Hospital kanban system implementation: Evaluating satisfaction of nursing personnel

In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 www.else ie .es/iedee Hospi al kanban sys em implemen a ion: E alua ing sa is ac ion o nu sing pe sonnel Víc o G. Aguila -Escoba a,∗, Sa ah Bou queb, Nicolás Godino-Gallegoc aDepa men o Financial Economics and Ope a ions Managemen , Facul y o Tou ism and Finance, Uni e si y o Se illa, A . San F ancisco Ja ie , 41018, Se ille, Spain bFacul y o Economics and Business, En e p ise Ins i u e, Neuchâ el Uni e si y, Rue A.-L. B egue 2 - 2000 Neuchâ el, Swi ze land cPu chasing, and Logis ic Dis ibu ion Depa men , Hospi al Uni e si a io Vi gen Maca ena y A ea de Se illa, A . D . Fed iani 3, 41009 Se ille, Spain a i c l e i n o A icle his o y: Recei ed 8 May 2014 Accep ed 19 Decembe 2014 A ailable online 3 Ma ch 2015 JEL classifica ion: I1 M11 Keywo ds: Heal h Kanban Two-Bin Sys em Nu sing Hospi al Logis ics a b s a c Li e a u e on heal hca e supply chain managemen has shown ha he kanban sys em can p o ide sig- nifican benefi s. Howe e , e y ew benefi s ha e been empi ically demons a ed and he ex en o each emains unknown. This s udy aims o measu e nu ses’ sa is ac ion wi h kanban sys ems in logis ics o medical consumables and assesses possible ad an ages and di e ences among use g oups h ough an anonymous su ey a Hospi al Uni e si a io Vi gen Maca ena o Se ille, Spain. T ea men o esponses included an explo a o y ac o analysis, and a CAPTCA analysis. The esul s showed a high le el o sa is- ac ion o each aspec o he kanban sys em. Mo eo e , i highligh ed he di e ences o opinion be ween g oups o indi iduals acco ding o wo kplace, nu sing uni s, job ca ego y, senio i y, age and kanban ain- ing. The explo a o y ac o analysis e ealed ha wo ac o s unde lie he collec ed assessmen s: he inhe en ad an ages o a kanban sys em, and he logis ics sys em pe o mance as a whole. Thus, hospi al manage s should p omo e he implemen a ion o kanban sys ems, since i inc eases nu ses’ sa is ac ion and p o ides significan benefi s. © 2015 AEDEM. Published by Else ie España, S.L.U. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). Implan ación del sis ema Kanban en Hospi ales: e aluación de la sa is acción de en e me ía Códigos JEL: I1 M11 Palab as cla e: Salud Kanban Sis ema de doble cajón En e me ía Logís ica Hospi ala ia e s u m e n La li e a u a sob e cadena de suminis o en sanidad ha mos ado que el sis ema kanban puede p o- po ciona impo an es en ajas a las o ganizaciones sani a ias. Sin emba go, muy pocas de ellas se han demos ado y alo ado empí icamen e. Es e abajo iene como obje i o medi la sa is acción de la en e - me ía con el sis ema kanban, alo a cada una de sus posibles en ajas e iden ifica posibles di e encias en e di e en es g upos de usua ios. El abajo ha consis ido en ealiza una encues a anónima en el Hospi al Uni e si a io Vi gen Maca ena de Se illa, Espa˜ na. El a amien o de las espues as incluye un análisis ac o ial explo a o io y un análisis CAPTCA. Los esul ados mues an un al o ni el de sa is ac- ción con cada aspec o del sis ema kanban y la exis encia de di e encias en e indi iduos según cen o de abajo, unidad de en e me ía, ca ego ía labo al, an igüedad, edad y o mación en kanban. El análisis ac o ial explo a o io se˜ nala la exis encia de dos ac o es subyacen es a las alo aciones ecogidas: las en ajas inhe en es del sis ema kanban y el endimien o del sis ema logís ico en su conjun o. Como con- clusión, los di ec i os hospi ala ios debe ían p omo e la implan ación de sis emas kanban ya que ello inc emen a la sa is acción de en e me ía y p opo ciona impo an es beneficios. © 2015 AEDEM. Publicado po Else ie España, S.L.U. Es e es un a ículo Open Access bajo la licencia CC BY-NC-ND (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/). ∗Co esponding au ho . E-mail add ess: [email p o ec ed] (V.G. Aguila -Escoba ). 1. In oduc ion The inc eased demand o heal h se ices, oge he wi h he economic eali ies, highligh s conce ns ega ding public heal h sys em sus ainabili y, making i necessa y o seek maximum h p://dx.doi.o g/10.1016/j.iedee.2014.12.001 1135-2523/© 2015 AEDEM. Published by Else ie España, S.L.U. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/ licenses/by-nc-nd/4.0/). Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. 102 V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 e ficiency in hei managemen . Pa ien ca e is suppo ed by a wide a ay o ac i i ies including in en o y managemen , pu - chasing, and he dis ibu ion o supplies o he poin o ca e. These ac i i ies, e e ed o by many as heal hca e logis ics, o supply chain managemen , aim o ensu e he deli e y o he igh p oduc s whe e and when needed, a he quali y and quan i y equi ed o p o ide se ices while p e en ing any s ock dis up ions. As hese logis ic se ices use app oxima ely 30–40% o he hospi al’s ope - a ing budge (Land y & Beaulieu, 2013), hese ac i i ies a e o en ci ed as a eas o imp o emen (Aguila -Escoba & Ga ido-Vega, 2013; Aguila -Escoba , Ga ido-Vega, & Godino-Gallego, 2013; Callende & G assman, 2010). Logis ics he e o e become no only a sou ce o sa ings in e ms o suppo se ices, bu may also ende clinical p o essionals mo e p oduc i e as hey can inc ease dedica ion o hei daily wo k. Va ious managemen ools ha e been used o imp o e logis- ic and supply chain se ices. In ecen yea s, he Lean app oach has expe ienced inc easing popula i y o he imp o emen o p ocesses. Lean aims o ocus on he alue desi ed by he end- cus ome , o elimina e was e by iden i ying he alue s eams o he o ganiza ion p ocesses, o main ain a con inuous flow o pull be ween p ocess s eps, and finally, o con inuously imp o e (Womack & Jones, 2003). Due o Lean’s success in he au omo- i e and manu ac u ing indus ies, a significan numbe o o he sec o s, including se ices indus ies such as heal hca e, s a ed o adop Lean p inciples (Aguila -Escoba & Ga ido-Vega, 2013; Domínguez-Machuca, González-Zamo a, & Aguila -Escoba , 2007a, 2007b; Spea , 2005). Lean p inciples ha e been p oposed o b ing conside able imp o emen s in heal hca e (Ja e , 2006). A ecen li e a u e e iew on Lean applica ions in heal hca e e ealed ha he mos common a eas o imp o emen included imeliness o se ice, cos educ ions, p oduc i i y enhancemen s, and se e