
f
out
of
poc
ket
heal
t
h
expe
n
d
i
t
u
re a
m
ong f
o
rm
al
sect
or em
pl
oy
ees i
n
Deb
r
eM
ar
k
o
s To
wn
we
re ver
y
tim
e
ly
.
2.
R
E
SEARC
H M
ETHOD
2.
1.
Stud
y design
A cr
oss
-
sect
i
o
nal
su
rvey
wa
s con
d
u
ct
ed t
o
assess t
h
e m
a
gni
t
u
de
of
out
of
poc
ket
ex
p
e
ndi
t
u
re a
n
d
associated fact
ors
.
2.
2.
Stud
y are
a
Th
is st
u
d
y
w
a
s cond
u
c
ted
du
r
i
n
g
Janu
ar
y-May 2
013
in
D
e
br
eMarko
s
Tow
n
.
D
e
br
eMar
ko
s is t
h
e
cap
ital o
f
East
Gojj
am
Ad
min
i
strativ
e Zon
e
; lo
cated
in
t
h
e
North
West of
Addis Abe
b
a a
t
a distance of
300
km
s and 26
5
km
sB
ahi
r
Dar,
respect
i
v
el
y
.
W
i
t
h
re
gar
d
t
o
t
h
e po
pul
at
i
o
n of t
h
e t
o
w
n
i
t
i
s
est
i
m
a
t
e
d t
o
be
1
076
84
o
f
wh
i
c
h
577
91
ar
e
fe
m
a
les an
d
498
93
ar
e m
a
les. Fr
o
m
p
o
p
u
l
atio
n
f
i
gu
r
e
m
e
n
tio
n
e
d
abov
e 42
743
f
e
m
a
les an
d
358
43
m
a
le p
o
p
u
latio
n
s
said to
h
a
v
e
b
e
en
t
h
e
yo
un
g pop
u
l
atio
n at th
e
pr
oductiv
e ag
e. [4
].
2.
3.
Study population and
samp
l
e
siz
e
determinati
on
St
udy
po
p
u
l
a
t
i
on
was f
o
rm
al
sect
or em
pl
oy
ees i
n
De
breM
ark
o
s t
o
w
n
an
d we
use
d
si
m
p
l
e
ran
d
o
m
tech
n
i
qu
e t
o
selected
stu
d
y
p
a
r
ticip
an
ts. Sa
m
p
le size is d
e
ter
m
in
ed
u
s
i
n
g
t
h
e for
m
u
l
a
f
o
r
sing
le popu
latio
n
p
r
op
or
tio
n. Becau
se t
h
er
e is
n
o
stud
y co
ndu
cted
i
n
D
e
b
r
eMar
ko
s t
o
wn
an
d in similar
setu
p on
t
h
is t
o
p
i
c, a
p
r
op
ortio
n of
50
% is tak
e
n b
a
sed
on
th
is t
h
e
sam
p
le size is 4
6
7
.
2.
4.
Data c
o
llection pr
ocedure
s
Th
e
d
a
ta was co
llected
fro
m
p
a
rticip
an
ts’
by th
e m
eans of
sem
i
-st
r
uct
u
re
d
quest
i
o
n
n
ai
r
e
. A t
o
t
a
l
of
46
7 re
sp
o
nde
n
t
s were co
ve
r
e
d d
u
ri
ng t
h
e
sur
v
ey
. T
h
i
s
sur
v
ey
col
l
ect
s heal
t
h
care
expe
n
d
i
t
u
re
da
t
a
for
v
a
ri
o
u
s
categories of treatm
e
n
t
lik
e
h
o
sp
italized
care,
o
u
t
patien
t
care,
b
i
rth
d
e
liv
ery, chron
i
c illn
ess an
d all
heal
t
h
e
x
pen
d
i
t
ure
pai
d
di
rec
t
l
y
by
h
o
u
s
eh
o
l
ds
of
f
o
rm
al
sect
or em
pl
oy
e
e
s i
n
t
h
e
st
u
d
y
area
an
d
pe
ri
od
.
In
t
h
i
s
st
udy
10
adeq
uat
e
l
y
t
r
ai
ned
dat
a
c
o
l
l
ect
ors a
nd t
w
o
sup
e
r
v
i
s
o
r
s w
e
re i
n
vol
ved i
n
t
h
e dat
a
c
o
l
l
ect
i
o
n
pr
ocess
.
