Healthcare systems and performance evaluation: comparison of performance indicators in v4 countries using models of composite indicators

133 3, XX, 2017 Business Adminis a ion and Managemen DOI: 10.15240/ ul/001/2017-3-009 In oduc ion Heal hca e sys ems ha e been p o iding clien s wi h se ices o an inc easingly highe quali y hanks o many posi i e de elopmen s in heal hca e, p ima ily he high le el o expe ise o heal hca e pe sonnel, imp o emen s in echnology, and speedie app oaches o ea ing illnesses. Howe e , he heal hca e sys em has s uggled wi h o he a ious nega i e aspec s: inapp op ia e s uc u ing, poo managemen , o ine ec i e i nancing. In ligh o hese nega i e aspec s and he suppo and de elopmen o he posi i e, hese sys ems should be managed p ope ly and hei pe o mance should be e alua ed (Hejduko á & Ku eko á, 2016b). Acco ding o Donabedian (1972), he de i ning goal o he heal hca e sys em is o imp o e he heal h o he popula ion. I heal hca e sys ems did no con ibu e o imp o ing heal h, we would choose no o ha e hem a all. The heal h o he popula ion should e l ec he heal h o indi iduals h oughou he cou se o hei li es and include bo h p ema u e mo ali y and non- a al heal h ou comes as key componen s. Many s udies highligh he ac ha heal h and heal hca e sys ems an impo an ole in he economy, see S oukal (2016) and poin o he link be ween social capi al and public goods (Son, 2016). In ega d o he sys em o heal hca e as a whole, esponsibili y o i s e ec i e ope a ion lies wi h he go e nmen o a gi en coun y. Public heal hca e policy mus hen de i ne s a egies and include heal hca e in public i nances on one hand and suppo p i a e ins i u ions in he a ea o heal hca e p o ision on he o he (WHO, 2000). Measu ing and e alua ing pe o mance in he i eld o heal hca e mus hen be included in public policy – o mo e see o example Wou e e al. (2010) o Ba ák (2010). In o de o measu e and e alua e pe o mance on he le el o heal hca e sys ems, many in e es ing indica o s ha e been c ea ed by e.g. he WHO, Eu os a , OECD Heal h S a is ics, o OECD Heal h Policy S udies – selec ed indica o s a e compa ed in he pape below. This pape begins wi h heo e ical esea ch on he issue and speci i es he basic aspec s o he selec ed heal hca e sys ems. The pape hen ocuses on p esen ing goals, me hodology, and da a, and also p o ides i s own model o composi e indica o s. The a icle con inues wi h a chap e de o ed o esul s and discussion on he opic be o e d awing i s i nal conclusion. 1. Li e a u e Re iew 1.1 Heal hca e Sys ems and Thei Pe o mance A high-quali y heal hca e sys em p o ides high- quali y se ices o all indi iduals who equi e hem. The basic building blocks o he heal hca e sys em di e om coun y o coun y; howe e , hese sys ems all need sou ces o i nancing, educa ed pe sonnel, and eliable in o ma ion (Hejduko á & Ku eko á, 2016a). Acco ding o Roeme ’s de i ni ion (2002), he heal hca e sys em is a combina ion o esou ces, o ganiza ion, i nances and managemen , which culmina es in he p o ision o se ices o a gi en popula ion. The Wo ld Heal h O ganiza ion o e s a di e en de i ni ion, HEALTHCARE SYSTEMS AND PERFORMANCE EVALUATION: COMPARISON OF PERFORMANCE INDICATORS IN V4 COUNTRIES USING MODELS OF COMPOSITE INDICATORS Pa lína Hejduko á, Lucie Ku eko á EM_3_2017.indd 133EM_3_2017.indd 133 7.9.2017 10:34:267.9.2017 10:34:26 134 2017, XX, 3 Ekonomika a managemen which desc ibes he heal hca e sys em as “all he ac i i ies whose p ima y pu pose is o p omo e, es o e and/o main ain heal h.” In p e ious yea