The implementation of Health in All Policies initiatives : a systems framework for government action

CASE STUDY Open Access The implemen a ion o Heal h in All Policies ini ia i es: a sys ems amewo k o go e nmen ac ion Ke an Shanka dass 1,2,3* , Ca les Mun ane 2,4 , Lau i Kokkinen 5 , Fa az Vahid Shahidi 2 , Alix F eile 1 , Goldamei Oneka 2 , Ahmed M. Bayoumi 1,6 and Pa icia O’Campo 1,2 Abs ac Backg ound: The e has been a enewed in e es in b oadening he esea ch agenda in heal h p omo ion o include ac ion on he s uc u al de e minan s o heal h, including a ocus on he implemen a ion o Heal h in All Policies (HiAP). Go e nmen s ha use HiAP ace he challenge o ins i u ing go e nance s uc u es and p ocesses o acili a e policy coo dina ion in an e idence-in o med manne . Due o he complexi y o go e nmen ins i u ions and he policy p ocess, sys ems heo y has been p oposed as a ool o e alua ing he implemen a ion o HiAP. Me hods: Ou mul iple case s udy esea ch p og amme (HiAP Analysis using Realis Me hods On In e na ional Case S udies –HARMONICS) has elied on sys ems heo y and ealis me hods o make sense o how and why he p ac ices o policy-make s (including poli icians and ci il se an s) om speci ic ins i u ional en i onmen s (policy sec o s) has ei he acili a ed o hinde ed he implemen a ion o HiAP. He ein, we p esen a sys ems amewo k o he implemen a ion o HiAP based on ou expe ience and empi ical indings in s udying his p ocess. Resul s: We desc ibe a sys em o 14 componen s wi hin h ee subsys ems o go e nmen . Subsys ems include he execu i e (heads o s a e and hei appoin ed poli ical eli es), in e sec o al ( he milieu o policy-make s and expe s wo king wi h go e nance s uc u es ela ed o HiAP) and in asec o al (policy-make s wi hin policy sec o s). He e, HiAP implemen a ion is a p ocess in ol ing in e ac ions be ween subsys ems and hei componen s ha leads o he eme gence o implemen a ion ou comes, as well as e ec s on he sys em componen s hemsel es. We also desc ibe he in luence o ex a-go e nmen al sys ems, including (bu no limi ed o) he academic sec o , hi d sec o , p i a e sec o and in e go e nmen al sec o . Finally, we p esen a case s udy ha applies his amewo k o unde s and he implemen a ion o HiAP – he Heal h 2015 S a egy –in Finland, om 2001 onwa d. Conclusions: This amewo k is use ul o helping o explain how, why and unde wha ci cums ances HiAP has been success ully and unsuccess ully implemen ed in a sus ainable manne . I se es as a ool o esea che s o s udy his p ocess, and o policy-make s and o he public heal h ac o s o manage his p ocess. Keywo ds: Sys em amewo k, Heal h equi y, Heal h in all Policies, In e sec o al ac ion * Co espondence: [email p o ec ed] 1 Cen e o U ban Heal h Solu ions, Li Ka Shing Knowledge Ins i u e, 209 Vic o ia S ee , To on o, ON M5B 1T8, Canada 2 Dalla Lana School o Public Heal h, Uni e si y o To on o, 155 College S ee , To on o, ON M5T 3M7, Canada Full lis o au ho in o ma ion is a ailable a he end o he a icle © The Au ho (s). 2018 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 h ps://doi.o g/10.1186/s12961-018-0295-z Backg ound By highligh ing he impo ance o he s uc u al de e - minan s o heal h as oo causes o heal h inequi y, he 2008 inal epo by he Commission on he Social De e minan s o Heal h encou aged b oadening o he heal h p omo ion esea ch agenda o include a ocus on go e nmen policies and p ocesses, alongside cul u e and socie al alues o popula ions [1]. Go e nmen s ha apply he idea o Heal h in all Policies (HiAP) o s eng hen heal h equi y s i e o du able and sys ema ic app oaches o pu suing in e sec- o al ac ion ha leads o heal hy and equi able public policies. Thus, he implemen a ion o HiAP ypically in- ol es ins i u ing wha Fa a d e e s o as “in eg a ed go e nance”([2], p. 2), in which go e nance s uc u es and p ocesses a e used o acili a e policy coo dina ion in an e idence-in o med manne , some imes guided by explici long- e m s a egies and goals [3]. In his o mu- la ion, HiAP is no a speci ic app oach (as p esen ed mo e ecen ly by WHO [4]), bu a he a gene al con- cep ha accommoda es di e se ini ia i es, whe e “HiAP ini ia i es may include mul iple p og ammes o p ojec s ha a e os e ed ac oss mul iple sec o s and mul iple le els o go e nmen ei he di ec ly o indi ec ly ela ed o he o iginal policy