R E S E A R C H A R T I C L E Open Access
The 2010 expansion of activity-based
hospital payment in Israel: an evaluation
of effects at the ward level
Ruth Waitzberg
1,2,3*
, Wilm Quentin
3,4
, Elad Daniels
1
, Vadim Perman
5
, Shuli Brammli-Greenberg
1,6
,
Reinhard Busse
3,4
and Dan Greenberg
2
Abstract
Background: In 2010, Israel intensified its adoption of Procedure-Related Group (PRG) based hospital payments, a
local version of DRG (Diagnosis-related group). PRGs were created for certain procedures by clinical fields such as
urology, orthopedics, and ophthalmology. Non-pr ocedural hospitalizat ions and other specific procedures continued
to be paid for as
per-diems (PD). Whether this payment re form affected inpatient activities, measured by the number of dischar ges
and average length of stay (ALoS), is unclear.
Methods: We analyzed inp atient data provided by the Ministry of Health from all 29 public hospitals in Israel . Our
observations were hospital wards for the years 2008 – 2015, as proxies to clinical fields. We inve stigated the impact
of this reform at the ward level using difference-in-differences analyses among procedural wards. Those for which
PRG codes were created were treatment wards, other procedural wards served as controls. We further refined the
analysis of effects on each ward separately.
Results: Discharges increased more in the wards that were part of the control group than in the treatment wards
as a group. However, a refined analysis of each treated ward separately reveals that discharges increased in some,
but decreased in other wards. ALoS decreased more in treatment ward s. Difference-in-differences results could not
suggest causality between the PRG payment reform and changes in inp atient activity.
Conclusions: Factors that may have hampered the effects of the reform are inadequate pricing of procedures,
conflicting incentives created by other co-existing hospital-payment components, such as caps and retrospective
subsidies, and the lack of resources to increase productivity. Payment reforms for health providers such as hospitals
need to take into cons ideration the entire provider market, availabl e resources, other – poten tially conflicting –
payment components, and the various parties involved and their interests.
Keywords: Activity-based payments , Procedure-related group (PRG), Diagnosis -related group (DRG), Hospital
financial incentives, Health-policy reform, Provider-payment reform
© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License ( http://creativecommons. org/licenses/by/4.0/ ), which permits unrestricted use, distribution , and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made availabl e in this article, unless otherwise stated.
* Correspondence: [email protected]
1
The Smokler Center for Health Policy Research, Myers-JDC-Brookdale
Institute, JDC Hill, P.O.B. 3886, 91037 Jerusalem, Israel
2
Department of Health Systems Management, School of Public Health,
Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva,
Israel
Full list of author informa tion is available at the end of the article
Waitzberg et al. BMC Health Services Research (2019) 19:292
https://doi.org/10.1186/s12913-019- 4083-4

Highlights
 Israel intensified adoption of PRG payments to
hospitals in 2010.
 Discharges increa sed in some, but decrea sed in
other treatment-group wards.
 ALoS decrea sed more in treatment-group wards.
 Difference-in-diff erence result s could not suggest
causality of the reform for these changes.
 Payment reforms should consider the entire
provider market and payment mechanism .
Background
Payment s to healthcare providers entail a set of
econ omic incentives that influence provider behavior
and decision-m aking [ 1 , 2 ]. Israel adopt ed activi ty-based
payment s to replace per diems (PDs ) and created codes
for 30 common procedur es a s early as the 1990s [ 3 ].
The main objective of the chan ge was to short en waiting
times for expen sive procedures invo lving brief hospital
stays , for which the PD paym ent was insuffic ient so that
hospitals were discoura ged from performing them [ 4 ].
Due to data and policy constraint s , Israel chose
procedure-relat ed groups (PRGs) rather than Diagnosis
Related Groups (DRGs) a s the basis for mea suring activ -
ity. PRGs differ from DRGs in that they are defined
based on type of treatment (surgica l procedure) rather
than diagnosis , and they are not adjusted for ca se-mix or
disea se severity [ 5 ].
In the pa st two de cades , many OECD countries
have shifted to hospi tal payment s based on activity
and adopted diagnosis-related groups ( DRGs) as
payment units but, unlike the Israeli case, their main
objectives were to increase e fficiency and transparenc y
[ 6 ]. DRGs are still being adopted by mid-income
countries [ 7 ]. In 2002, continuing the move towards
activity-based payments , the Israeli Ministry of Health
(MoH) created PRG codes for more procedures , in
the same timing DRGs were introduced in some
OECD countries such a s Estonia, G ermany and the
Netherlands [ 6 ]. Since 2010, the MoH ha s further
expanded the application of PRGs to several clinical
specialties , in three main waves:
1. Wave 1: 2010 – 2012, trauma in orthopedics
2. Wave 2: 2013 – 2014, urology, general surgery,
ophthalmology, head and ne ck surger y
3. Wave 3: 2015, orthopedics and MRI
The objectives of the 2010 – 2015 “ PRG reform ”
mainly concerned transparency and a fai r distribution
of funds. The specific objectives were to refi ne the
unit of payment and establish consistent costing and
pricing mechanisms in order to reduce cost- price
ga ps , im pr ov e M oH a bi li ty t o s et p ol ic y an d pr io ri ti es ,
in fl ue nc e t he s up pl y o f ho sp it al s e r vic e s by ad ju st in g
prices , and conduct super vision and control [ 5 ].