s, his de i ni ion was expanded by he Wo ld Bank (2007) o include he p e en ion o domes ic po e y due o illness. Acco ding o he de i ni ion gi en by Plsek and G eenhalgh (2001), he heal hca e sys em is a complex adap i e sys em ha plays a signi i can ole in in l uencing heal hca e sys ems p ima ily in p oducing be e esul s in he i eld o heal h. In e ms o long- e m sus ainabili y, he heal hca e sys em mus ha e a clea concep . Acco ding o Kelley and Hu s (2006a), he concep ual amewo k o he cu en heal hca e sys em mus include he ollowing aspec s: e ec i eness, sa e y, esponsi eness, accessibili y, equi y and e i ciency. O he au ho s, e.g. Kacíř (2010), claim ha he ollowing c i e ia mus be ul i lled in o de o a heal hca e sys em o p ope ly unc ion: a ailabili y, quali y, easonable i nancial bu den, e ec i eness, equal access, and social accep abili y. Sassi e al. (2002) s a e ha he pe o mance o heal hca e sys ems is e alua ed as he pe o mance o indi idual ac i i ies in a heal hca e sys em. The WHO (2006) p o ides in e es ing insigh in o he heal hca e sys em in i s de i ni ion o he six pilla s o he heal hca e sys em – see Tab. 1. As is e iden in Tab. 1, he heal hca e sys em is p ima ily based on economic, social, legal, and poli ical amewo ks. We can also summa ize his by s a ing ha public policy plays he p ima y ole in each o he pilla s lis ed abo e. Measu ing and e alua ing he pe o mance o heal hca e sys ems on a na ional le el equi es analysis o he ela ionship be ween pe o mance and public alues – see Hood (1991) and his discussion o ideas known as „New Public Managemen “. Bouckae and Halligan (2008) de i ne pe o mance on h ee le els: mic o, middle, and mac o. I we ocus on he mic o le el acco ding o hese au ho s, he pe o mance o a gi en Pilla s o he heal hca e sys em Speci i ca ion 1. Se ice deli e y Supply Demand In as uc u e Managemen P o ec ion and quali y 2. Heal h wo k o ce Na ional employmen policy Legal amewo k No ms S anda ds 3. In o ma ion In o ma ion sys ems Moni o ing sys ems Global in o ma ion s anda ds 4. Heal h p oduc s and echnologies Policy Public ende s No ms Acces and quali y 5. Financing S uc u e o public i nance policy in he i eld o heal hca e Da a on heal hca e expendi u es Calcula ions 6. Leade ship/go e nance Sec o policies Moni o ing and egula ion In e en ion Ha moniza ion and uni i ed me hods Sou ce: own, based on WHO (2006) Tab. 1: Six basic pilla s o he heal hca e sys em EM_3_2017.indd 134EM_3_2017.indd 134 7.9.2017 10:34:277.9.2017 10:34:27 135 3, XX, 2017 Business Adminis a ion and Managemen si ua ion is e alua ed. On he middle le el, he au ho s mos ly poin o he pe o mance o a ce ain b anch (e.g. heal h ca e). On he mac o le el, discussions a e ypically led on he pe o mance o a coun y, e.g. coun ies o he OECD o EU. As s a es K echo ská (2014), e e y en i y should choose an app oach and a pe o mance e alua ion me hod ha will bes co espond o hei needs, goals and cu en si ua ion. App oaches o measu ing and e alua ing pe o mance in heal hca e ha e begun o expand signi i can ly in connec ion o he ins a emen and implemen a ion o heal hca e e o ms and changes in a i udes owa d heal h in de eloped coun ies (Há a & Maško á- Hanušo á, 2009a). These new app oaches and concep s s em p ima ily om sup ana ional ins i u ions, p ima ily he OECD, WHO, Eu opean Obse a ion, o he UN. In e alua ing he pe o mance o heal hca e sys ems, i is impo an o conside he ac ha no only he i nancial aspec s a e e alua ed; simul aneously, a g owing emphasis is also being placed