al quali y aspec s in ol ing educ ion in e o s, and imp o ed s a and pa ien sa is ac ion (Mazzoca o, Sa age, B ommels, A onsson, & Tho , 2010). The Lean app oach can be iewed h ough wo pe spec i es (Hines, Holweg, & Rich, 2004): a a s a egic le el, ocusing on p in- ciples and o ganiza ional cul u e o long-las ing ou comes; and, a an ope a ional le el, ocusing on ools and echniques. Va ious echniques ha e been linked o he Lean app oach, such as jus -in- ime, esou ce educ ion, imp o emen s a egies, de ec con ol, s anda diza ion, and scien ific managemen p ac ices (Pe e sen, 2009). Acco dingly, one o he mos common and widely sp ead Lean ools is kanban (Adams & Ruiz-Ulloa, 2003). The Japanese wo d kanban, meaning “signboa d”, is a sys em ha allows o he managemen o he o e all supply chain by s a egically and ope a ionally linking p oduc ion demands and he managemen o supplies. Kanban aces i s oo s o he ea ly days o he Toyo a P oduc ion Sys em when Taiichi Ohno, a Vice P esiden o Toyo a, de eloped kanban ca ds o implemen jus in ime (pull) p oduc ion and minimize wo k in p og ess (WIP) by means o a sim- ple isual ool. The kanban c ea es a “pull” ma e ial flow equi ing employee pa icipa ion o con ol and imp o e p ocesses be ween wo ks a ions. The idea behind he kanban concep is ha wo k- s a ions p oduce/deli e desi ed componen s only when needed, hanks o a isual signal in he o m o he ecep ion o a ca d, box, o emp y con aine (Sugimo i, Kusunoki, Cho, & Uchikawa, 1977). Thus, kanban sys ems no mally c ea e cos educ ion no only by elimina ing was e, bu also by being mo e esponsi e o change, acili a ing quali y con ol, and gi ing impo ance, us and sup- po o he employees unning he p ocesses (Chalice, 2007). In hospi als, ma e ial dis ibu ion no mally ollows a pe i- odic in en o y sys em whe e ma e ial is eplenished in ba ches. Some hospi als use exchange ca s, o pa le el, bu usually ma e ial dis ibu ion is a demand-based o de ing sys em whe e nu sing s a conduc egula in en o y coun s combined wi h consump ion es ima es. P oduc s iden ified as low in in en o y a e no ed on a equisi ion o m ha is hen o wa ded, ei he manually o elec onically, o he ma e ial managemen depa men . Based on his equisi ion, equi ed supplies a e picked and shipped o he app op ia e wa ds. Unde his sys em, nu sing pe sonnel a e o en assigned he ask o pu ing away he supplies in he s o age uni s (Land y & Beaulieu, 2010). Kanban s a ed o appea in heal hca e in he la e 1980s h ough he de elopmen o a wo-bin sys em (also called no-coun ), by Danish and Du ch companies (Land y & Philippe, 2004), o medical supplies.1Thus, he quo a o each i em in his double- bin sys em is e enly dis ibu ed in o wo compa men s. When one compa men is emp y, he nu ses use he second (o backup compa men ), and iden i y ha a bin has been emp ied. Ma e- ial handle s no mally conduc ounds o he nu sing uni s o be eplenished acco ding o a fixed schedule. Ra he han d awing on hei expe ience only and “eyeballing” he ma e ials as wi h pa le el, he kanban ca d ini ia es eplenishmen o a p ede e mined quan i y (Leone & Rahn, 2010). Thus, ma e ial handle s can sim- ply scan he kanban ca ds and ans e he eques s o he ma e ial managemen ’s in o ma ion sys em. Fo i ems s o ed in he cen al wa ehouse, a pick lis is gene a ed om he ma e ial manage- men ’s in o ma ion sys em. Fo di ec pu chases (i ems sou ced ex e nally), a equisi ion is ansmi ed o supplie s. Finally, ma e- ial handle s deli e medical supplies di ec ly in o he emp y bins, he eby ensu ing s ock o a ion in each uni . In a ew hospi als, he scanning p ocess (in he o m o eplenishmen -o de ing ounds) has been elimina ed wi h he use o RFID-enabled kanban ca ds, which c ea e an au oma ic eplenishmen eques once he ca d is placed on he eade boa d (Benda id, Boeck, & Philippe, 2010). In hospi als, kanban has been used o manage medical supplies, commonly dispensed d ugs, o fice supplies, linen and o he com- modi ies (Land y & Philippe, 2004; Pe sona, Ba ini, & Ra ele, 2008). Fo many hospi als, kanban eplenishmen sys ems ha e gi en be e in en o y managemen esul s han adi ional equisi ion- based me hods using o de on eques , exchange ca s, pa le el, o he mo e expensi e au oma ed cabine s (G aban, 2008; Land y & Beaulieu, 2010; Land y, Blouin, & Beaulieu, 2004; Leone & Rahn, 2010). Land y and Philippe (2004) ound mul iple benefi s o he wo-bin (kanban) sys em in hei s udy o a ious eplenishmen sys ems in hospi als. The o de ing p ocess was as e , as he e we e buil -in decision ules, no coun ing was equi ed, and only emp y compa men s we e scanned (o e ie ed) by ma e ial handle s, he eby igge ing he eplenishmen p ocess. Fu he mo e, he use o kanban sys ems d as ically educed he ma e ial managemen wo kload o he nu sing s a , and e en elimina ed i when ma e- ial handle s we e employed; he eby pe mi ing nu ses o ocus on ca e a he han on ma e ial managemen ac i i ies. Mo eo e , as ma e ial handle s would spend less ime in he nu sing uni s and educe p oduc handling, he e was less chance o dis up ion in clinical ac i i ies. Addi ional benefi s, linked o he u iliza ion o wo-bin sys ems and poin -o -use in o ma ion, include he buil -in s ock o a ion and lowe a e age in en o y le els which, in u n, educed he numbe o expi ed p oduc s (Land y & Philippe, 2004). Ano he s udy o kanban implemen a ion, in a hospi al se ing, showed conside able sa ings in e ms o eplenishmen ac i i ies (Land y e al., 2004). In his kanban s udy, a ma e ial handle , in only 35 min, would ypically scan p oduc s in 18 s ocking loca- ions sp ead ac oss nine wa ds, which was ou o se en imes as e han in adi ional sys ems. In summa y, se e al benefi s 1Al hough double-bin is no he only possible way o implemen kanban in hos- pi al logis ics (Heinbuch, 1995; Pe sona e al., 2008; Pa e son, 2012) i is he mos widely used applica ion o kanban. The e o e, in his a icle we use he e ms double- bin, wo-bin, and kanban sys em in e changeably. Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 103 linked o kanban usage in heal hca e se ings ha e been ound. Among hese we can highligh : accele a ion o he o de ing p o- cess, be e o ganiza ion o s o age, ewe expi ed i ems, and less wo k o he nu sing pe sonnel (G aban, 2008; Kho ajia, Fa is, & Haas, 2009; Land y & Beaulieu, 2010; Land y e al., 2004; Land y & Philippe, 2004; Pe sona e al., 2008). All hese benefi s exe a posi i e influence on supply chain pe o mance and should lead o g ea e wo k sa is ac ion o nu ses. One way o measu e his sa - is ac ion could be o collec opinions abou po en ial benefi s and o e all sa is ac ion o he nu sing s a owa ds kanban sys ems. Se e al au ho s belie e ha in e nal cus ome sa is ac ion should be included in he p ima y indica o s o supply chain pe o mance (Aguila -Escoba e al., 2013; Gunaseka an, Pa el, & McGaughey, 2004; Heikkilä, 2002; O o & Ko zab, 2003). Addi ionally, se e al s udies (Heinbuch, 1995) ha e shown ha ma e ials managemen should ocus on in e nal cus ome s o he hospi al, such as nu ses, doc o s, and adminis a i e s a . They e en sugges ed he need o logis ics manage s o p o e ha in e nal cus ome demands we e me and sa is ac ion e alua ed (Heinbuch, 1995). Some au ho s ha e sugges ed and es ed (Mu ells, Clin on, & Robinson, 2005; Penz, S ewa , D’A cy, & Mo gan, 2008) ha he ailu e o ha e adequa e supplies influences he wo k sa is ac ion o he nu s- ing s a since his ailu e p e en s hem om p o iding quali y ca e. Howe e , o he bes o ou knowledge, e y li le li e a u e has ocused on e alua ing, analysing and empi ically p o ing ha nu sing pe sonnel a e sa isfied wi h kanban implemen a ions and ha his sys em p o ides he expec ed esul s in supply manage- men . Consequen ly, his pape seeks o u he in es iga e nu ses’ sa is ac ion and e alua e po en ial ad an ages o kanban sys ems in hospi al supply chains. Addi ionally, wo complemen a y objec- i es a e s udied: (1) o see whe he he e a e sa is ac ion le el di e ences be ween use g oups; and (2) o de e mine whe he he o iginal 12 a iables used in he s udy can be educed o only a ew ac o s (2 o 3). To achie e hese goals, we conduc ed a su ey o nu sing s a a AHVM. 2. Me hodology 2.1. Case s udy and da a collec ion This s udy was conduc ed in a 955-bed hospi al g oup, Hospi al Uni e si a io Vi gen Maca ena y Á ea (AHVM), loca ed in Se ille, Spain be ween Decembe 2012 and Ma ch 2013. A ha ime, he AHVM had 2 hospi als, 3 specialized cen es, and 27 ope a ing he- a es. In 2006, he AHVM de eloped a comp ehensi e logis ics plan ha included he c ea ion o a logis ics pla o m in 2007 and he implemen a ion o a kanban sys em (Aguila -Escoba & Ga ido- Vega, 2013). P io o he implemen a ion o he kanban sys em, he nu s- ing supe iso in each depa men ook ca e o o de ing ma e ial using a p ocu emen in o ma ics sys em ha ansmi ed eplen- ishmen eques s o he logis ics and pu chasing depa men . On a e age, hey would place ma e ial o de s once a week. The e was no s anda dized p ocess; each nu sing supe iso would o de wi h a iable equencies and quan i ies. Howe e , i was ag eed ha he o de quan i y should no exceed one mon h’s consump ion o i ems ha had no loca ion in he cen al s o e (Type 1), and 15 days o i ems ha did ha e a loca ion (Type 2). The financial managemen depa men supe ised and au ho ized each o de . The implemen a ion o he modula double-bin (kanban) shel ing uni s comple ely changed hei eplenishmen sys em. This new e gonomical sys em p o ided he possibili y o easy access o bo h fi s bin and second bin loca ions, o configu able size. Each fi s bin was iden ified wi h a label ha included c i ical in o ma ion such as he loca ion numbe , p oduc name, code, quan i y and a ba code. The ba code included he p oduc code, he loca ion and cos cen e. The in o ma ion con ained on he label was unique o each i em and s anda dized wi hin he hospi al g oup. The second bin was iden ical o he fi s bin. Once a bin was emp y, he nu s- ing pe sonnel was esponsible o u ning he fi s bin label om he whi e o he ed side o indica e a eplenishmen need, and p oceeded o use he ma e ial loca ed in he second bin. The logis- ics cle ks we e now esponsible o he eplenishmen o de s in he uni s. They passed h ough each uni acco ding o a specific schedule. The logis ic cle ks comple ed hei ounds by scanning all he ed ags in he uni s and he associa ed in o ma ion was di ec ly uploaded o he sys em ia Wi-Fi. This c ea ed eplenish- men o de s o he cen al ma e ial s o e. O de s we e now mo e equen , wice ins ead o once pe week, he eby gene a ing he need o o e b eak-bulk picking. In cases o excep ionally high usage, he sys em allowed addi ional eques s which p e en ed po en ial s ock-ou s. The o de s we e p epa ed and deli e ed he same day by logis ic cle ks. The ma e ials we e b ough o he uni s, unboxed and placed in o he bins. The cle ks would also mo e he emaining ma e ial om he second o he fi s bin and eplenish he second bin; he eby ensu ing s ock o a ion was comple ed. Once his las s ep was accomplished, he logis ic cle ks u ned he labels o show he whi e side o indica e ha he i em had been eplenished. The whole p ocess, om he ba code eading o he eplenishmen , no mally ook unde 24 h. The AHVM’s finance and adminis a ion manage o de ed he p ocu emen and logis ics dis ibu ion depa men o implemen he kanban sys em in 36 s o es. The p ojec s a ed in ea ly Feb ua y 2010 wi h a fi s pilo implemen a ion and ended in la e Augus 2010 wi h ac i i ies such as analysis, planning, so wa e de elopmen , and p ocu emen . F om mid-No embe 2010 o mid-No embe 2011, he kanban sys em was implemen ed in he emaining 35 s o es wi h an a e age o 3 implemen a ions pe mon h (one e e y 10 days). Cu en ly, he AHVM has implemen ed he kanban sys em in 60% o hei s o es (i.e. 36 o a o al o 60 s o es). The main c i e ia o de e mine whe he he s o e was o ecei e a kanban sys em, we e he exis ence o a minimum s o age space o 5 m2and ha he s o e con ained medical consumables. Table 1 shows he kanban s o es ha ha e been implemen ed, he implemen a ion da e, he numbe o i ems wi hin he kanban sys em, he pe cen age o i ems in he kanban sys em on he o al amoun o i ems, and finally he pe cen age ep esen ing he alue o i ems in he kanban sys em on he o al s ock alue in he yea 2013 (weigh ed a e age alue in eu os). The a e age s o e size, o he 36 implemen ed kanban sys ems, was o 17.5 m2, and he median (which is ep esen a i e o a ypi- cal s o e size) was o 14.5 m2. No all medical consumables used in he uni s we e wi hin he kanban sys em. O a o al o 8583 logis ic loca ions ( he e is only one logis ic loca ion pe s ock-keeping uni (SKU) in each s o e uni ), 6353 (74.02% o he o al) co esponded o p oduc s on he kanban (double-bin) shel es, and 2230 (25.98%) on con en ional shel ing ha was managed using a adi ional eplenishmen sys em. The annual i em consump ion o hese 36 uni s in 2013 was o D 5,365,644 and he p opo ion ound wi hin he kanban sys- em was o D 2,531,511 (47.18%). This is an os ensibly low a e age conside ing he o e all consump ion as a whole due o he high economic alue o ce ain su gical p oduc s ha a e nei he in he kanban sys em no ha e a loca ion in he cen al o uni s o e. These p oduc s a e expensi e and a e he e o e only o de ed on demand. Acco dingly, he e was li le in e es in inco po a ing i ems wi h low u no e a es in o he kanban sys em as his would only ha e inc eased he in en o y and he numbe o expi ed goods. The in es men made by FEDER, Funds o he Eu opean Union, ollowing hese 36 implemen a ions o modula double-bin shel - ing uni s was o D 186,408.2, ep esen ing an a e age cos o D 5178 pe s o e. Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. 104 V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 Table 1 Desc ip ion o he a ious s o es ha had implemen ed kanban by Ap il 2014. S o e Ins alla ion da e Su ace (m2) No. o kanban i ems % o i ems in kanban (on o al i ems) % o i em alue in kanban (on o al alue in D ) Clinical managemen uni Ope a ing Room, G ound Floo 20/08/2010 75 254 50.70% 14.97% Gene al Su ge y T auma ology O hopedic Su ge y Anes hesia and Su gical Uni O ola yngology Oph halmology 8 h Floo , A Wing 19/11/2010 8.86 222 92.89% 97.65% In e nal Medicine 8 h Floo , B Wing 26/11/2010 14.49 193 83.19% 93.26% In e nal Medicine 8 h Floo , C Wing 03/12/2010 17.64 176 86.27% 93.12% Diges i e 8 h Floo , D Wing 17/12/2010 25.11 175 76.09% 31.43% Hema ology Oncology 7 h Floo , A Wing 14/01/2011 8.86 193 80.42% 90.33% In e nal Medicine 7 h Floo , B Wing 21/01/2011 14.49 194 88.18% 92.81% In e nal Medicine 7 h Floo , C Wing 28/01/2011 17.64 187 82.02% 82.90% Neu ology 6 h Floo , A Wing 04/02/2011 8.86 153 53.31% 51.57% Pedia ics 7 h Floo , D Wing 11/02/2011 25.11 221 75.68% 66.75% Neph ology 6 h Floo , C Wing 25/02/2011 17.64 192 67.84% 38.91% Pedia ics 6 h Floo , D Wing 04/03/2011 25.11 181 79.74% 62.38% Pedia ics 6 h Floo No h, P ema u e babies 11/03/2011 3.87 112 62.92% 53.59% Neona ology 5 h Floo , A Wing 18/03/2011 8.86 200 74.07% 59.34% O ola yngology Maxillo acial Su ge y 5 h Floo , B Wing 25/03/2011 14.49 200 81.97% 93.74% In e nal Medicine 5 h Floo , C Wing 01/04/2011 17.64 184 82.88% 96.57% Oph halmology In ec ious Diseases and Mic obiology De ma ology 4 h Floo , A Wing 08/04/2011 8.86 120 81.08% 85.53% Obs e ics and Gynaecology 4 h Floo , B Wing 15/04/2011 14.49 204 88.31% 95.78% Neu o Su ge y Plas ic Su ge y 4 h Floo , C Wing 03/05/2011 17.64 119 86.23% 95.57% Obs e ics and Gynaecology 4 h Floo , D Wing 13/05/2011 25.11 176 85.85% 95.53% Obs e ics and Gynaecology 4 h Floo , Deli e y oom 20/05/2011 19.68 153 69.23% 55.11% Obs e ics and Gynaecology Sma Ope a ing Room, 4 h Floo 27/05/2011 3.65 30 58.82% 65.70% Gene al Su ge y Ope a ing oom, 4 h Floo 27/05/2011 10.15 244 57.28% 22.10% U ology Pedia ic Su ge y Obs e ics and Gynaecology Anes hesia and Su gical Uni 3 h Floo , A Wing 03/06/2011 8.86 189 78.42% 92.94% Ca diology 3 h Floo , B Wing 10/06/2011 14.49 217 74.83% 88.41% Gene al Su ge y 3 h Floo , C Wing 17/06/2011 17.64 240 76.19% 47.83% Gene al Su ge y 3 h Floo , D Wing 24/06/2011 25.11 172 74.46% 74.54% Ca dio ascula Su ge y 2nd Floo , B Wing 16/08/2011 14.49 164 83.25% 91.35% U ology 2nd Floo , D Wing 02/09/2011 25.11 201 86.27% 97.30% T auma ology 2nd Floo , A Wing 12/09/2011 8.86 198 79.52% 70.58% Tho acic Su ge y Pneumology Endoscopy Uni 19/09/2011 12.65 105 47.09% 5.90% Diges i e 2nd Floo , C Wing 26/09/2011 17.64 183 84.33% 97.15% T auma ology Diu nal ca e Uni , 1s Floo 10/10/2011 8.86 72 65.45% 12.18% In e nal Medicine In ec ious Diseases and Mic obiology Oncology In ensi e Ca e Uni 25/10/2011 52.04 257 61.34% 57.01% C i ical Ca e and Eme gency Psychia ic hospi aliza ion, B and C Wings 08/11/2011 12.22 124 75.61% 73.82% Men al Heal h Hemodialysis Uni , G ound Floo 14/11/2011 9.89 148 83.62% 85.23% Neph ology To al 631.11 6353 74.02% 47.18% 2.2. Su ey The su ey was de eloped o explo e and measu e he a i- ous aspec s ha define sa is ac ion le els in he kanban sys em. The aspec s, selec ed o be alued, we e based on a sa is ac ion su ey in heal hca e logis ic se ices (Aguila -Escoba e al., 2013) and adap ed o he kanban sys em wi h he addi ion o ou new ques ions. The da a collec ion ook place be ween Feb ua y 2013 and Ap il 2013 in all uni s ha had comple ed hei kanban imple- men a ion (26 di e en depa men s dis ibu ed be ween he San Laza o and Maca ena Hospi als). I comp ised gene al in o ma ion on he esponden , including he name o he uni in which hey wo ked, hei job ca ego y, senio i y ange, age ange, whe he hey had ecei ed aining on kanban sys ems, and whe he hey had wo ked in he hospi al p io o he implemen a ion o he kan- ban sys em. The su ey eques ed hei esponses o a lis o 12 s a emen s (see he su ey in Appendix) using a 10-poin Like scale ( anging om “comple ely disag ees” o “ o ally ag ees”). The a ge popula ion using he kanban sys em in his hospi al was o 1123 people (nu sing s a ). By using simple andom samp- ling om a fini e popula ion, we ob ained a o al o 208 su eys, gi ing us a esponse a e o 18.52%. The sampling e o was o ±6.1% o a confidence le el o 95%. Wi h he da a ob ained om hese su eys, a desc ip i e and explo a o y ac o analysis was conduc ed including mul iple means compa isons (ANOVA and S uden ’s - es ) using SPSS 22. P io o he comple ion o he con as s, we pe o med Le ene’s es o he equali y o a iances be o e ca ying ou he S uden ’s Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 105 - es o he equali y o means and ANOVA analysis. Due o he es ou come, he B own–Fo sy he es was chosen ins ead o he classic ANOVA analysis, since i p esen s obus ness o unequal a iances. Following he esul s, i was decided ha he Tukey o Games–Howell es s we e app op ia e o ca y ou he con as s (o compa isons) a pos e io i. In o de o g oup he a iables in o ep esen a i e ac o s, we conduc ed a Fac o ial Analysis. Subse- quen ly, a Ca ego ical P incipal Componen s Analysis (CAPTCA) was ca ied ou o confi m he p e ious esul s. 3. Findings We now p esen he esul s and analysis s a ing wi h he cha - ac e is ics o ou sample, he sa is ac ion le els o hospi al s a wi h he kanban sys em, a desc ip i e analysis o sa is ac ion le - els, sa is ac ion le el compa isons be ween di e en g oups, an explo a o y ac o analysis, and a CAPTCA analysis. 