The
d
a
t
a
col
l
ect
ors
were
gra
d
uat
e
s of sec
o
n
d
a
r
y
sch
ool
an
d
wh
o co
ul
d
wri
t
e
and
fl
ue
nt
l
y
speak t
h
e
l
o
cal
l
a
ng
uage
. Su
per
v
i
s
o
r
s
and t
h
e pri
n
ci
pal
i
nvest
i
g
at
or c
h
eck
on
dai
l
y
basi
s fo
r t
h
e com
p
l
e
t
e
ness,
accuracy, a
nd
clarity of the que
stionn
aire
. Any error, ambiguity, incomple
teness,
or otherwise encountere
d
was
add
r
essed
o
n
th
e fo
llo
wi
ng
d
a
y
b
e
fo
re startin
g
th
e n
e
x
t
d
a
y
activ
ities.
2.
5.
Da
ta
quality
and anal
ysis
Th
e qu
estionnair
e w
a
s car
efu
lly tr
an
slated in
to
A
m
h
a
r
i
c b
y
ex
p
e
r
t
s,
with
g
r
eat em
p
h
a
sis g
i
v
e
n
to
local vocabularies. The
questio
n
n
ai
re
i
s
al
so bei
n
g pre
-
t
e
st
ed on p
r
i
v
ate
organizations. T
h
e c
o
llected da
ta wa
s
entere
d into a com
puter by using
E
p
i-Data versi
on
3.
1 and analyzed
using S
PSS
versi
on
16 for windows
.
Possi
bl
e ass
o
c
i
at
i
on bet
w
een
o
u
t
o
f
poc
ke
t
heal
t
h
e
xpe
n
d
i
t
u
re
an
d i
t
s
pre
d
i
c
t
o
rs a
r
e anal
y
zed
by
b
o
t
h
biva
riate and
m
u
ltivariate analysis. The c
u
t
of
f
poi
nt for st
atistical significance is
P<
0.05.
2.
6.
Ethical c
o
nsiderati
ons
Ethical clearance was
obtaine
d
from
Debre
M
arkos
Un
i
v
er
sity
; furthe
r
pe
rm
ission letter also sec
u
re
d
from
each formal sector institutions
in De
breMa
r
kos town. Inform
ed co
nsent was al
so obtaine
d
from
each
stu
d
y
p
a
rticip
an
ts. Priv
acy and
co
nfid
en
tialit
y are also m
a
in
tain
ed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-88
06
IJPHS Vol. 4, No. 4, D
ecem
ber 2015
:
332 – 337
33
4
3.
R
E
SU
LTS AN
D ANA
LY
SIS
3.
1.
Socio dem
o
gr
aphic c
h
ar
ac
teristics
A to
tal o
f
467
stu
d
y
p
a
rticip
an
ts respon
ded
to
th
is stu
d
y
with
5
%
no
n-resp
on
se rate. The
m
ean
ag
e
o
f
t
h
e stud
y p
a
r
ticip
an
ts w
a
s
4
1
year
s, SD
(±
11
.13
)
.
Th
e ag
e of
stud
y p
a
rticip
an
ts r
a
ng
ed
b
e
t
w
een
18
an
d
65
.
Maj
o
rities were in
b
e
tween
25
and
44
,
2
5
8
(5
5.2%). m
o
st 3
8
0
(81
.
4
%
) are g
r
ad
u
a
tes
o
f
h
i
gh
er ed
u
cati
o
n
an
d
maj
o
rity are Ortho
dox
Christia
n
whi
c
h acc
o
unt
s
4
4
6
(
9
5.
5
%
),
In t
h
e st
udy
fa
m
i
ly
si
ze vari
e
s
i
n
bet
w
ee
n
1
and
1
2
a
n
d t
h
e
d
o
m
i
nant
num
ber
2
6
3
(
5
6.
3
%
) are
i
n
t
h
e
cat
ego
r
y
of
fa
m
i
ly
si
ze whi
c
h i
s
l
e
ss t
h
an
or e
qual
t
o
4,
t
h
e rem
a
i
n
i
ng
20
4 (
4
3.
7)
wh
ere i
n
t
h
e cat
e
g
o
r
y
o
f
fam
i
l
y
si
ze wh
i
c
h i
s
>4 m
ont
hl
y
i
n
com
e
of t
h
e st
udy
h
o
u
s
e
hol
ds va
ri
es f
r
om
300 t
o
4
0
0
0
,
wi
t
h
a m
e
an o
f
17
3
6
.
7
4
S
D
(
±
64
6.
1
9
)
.