on non- i nancial cha ac e is ics – o mo e, see e.g. Robe s e al. (2003), P eke e al. (2000), Donabenian (1982), o Smi h e al. (2008). I we s a e he basic goals o he ins i u ions and au ho s lis ed abo e, we can claim ha he WHO (2000) de i nes h ee goals in e alua ing he pe o mance o a heal hca e sys em. These a e: heal h, esponsi eness, and ai i nancing. On he con a y, he OECD (2004) and Kelly and Hu s (2006a; 2006b) app oach he issue o heal hca e sys em e alua ion in a b oade sense and ci e addi ional ac o s, e.g. sa is ac ion o he pa ien , adequacy, o imeliness o ca e. In e na ional compa isons o he pe o mance o heal hca e sys em discuss many au ho s – see Ba ák (2012), Dlouhý (2009), Hadad e al. (2013) o Lawson e al. (2012). The e exis many indica o s which a e used o measu e and e alua e he pe o mance o heal hca e sys ems. The WHO lis s one le el o e alua ion as “be e heal h” linked o he i s o he h ee goals lis ed abo e (WHO, 2000). Acco ding o he Uni e si y o Ljubljana (2015), wo ypes o indica o s exis on his le el: indica o s o chances o li ing a ull li e and indica o s o los yea s o li e. O he app oaches use he ollowing indica o s: numbe o doc o s and nu ses, educa ion o he heal hca e pe sonnel, in an mo ali y, o e weigh ness and obesi y, o consump ion o alcohol (OECD, 2016). In he a icle, Hejduko á (2015) uses o he indica o s o e alua ing pe o mance, o ins ance heal h ca e expenses o he numbe o gi en de ices o equipmen used in medicine. 1.2 Heal hca e Sys ems in V4 Coun ies As was no ed abo e, e o ms play an impo an ole in cu en heal hca e policy and hus in heal hca e sys ems. Among V4 coun ies, hese e o ms a e simila o one ano he and show coope a ion be ween he e o ming ins i u ions, law i ms, and en i ies on he i nancial ma ke p ima ily in a eas o mode nizing heal hca e acili ies wi h a s ong ocus on he ma ke (P eke & Ha ding, 2004; 2007). All ou V4 coun ies sha e simila i ies, such as ecen his o ical e en s, bu he e a e also di e en socio-economics and poli ical di e ences in Viseg ad G oup (Pio owicz, 2015). The heal hca e sys em in he Czech Republic is cha ac e ized by ex ensi e u iliza ion o heal hca e goods and se ices (Ma in, 2016). The Czech Republic is a V4 coun y ha has gone h ough many e o ms in ecen yea s in he i eld o heal hca e. These e o ms ha e deal no only wi h he heal hca e sys em i sel , bu also wi h he i elds o public i nance and p i a e law. The OECD (2008b) desc ibes he heal hca e sys em o he Czech Republic in a wholly posi i e manne , bu poin s ou some possible isks: “c eam skimming” by heal h insu ance companies, asymme ic in o ma ion be ween he managemen o heal hca e companies and poli icians, possible de e io a ion in he quali y o heal hca e as a esul o he g owing numbe o heal hca e acili ies, o issues wi h he ins a emen and subsequen cancella ion o doc o ees. Simila ly o o he V4 coun ies, he Hunga ian heal hca e sys em has been in l uenced by he coun y’s ans o ma ion om a ansi i e o ma ke economy. Acco ding o he OECD (2008c), isks linked o economic, i scal, o ma ke -o ien ed e o ms (co-pays, compe i ions in public heal hca e) can ha e e y nega i e impac s on he heal hca e sys em. “On he o he hand, Hunga y is a a ge coun y o c oss-bo de heal h ca e, mainly o den al ca e bu also o ehabili a i e se ices, such as medical spa ea men . The heal h indus y can hus be a po en ial s a egic a ea o economic de elopmen and g ow h.” EM_3_2017.indd 135EM_3_2017.indd 135 7.9.2017 10:34:277.9.2017 10:34:27 136 2017, XX, 3 Ekonomika a managemen (Eu opean Obse a o y on Heal h Sys ems and Policies, 2011, p. 20) As Wóyciska and G abowski (2007) no e, Poland is a V4 coun y ha has no been success ul in making a sa is ac o y shi o mee wel a e-s a e ends and in doing so c ea e a model o mode n social secu i y. In he a ea o heal hca e, Poland’s la ges p oblems lie in i s o e all expendi u es on heal hca e and he s a e o i s ci izens’ heal h (Wa son, 2006; Kuszewski & Ge icke, 2005). Acco ding o a s udy c ea ed by Há a and Maško á-Hanušo á (2009a), p oblems in Polish heal hca e a e also e iden in he i eld o medical ees, a na ional heal h insu ance und, disag eemen s in poli ics, and o he s. Acco ding o he OECD (2008a), one o Poland’s majo p oblems lies in he ac ha he coun y has one o he highes deg ees o income inequali y in he OECD; o o he in o ma ion see – Pa olini and Guillén (2013) o Cos a-Fon and G ee (2016). Slo akia is he i s V4 coun y o ha e ins a ed wide-scale e o ms in he i eld o heal hca e. Despi e his ac , howe e , Slo akia s ill lags behind i s V4 pa ne s in he e ec i eness o i s heal hca e. He e we can men ion he highe mo ali y o i s sys em, which is 15% highe in compa ison o o he V4 coun ies. Cha ac e is ic o Slo akia is also he high use o medicine and high a es o medical isi s. Highe cos s in compa ison wi h he o he V4 coun ies in he a ea o diagnos ics and labo a o y wo k a e also e iden (Zd a o nický deník, 2016). Despi e i s ini ial e o ms, Slo akia has been s ongly in l uenced by disag eemen s in poli ics, o example he es ablishmen and cancella ion o medical ees, a mo a o ium on ans o ming he legal s a us o heal hca e ins i u ions, and so on. (Ve hoe en e al., 2007). Acco ding o he OECD (2016a), all V4 coun ies show less expendi u e on heal hca e han he OECD a e age, which is also con i med by conclusions in he s udies by Há a and Maško á-Hanušo á (2009b) o OECD (2016b). We can also claim ha he V4’s heal hca e sys ems con inued o be unde i nanced in compa ison o he OECD a e age (OECD, 2016b; 2015). In addi ion o he ac ha hese coun ies a e s ill acing challenges wi h hei ans o ma ion om a ansi i e economy o a ma ke economy, he s uc u e o heal hca e sys ems in V4 coun ies is also in l uenced by demog aphic de elopmen s, i scal p oblems, e o s owa d cons an economic g ow h, and globaliza ion. Basic p oblems in he heal hca e sys ems o he V4 can be ound in he o ien a ion o indi idual sys ems on p ima ily economic goals, whe eas in e na ional ends a e mo e ocused on non- economic and non- i nancial aspec s. Measu ing and e alua ing he pe o mance o heal hca e sys ems can be seen as one o he mos impo an aspec s o e ec i e heal hca e managemen . In s uc u ing and c ea ing a me hod o he pe o mance e alua ion o heal hca e sys ems, i is necessa y (i a all possible) o ocus on objec i ely measu able indica o s. Thus, i is impo an o de e mine and e alua e he capabili y o in e p e ing esul s o measu emen . Such measu emen p o ides o compa isons o pe o mance o be made in ime and in e ms o selec ed egional uni s (e.g. on he le el o a coun y’s indi idual egions, o on a na ional le el). Scien i i c s udies ha e paid only li le a en ion o hese measu able/quali a i e o ms o pe o mance e alua ion o heal hca e sys ems, hus p o iding an oppo uni y o us o ca y ou esea ch and c ea e a composi e indica o o pe o mance e alua ion in his a ea. 2. Me hodology and Resea ch Da a The goal o his pape is o compa e and e alua e he pe o mance o heal hca e sys ems in V4 coun ies using selec ed indica o s o heal hca e sys ems and de e