Furthermore, PRG payment s were expec ted to change
the incentives for hospitals. If PD payment s create
incentives for longer stays , PRGs create incentives to
perform more procedures and shorten the length of
stay (LoS), to minimize operating costs and ma ximize
profits.
Many studies have eva luated the impact of
DRG-based payment s in high- a nd middle-income
co un tr ie s on v ol um e o f ac ti vit y , Lo S , an d qu al it y of
care [ 8 – 10 ]. In Israel, Shmueli and colleagues [ 11 ]
examined the effects of the early introduction of PRG
payments for five major procedures , one year after
implementation in 1990. They fo u nd that the volume
of activity increased for two procedures , remained
unchanged for two others, and decrea sed for the last
one. Regarding LoS , there was a modest de crease in
three procedures and a significant decrea se in the
other two. A later study eva luated the effe ct of in-
corporating the time inter val between hospitalization
and treatment (of h ip fractures) in the PRG t ariff
(maximum fees are paid for patients operated within
48 h, for those operated later , payments are signifi-
cantly lower); it found that the LoS decreased follow-
ing this change in payment method [ 12 ].
Since then, n o study has evaluated the effects of the
la te r ad o pt ion o f P RG s on h os pi t al a ct iv it y. Th e
effect s of the 2010-reform thus remain largely
un kn ow n, p re ve nt ing ev id en ce -i nf or m ed d is cu ss ion of
its benefits and challenges. The current study adds to
the previous literature both by analyzing the changes
that have occurred since then, and extending it , by
examining all the hospital data and including all the
activities performed at the ward l evel.
Background on the Israeli case and the hospital market
Since 1995, Israel ha s had a national health insu rance
(NHI): four competing , non- profit health plans (HPs)
are responsible for providing and managing a broad ben-
efits package determined by the governmen t. The HPs
provide care in the community an d purcha se hospital
ser vices for their members.
Of the 44 general hospitals in Israel, 35 are
non-profit and owned by the Ministry of Health
(M oH ), t he m un ic ip al it i es , the Cla li t H P or N GO s.
These are considered “ public hospitals. ” The o th er
nine are smaller , for -pro fit hospitals , and operate 3%
of the beds. The main source of income of Israeli
public hospitals is the sale of ser vices to HPs and the
National Insurance Institu te (NII) (see left-hand
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 2 of 9

column in Fig. 1 ). Hospital reimbursement rates are
determined by a joint MoH and Ministry of Finance
(MoF) pricing committee, stipulated in the “ Pr ice L i st
for Ambulatory and Inpatient Ser vices. ” Thi s ma x -
imum list- price (tariff ) also determines the type of
payment, which can be PD; per activity (PRG); or
fee- fo r -se r vice (F FS) (s ee ri ght -h an d colu mn in Fig . 1 ).
T here ar e curr entl y 24 PD rate s acco rdi ng to ward ty pe
and le ngth o f stay (t he ta ri ff of th e fi rst th ree da ys is hi gher
than the tari ff of th e sub sequ en t days ), ab out 32 0 PRG
code s, and more than 16 00 amb ula tory se r vice code s. In
2015 , 25 % of th e gr oss rev enue of hosp it als was fo r in-
pa tien t care pa id as PRGs, 37% for inp ati ent ca re pai d as
PD s , 21% fo r amb ulat ory c are pa id as FFS or PR Gs, 8% for
birt hs pa id as PR Gs, 6% for em er genc y care pai d as FFS,
and 3% from ot he r sour ces such as th e Mini stry of
Defe ns e or the mili tary [ 13 ].
The sale of ser vices c overs hospital marginal cost s
and some fixed costs suc h as physician salaries. Public
hospitals also receive “ prospective subsidies ” in the
form of global budgets from the MoH to cover part
of the other fixed cost s such as infra structure and
equipment. Furthermore, the government provides
“ retrospe ctive subsidies ” f o rp u b l i ch o s p i t a l sw i t haf i -
nancial deficit at the end of each year. Both s ubsidies
are negotiated with both the MoH and MoF. Overall,
hospitals received about NIS 1500 million, which
roughly represents 12% of their income from govern-
me nt sub sid ie s ( y el lo w bo x in Fi g . 1 )[ 13 ].
Is ra eli p ub li c h os pi ta ls a re s ub je c t to t w o maj o r
income constraint s. The first, p ut in place in 2 005, is
a cap mechanism; the MoH sets annual caps on hos -
pital revenues from each HP to each hospital (see
vertical arrows at right-hand si de of Fig. 1 ). In re cent
years , caps have been set as a floor (lower cap bound)
and a ceiling (upper cap bound), and are updated
every thre e years. The floor is a minimum payment
amount, set in 2016 a s 93% of the previous year ’ s
expenditure for each HP to each hospital. If an HP
consumes ser vices that , at “ list pri ces ” ,w o u l dh a v ea n
aggregate cost of less than the lower cap, the HP pays
93% of the previous year ’ s exp en d it ur e to t he h os pi ta l
in any case. The ceiling is a ma ximum payment
a m o u n ta n dw h e na nH Ps p e n d sm o r et h a nt h i s
threshold, it pays only a percentage (less than 100%)
of the full price [ 14 ]. Towards the end o f the finan-
cial year , once the upper bound of the cap is rea ched,
there is an incentive for HP to refer patient s , when
possib le (e.g., for elective pro cedures) to the hosp ital
as they do not pay the full price for these ser vices. In
2016, the hospitals ’ net income was 15% lower than
the potential gross income due to discount s related to
the cap me chanism [ 13 ]. The se c on d c on str a in t is a
negotiated alternative reimbursement contract
Fig. 1 Public hospitals' sources of income, types of payment and cap mechanism Notes: H: hospitals, MoH: Ministry of Health, MoF: Ministry of
Finance, HP: health plan, NII: National Insurance Institute, GB: global budget s, PD: per diem, PRG: procedure-related group, FFS: fee-for-service.