3.1. Sample cha ac e iza ion As p e iously s a ed, he cha ac e is ics o he sample o he s udy we e based on he 208 su eys ga he ed a AHVM. We ob ained 16 su eys (7.7%) om he San Laza o Hospi al and 192 (92.3%) we e ob ained in he Maca ena Hospi al. When analysing he a ious uni s ha had esponded, 59.1% o he esponden s wo ked in a clinical uni and 40.9% in su gical se ices o in ensi e ca e. The job ca ego ies ep esen ed wi hin he 205 esponden s we e: 12 head nu ses (5.9%), 98 nu ses (47.8%), 94 auxilia y nu ses (45.9%), and 1 (0.5%) in ano he ca ego y. The senio i y analysis showed ha 114 o he esponden s had been wo king in he hos- pi al o mo e han 20 yea s (57.3%), 38 people had be ween 15 and 19 yea s o se ice (19.1%), 29 people had be ween 10 and 14 yea s (14.6%), 16 people be ween 5 and 9 yea s (8%), and 2 people had less han 5 yea s o senio i y (1%). Nine o he 208 esponden s did no answe he senio i y ques ion. Mo eo e , when asked whe he hey had wo ked in he hospi al p io o he implemen a ion o he kanban sys em, he esul s indica ed ha 164 esponden s had (82%) and 36 had no (18%) wo ked in he hospi al p io o he implemen a ion. In ela ion o he esponden ’s age, 12 wo ke s we e 60 yea s old o o e (5.9%), 78 we e be ween 50 and 59 yea s old (38.2%), 81 we e be ween 40 and 49 yea s old (38.9%), 31 we e be ween 30 and 39 yea s old (14.9%), 2 we e unde 30 yea s old (1%), and finally 4 wo ke s did no espond o he ques ion. These esul s combined wi h senio i y demons a e a wo ying sign o ageing and non- enewal o nu sing in he Spanish public sys em. Las ly, we asked abou kanban aining, an aspec o impo ance o he p ope use o his sys em. In his ega d, 140 esponden s epo ed ha ing ecei ed aining (69.3%) compa ed o 62 ha indica ed o he wise (30.7%), while 6 did no espond o he ques ion. 3.2. Desc ip i e analysis o he hospi al s a sa is ac ion wi h he kanban sys em As s a ed in Sec ion 2.2, he esponden ’s sa is ac ion le el wi h he kanban sys em was analysed h ough 12 alua ion ques- ions. The fi s 10 ques ions we e o mula ed as a compa ison be ween he kanban sys em and he ini ial eplenishmen sys em. Thus, hese ques ions could only be answe ed by hose who wo ked in he hospi al p io o he kanban implemen a ion. Ques ion 11 enqui ed whe he he majo i y o he p oduc s we e ound in he double-bin sys em and ques ion 12 made a gene al sa is ac ion assessmen . Table 2 shows he numbe o esponses (N), means and s anda d de ia ions (S d. de .) co esponding o each o he 12 alua ion ques ions. Table 2 Main desc ip i e s a is ics o he alua ion ques ion. Aspec o he kanban sys em N Mean S d. de a (1) Fas e 196 6.01 2.344 (2) Mo e eliable 190 5.48 2.352 (3) Fewe di e ences 194 6.17 2.334 (4) Fewe s ock-ou s 197 5.30 2.605 (5) Be e phone a endance 153 5.84 2.301 (6) Less wo k 194 6.96 2.647 (7) Mo e o de 196 7.53 2.208 (8) Less space 190 7.47 2.072 (9) Fewe expi ed i ems 185 7.86 1.697 (10) Mo e sa isfied han be o e 198 6.81 2.610 (11) Mos p oduc s in double-bin sys em 203 7.78 1.964 (12) Deg ee o o e all sa is ac ion 206 6.94 2.365 aS anda d de ia ion. Table 2 also p esen s he high a e age (6.94) o he o e all sa - is ac ion assessmen linked o he kanban sys em implemen a ion. In e es ingly, ano he su ey on logis ics sys em sa is ac ion in he same hospi al ga e a global sa is ac ion assessmen alue o 6.7 which was al eady conside ed a high sco e by he au ho s (Aguila - Escoba e al., 2013). Mo eo e , in a simila wo k by Newell, S einme z-Mala o, and Van Dyke (2011), an a e age sa is ac ion o 3.1 on a scale o 0–5 was ob ained. Likewise, he emaining su ey ques ions p esen ed conside - ably high esponse means. The aspec mos alued by esponden s was ha kanban sys ems gene a ed ewe expi ed i ems compa ed o con en ional sys ems (wi h an a e age alue o 7.9). In addi- ion, he su ey ques ions ha ga he ed an a e age o o e 6.9 we e hose ha s a ed ha kanban sys ems made s o es mo e o de ly (7.53), equi ed less s o e space (7.47), and esul ed in less wo k o nu sing pe sonnel (6.96). Thus, ques ion 10 compa ing kanban wi h he p e ious sys em eached a high a e age alue o 6.81. Th ee ques ions p esen ed an a e age below 6.2, which we e being he ques ions conce ning educed s ock-ou s (wi h a mean alue o 5.3), inc eased eliabili y (5.48) and imp o ed phone a endance (5.84). Mo eo e , i was obse ed ha he s anda d de ia ions we e always less han hal o he co esponding a e age, and hence i can be concluded ha dispe sions we e ela i ely small. 3.3. Sa is ac ion-le el compa isons be ween di e en g oups In o de o de e mine whe he demog aphic a iables, consid- e ed in he s udy, had an impac on he le el o sa is ac ion, we ca ied ou a ious con as s o es he di e ences be ween g oups. Table 3 summa izes he sa is ac ion analysis including he compa - ison be ween g oups o each aspec . The las column shows he significan di e ences be ween g oups, he mean alue (shown in pa en heses, on a scale o 0–10), he di e ence di ec ion (> highe , < smalle ) and he s a is ical significance le el (*, **). In ela ion o he cen e (hospi al loca ion) whe e he nu ses wo k, we applied a S uden ’s - es compa ing he means o independen samples. The e we e significan di e ences be ween ques ions (3) Fewe di e ences, (5) Be e phone a endance, and (12) Deg ee o o e all sa is ac ion, wi h a cons an highe alua ion in he San Laza o Hospi al compa ed o ha o he Maca ena Hos- pi al. The eason o his highe alua ion may be linked o he ac ha he fi s kanban s o e was implemen ed in he newly opened San Laza o Hospi al ope a ing ooms, he eby eplacing se e al small ou da ed s o es. We hen applied a S uden ’s - es by compa ing he means o independen samples o in o man s om bo h cen es g ouped in o wo wo king uni ca ego ies: su gical/in ensi e ca e uni s and clinical uni s. As seen in Table 3, 7 o he 12 su ey ques ions showed significan di e ences. Fu he mo e, all aspec s showed Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. 