mine he s a e o he Czech Republic’s heal hca e sys em in compa ison o o he V4 coun ies. This pape will a emp o answe he ollowing ques ions: Wha is he Czech Republic’s posi ion in e ms o heal hca e indica o s in compa ison o o he V4 coun ies? Wha esul s do he selec ed indica o s show and how can we in e p e hem? Resea ch o li e a u e and selec ed s a is ical indica o s p o ided he answe s o hese ques ions. Da a o his pape was mainly aken om OECD Heal h Policy S udies. The heal hca e sys ems s udied in his pape we e chosen due o he ac ha V4 coun ies a e ansi i e economies o he CEE wi h a poo e s a e o heal h among o hei gene al popula ion han in de eloped coun ies o he OECD o EU. In addi ion, hese coun ies sha e ce ain cha ac e is ics – hei sys ems ha e long been unde i nanced and e o ms in hese coun ies ha e mos ly been gea ed owa d economic goals. As Há a and Maško á-Hanušo á (2009b) EM_3_2017.indd 136EM_3_2017.indd 136 7.9.2017 10:34:277.9.2017 10:34:27 137 3, XX, 2017 Business Adminis a ion and Managemen poin ou , howe e , in e na ional discou se has helped o de elop b oade concep s o long- e m sus ainabili y wi h an emphasis on o e all e ec i eness, quali y o p o ided se ices, and human igh s wi h ega d o heal h. The calcula ion o composi e indica o s is some solu ion ha p o ides he abili y o cap u e a complex phenomenon. The au ho s ealize ha a ce ain le el o abs ac ion will always occu when ying o quan i y phenomena using composi e indica o s. In composi e indica o , he deg ee o abs ac ion is mul iplied because i is based on mul iple indica o s. This disad an age, howe e , is balanced by he ac ha he composi e indica o e l ec s ac by means o se e al sub-indica o s and includes di e en , o en con adic o y, aspec s o he phenomenon. The e is no only one co ec CI app oach, so he e is no one igh solu ion o his ask, so he au ho s o e h ee me hods (models) o CI calcula ion. Th ea s associa ed wi h he use o composi e indica o s a e (i) he isk o inco ec calcula ion, and (ii) he possibili y o simple in e p e a ion can lead o oo simpli i ed and hus inco ec conclusions. The au ho s ied o elimina e hese h ea s by s udying app op ia e li e a u e and using hei expe ience wi h s a is ics and indica o s. The au ho s a e awa e ha using he same weigh s o componen s he e is a isk ha ce ain a eas (one pilla o componen ) will be assigned a g ea e weigh by being ep esen ed by a la ge numbe o indica o s in CI. This may ha e a s onge impac on o e all CI han i does wi h he in ended CI and i s heo e ical amewo k. The same weigh s a e app op ia e i he co ela ion be ween he indica o s does no indica e he double coun ing o he same aspec o he measu ed phenomenon due o he edundan sub-index. The use o he same weigh s as e idenced by he p esence o a la ge numbe o pa ial indica o s, i.e. 50-100 pa ial indica o s (Gulliksen, 1950). In he case o a la ge numbe o indica o s, i is di i cul o adequa ely de e mine he weigh s o each indica o , and he mo e complex weighing scheme would make i di i cul o in e p e he impac o he indi idual sub-indica o s. Simila ly, i we ha e only a small se o da a ha consis s o less han 30 obse a ions, he same weigh s a e app op ia e (Raju e al., 1999). Hopkins (1991) says ha i i is impossible o ob ain a gene al consensus o weigh de e mina ion, he simples solu ion is he bes . The au ho s e en ually applied linea weighing, his p ocess is u he suppo ed by he esul s in he empi ical pa , he esul s a e consis en wi h esul s in o he s udies (see Dlouhý, 2016a; Dlouhý, 2016b). A