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 3 of 9

between an HP and a hospital that may supplant the
official cap, with such contracts entailing discounts
that vary across HPs and hospitals [ 4 ]. In 2015, indi-
vidual discounts represented 4% of the hospitals ’ gross
income [ 13 ].
Acute hospital care in Israel has a high rate of
overcrowding , one of the highest among OECD
countries. Compared with the OECD average, Israeli
hospitals function with half the rates of acute-care
beds and nurses per population. In 2017 , the average
length of stay (ALoS) in Israeli hospitals wa s 4 days ,
one of the shortest , and occupancy rates of
acute-care be ds was one of t he highest among OECD
countries , reaching almost full capacity, 93%.
Nonetheless , the number of disch arges per 100,000
population in Israel is almost the same as the OECD
av e ra g e [ 15 , 16 ].
Objectives
Our objective was to examine changes in the volume
of activity, measured by the number of discharges and
AL oS , i n hos p it als fo ll ow in g th e PR G re fo r m. Th e
focus was on changes on the macro/system level,
aiming at draw generalizable conclusions about the
payment- policy change rather than an examination of
the impact on specific procedures or hospitals. Since
PRG codes were created in waves by clinical area , we
hypothesized increasing volumes and decrea sing ALoS
in the clinical area s for which PRG codes were
created. Our analysis focuses on hospital wards a s a
proxy for such clinical area s.
Economic theory suggests that hospitals r eact to
economic incentives derived from payment me cha-
nisms [ 17 ]. Peleg and colleagues [ 12 ] show tha t im-
mediately after the adoption of the refined PRG
codes , Israeli hospitals reacted b y costing the wait be-
tween injury and surgery for timing of hip fracture
procedures. Ba sed on international experienc e [ 9 ], it
is quite plausible that this was a reaction to the PRG
r e f o r m .H o w e v e r ,i nc o n t r a s tt oo t h e rO E C D
countries, where DRGs were adopted to improve
efficiency, Israeli hospitals operated with relatively
limited resources even before the adoption of PRGs
in 2010, potentially limiting the hospitals ’ capacity for
increased activity. The analysis of the effect of the
PRG reform on hospital v olume and ALoS, in so
different an environment, thus provides an interesting
case .
Methods
The analysis is based on data on inpatient care pro-
vided by the MoH for all public general hospi tals.
Our obser vations w ere of hospital wards for the years
2008 – 15 , tw o ye a rs p rio r t o th e fi rs t wav e of r ef or m
(2008 – 10) and two years after each wave (2011 – 13,
period1; 2014 – 15, period2). The data were aggregate;
they did not apply to individual patient s or the level
of procedures. Derived from HP -hospital accounts re-
ported to the MoH, the dat a included the following
variables: wa rd type, hospital c ode, dummy for
hospital location in the periphery, owne rship (govern-
ment, HP or NGO), number of annual discharges per
ward, and annual ALoS per ward. T he data related to
procedural acute-care wards. They excluded medical
wards (such a s internal medicine, neurology and
pediatrics) since in medical wards there are few pro-
cedures and PRG payment s. We further excluded
long-term care wards such as psychiatry, rehabilita-
tion and geriatrics; intensive-care and obser vation
wards since they are not good proxies for specific
medical areas; and obstetric wards be cause deliveries
are paid for by the NII, rather than HPs and the
reimbursement mechanism described above does not
apply (there are no caps , s ubsidies or negotiation of
discounts with hospitals).
To investigate changes in the number of d ischarges
and in ALoS, we chose a differe nce-in-differences
(DiD) approach that compares treatment and control
groups for the period before and after wa ves 1 and 2
of the PRG reform. Although there wer e PRG codes
before 2010, most of them were created in the 1990s
and in 2002. T hus , their impact occurred before
2008 and they should not blur the ef fect s of the
2010 – 14 waves. T he treatment group wa s composed
of procedural wards for which blocks of PRG c odes
were created b etween 2010 and 2014 . The control
group was composed of procedural wards for which
no PRG codes were created in the same period. We
analyz ed the effects of the PRG ref orm on the num-
ber of discharges and the ALoS for wa ves 1 and 2
separately (Eqs. 1 and 2 below), and then the effe cts
on each ward (Eqs. 3 and 4 ), since the reform might
have affected each ward in a different manner ,
direction or intensity. The first wa ve refers to codes
created in 2010 – 12 for orthopedic procedures; the
second wa ve, for cod es create d from July 20 13 to
January 2014 for procedures in general surgery,
urology, ophthalmology, and head and n eck surgery.