M
a
jo
ri
t
y
o
f
t
h
e
s
t
udy
part
i
c
i
p
a
n
t
s
35
7(
7
6
%)
earne
d
i
n
bet
w
een
10
0
0
-
2
5
0
0
B
i
rr
(Table 1
)
Table 1. Soci
o-Dem
ogra
phic Characteristics
of
stud
y
p
a
r
ticip
an
ts, D
e
b
r
eMar
ko
s,
201
3
3.
2.
Additi
on
al attributes of
s
t
u
d
y par
t
icipants
Inform
atio
n
abo
u
t
h
ealth
statu
s
o
f
th
e ci
v
il serv
an
ts
v
i
sited
h
ealth
institu
tio
n
at least
o
n
e m
e
m
b
er of
th
e h
o
u
s
eho
l
d h
a
d
chron
i
c illn
ess, d
e
b
t
(p
resen
ce of an
y h
o
u
s
eh
o
l
d
me
m
b
er who
h
a
d
an
y d
e
b
t
), p
a
i
d
ed
u
cation
a
l fee at least fo
r on
e m
e
m
b
er o
f
th
e ho
u
s
eho
l
d
,
p
r
ed
o
m
in
an
tly u
s
ed
h
e
alth
in
stitu
tio
n
(private or
pu
bl
i
c
),
ha
d a
h
ouse
(
s
)
un
de
r co
nst
r
uct
i
o
n
ow
ne
d
by
any
m
e
m
b
er of t
h
e h
ouse
h
ol
d a
n
d
m
ount
of a
n
n
u
a
l
avera
g
e health care
e
xpe
nditure.
In
stud
y hou
seh
o
l
d
s
wh
ere t
h
ere
were at l
east o
n
e m
e
mb
er
of th
e
ho
useho
l
d
with
chron
i
c illn
ess
accounted for 457
(97.9%) of
the
partic
ipa
n
t
s
. Study hous
e
hol
ds who we
re
paying e
d
uca
tional fee for at leas
t
one
m
e
m
b
er of the
house
h
old accounte
d
for
451(96.9%
)
Study
participa
n
ts who
pr
edominantly use
d
private
health i
n
stitution accounted
for
455
(97.4%) a
n
d who
predom
inantly use
d
public he
alth care i
n
stitutions
accounted
for 12 (2.6%).
Study
households
with at least
one
m
e
m
b
er
of the
house
h
old on t
h
e proce
ss of
b
u
ild
i
n
g an
y
h
o
u
s
e accoun
ted
for
1
1
4
(
2
4
.
4
%
) and
t
h
ose wh
o w
e
r
e
n
o
t
on
th
is
pro
cess accou
n
t
ed
353
(7
5.
6%
) of
t
h
e part
i
c
i
p
a
n
t
s
.
Study
house
h
olds whos
e annual avera
g
e
he
alth care
exp
e
n
d
iture less than
3
%
o
f
t
h
eir in
co
m
e
were
accounted
118 (25.3%), whose expe
nditu
re
in betwee
n
3-6% we
re acc
ount
e
d
86 (18.4%),
7-12% acc
ounted
75
(1
6.
1%
) an
d t
h
ose w
h
o s
p
end
≥
12% t
h
eir inc
o
m
e
accounts 188
(40.3%
).House
hol
ds whose
s
o
urce of fund
fo
r
out
o
f
poc
k
e
t
heal
t
h
e
x
pen
d
i
t
u
re
w
h
ere
f
r
o
m
hou
seh
o
l
d
s
i
n
com
e
acco
u
n
t
s
38
5
(8
2.
4%
), t
hose
t
h
e
r
e f
u
n
d
whe
r
e
from
hous
ehold savi
ng accounts
28
(6%
)
,
those
from
selling of fam
ily asset accounts
10
(2.1%
)
, those
from
aid accounts
4 ( 0.9%) a
n
d those t
h
er
e
expe
nditure
where
from
lo
an accounts 40
(8.6%).
Variable
Categor
y
No
Percent
Age
18-
24
25-
44
45-
65
Total
39
258
170
467
8.
4
55.
2
36.
4
100
Religion
Sex
Or
thodox Chr
i
stian
Mu
sli
m
Pr
otestant Chr
i
stian
Catholic
Male
446
13
7
1
300
95.
5
2.
8
1.
5
0.
2
64.
3
E
t
hnicity
Am
ha
r
a
Tig
r
e
Oro
m
o
Gur
a
ghe
456
4
5
2
97.