composi e indica o o he pe o mance o he heal hca e sys em was c ea ed. The main ad an age o using composi e indica o s is ha hey allow us o summa ize and exp ess complex phenomena in o one indica o . In ou case, he indica o exp esses he pe o mance o he heal hca e sys em. The composi e indica o enables us o c ea e a ela i ely simple compa ison o a selec ed uni (e.g. coun y, egion, company, e c.) in ime o be ween indi idual uni s. Acco ding o Booysen (2002), ano he ad an age o composi e indica o s is hei l exibili y in he sense ha compu a ions o me hodology can be easily changed (selec ion o indica o s, no maliza ion, weigh ing sys ems, agg ega ion). These changes, howe e , a e made a he expense o he compa abili y o indica o s in ime. This p oblem can be pa ially deal wi h ia e ospec i e e ision acco ding o new me hodology. Da a om he OECD Heal h Policy S udies was used o cons uc he composi e indica o . The pe iod o ime chosen o he s udy was en yea s, i.e. om 2005 o 2014. Based on expe selec ion, en indica o s we e chosen and he assump ion was made ha indica o s and hei alues e l ec he pe o mance o heal hca e sys ems in he indi idual membe s a es o he V4. Indica o s we e di ided in o wo pilla s (Pilla I and II) based on how posi i ely o nega i ely hey con ibu e o heigh ening he pe o mance o heal hca e sys ems. Six o en selec ed indica o s we e included in Pilla I and a e in di ec dependence on he pe o mance o he heal hca e sys em. The emaining ou ha e been included in Pilla II, which a e in indi ec dependence on he pe o mance o he heal hca e sys em. A summa y o he selec ed a iables is lis ed in Tab. 2. A mean was calcula ed o each indica o du ing he gi en pe iod. Indica o ID_8 shows a ailable da a o e he cou se o he whole pe iod only o he Czech Republic. The mean o indica o ID_8 o he emainde o coun ies was calcula ed only o 2009 and 2014. As ega ds indica o ID_10, comple e da a was a ailable only in he case o Poland; in he Czech and Slo ak Republic’s case, in o ma ion EM_3_2017.indd 137EM_3_2017.indd 137 7.9.2017 10:34:277.9.2017 10:34:27 138 2017, XX, 3 Ekonomika a managemen was published in wo-yea in e als. Missing alues we e hus subs i u ed by he alue o he a i hme ic mean and, in Hunga y’s case, da a was a ailable only o he pe iod be ween 2005 and 2009. The me hod o linea eg ession was selec ed o he impu a ion o absen alues. Values we e impu ed based on an es ima e o he eg essi e axis by using he leas squa es me hod. Mean alues o indi idual indica o s a e p esen ed in Tab. 3. Based on he selec ed indica o s, composi e indica o s we e cons uc ed. These indica o s a e mean o e l ec he pe o mance o heal hca e sys ems o he V4 membe s a es. Th ee me hods o modeling hese indica o s we e used in o de o es ablish he compa ison o heal hca e sys em pe o mance. These h ee me hods a e: (i) using he a ea o a ada cha , (ii) de e mining pe o mance based on o de , and (iii) de e mining pe o mance based on dis ance om he e e ence uni . These h ee me hods o de e mining he pe o mance o heal hca e sys ems allow us o compa e he si ua ions o indi idual heal hca e sys ems in e ms o pe o mance. 