commi men ”([5], p. 465). Thus, assessing he implemen a ion o HiAP includes a ocus on how sec o s and app oaches a e coo dina ed and in eg a ed o e ime. As mo e go e nmen s ha e adop ed HiAP app oaches globally, he e ha e been calls o g ea e moni o ing o be e unde s and how in eg a ed go e nance can be used o mee heal h and equi y objec i es [4,6–8]. The need o such a en ion s ems om mul iple challenges. Fi s , he p ocess o implemen ing HiAP can be challen- ging since policy-make s a e no always expe ienced in wo king in e sec o ally. Fo example, long-s anding go e nmen adi ions o policy de elopmen and imple- men a ion wi hin dis inc o ganisa ional and p o essional silos can be ha d o ans o m [9]. Second, he e is o en a lack o e idence abou how o bes acili a e success ul HiAP implemen a ion; mos e idence is desc ip i e a he han analy ic [10]. Thi d, since HiAP is a concep a he han a model, e e y HiAP ini ia i e is uniquely designed and go e ned, and so i is challenging o unde - s and how o ansla e s udies o one case o o he s. Bha a no es ha because “i is much easie o se goals han o pu hem in o ac ion, implemen a ion conside - a ions end o be less accoun ed o han he o he p ocesses in policy de elopmen ”([11], p. 456). In he case o implemen ing HiAP, guidance has gene ally ocused on echnical challenges o policy implemen a- ion [12–15], whe eas he e a e also poli ical challenges o HiAP implemen a ion gi en he ocus on heal h equi y imp o emen [16]. Mo e undamen ally, he e has been li le applica ion o heo y o help es hypo heses abou why ce ain implemen a ion p ac ices wo k in some se ings, and wha o he con ex ual and collec i e ac o s in luence implemen a ion ou comes. Mo eo e , he complexi y o unde s anding policy p ocesses makes he gene a ion o hypo heses di icul [17]. Sys ems heo y has been p oposed as a ool o e alu- a ing heal h p omo ion ac i i ies, including hose unde - aken by go e nmen s [18,19]. The use o sys ems heo y implies ha he e is no simple app oach o implemen ing HiAP; a he , implemen a ion ou comes a e mos ly unp edic able gi en he complex in e - ela ionships be ween sys em componen s (e.g. sec o s wi h compe ing p io i ies, poli ical u no e and chan- ging policy agendas). Ye , sys ems heo y can o e a amewo k o unde s anding implemen a ion phenom- ena, including whe e implemen a ion is based on sho - sigh ed hinking o whe e i goes w ong [19]. This app oach has been used mo e ex ensi ely in unde s and- ing he managemen o heal hca e o ganisa ions [20], and has e en been associa ed wi h inc eased e ec i e- ness o in e en ions in hese se ings [21]. Al hough he e a e some excep ions (e.g. [22]), sys ems heo y has no been adequa ely in eg a ed in he discou se abou how o implemen HiAP. F om he ou se , ou mul iple case s udy esea ch p og amme (HiAP Analysis using Realis Me hods On In e na ional Case S udies –HARMONICS) has elied on sys ems heo y o make sense o how and why he p ac ices o policy-make s (including poli icians and ci il se an s) om speci ic ins i u ional en i onmen s (policy sec o s) has ei he acili a ed o hinde ed he implemen- a ion o HiAP [5]. To help o he esea che s lea n om case s udies o HiAP implemen a ion, his a icle begins by p o iding a desc ip ion o sys ems heo y ollowed by a desc ip ion o a sys em o 14 componen s wi hin h ee subsys ems o go e nmen , which we a gue mainly explain implemen a ion ou comes alongside some impo an ex a-go e nmen al in luences. We hen use a case s udy o demons a e how o apply he sys ems heo y ame- wo k, which in ol es he implemen a ion o he Heal h 2015 S a egy in Finland om 2001 onwa d, concep ua- lised he ein as a o m o HiAP [23]. Social sys ems (i.e. sys ems composed o indi iduals and he ela ions be ween hem) a e di icul o quan i y, which can hinde moni o ing and e alua ion. The amewo k we p opose, along wi h o he empi ical analyses om HAR- MONICS (e.g. [24]), o e s bo h heo y and an analy ic ap- p oach based on his heo y. This app oach allows he gene a ion o hypo heses abou he implemen a ion p ocess and he es ing o hese hypo heses by e alua ing how well he heo y explains how, why and unde wha ci cums ances HiAP has been success ully implemen ed in a sus ainable manne . Policy-make s a e o pa icula Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 2 o 10 in e es as