6
0.
9
1.
1
0.
4
Fa
m
i
l
y
size
≤
4
>4
263
204
56.
3
43.
7
Education
Higher
E
ducation
10-
12
Co
m
p
lete pr
i
m
ary
E
duc.
I
n
co
m
p
lete pr
i
m
ary
E
duc.
To
tal
380
64
11
8
467
81.
4
13.
7
2.
4
1.
7
100
M
onthly
In
co
m
e
<651
651-
99
9.
999
1000-
2
500
>2500
22
18
357
70
4.
7
3.
9
76
15
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Magnitude of Out of
P
o
cket Health
Ex
penditures
and
As
s
o
ciated F
a
ctor
s Among
.... (M
ekuane
n
et Ge
r
e
mew)
33
5
Table 2. Factors
associate
d
w
ith
ou
t
o
f
po
ck
et h
ealth
car
e
exp
e
nd
itur
e
.
D
e
br
e Mark
s, 201
3
3.
3.
Fact
ors
as
soci
ate
d
w
i
th
ou
t
of
po
cket
he
al
th c
a
re e
x
pe
n
d
i
t
ure
To find out factors ass
o
ciated w
ith
ou
t o
f
p
o
c
k
e
t h
ealth
care ex
p
e
nd
iture b
i
v
a
riate an
d
m
u
lt
iv
ariate
anal
y
s
i
s
was d
one c
o
n
s
i
d
eri
n
g soci
o dem
o
g
r
ap
hi
c vari
a
b
l
e
s, heal
t
h
st
at
us
of t
h
e h
o
u
se
h
o
l
d
(
w
i
t
h
o
r
w
i
t
hout
ch
ron
i
c illn
ess),
d
e
b
t
on
an
y
o
f
t
h
e
h
o
u
s
eho
l
d
,
hou
se on
co
nstru
c
tion
own
e
d
b
y
an
y
h
o
u
s
eho
l
d
m
e
m
b
er,
ed
u
cation
a
l fee for at least o
n
e
m
e
m
b
er o
f
th
e ho
u
s
eho
l
d
an
d
p
r
edo
m
in
a
n
tly u
s
ed
h
ealth
in
stitu
tio
n
(priv
a
t
e
or
public) as an inde
pe
ndent variable and avera
g
e
out of
poc
ket heal
th care ex
pen
d
i
t
u
re as a de
pen
d
e
n
t
vari
a
b
l
e
.
Income
:
Th
e
m
u
l
tiv
ariate an
alysis show
ed
th
at
in
co
m
e
of th
e hou
seh
o
l
d
,
h
ealth statu
s
of t
h
e
h
o
u
s
eho
l
d
,
pred
o
m
in
an
tly used
h
ealth
in
stitu
tio
n
,
e
d
u
c
atio
n
a
l fee and d
e
b
t
foun
d to
h
a
v
e
statistically
sig
n
i
f
i
can
t
asso
ciatio
n
w
ith
o
u
t
o
f
p
o
c
k
e
t
h
ealth
exp
e
n
d
i
tu
r
e
. Resp
onden
t
s wh
ose inco
m
e
w
a
s <6
51
Bir
r
whe
r
e
10.34 t
i
m
e
s
m
o
re likely to spend
for healt
h
care than t
hose
whose i
n
com
e
greate
r
tha
n
2500
A
O
R
=1
0.34
,
95
% CI
(1
.82
,
58
.7
8)
Respon
den
t
s wh
o
s
e i
n
co
m
e
w
a
s in
th
e r
a
ng
e of
Bir
r
6
5
1
to
9
9
9
.
99
w
h
er
e
1
1
.01
tim
es
mo
re lik
ely to
sp
end fo
r th
eir h
ealth car
e
than t
hose
whos
e inc
o
m
e
whe
r
e
greater tha
n
Birr
g
r
eater
th
an
250
0
on
AO
R=
11
.0
1, 9
5
%CI
(1
.3
9, 8
6
.5)
.
Resp
ond
en
ts
who
s
e in
co
m
e
was in
th
e
rang
e of Birr
1
000
to
250
0
wh
ere 3
.
14
ti
m
e
s
m
o
re lik
ely
to
spe
nd
fo
r t
h
ei
r
ho
use
hol
d he
al
t
h
care t
h
a
n
t
hose
wh
ose i
n
com
e
was grea
t
e
r t
h
an B
i
r
r
2
5
0
0
on
AOR
=
3.
1
4
,
95
%C
I,
(
1
.