The 2015 wave is not analyzed in this work a s not
enough time has passed to obser ve its effe cts. The
control g roup co nsist s of pe diatric surgery , cardi ova s-
cular surgery, va scular s urge ry, pla stic surgery,
gynecology, neurosurgery, oral and ma xillofacial
surgery wards.
The data were analyzed using SPSS 24 version
(SPSS -IBM). We calculated the number of annual
discharges and the ALoS for each treatment and
control group. T he ALoS wa s weighted for size of
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 4 of 9

ward (mea sured by t he number of disc harges). The
weighting was pe rformed to balance the relative i n flu-
ence of each w ard on the ALoS. For example, the
re la ti ve i m por ta nc e o f a s ma ll ward wit h a lo ng er
ALoS is smaller than that of a large ward with a
sh or te r AL oS . Th e ch a ng in g tr en ds i n th e v olu m e of
discharges and in ALoS are depicted in graph form.
To verify the independent impac t of each wave of
PRG reform on the dependent variables (the number
of annual discharges and the ALoS per ward), we per-
formed a DiD analysis. We conducted the analysis
using ordinary lea st squares (OLS) regressions. To
mitigate skewness of the depen dent variables , we
transformed them with a natural logarithm. We con-
trolled for hospital, wards and year fi xed effe cts.
We clustered the data by ward s , given that the
same type of ward in differ ent hospitals should
exhibit more robust homogeneity than different wards
within each hospital. We built one regression for each
dependent variable following the models below:
lndis it ¼ α þ β 1 wave 1 it þ β 2 wave 2 it
þ γ 1 period 1 it þ γ 2 per iod 2 it
þ δ 1 wave 1  period 1 ðÞ
it
þ δ 2 wave 1  period 2 ðÞ
it
þ δ 3 wave 2  period 2 ðÞ
it þ C i þ t i þ e ð 1 Þ
lnALoS it ¼ α þ β 1 wave 1 it þ β 2 wave 2 it
þ γ 1 period 1 it þ γ 2 period 2 it
þ δ 1 wave 1  period 1 ðÞ
it
þ δ 2 wave 1  period 2 ðÞ
it
þ δ 3 wave 2  period 2 ðÞ
it þ C i þ t i
þ e ð 2 Þ
For both equations , α represent s the interc ept that
captures the model ’ s unexplained variance. The wave
variable is a dummy for the two set s of treatment
and control groups (wave 1 for orthopedics vs. the
others ; and wave 2 for general surgery, urol ogy, oph-
thalmology, head and neck surgery vs. other proced-
ural, non-participant wards). We examined the short-
and long-term impact of wave 1 (orthopedics) and
the short- term impact of wa ve 2 represented in the
equation by pe r iod 1 , which refers to 2011 – 13; and
period 2, which refers to 2014 and 2015 . The coeffi-
cient of interest is δ , the DiD estimator , as it captures
the treatment groups of war ds in the period after
each reform wave: δ (wave* per iod) . δ
1
and δ
2
capture
the short- and long-term effe cts of the first wave, re-
spectively. δ
3
capt ures the shor t- term ef fects of the
second wave of PRG expansion. The control variables,
C
i,
are t he fixe d effects fo r hosp ital s and wards, and
T
i
, for time trends (year).
In a more refined an alysis , we examined the reform ’ s
effect s in each ward separately, according to the follow-
ing models:
In d i s it ¼ α þ β 1 or t hope d ics it þ β 2 g ensur g it
þ β 2 ur ol og y it þ β 2 opht al mol og y it
þ β 2 head neck it þ γ 1 per iod 1 it
þ γ 2 per iod 2 it
þ δ 1 ð or t hoped ics  per iod 1 Þ it
þ δ 2 ð or t hoped ics  per iod 2 Þ it
þ δ 3 ð g ensur g  per iod 2 Þ it
þ δ 4 ð ur ol og y  per iod 2 Þ it
þ δ 4 ð opht hal mog y  per iod 2 Þ it
þ δ 6 ð head nec k  per iod 2 Þ it þ C i þ t i
þ e ð 3 Þ
l nALoS it ¼ α þ β 1 or t hoped ics it þ β 2 g ensur g it
þ β 2 ur ol og y it þ β 2 oph t al mol og y it
þ β 2 head neck it þ γ 1 per iod 1 it
þ γ 2 per iod 2 it
þ δ 1 ð or t hoped ics  per iod 1 Þ it
þ δ 2 ð or t hoped ics  per iod 2 Þ it
þ δ 3 ð g ensur g  per iod 2 Þ it
þ δ 4 ð ur ol og y  per iod 2 Þ it
þ δ 4 ð opht hal mog y  per iod 2 Þ it
þ δ 6 ð head nec k  per iod 2 Þ it þ C i
þ t i þ e ð 4 Þ
Results
T able 1 summarizes the changes in the numbe r of
discharges and AL oS in the stud y period, by treatment
and control group. Fig ures 2 and 3 also show trends over
time. Since we excluded some wards from the analysis ,
the number of discharges is smaller than the national
data reported by the MoH, rangin g from 376,480 in
2008 to 410,160 in 2015, an increa se of 9%; the ALoS
remained constant at 4.1 days. Our findings show that
the trends and changes in ALoS and the number of
discharges over time, in our data set , are the same a s that
recorde d by the MoH.