3. Models o Composi e Indica o s The i s model is p ima ily based on a g aphic ep esen a ion o he composi e indica o . Fo hese pu poses, a ada o “spide web” cha is used. This cha is a basic analy ical ool ha allows us o isualize se e al indica o s in one cha and hus o e s a isualiza ion o he a iable’s a ea (in ou case he pe o mance o he heal hca e sys em). Based on he size o he web’s a ea, we can hen easily compa e he heal hca e sys ems be ween gi en coun ies. Pilla Indica o I ID_1 Numbe o doc o s pe 1,000 people (in %) ID_2 Numbe o nu ses pe 1,000 people (in %) ID_3 Li e expec ancy a bi h in e ms o he o e all popula ion (in yea s) ID_4 Heal hca e expendi u es in e ms o GDP (in %) ID_5 P e alence o examina ion echnology: Magne ic esonance imaging machines pe 1,000,000 people (numbe o machines) ID_6 P e alence o examina ion echnology: Compu ed omog aphy scanne s pe 1,000,000 people (numbe o machines) II ID_7 Alcohol consump ion pe pe son among people o e 15 yea s o age (in li e s) ID_8 Consump ion o obacco p oduc s, % o popula ion o e 15 yea s o age who a e daily smoke s (%) ID_9 In an mo ali y, numbe o dea hs pe 1,000 li e-bo n indi iduals (in numbe o dea hs) ID_10 Thi y-day mo ali y a e admission o hospi al o AMI based on admission da a (Age-sex s anda dized a e pe 100 pa ien s) Sou ce: own based on OECD (2016a) Tab. 2: Selec ed indica o s o he pe o mance o heal hca e sys ems EM_3_2017.indd 138EM_3_2017.indd 138 7.9.2017 10:34:277.9.2017 10:34:27 139 3, XX, 2017 Business Adminis a ion and Managemen Based on he ada cha , he heo e ical exp ession o pe o mance can be exp essed as he di e ence be ween he sizes o he indi idual a eas o he web. (1) whe e Aj ep esen s he di e ence be ween he a ea o he spide web eached in Pilla I (PI) and Pilla II (PII). The size o he a eas PI and PII is depic ed in he ollowing wo i gu es. Loga i hmic ans o ma ion o he mean alues o indi idual indica o s ( ) was used o he isualiza ion. I is o mally possible o alues Aj o calcula ed pe o mance based on he ada cha o ake on nega i e alues, bu de e mining pe o mance in his manne in ou case me ely allows us o compa e he selec ed uni s. I is no possible o s a e whe he a nega i e alue indica es a dec ease in pe o mance in ime and ice e sa. The i gu es hemsel es a e a pa o he ollowing chap e on Resul s and Discussions. I should also be s a ed ha he cha s in Fig. 1 encompass six indica o s whose ising alues should ha e a posi i e e ec on he pe o mance o he heal hca e sys em o , in o he wo ds, he highe he pe o mance o he heal hca e sys em, he la ge he a ea o he “spide web”. On he con a y, Fig. 2 shows ou indica o s whose ising alues should dec ease he pe o mance o he heal hca e sys em. In o he wo ds, a smalle a ea o he web can be expec ed among heal hca e sys ems o highe pe o mance. The second model o compa ing he pe o mance o he heal hca e sys em is he model o a calcula ed sco e based on o de . In o he wo ds, his is a si ua ion in which da a was no malized using o de in e ms o indi idual coun ies (j). (2) whe e is he o iginal alue o he indica o i o he compa ed uni j exp essed as he mean o e he selec ed pe iod o ime. This means ha , o he nex analysis, i is possible o use only o dinal in o ma ion on he o de o he gi en uni . Thus, no conclusions can be d awn on he ela i e di e ence be ween uni s. On he o he hand, he me hod is easy o comp ehend, i is no in l uenced by emo e obse a ion, and i uni i es he a iance and ange o all indica o s. The mean om he o de uses a a ing om The Medica e S udy on Heal hca e Pe o mance ac oss he Uni ed S a es (Jencks e al., 2003). In Pilla I, indica o s wi h he highes alues eached he bes placing. Con a y o his, in Pilla II he index alues we e placed in descending o de , i.e. coun ies wi h he lowes indica o alue eached he bes placing. Fu he mo e, he mean o de o he lis was also calcula ed: CZ HU PL SK ID_1 3.61 3.02 2.20 3.30 ID_2 8.04 6.19 5.26 6.02 ID_3 77.56 74.52 76.24 