9
6
,
5.
03
).
Heal
t
h
st
atu
s
:
In
stud
y ho
u
s
eh
o
l
d
wh
o
h
a
ve at least o
n
e
me
m
b
er with
ch
ron
i
c illn
ess wh
ere 6
.
04
ti
m
e
s
m
o
re l
i
k
e
ly to
sp
en
d
for th
eir hou
sehold
h
ealth
care t
h
an
tho
s
e
who
d
i
d
no
t h
a
v
e
ch
ron
i
c illn
ess AOR=
6.
04
,
95
% C
I
(
1
.
0
8
,
33
.8
9
)
.
Predominately used he
alth care instituti
o
n:
St
u
d
y
h
ous
ehol
ds w
h
o w
e
re p
r
ed
om
i
n
ant
l
y
vi
si
t
e
d
p
r
i
v
ate h
ealth
care in
stitu
tions wh
ere
11
.7
3
ti
m
e
s
m
o
re lik
ely to
sp
end
fo
r t
h
eir
h
o
u
s
eh
o
l
d
h
e
alth
care th
an
th
o
s
e who
p
r
edo
m
in
an
tly u
s
ed pu
b
lic
h
ealth
care
in
stitu
tion
s
on
AOR= 11
.7
3, 9
5
%
CI
(2
.2
1, 6
2
.28).
Educatio
na
l fee
:
St
udy
h
o
u
s
eh
ol
ds w
h
o
were
pay
i
ng
f
o
r e
ducat
i
o
n f
o
r at
l
east
on
e ho
use
hol
d
m
e
m
b
er where
3.3
9
t
i
m
e
s
more l
i
k
el
y
t
o
spen
d f
o
r t
h
ei
r ho
use
h
ol
d heal
t
h
care t
h
an t
h
ose w
ho
di
d n
o
t
pa
y
any educationa
l fee
on AOR
=
3
.
39
,
95
% C
I
(1
.1
, 10
.4)
.
Debt
:
St
u
d
y
h
ous
eh
ol
ds
w
h
o
ha
d a
n
y
debt
t
o
be
pai
d
w
h
e
r
e 2.
9
8
t
i
m
es
m
o
re
l
i
k
el
y
t
o
s
p
end
f
o
r
t
h
ei
r
ho
use
h
ol
d heal
t
h
care
t
h
a
n
t
h
ose wh
o di
d no
t
have
any
de
bt
o
n
AOR
= 2.
9
8
, 9
5
%
C
I
(
1
.
1
2, 7.
9
8
)
.
Source
s of
fu
nd for
ou
t of
pocke
t
he
alth
expen
d
iture
:
Based
on
th
e data, stu
d
y
p
a
rti
c
ip
an
ts sp
en
t
all these healt
h
ca
re e
xpe
nditures
from
different s
o
urces
like from
cash
reve
nue (CR
)
accounts
83%, from
ho
use
h
ol
d sa
vi
ng
(
H
S
)
6
%
,
fr
om
sal
e
s of fa
m
i
ly
asset
like jewels
, house,
car, etc. a
ccounts 2%,
from
loan (L
)
9% a
n
d f
r
o
m
gi
ft
(G
)
1%
(Fi
g
ure
1
)
.
Out
Of Poc
k
et
He
alth Expenditu
re
Variable/Ques
t
ion
Categor
y
Aver
age annual health
expenditur
e
AOR COR
(
95%CI
)
AOR
<3%
≥
3%
Inco
m
e
< 651 Bir
r
2
20
10.
34
0.
51
(
1
.
82,
58.
78)
651-
99
9.
999 Bir
r
1 17
11.
01
(
1
.
39,
86.
8)
1000-
2
500 Bir
r
89
268
3.
14
(
1
.
96,
5.
03)
>2500 Bir
r
26
44
1.
0
1.
0
Presence
of ch
ronic
illness
Yes
110
347
6.
05
0.
079
(
1
.
08,
33.
89)
NO
8
2
1.
0
1.
0
P
r
e
d
om
i
n
a
n
t
l
y
used health
insti
t
ution
Private
108
347
11.
72
0.
062
(
2
.
2
,
62.
28)
Public
10
2
1.
0 1.
0
Educational Fee
Yes 110
341
3.
39
0.
323
(
1
.
1
,
10.
43)
No 8
8
1.
0 1.
0
Debt
Yes
105
336
2.
98
3.
2
(
1
.
12,
7.
96)
No
12
12
1.
0
1.