106 V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 Table 3 Compa isons be ween g oups. Fac o Aspec o he kanban sys em S a is ical alue Significan di e ences be ween g oups Cen e 3) Fewe di e ences (13.887) = 3.486, p = 0.004aSan Láza o Hospi al (7.64) > Maca ena Hospi al (6.08)** 5) Be e phone a endance (151) = 2.064, p = 0.041bSan Láza o Hospi al (7.57) > Maca ena Hospi al (5.75)* 12) Deg ee o o e all sa is ac ion (20.852) = 2.129, p = 0.045aSan Láza o Hospi al (7.81) > Maca ena Hospi al (6.86)* Wo king uni s 1) Fas e (194) = 2.827, p = 0.005bSu gical and in ensi e ca e uni s (6.58) > Clinical uni s (5.63)** 3) Fewe di e ences (192) = 2.102, p = 0.037bSu gical and in ensi e ca e uni s (6.61) > Clinical uni s (5.89)* 7) Mo e o de (186.783) = 2.247, p = 0.026aSu gical and in ensi e ca e uni s (7.94) > Clinical uni s (7.25)* 9) Fewe expi ed i ems (180.302) = 2.635, p = 0.009aSu gical and in ensi e ca e uni s (8.24) > Clinical uni s (7.61)** 10) Mo e sa isfied han be o e (188.194) = 2.262, p = 0.025aSu gical and in ensi e ca e uni s (7.3) > Clinical uni s (6.48)* 11) Mos p oduc s in double-bin (199.368) = 2.345, p = 0.020aSu gical and in ensi e ca e uni s (8.14) > Clinical uni s (7.52)* 12) Deg ee o o e all sa is ac ion (202.805) = 2.827, p = 0.005aSu gical and in ensi e ca e uni s (7.46) > Clinical uni s (6.57)** Job ca ego y 6) Less wo k F(2, 121.834) = 12.599, p = 0.000cNu se assis an (7.55) > Nu se (6.18)** Head nu ses (8.58) > Nu se (6.18)** 7) Mo e o de F(2, 189) = 3.439, p = 0.034dNo significan di e ences be ween g oups Senio i y 1) Fas e (96.714) = 2.888, p = 0.005aLess han 15 yea s (6.73) > 15 yea s o mo e (5.74)** 6) Less wo k (152.912) = 3.583, p = 0.000aLess han 15 yea s (7.84) > 15 yea s o mo e (6.69)** 7) Mo e o de (138.22) = 2.492, p = 0.014aLess han 15 yea s (8.07) > 15 yea s o mo e (7.38)* 12) Deg ee o o e all sa is ac ion (105.566) = 2.449, p = 0.016aLess han 15 yea s (7.56) > 15 yea s o mo e (6.75)* Age o esponden 1) Fas e (190) = 2.494, p = 0.013bUnde 40 o o e 60 yea s old (6.79) > Be ween 40 and 59 yea s (5.77)* 2) Mo e eliable (184) = 2.625, p = 0.009bUnde 40 o o e 60 yea s old (6.35) > Be ween 40 and 59 yea s (5.23)** 3) Fewe di e ences (188) = 2.347, p = 0.020bUnde 40 o o e 60 yea s old (6.9) > Be ween 40 and 59 yea s (5.93)* 4) Fewe s ock-ou s (191) = 2.833, p = 0.005bUnde 40 o o e 60 yea s old (6.28) > Be ween 40 and 59 yea s (4.99)** 8) Less space (125) = −2.709, p = 0.008bUnde 40 o 60 o e 60 yea s old (8.16) > Be ween 40 and 59 yea s (7.11)** T aining 3) Fewe di e ences (186) = 2.554, p = 0.011bYes (6.39) > No (5.43)* 6) Less wo k (79.025) = 2.641, p = 0.010aYes (7.28) > No (6.04)* 7) Mo e o de (188) = 2.125, p = 0.035bYes (7.73) > No (6.98)* 8) Less space (184) = 3.023, p = 0.003bYes (7.76) > No (6.75)** 10) Mo e sa isfied han be o e (190) = 2.322, p = 0.021bYes (7.06) > No (6.09)* 12) Deg ee o o e all sa is ac ion (199) = 2.376, p = 0.018bYes (7.15) > No (6.30)* aCon as o means no assuming equal a iances acco ding o Le ene’s es . bCon as o means assuming equal a iances acco ding o Le ene’s es . cANOVA no assuming equal a iances wi h B own–Fo sy he obus s a is ical, and compa ison be ween g oups wi h Games–Howell s a is ical. dANOVA assuming equal a iance and mul iple compa isons wi h Tukey’s HSD. *p < 0.05. ** p < 0.01. highe alues o he su gical and in ensi e ca e compa ed o he clinical esponden s. Mo e specifically, he e was high o e all sa is- ac ion (ques ion 12) o he kanban sys em in In ensi e Ca e uni s, Gene al Su ge y and T auma ology wi h sco es o 8.6, 8.29 and 7.30. In con as , o he clinical esponden s, he o e all sa is ac ion o nu sing s a in De ma ology, Gas oen e ology and Endoscopy was o 4.75, 4.75 and 5.33, which g ea ly dec eased he g oup a e age. Conce ning he job ca ego y, mean compa isons be ween g oups showed significan di e ences in wo aspec s (see Table 3): (6) less wo k and (7) mo e o de . In bo h aspec s, nu ses pu lowe alua- ion sco es compa ed o he head nu ses. Addi ionally, o ques ion 6, whe e kanban is said o gene a e less wo k, auxilia y nu ses ga e a highe alua ion han nu ses. The case s udy showed ha bo h he head nu ses and auxilia y nu ses enjoyed he g ea es ime- sa ing. Head nu ses elimina ed in en o y coun s and o de o ms om hei wo kload. Mo eo e , unpacking and placing me chan- dise we e no longe he esponsibili y o auxilia y nu ses. When analysing he da a in ela ion o senio i y in he wo kplace, we simplified he analysis by sepa a ing he esponden s in o wo g oups: hose wi h less han 15 yea s o expe ience and hose wi h 15 o mo e. As shown in Table 3, he independen sample S uden ’s - es ha compa es he means showed ha he e we e significan di e ences in ou aspec s, which we e always highe in he g oup wi h less senio i y. Simila ly o he senio i y analysis, we dis inguished esponden s be ween wo age g oups: hose unde 40, and hose 60 yea s old o mo e (“younges and oldes ”); and hose be ween 40 and 59 yea s old. We compa ed means using an independen sample S uden ’s - es . As shown in Table 3, he e we e significan di e ences in fi e aspec s. In each case he a e age a ing was always highe in he “younges and oldes ” g oup. I may also be ha he middle age g oup is mo e c i ical owa ds he heal h sys em as a whole. Finally, wi h ega d o hose ha had ecei ed aining (Yes) o no (No), we compa ed he means using a S uden ’s - es (see Table 3) and ound ha he e we e significan di e ences in six a iables. All aspec s showed a highe a e age a ing o he g oup ha had ecei ed aining. This esul emphasized he impo ance o p o iding aining on kanban sys em p ocedu es o ensu e good pe o mance and high nu sing sa is ac ion wi h he sys em. 3.4. Explo a o y ac o and CAPTCA analysis In o de o analyse all he i ems wi hin ou “sa is ac ion le el” cons uc , he 12 aspec s we e subjec ed o an explo a o y ac- o analysis. Bo h he Kaise –Meye –Olkin es , which ga e a alue o 0.907 abo e he ecommended 0.6 (Tabachnick & Fidell, 2007), and he Ba le es o sphe ici y, which was s a is ically signifi- can (p = 0.000), indica ed ha ou sample was app op ia e o his ype o analysis. Mo eo e , ou sample size su passed he ecom- mended minimum o 10 obse a ions pe i em o hese ypes o es s (Nunnally, 1978). The p incipal componen analysis showed wo ac o s when applying eigen alues g ea e han 1. The fi s