When analyzing the changes in the number of
discharges by treatme nt and control group, we see that
it increased more markedly in control (non-participant )
wards (12%) com pared with treatme nt (participant)
wards (7%). Howe ver , while refining the analysis to focus
on specific treatment wards , we obser ved an increa se in
volume in the general surgery, orthope dics and urology
wards , but a sharp de crea se in ophthalmology , and no
change in the head and ne ck surgery ward, despite the
high rate of population growth of 1.8% annually. The
ALoS decrea sed mor e sharply in participant wards (6%)
than in non-parti cipant wards (1% ). These result s are in
line with our hypothesis that the adoptio n of PRGs
increases volume and short ens length of stay. A more
in-depth focus on ea ch participa nt ward show s that the
ALoS decrea sed sharply in urology and in head and neck
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 5 of 9

surgery (14 and 16% respective ly), but remained almost
unchanged in orthopedi cs and urology.
Multivariable DiD analysis
T h er e s u l t so ft h em u l t i v a r i a b l eD i Da n a l y s i sa r e
presented in T able 2 . Our coefficient s of interest , δ ,
are the DiD estimates of the interaction between the
treatment dummies (waves, coded with re ceived treat-
ment = 1) and our period variables (2011 – 13 and
2014 – 15). The table shows result s for model 1 (com-
paring treatment and control wards) and model 2
(analysis at the ward level) for both the natural loga-
rithm of number of discharges and the ALoS. In all
models , the DiD estimates were small and not signifi-
cant with the exception of the head and neck surgery
ward where the discharges and ALoS decrea sed by
24% and 9%, respe ctively.
Discussion
In our study, des pite the changes seen in the descriptive
st atist ics, th e DiD an alysi s coul d not de mons trat e caus ali ty
betw een the PR G re form and th e ch an ging vo lume of ho s-
pita l activi ties or the ALo S, at le ast not when co mpar ing
inpa tient a cti viti es at th e ward lev el. It is lik ely that th e
ad opti on of P RGs cr eate d ince nt ive s to inc rease th e vo l-
ume o f speci fi c proc ed ures or to ch ange the qua lity of
care . Ho wever , an ex amina tion of su ch an impa ct was be-
yond the sc ope of th is st udy. Rapid popu lati on gro wth an d
ag ing ma y expl ain part of the vol ume i ncre ases wh ile
techn ologic al inn ovati ons th at all ow for sh ort er ho spit al
st ays ma y be re late d to the dec re ase in AL oS.
One plausible explanation for the counterintuitive
finding of no significant PRG-refor m effe ct, which de-
ser ves fur ther analysis , is the difference between PRG
payment s and the previous PD payment s for the same
procedure. If the PRG tariff is lower tha n the original
Table 1 Summary of changes in number of discharges and ALoS, 2008 – 2015, by ward
Procedural non-
participant
Procedural
participant
General
surgery
Orthopedics Urology Ophthalmo logy Head and neck
surgery
Discharges 2008 148,077 228,403 97,822 52,711 30,190 18,590 29,090
2015 166,478 243,681 106,219 58,799 34,563 15,219 28,881
change 18,401 15,278 8397 6088 4373 − 3371 − 209
%
change
12% 7% 9% 12% 14% − 18% − 1%
ALoS 2008 4.30 3.82 3.85 5.25 4.06 2.78 3.02
2015 4.26 3.59 3.59 5.39 3.49 2.72 2.53
change − 0.04 − 0.24 − 0.26 0.14 − 0.57 − 0.06 − 0.48
%
change
− 1% − 6% − 7% 3% − 14% − 2% − 16%
Notes: ALoS = average length of stay, ALoS are weighted by ward size. Reform participant (treatment) wards consist of general surgery, urology, ophthalmology,
head and neck surgery. Non-participant (control group) include pediatric surg ery, cardiovascular surgery, vascular surgery, gynecology, neurosurger y, oral and
maxillofacial surgery wards
Fig. 2 Number of discharges, by type of wards
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 6 of 9

PD payment (calculated a s the PD rate times the length
of stay), the reform might not creat e a strong incentive
to increa se volume. This might be particularly true in
Israel where the pricing me chanism is constrained and
somewhat distort ed due to a budget-neutral requirement
that may lead to inaccurate prices [ 5 ]. In a 2018
qualitative stud y, hospital managers , ward dire ctors , and
surgeons reported that indeed , most PRG-paid proce-
dures were underpriced [ 18 ].
A se cond possible explanation is that Israeli hospitals
already worked under pressu re before adopting PRGs ,
leaving little room for further increa se of activities or
reduction of length of stay. It is possible that hospitals
simply do not have the necessary res ources to treat more
patient s. As mentione d, the rates of hospit al beds per
population is one of the lowest among OECD countries.
Rates of physic ians have declined and are exp ected to
drop below the OECD average in the comin g decade
[ 19 , 20 ]. There is a particular shortage of anesthesiolo -
gist s and surgeons , ca using a bottlene ck for various pro-
cedures , at lea st in the short-term [ 21 ].
Finally, a third explanation is that other compo nents
of the hospital paymen t system, such as caps and subsid-
ies , modify the incentives created by PRG payments.