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-88
06
IJPHS Vol. 4, No. 4, D
ecem
ber 2015
:
332 – 337
33
6
Fi
gu
re 1.
S
o
u
r
ce
o
f
f
u
nd
f
o
r
out
o
f
poc
ket
h
eal
t
h
ex
pe
ndi
t
u
re of st
u
d
y
pa
rt
i
c
i
p
ant
s
, De
bre
M
ark
o
s, 2
0
1
3
3.
4.
Discussion
The m
e
di
an of
out
o
f
p
o
cket
heal
t
h
ex
pen
d
i
t
u
re of t
h
i
s
st
udy
wa
s ab
out
8.2
6
%
of t
o
t
a
l
hou
seh
o
l
d
incom
e
with [(IQR)
, 2.
9
8
, 2
2
.
7
1
)
]
.
The g
o
v
e
r
nm
ent of Et
hiopia wa
nts to cove
r f
o
rm
al
sector em
ploy
ees with
soci
al
heal
t
h
i
n
su
ra
nce.
The
am
ount
o
f
m
oney
p
r
op
ose
d
t
o
be
prem
i
u
m
fo
r t
h
i
s
i
n
s
u
ra
nce i
s
3%
o
f
m
ont
hl
y
salary (12
)
. This
m
ean
s th
e
mag
n
itu
d
e
o
f
o
u
t
of po
ck
et ex
p
e
nd
itu
re in th
is stu
d
y
is
m
o
re th
an
two ti
mes o
f
t
h
e p
r
op
ose
d
p
r
em
i
u
m
.
In
th
is stud
y in
co
m
e
o
f
th
e h
o
u
s
eho
l
d
was fo
und
to
b
e
statistica
lly
sig
n
i
fican
t with
ou
t o
f
po
ck
et
heal
t
h
ex
pe
n
d
i
t
u
re. R
e
s
p
on
de
nt
s w
h
o
s
e i
n
c
o
m
e
was l
e
ss t
h
an t
h
e
fi
rst
i
n
c
o
m
e
gro
up
(<
65
1 B
i
r
r) a
r
e
10
.3
4
ti
m
e
s
m
o
r
e
lik
ely to
sp
end
than
tho
s
e who
s
e in
co
m
e
w
h
er
e
g
r
eater on
AOR = 10
.34
,
95% CI
(
1
.82
,
58.75
).
Resp
ond
en
ts wh
o
s
e in
co
m
e
g
r
oup
is i
n
b
e
tween
Birr
6
51-
99
9.999
w
h
er
e
1
1
.01
tim
es
mo
r
e
lik
ely t
o
spe
nd t
h
an t
h
ose whose i
n
come is greater on AOR= 11
.01, 95
% CI
(
1
.39, 86
.8
5)
.Respon
d
e
n
t
s
w
h
o
s
e i
n
co
m
e
was in
t
h
e range o
f
Birr
1
000
to
25
00
wh
ere
3
.
1
4
tim
es
m
o
re lik
ely to
sp
end
th
an
tho
s
e
wh
o
s
e in
co
m
e
g
r
eater
th
an Bir
r
250
0 on
AO
R=
3
.
14
,
95
% C
I
(
1
.9
6,
5
.
0
3
)
.
Th
is m
a
y b
e
b
ecause of
ho
u
s
eho
l
d
s
f
r
o
m
lo
w
i
n
co
me
l
e
vel
or
gr
o
up
i
s
obl
i
g
e
d
t
o
e
xpe
n
d
hi
gh
pr
o
p
o
r
t
i
o
n o
f
t
h
ei
r
i
n
com
e
t
o
cov
e
r t
h
ei
r m
e
di
cal
cost
. Heal
t
h
s
t
at
us
o
f
th
e
ho
usehold
s
, t
h
o
s
e
wh
i
c
h
h
a
s at least
on
e m
e
m
b
er
o
f
th
e
ho
usehold
with ch
ro
n
i
c illn
ess
wh
ere 6
.
04
t
i
m
e
s
m
o
re l
i
k
el
y
t
o
spen
d t
h
an t
h
ose w
h
o
di
d
not
h
a
ve
o
n
A
O
R
=
6
.
0
5
,
95
% C
I
(
1
.
0
8
,
33
.8
9
)
. T
h
i
s
coul
d be
th
o
s
e
ho
u
s
eho
l
d
s
with
chron
i
c illn
ess are mo
re
frequ
en
tly v
i
sit h
ealth
care in
stitu
tio
n
s
t
h
an
t
h
ose who
d
i
d
n
o
t
have
.