ac o (which included he aspec s con ained in ques ions 6, 7, 8, 9, 10 and 11) accoun ed o 56.93% o he a iance. The second ac o (which included ques ions 1, 2, 3, 4 and 5) accoun ed o 11.93% o he Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 107 Table 4 P incipal componen analysis ( wo ac o s). Fac o loadings Commonali ies Fac o 1 Fac o 2 1) Fas e 0.341 0.732 0.652 2) Mo e eliable 0.190 0.856 0.769 3) Fewe di e ences 0.499 0.705 0.745 4) Fewe s ock-ou s 0.236 0.836 0.755 5) Be e phone a endance 0.149 0.734 0.560 6) Less wo k 0.584 0.413 0.511 7) Mo e o de 0.839 0.273 0.778 8) Less space 0.843 0.171 0.739 9) Fewe expi ed i ems 0.840 0.161 0.732 10) Mo e sa isfied han be o e 0.751 0.475 0.790 11) Mos p oduc s in double-bin 0.645 0.201 0.457 12) Deg ee o o e all sa is ac ion 0.648 0.596 0.776 Ex ac ion me hod: P incipal Componen Analysis. Ro a ion me hod: Va imax wi h Kaise No maliza ion. Ro a ion con e ged in h ee i e a ions. Alpha de C onbach: Fac o 1 = 0.876, Fac o 2 = 0.893: a iance explained = 68.862%. The highes loading o each i em (excep 12) is indica ed by bold cha ac e s. a iance. The e o e, hese wo componen s oge he accoun ed o 68.86% o he explained a iance. As seen, ques ion 12, being he gene al sa is ac ion, could be included in ei he he fi s o he second ac o . Table 4 p esen s he ac o loadings and communali ies o he wo- ac o solu ion. The fi s ac o appea s o g oup aspec s ela ed o he kanban sys em and i s inhe en ad an ages such as: less nu sing wo k, mo e o ganized s o age, less s o age space, ewe expi ed i ems, inc eased in e nal cus ome sa is ac ion, and inco po a ion o a majo i y o p oduc s in he sys em. In ou iew, he second ac o g ouped o he a iables ha we e mo e ele an o he pe o - mance o he AHVM hospi al logis ics sys em. Indeed, his ac o included he ollowing aspec s: as e and mo e eliable deli e ies, ewe di e ences, ewe s ock-ou s, and be e phone a endance. We measu ed he su ey’s in e nal consis ency wi h hese wo ac o s and ob ained a C onbach’s alpha o 0.876 o he fi s ac o and 0.893 o he second ac o . These alues a e high (close o 1), he eby demons a ing he high eliabili y o he measu emen scale used in his su ey. In o de o s a is ically con as hese esul s, a Ca ego ical P incipal Componen s Analysis (CATPCA) was pe o med g ouping a iables in wo ep esen a i e ac o s. Fig. 1 shows a wo- dimensional space and he componen loadings o each o he a iables. The obse ed clus e ing is explained wi h he angles be ween ec o s showing ha he smalle he angle be ween a pai o ec o s, he g ea e he associa ion be ween he a iables ep e- sen ed. Thus, he fi s fi e a iables ( om 1 o 5) a e g ouped on one side, and iden i y he ad an ages o he kanban sys em i sel , and he o he six a iables ( om 6 o 11) on he o he side iden i y he pe o mance o he logis ics sys em. Ques ion 12 is a esul o bo h, he eby indica ing ha sa is ac ion o nu sing s a is influenced by bo h sa is ac ion wi h he kanban sys em and he logis ics sys em pe o mance. 4. Discussion This pape aimed o u he in es iga e nu sing pe sonnel sa is- ac ion owa ds kanban sys ems in hospi al supply chains. Ou main objec i e was o measu e nu sing pe sonnel sa is ac ion le els and see whe he hey posi i ely o nega i ely alued implemen a ion o kanban sys ems o sol e logis ic p oblems. Ou findings showed high o e all le els o sa is ac ion wi h he kanban sys em (6.94 ou o 10). Mo eo e , he new kanban sys em ecei ed highe sa is ac- ion le els han he p e ious eplenishmen sys em (6.81). Indeed, 0.8 0.6 0.2 0.0000 –0.2 –0.4 0.0 0.2 0.4 0.6 0.8 1.0 Dimension 1 2) Mo e eliable 1) Fas e 4) Fewe s ocks-ou s 4) Fewe di e ences 12) O e all sa is ac 10) Mo e sa is ied 6) Less wo k 7) Mo e o de 8) Less space 11) Mos p oduc s in 9) Less obsolescence Dimension 2 0.4 Componen loadings o ec o -quan i ied a iables 5) Be e phone a e Fig. 1. Componen loadings g aph. he su ey showed a posi i e app ecia ion in all aspec s e alua ed, since all alua ions emain abo e 5, on a 0–10 Like scale. These esul s confi m ha he kanban sys em ecei ed a posi i e e alua- ion and his has a posi i e influence on job sa is ac ion on he pa o nu sing s a because i enables hem o p o ide quali y se ices o pa ien s (Mu ells e al., 2005; Penz e al., 2008). Ou fi s complemen a y objec i e was o e i y whe he he e we e di e ences in he le el o sa is ac ion be ween use g oups. In his ega d, he wo k cen es (specific hospi al), wo k uni , he job ca ego y, he senio i y, and he age o he esponden and aining we e ound o be de e minan s o opinion. Ou second complemen a y objec i e was o de e mine which ac o s, p incipal componen s, o dimensions o sa is ac ion, could educe he o iginal a iables used in he s udy. Acco ding o he p incipal componen analysis, all he ques ions in he su ey we e influenced by wo ac o s ha could be called inhe en ad an ages o a kanban sys em (less wo k, mo e o de , less space, ewe expi ed i ems and inc eased cus ome sa is ac ion) and logis ics sys em pe - o mance ( as e , mo e eliable, ewe di e ences, ewe s ock-ou s, and be e phone a endance). The hospi al logis ics depa men p o ided suppo se ices o he kanban sys em. In e es ingly, nu sing pe sonnel de e mined ha aspec s 1, 2, 3, 4 and 5 we e less dependen on he kanban sys em bu a he on he logis ics sys em. Mo eo e , manage s ha implemen ed a kanban sys em could imp o e nu sing s a sa is ac ion, especially ha ela ed o aspec s 6, 7, 8, 9 and 10. The kanban sys em p omo ed a majo inc ease in sa is ac ion in compa ison wi h he con en ional sys- em, and bo h ac o s (kanban implemen a ion and good logis ics sys em) imp o ed nu sing s a sa is ac ion. The e o e, based on he di e ence in alua ion o he i ems and on he explo a o y anal- ysis, we deduce ha he main ad an ages o he kanban sys em o heal hca e logis ics a e ha : nu sing wo k linked o ma e ial managemen is minimized (ques ion 6), s o es a e mo e o de ly (ques ion 7), less space is equi ed in s o es (ques ion 8), and he e a e ewe expi ed i ems (ques ion 9). These esul s a e consis- en wi h hose ob ained by o he au ho s (G aban, 2008; Kho ajia e al., 2009; Land y & Beaulieu, 2010; Land y e al., 2004; Land y & Philippe, 2004; Pe sona e al., 2008). Howe e , he abo e au ho s eached hei conclusions by di e en means o hose p oposed in Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. 