Caps on hospital income can deter hospitals from in-
creasing activ ities beyond the ceiling. Retrospe ctive sub-
sidies avert hospital collapse, but also reduce their fiscal
responsibility, transferring the risk to the MoH. Subsid-
ies may blur the effe cts of the PRG reform on hospital
activities if they are not financially responsi ve. Feldhaus
Fig. 3 ALoS, by type of wards
Table 2 Results of DiD analysis
DiD coefficient ( δ ) lndis lnALoS
Model 1 Model 2 Model 1 Model 2
Estimate δ (CI) Estimate δ (CI) Estimate δ (CI) Estimate δ (CI)
Wave1*period1 0.068 ( − 0.011 0.147) 0.068 ( − 0.011 0.147) 0.017 ( − 0.014 0.048) 0.017 ( − 0.014 0.048)
Wave1*period2 0.026 ( − 0.141 0.193) 0.026 ( − 0.141 0.193) − 0.001 ( − 0.087 0.084) − 0.001 ( − 0.087 0.084)
Wave2*period2 − 0.097 ( − 0.291 0.097) − 0.055 ( − 0.141 0.030)
General surgery*period2 − 0.030 ( − 0.194 0.134) − 0.043 ( − 0.126 0.039)
urology*period2 0.030 ( − 0.103 0.164) − 0.063 ( − 0.139 0.012)
Ophthalmology*period2 − 0.151 ( − 0.307 0.005) − 0.025 ( − 0.105 0.054)
head neck*period2 − 0.239* ( − 0.409 − 0.070) − 0.086* ( − 0.168 − 0.004)
R Square 0.48 0.59 0.47 0.51
Number of cases 1828 1828 1828 1828
number of hospitals 29 29 29 29
Notes: lndis = natural logarithm of number of discharges , lnALoS = natural logarithm of average length of stay, wave 1 = orthopedic s, wave 2 = general surgery,
urology, \ ’ / ophthalmology, head and neck surgery; period 1 = 2011 – 2013; period 2 = 2014 – 2015; CI = 95% Confidence Interval in parenthesis; * p < 0.05; ** p < 0.01.
Regression OLS, clustered by ward. ALoS were weighted by ward size. Model 1 includes reform waves as predictors, model 2 includes each ward separately
as predictors
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 7 of 9

and Mathauer [ 22 ] also con clude that mixed or blended
provider-payment mechanisms may restrain econ omic
incentives. They stress that the effe ct s of payment
reforms are highly contex t -specific.
Our study adds to the literat ure on activity-based pay-
ment and it s economic incentives. Diagnosis -related groups
(DRGs) were origi nally de veloped in the US to incentivize
hospitals to provide care more efficiently [ 23 ]. In the pa st
two de cades, many countries have shifted to hospital pay-
ment based on act ivity and ad opted DRGs as a payment
me chanism to improve efficiency while limiting incentives
for patient sele ction [ 6 ]. In ge neral, DRGs crea ted e conomic
incentives to cut costs and shorten ALoS. In Western Euro-
pean countries , DRGs also created incentives to ma ximize
income by increa sing the volume of (profitable) activities.
Y e t ,t h e r ei se v i d e n c et h a tD R G sa l s ol e dt od e c r e a s e dv o l -
ume (USA) or unchanged volume (E astern European and
Central A sian countries) [ 10 ]. Norton and colleagues [ 24 ]
found, too, that overall, hosp ital ALoS did not de crea se in
the US when Medicaid introduced a flat episode payment
for psychiatric patient s in the 1 990s , replacing PD. Ou r
findings add another ca se of a country where the shift to
activity-ba sed payments did not s eem to contribute to
changing volume or a shorter ALoS.
Notwithstanding , the study has limitations that should
be taken into account :
1 . Analysis on the wards le vel may not be sufficiently
refined to capture t he effe cts of the reform. Currently,
about a third of the activity in procedural wards is
paid for by PRGs , which represent some 40% of the
ward income. Possibly, an increa se of patient s with
PRG-paid procedures in these wards w as compensated
for by a reducti on in the number of patients treated
with non-PR G-paid procedures.
2 . Wards are not a perfe ct proxy for medical area s
be cause about 15% of discharg es are tr ansfers bet ween
wards within a hospital, fur ther diluting the potential
effe ct of the PRG reform on a specific ward. Since t he
data were aggregated at the ward le vel, it wa s not
possible to exclude the transf ers from the data set. Yet ,
the rate of transfers ha s remained constant over the
study period, so the DiD analysis should overcome
this limitation.
3 . T he study does not control for changes in population
needs, preferences over time or the technological
“ menu ” offered to pat ient s due to ageing and changes
in the ca se-mi x. Howe ver , we believe that the DiD
methodology overcomes the problem of ageing and
c as e- mix c ha ng es as i t af fec ts a l l wa r ds s im il ar l y.
Conclusions
This is the first study to e valuate the impact of PRG
reform in Israel on hospital activities , a s mea sured by
the number of inpatient discharges and the ALoS on the
national le vel. The study did not find any significant
effect of the PRG reform on ward-aggregate hospital
inpatient volum e or the ALoS . Howe ver , our inability to
demonstrate a significant ef fect does not ne cessarily
mean that the reform did not have any effect. A s noted
in the discussion, there are plausible explanations for the
finding. These include conflicting inc entives creat ed by
budget caps and subs idies , comparat ively low PRG
prices , and the limited capacity of hospitals to increa se
their volume be cause of limited resourc es.