In
th
is stud
y p
r
edo
m
in
an
tly
v
i
sited
p
r
iv
at
e h
ealth
care in
stitu
tio
n
where statistical
l
y
sig
n
i
fican
t
,
th
o
s
e
h
o
u
s
eho
l
d
s
who
were predo
m
in
an
tly
v
i
sited
prim
ar
y h
ealth
care in
stitu
tio
n
s
where 11
.7
3
tim
e
s
m
o
re
lik
ely to
sp
end for h
ealth
care th
an
tho
s
e
who
predo
m
in
an
tly u
s
ed
p
u
b
lic h
ealth
care i
n
stitu
tio
n
.
Th
is may b
e
d
u
e
to
t
h
e
fact
th
at priv
ate
h
e
alth
in
stitu
tio
ns h
a
v
e
h
i
gh
serv
ice cost th
an
p
u
b
lic h
ealt
h
i
n
stitu
tio
n
s
.
A
ccord
ing
to th
is stud
y,
study h
o
u
s
eho
l
d
s
w
h
o
w
e
re
p
a
yin
g
an
y ed
u
cat
io
n
a
l
f
ee
w
e
r
e
3
.
39
tim
es
m
o
re l
i
k
el
y
t
o
spe
nd
fo
r t
h
ei
r
ho
use
h
ol
d
he
al
t
h
care t
h
a
n
t
hose
w
ho
di
d
not
pay
.
T
h
i
s
i
s
m
a
y
be due
t
o
t
h
e
fact th
at as exp
e
nd
itures
for
d
i
fferen
t
activ
ities in
creas
es,
p
e
op
le
b
eco
m
e
stressed
an
d th
eir
h
ealth
cond
itio
n
mig
h
t
b
e
d
i
srup
ted
.
As a
result, th
ey wo
u
l
d
b
e
forced to
v
i
sit h
ealth
care i
n
stitu
tio
n
s
.
St
udy
h
o
u
seh
o
l
ds wh
o ha
d any
debt
t
o
be
pai
d
w
h
ere
2.
98 t
i
m
es
m
o
re l
i
k
el
y t
o
spen
d fo
r t
h
ei
r
ho
use
h
ol
d heal
t
h
care t
h
an t
h
ose w
ho
di
d n
o
t
have any
debt
on A
O
R
=
2.
9
8
, 9
5
% C
I
(
1
.
1
2, 7
.
9
8
)
.
Thi
s
i
s
m
a
y
b
e
du
e to
th
e
fact th
at as exp
e
nd
itures fo
r
d
i
fferen
t
activ
ities in
creases,
p
e
op
le b
e
co
me stressed
an
d th
eir
h
ealth
co
nd
ition
m
i
g
h
t
b
e
d
i
sru
p
t
ed
.
As a resu
lt, th
ey
wou
l
d b
e
fo
rced
to v
i
sit h
ealth
care i
n
stitu
tio
n
s
.
4.
C
O
N
C
L
U
S
ION
A
N
D
R
E
COMM
ENDATION
The study tries to assess
m
a
gnitude
of
out of pocket health care e
xpe
nditure a
nd ass
o
ci
ated factors
am
ong st
u
d
y
h
ous
eh
ol
ds a
nd
t
hose w
h
o wer
e
i
n
t
h
e l
o
w i
n
com
e
l
e
vel
wh
ere spe
nd m
u
ch m
o
re pro
p
o
rt
i
on o
f
th
eir in
co
m
e
. Mo
re th
an
75% o
f
stud
y p
a
rticip
an
ts in
th
i
s
stu
d
y
sp
en
d
≥
3%
of their i
n
com
e
for hea
lth care
service that means, t
h
ey were spend m
u
ch
m
o
re tha
n
t
h
e go
ve
rnm
e
nt
ann
o
unce
s
t
o
pay
fo
r f
o
rm
al sect
or
e
m
ployees to
be the m
e
m
b
er of social
health ins
u
ra
n
ce
but
t
h
ey
d
o
n
o
t
ha
v
e
heal
t
h
i
n
s
u
ra
nce c
ove
ra
ge.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PH
S I
S
SN
:
225
2-8
8
0
6
Magnitude of Out of
P
o
cket Health
Ex
penditures
and
As
s
o
ciated F
a
ctor
s Among
.... (M
ekuane
n
et Ge
r
e
mew)
33
7
The st
udy
s
h
o
w
ed t
h
at
i
n
c
o
m
e
of t
h
e
h
ous
ehol
d,
heal
t
h
s
t
at
us o
f
t
h
e
ho
use
hol
d at
l
east
one m
e
m
b
e
r
o
f
t
h
e
h
o
u
s
ehold
with or
witho
u
t
chron
i
c illness,
p
r
ed
o
m
in
an
tly u
s
ed
health
in
stitu
tion
eit
h
er
priv
ate
o
r
pu
b
lic
h
ealth
in
stitu
tio
n
s
, edu
cational fee at least
for on
e m
e
m
b
er of th
e fam
i
l
y
an
d
d
e
b
t
on an
y
m
e
m
b
er o
f
the
ho
use
h
ol
d f
o
u
nd t
o
ha
ve s
t
at
i
s
t
i
c
al
l
y
si
gni
fi
cant
ass
o
ci
at
i
on wi
t
h
ou
t
of p
o
c
k
et
heal
t
h
ex
pe
nd
i
t
u
re.