108 V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 his pape . To he bes o ou knowledge, his is he fi s s udy ha links nu sing opinion wi h hese ad an ages o he kanban sys em. Mo eo e , we belie e ha he esul s would no ha e been as success ul i he implemen a ion o kanban had no been p e- ceded by a se ies o changes h oughou he hospi al logis ics sys em. These changes include he c ea ion o a logis ics pla o m, he implemen a ion o a new wa ehouse managemen sys em, and he use o new adio equency e minals connec ed wi h he main compu e sys em (Aguila -Escoba & Ga ido-Vega, 2013). In ac , as ecognized in manu ac u ing implemen a ions o he kan- ban sys em (Domínguez Machuca, Ga cía González, Domínguez Machuca, Ruiz Jiménez, & Ál a ez Gil, 2003; Womack, Jones, & Ross, 1990), hospi al implemen a ion also equi es changes in he logis ics sys em ha enable an adequa e esponse o issues such as he main enance o high le els o se ice quali y, elimina ion o e o s in he p ocess, educ ion o ime, and assu ance o eliable deli e ies. Such a s uc u ed en i onmen is essen ial in ope a ing a kanban sys em and can e en be conside ed an ex ension o he sys em (Sugimo i e al., 1977). In ac , he kanban sys em en ails sending supplies om he cen al wa ehouse mo e o en and in smalle amoun s (Pe sona e al., 2008). In he s udied case, he kan- ban sys em igge s eplenishmen om he s o e wice pe week while he uni s in he adi ional sys em we e mainly eplenished only weekly. This new sys em ep esen ed mo e han 1500 daily eplenishmen lines in Oc obe 2011. The subsequen inc ease o wo kload in he cen al wa ehouse esul ed in ce ain di ficul ies in main aining se ice le els (Aguila -Escoba & Ga ido-Vega, 2013). Howe e , he su ey esul s showed ha hese ini ial di ficul ies had been o e come a he ime o ou su ey (in 2013) which is eflec ed in he esul s o he sa is ac ion e alua ion o nu ses wi h he kanban sys em and he en i e logis ics sys em. In conclusion, we ound ha he kanban sys em has p omo ed a se ies o e o ms in he logis ics sys em, and he logis ics sys em has con ibu ed owa ds he success o he kanban sys em. 5. P ac ical implica ions The high le els o sa is ac ion and he benefi s de i ed om using kanban sys ems, as ou lined in his pape , could encou age heal h adminis a o s and manage s owa ds hei implemen a- ion. Addi ionally, since nu sing sa is ac ion has been linked wi h highe pe o mance (Legga , Ba am, Casimi , & S an on, 2010), heal h adminis a o s and manage s should conside ha hese esul s empi ically demons a e nu sing sa is ac ion owa ds kan- ban sys em implemen a ion. Indeed, as kanban sys ems p o ide benefi s and sa is ac ion o nu sing pe sonnel, implemen a ion isks could po en ially be educed. Fu he mo e, manage s should conside di e ences in he le el o sa is ac ion be ween use g oups since his le el could indica e he need o adap app oaches o be e fi hei equi emen s: e.g. he impac o a kanban imple- men a ion in a su gical o in ensi e uni could di e om ha in a clinical uni . We highligh he impo ance o ca ying ou a aining p o- g amme as his significan ly inc eased he alua ion o all aspec s conside ed in his s udy. Ano he impo an implica ion was ha hose nu ses who pe cei ed hei wo k uni s as being mo e pa ien -cen ed we e sig- nifican ly mo e sa isfied wi h hei jobs since hey had a di ec e ec on pa ien ca e (Ra he & May, 2007). Acco dingly, kanban sys- ems con ibu ed owa ds making nu ses mo e pa ien -cen ed by lea ing non-ca e ac i i ies o logis ics pe sonnel, he eby ensu - ing hey had he ma e ial necessa y o p ope ca e. Indeed, ou wo k demons a ed ha nu sing pe sonnel needed less ime o logis ic ac i i ies wi h a kanban sys em han wi h a con en ional sys em. Thus, implemen ing pa ien -cen ed p ac ices, such as kan- ban sys ems, could esul in imp o ed nu se sa is ac ion ollowed by imp o ed pa ien ca e, and o ganiza ional ou comes. Mo eo e , in he medium e m, his ime-sa ing can educe pe sonnel cos s, he eby o se ing he in es men o D 5178 pe s o e. 6. Resea ch limi a ions and sugges ions o u u e esea ch This pape has ou lined a ious limi s. Fi s , one o he au ho s ac ed as an economic-adminis a i e di ec o , esponsible o hos- pi al logis ics, in his hospi al g oup o much o he pe iod conside ed (Augus 2006–Sep embe 2010), and a second au ho ac ed as head o pu chasing and logis ics depa men , which may ha e in oduced a bias in o he analysis. Howe e , o pa ly com- pensa e o his bias, a hi d in es iga o was in ol ed in he case s udy and a igo ous esea ch me hodology was applied. The second limi a ion is linked o he ac ha ou esea ch me hodology collec ed da a using a single case s udy. Indeed, we measu ed employee sa is ac ion owa ds kanban sys ems in a single hospi al g oup (AHVM), which makes us ques ion he ans- e abili y o he esul s o o he hospi als employing a kanban sys em. Howe e , ou esul s a e alid, om an analy ical poin o iew, since hey p esen an example ha ein o ces and de el- ops he exis ing heo y conce ning he ad an ages o using kanban sys ems in heal hca e. Simila s udies on kanban sys ems in hospi als, including c oss- case- ype s udies, should be conduc ed o u he alida e ou findings. Addi ionally, he wo ac o s ound in he p incipal compo- nen analysis (ad an ages o a kanban sys em and logis ics sys em pe o mance), and he a iables ep esen ed wi hin each, should be in es iga ed and confi med h ough u he esea ch in o heal h- ca e se ings. Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed. V.G. Aguila -Escoba e al. / In es igaciones Eu opeas de Di ección y Economía de la Emp esa 21 (2015) 101–110 109 APPENDIX. Kanban su ey Re e ences Adams, S. G., & Ruiz-Ulloa, B. C. (2003). An in es iga ion o pe sonnel issues a ec - ing Kanban pe o mance: A case s udy. Enginee ing Managemen Jou nal, 15(4), 19–27. Aguila -Escoba , V. G., & Ga ido-Vega, P. (2013). Ges ión Lean en logís ica de hos- pi ales: es udio de un caso. Re is a de Calidad Asis encial, 28(1), 42–49. Aguila -Escoba , V. G., Ga ido-Vega, P., & Godino-Gallego, N. (2013). Mejo ando la cadena de suminis o en un hospi al median e la ges ión Lean. Re is a de Calidad Asis encial, 28(6), 337–344. Documen downloaded om h p://www.else ie .es, day 21/08/2015. This copy is o pe sonal use. Any ansmission o his documen by any media o o ma is s ic ly p ohibi ed.