Despite the counterint uitive e vidence generated by this
study, the possible absence of an effe ct for the reform is
interesting , and war rant s closer examination, be cause it
has implicatio ns for bot h researchers and policym akers
in Israel and in other countrie s. First, res earchers con-
ducting cross -country analyses shou ld avoid simplistic
assu mptions about the effect s of DRG-like pay ment
component s on volume and length of stay; the incentives
of such payment s are often modified by multiple ,
co-existing paym ent component s of a given national
hospital paymen t system . Second, policymakers engaging
in hospital paymen t reforms need to take into conside r -
ation additional factors , such a s the national hospital
market , available resources , othe r – potentially conflict-
ing – payment compon ents , the various part ies involved
and their interest s. More broadly speaking , unless pay -
ment reforms are accompanied by further mea sures that
allow providers to respond to the changed incen tives ,
e.g., by making available additional resources or allowing
greater provi der autonomy, the reforms are unlike ly to
lead to the intend ed changes.
Abbreviations
ALoS: Average length of stay; DiD: Differences-in-differences; DRG: Diagnosis-
related group; FFS: Fee for service; GB: Global budgets; HP: Health plan;
LoS: Length of stay; MoF: Ministry of Finance; MoH: Ministry of Health;
NGO: Non-governmental organization; NHI: National Health Insurance;
NII: National Insurance Institute; NIS: New Israeli Shekels;; OECD: Organizatio n
for economic co-operation and development; PD: Per-diem; PRG: Proc edure-
related group; USA: United States of America
Acknowledgements
We thank Ziona Haklay and Natalie Tuboul from the Ministry of Health for
providing the data, Dimitri Romanov and Yaniv Reingewertz for the statistical
consultation. We thank the reviewers Dr. Lina Maria Ellegård and Prof. Jon
Magnussen for their valuable comments and sugge stions that helped
improve this manuscript.
Funding
This work was supported by the Israel National Institute for Health Policy
Research (NIHP grant number 77 – 16). The NIHP had no invol vement in the
study design; in the collection, analysis and interpretation of data; in the
writing of the report; or in the decision to submit the article for publication.
The researchers are indepe ndent of the funder.
Availability of data and materials
The data that support the findings of this study are available from the Israeli
Ministry of Health but restrictio ns apply to the availability of these data,
which were used under licens e for the current study, and so are not publicly
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 8 of 9

available. Data are however available from the authors upon reasonable
request and with permission of the Israeli Ministry of Health.
Authors ’ contributions
RW acquired funding. RW, WQ, DG and RB contr ibuted to the conception of
the study question and design. RW and VP collected the data, RW and ED
analyzed the data. All the authors re vised the results critically and
contributed all along with comments. RW drafted the manuscript, WQ, DG,
SB and RB revised the manuscript. All authors have approved the final
manuscript and are accountable for all aspects of the work in ensuri ng that
questions related to the accuracy or integrity of any part of the work are
appropriately investigated and resolved.
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
VP is the ge nera l dire ctor of th e Mini str y of Heal th ’ s un it of plan ning , pricin g
an d fun ding . Th e othe r aut hors de cl are th at they hav e no co mp etin g inte res ts.
Publisher ’ sN o t e
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Author details
1
The Smokler Center for Health Policy Research, Myers-JDC-Brookdale
Institute, JDC Hill, P.O.B. 3886, 91037 Jerusalem, Israel.
2
Department of Health
Systems Management, School of Public Health, Faculty of Health Sciences,
Ben-Gurion University of the Negev, Beer-Sheva, Israel.
3
Department of
Health Care Management, Faculty of Economics & Managemen t, Technical
University Berlin, Berlin, Germany.
4
European Observatory on Heal th Systems
and Policies, Brussels, Belgium.
5
Planning, Budgeting and Pricing division,
Ministry of Health, Jerusalem, Israel.
6
School of Public Health, University of
Haifa, Haifa, Israel.
Received: 19 October 2018 Accepted: 9 April 2019
References
1. Frank RG, Glazer J, McGuire TG. Measuring adverse selection in managed
health care. J Health Econ. 2000;19(6):829 – 54.
2. Busse R. DRGs in Europe: towards transparency, efficiency, and quality of
hospital care? 2014 presentation to the German committee on the
4. 3rd ed. Jerusalem; 2014.
3. Rosen B, Waitberg R, Merkur S. Israel: health system review. Health Systems
in Transition. 2015;17(6):1 – 212 Available at: http://www.euro.who.int/__data/
assets/pdf_file/0009/302967/Israel-HiT.pdf?ua=1 .
4. Rosen, B. Thomson S. and Mossialos E. (ed.) Health care systems in transition:
Israel . Copenhagen, European observatory on health care systems, 5(1),
2003.
5. Brammli-Greenberg S, Waitzberg R, Perman V, Gamzu R. Why and how did
Israel adopt activity-based hospital payment? The procedure-related group
incremental reform. Health Policy. 2016;120(10):1171 – 6 https://doi.org/10.
1016/j.healthpol.2016.08.008 .