U
n
d
e
r
s
tand
ing th
e
p
r
op
or
tion
o
f
ou
t
o
f
pock
e
t ex
p
e
nd
itur
e
for
health c
a
re se
rvices
a
n
d ass
o
ciated
factors
m
u
st b
e
fu
nd
amen
tal fo
r bo
t
h
th
e gov
ern
m
en
t o
f
Et
h
i
op
ia
and form
al sector em
ployees to apply socia
l
health
i
n
su
rance
s
. B
a
sed
o
n
t
h
e
f
i
ndi
n
g
s
an
d c
oncl
u
si
o
n
s
o
f
t
h
e st
udy
t
h
e f
o
l
l
o
wi
ng
r
ecom
m
e
ndat
i
o
ns a
r
e
fo
rwa
r
ded;
I
n
th
is stud
y
mo
st o
f
ou
t of
po
ck
et h
ealth
car
e
e
xpe
nses c
o
vere
d by
o
u
t
o
f
reg
u
l
a
r i
n
c
o
m
e
. Theref
ore
t
h
e
go
ve
rnm
e
nt
of
Et
hi
o
p
i
a
sh
o
u
l
d
t
o
gi
ve e
d
u
c
a
t
i
on a
n
d
enc
o
u
r
agem
ent
fo
r
f
o
rm
al
sect
or e
m
pl
oy
ees.
There
we
re
a bi
g di
f
f
ere
n
ce bet
w
ee
n ho
use
hol
d out
of p
o
c
ket
heal
t
h
car
e
ex
pen
d
i
t
u
r
e
and
soci
al
heal
t
h
i
n
su
rance c
o
st
t
o
be pai
d
b
y
form
al
sector em
pl
oy
ees. Therefore social sect
or em
pl
oy
ees sh
o
u
l
d
b
e
encourage
d
t
o
use s
o
cial healt
h
insurance
.
Mo
r
e
r
e
sear
ch
is n
e
ed
ed on ou
t of
pocket
health care
expe
nditure a
n
d
ass
o
ciated factors
as
the
r
e was
no
su
ff
icien
t
study.
ACKNOWLE
DGE
M
ENTS
We are
ve
ry
grate
f
ul t
o
the
DebreMarkos
po
ly-techn
i
qu
e
Ad
m
i
n
i
strativ
e fo
r gran
tin
g u
s
to
comm
unicate to different departm
e
nts.
Our
special thanks
and si
ncere a
p
preci
at
i
o
n al
so
go t
o
su
per
v
i
s
ors
,
data collectors, study pa
rticipants
a
n
d teache
r
s for c
ont
ribution t
o
the
s
u
cc
ess of t
h
e
data
collection.
REFERE
NC
ES
[1]
World Summit on Sustainab
l
e D
e
velopment
,
“Johannesburg”, Gen
e
va, pp. 1-40, 20
02.
[2]
Health Care Fin
a
ncing Team
, Planning and Prog
ramming Depa
rtment, Federal Ministr
y
of
Health
,
Ethiopia'
s Thir
d
National Health
Accounts, 2
006.
[3]
Health
Car
e
Fin
a
ncing
Reform in Eth
i
opia, “Improvi
ng Quality
and Equity
”, Add
i
s Ababa, Ethiop
ia, 2012.
[4]
Ethiopia CSA,
EthioInfo
,
2007
. Retriv
ed
from; http://www.csa.gov.et/
.
Evaluation Warning : The document was created with Spire.PDF for Python.