75.47 ID_4 6.80 7.47 6.21 7.32 ID_5 5.70 2.82 4.13 6.20 ID_6 14.03 7.36 12.63 14.20 ID_7 11.81 11.78 10.21 10.46 ID_8 22.92 26.15 23.25 21.20 ID_9 2.84 5.31 5.27 5.92 ID_10 7.41 13.27 6.22 9.02 Sou ce: own based on OECD (2016a) Tab. 3: Summa y o mean alues o indi idual indica o s EM_3_2017.indd 139EM_3_2017.indd 139 7.9.2017 10:34:277.9.2017 10:34:27 140 2017, XX, 3 Ekonomika a managemen (3) Fo he hi d model o compa ing pe o mance o heal hca e sys ems, sco e calcula ion based on dis ance om he e e ence uni was selec ed. The dis ance om he e e ence uni was used o no malize da a, and he bes uni was used as a e e ence uni in Pilla I and he wo s uni as a e e ence uni in Pilla II. A e e ence alue was selec ed in each indica o and an indica o alue was exp essed as he a io o he e e ence alue (which is 100%). (4) whe e is he o iginal mean alue o he indica o and he compa ed uni j and is he alue o e e ence uni o indica o i. I is clea ha i he maximum alue is chosen as he e e ence uni , hen he alues o he no malized indica o s a e lesse o equal o one – his applies o Pilla I. I he wo s uni is chosen as he e e ence alue, he no malized da a is always g ea e o equal o one – his applies o Pilla II. This me hod p ese es he ela i e di e ences be ween uni s. In o de o es ablish ou own composi e indica o , a so-called “linea composi e indica o ” ela ed o expendi u es on heal hca e was used – o mo e see OECD (2002). The ma hema ical exp ession o his selec ed indica o is shown in he ollowing equa ion: (5) exp esses he pe o mance o he heal hca e sys em o he compa ed uni j. Fu he mo e, ep esen s he mean alue o expendi u es on heal hca e o he compa ed uni j. In addi ion, α ep esen s he weigh o he pilla in he composi e indica o o pe o mance . The e a e many ways o de e mine weigh s o indi idual sepa a e indica o s c ea ing an agg ega ed indica o . In ou case, symme ic weigh s we e c ea ed, i.e. α akes on he alue o 0.5. In he e en ha we only ha e a small se o da a, Raju e al. (1999) chooses o use he same weigh s. Hopkins (1991) also claims ha i i is impossible o gain common ag eemen o de e mining weigh s, he simples solu ion is he bes solu ion, i.e. also se ing he same weigh s. 4. Resul s and Discussion Fo all h ee sugges ed me hods o de e mining he composi e indica o , hei alues (Aj, Sj a g; Ej a g) we e calcula ed wi h esul s p esen ed in Tab. 4. These show ha he Czech Republic eaches he bes esul s in all cases in compa ison wi h o he membe s a es o he Viseg ad Fou . Acco ding o esul s o composi e indica o s, he second highes - pe o ming heal hca e sys em is he Slo ak Republic, while he wo s is Hunga y. I we compa e coun ies based on he alues o indi idual Pilla s I and II, we each almos exac ly he same conclusions as in he case o compa ing alues based on composi e indica o s. Fig. 1 shows ha he Czech Republic akes up he la ges a ea in Pilla I, wi h Slo akia in second place, Poland in hi d, and Hunga y in las . I we compa e a eas eached in Pilla II, we CZ HU PL SK PI13.59 10.79 10.67 13.04 PII 9.02 12.10 9.91 10.84 Aj4.57 -1.30 0.75 2.20 Sj a g 1.90 3.20 2.70 2.20 II ij 5.82 4.54 4.63 5.62 III ij 6.02 4.12 5.54 4.83 Ej a g 3.09 2.15 2.78 2.62 Sou ce: own calcula ion Tab. 4: Summa y o esul s o composi e indica o s EM_3_2017.indd 140EM_3_2017.indd 140 7.9.2017 10:34:277.9.2017 10:34:27 141 3, XX, 2017 Business Adminis a ion and Managemen Fig. 1: Visualiza ion o Pilla I Sou ce: own Fig. 2: Visualiza ion o Pilla II Sou ce: own EM_3_2017.indd 141EM_3_2017.indd 141 7.9.2017 10:34:287.9.2017 10:34:28