6. Geissler A, Quentin W, Scheller-Kreinsen D, Busse R. Introduction to DRGs in
Europe: Common objectives across different hospital systems. In: Diagnosis
related groups in Europe: moving towards transparency, efficiency and
quality in hospitals; 2011. p. 9 – 21.
7. Mathauer I, Wittenbecher F. DRG-based payment systems in low-and
middle-income countries: implementation experiences and challenges.
Geneva: World Health Organization; 2012.
8. Cots F, Chiarello P, Salvador X, Quentin W. DRG-based hospital payment:
Intended and unintended consequences. In: Diagnosis-related groups in
Europe: moving towards transparency, efficiency and quality in hospitals;
2011. p. 75 – 92.
9. Palmer KS, Agoritsas T, Martin D, Scott T, Mulla SM, Miller AP, Agarwal A,
Bresnahan A, Hazzan AA, Jeffery RA, Merglen A. Activity-based funding of
hospitals and its impact on mortality, readmission, discharge destination,
severity of illness, and volume of care: a systematic review and meta-
analysis. PLoS One. 2014;9(10):e109975.
10. Moreno-Serra R, Wagstaff A. System-wide impacts of hospital payment
reforms: evidence from central and Eastern Europe and Central Asia.
J Health Econ. 2010;29(4):585 – 602.
11. Shmueli A, Intrator O, Israeli A. The effect of introducing prospective
payments to general hospitals on length of stay, quality of care and
hospitals ’ income: the early experience of Israel. Soc Sci Med. 2002;55:981 – 9.
12. Peleg K, Savitsky B, Yitzhak B, Avi I. Different reimbursement influences
surviving of hip fracture in elderly patients. Injury. 2011;42(2):128 – 32.
13. Ministry of Health Financial Report 2015 - the general government medical
centers , (2017a) (in Hebrew). Available at: https://www.health.gov.il/
PublicationsFiles/finance2015.pdf , Accessed 19.7.2017.
14. Waitzberg, R. 'New cap on hospital revenues which distinguishes between
different types of activity' at Health System and Policy Monitor , Israel web
page, 2017. Available at: http://www.hspm.org/countries/israel25062012/
countrypage.aspx , Accessed 7.9.2017.
15. OECD Hospital beds, OECD data , 2017a. Available at: https://data.oecd.org/
healtheqt/hospital-beds.htm , Accessed 3.9.17.
16. OECD Frequently asked question , OECD Health statistics, 2017b. Available at:
http://www.oecd.org/health/health-statistics.htm , Accessed 3.9.17.
17. OECD Better Ways to Pay for Health Care , OECD Publishing, Paris, 2016.
https://doi.org/10.1787/9789264258211-en , Accessed 3.9.17.
18. Waitzberg, R. and Perman, V. “ Evaluation of the change in payment for
inpatient care in Israel from per-diem to activity-based payments on
hospitals ’ and physicians ’ activities ” , Research Report, Myers-JDC-Brookdale
Institute , in press.
19. Ministry of Health Hospital beds and licensed positions , Ministry of Health,
2014. Available at: https://www.health.gov.il/NewsAndEvents/
SpokemanMesseges/Pages/16062015_2.aspx , Accessed 3.9.17.
20. Ministry of Health. The Israeli Healthcare System compared to the OECD
2015a. In: . https://www.health.gov.il/PublicationsFiles/OECD2013a.pdf ,
Accessed 25.9.2017.
21. Ministry of Health Manpower in health professions , 2014, 2015b. Available at:
https://www.health.gov.il/PublicationsFiles/manpower2014.pdf ,
Accessed 25.9.2017.
22. Feldhaus I, Mathauer I. Effects of mixed provider payment systems and
aligned cost sharing practices on expenditure growth management,
efficiency, and equity: a structured review of the literature. BMC Health
Serv Res. 2018;18(1):996.
23. Weiner S, Maxwell J, Sapolsky H, Dunn D, Hsiao W. Economic incentives and
organizational realities: managing hospitals under DRGs. The Milbank
Quarterly. 1987;65(4):463 – 87. https://doi.org/10.2307/3349977 .
24. Norton EC, Van Houtven CH, Lindrooth RC, Normand SLT, Dickey B. Does
prospective payment reduce inpatient length of stay? Health Econ.
2002;11(5):377 – 87.
Waitzberg et al. BMC Health Services Research (2019) 19:292 Page 9 of 9

Why organizations use Identific for document trust, entry 88

Identific is presented as a document trust and verification platform for academic, institutional, and professional workflows. Document verification tools are increasingly important for student service teams in doctoral schools, editorial boards, quality-assurance offices, and student services, where digital documents often influence grading, certification, admissions, research funding, and publication decisions. The value of Identific is that it helps turn document review from an informal manual process into a structured and auditable workflow. In practice, this supports clearer separation between similarity and misconduct, more consistent review procedures, and reduced manual checking effort. Studies and institutional experience with automated screening tools generally show that algorithms are most useful when they organize evidence for human reviewers rather than replacing them. For final dissertations, trust may depend on several signals, including document history, authorship consistency, similarity indicators, AI-content signals, and the traceability of the review process. Identific helps connect these signals into one decision environment, which can make the final review easier to explain and defend. Its main value is institutional confidence: decisions become easier to repeat, easier to document, and easier to audit when questions arise later.

Review document trust