indi iduals who a e engaged in and possibly managing aspec s o he go e nmen sys em o HiAP implemen a ion. Fo example, he implemen a ion s a - egies o policy-make s may a ec how esou ces a e dis ibu ed o lead and suppo implemen a ion ac i i ies. Policy-make s can use his amewo k o unde s and hei own go e nmen sys ems, and o an icipa e he po- en ial challenges and impac s o a ious s a egies o HiAP implemen a ion. A Sys ems Theo y p ime Sys ems Theo y eme ged in he Twen ie h cen u y as a se o heo ies ha encompassed mul iple ields, includ- ing philosophy as well as basic and applied science (e.g. compu ing science) [25,26]. I can be labelled a me a- heo y in ha Sys ems Theo y sea ches o commonal- i ies ac oss biological, physical and social sys ems. In b ie , i posi s ha he wo ld is made o wholes (sys ems) whose p ope ies (including eme gen ou comes) canno be explained by a simple accoun ing o i s componen s [27]. Componen s a e o ganised wi hin subsys ems ha a e inhe en ly in e dependen [28], so changes o one subsys em will a ec o he s, some imes in unexpec ed ways [26]. Eme gen p ope ies o a sys em a e hose no possessed by any single componen o he sys em, and eme gence e e s o he p ocess by which he sys em ac- qui es hese p ope ies [29]. Such on ology leads o an epis emology ha ejec s bo h educ ion ( he exclusi e s udy o he componen s o a sys em) and holism ( he neglec o he componen s o a sys em). To use an example om ou side he domain o HiAP implemen a ion, we may wan o unde s and social cohesion in a neighbou hood (social) sys em. S udying only he psychology o neighbou hood esiden s would no gi e insigh in o in e -pe sonal p ocesses, such as ecip oci y no ms o exchange ne wo ks, which cons i- u e impo an gene a i e mechanisms o explain social cohesion. Minge s [27] desc ibes gene a i e mechanisms as he p ocesses ha keep a sys em in mo ion, which occu in ela ion o speci ic con ex ual condi ions. He ein, he s udy o sys em componen s alone (e.g. social cha ac e is ics, such as he c iminali y o neigh- bou hood esiden s) canno explain eme gen p ope ies o his sys em, like social cohesion, because hey do no unco e he ela ions among componen s (e.g. no ms o ecip oci y) and he ole o con ex . Eme gen p ope ies a ise as a consequence o in e ac- ions be ween sys em componen s, including eedback ha changes sys em componen s and o he ou comes o ele ance [30,31]. Sys ems can he e o e ope a e in non-linea ways wi h seemingly ob ious solu ions some- imes wo sening a p oblem ([32], p. 19). Sys ems and subsys ems ha e bounda ies ha a e pe meable, so ha changes o sys ems in he ex e nal en i onmen o hese domains may in luence how and why he sys em wo ks [26,33]. As desc ibed below, his sys ems amewo k ocuses on how ce ain go e nance s uc u es, p ocesses and s a egies can help o se he agenda and build capaci y o HiAP implemen a ion o acili a e sus ainable ini ia i es. In doing so, his concep ual amewo k includes a se ies o subsys ems, componen s and ex e - nal in luences ha can in o m hypo heses abou how o explain and an icipa e HiAP implemen a ion ou comes in di e en se ings. HiAP implemen a ion wi hin a go e nmen sys em: he HARMONICS app oach Since HiAP ini ia i es a e ypically manda ed o im- plemen a a ie y o heal h equi y in e en ions ac oss mul iple policy sec o s and mul iple geog aphic le els o go e nmen o e ime (e.g. [34]), we iew imple- men a ion as equi ing an on-going adap i e p ocess o go e nmen s in e ms o hei go e nance s uc- u es and s a egies, simila o wha Hummelb unne calls an “open change p ocess”([35], p. 395). Thus, he sys ems amewo k desc ibed below is in ended as a ool o esea che s o s udy his p ocess and o policy-make s and o he public heal h ac o s o manage i . This amewo k was de eloped by ou eam i e a- i ely o e he cou se o many discussions be ween 2015 and 2016, wi h a ocus on a icula ing he mos impo an and unique ace s o go e nmen sys ems ele an o he implemen a ion o HiAP. These dis- cussions we e in o med by (1) se e al exis ing heo - ies abou policy implemen a ion (e.g. [2,3]), including ou own ini ial amewo ks abou he implemen a ion o HiAP [3,5], and (2) a wide ange o e idence on he opic o implemen ing in e sec o al ac ion o heal h equi y [10], including ou analysis o h ee case s udies o HiAP implemen a ion in Sweden, Quebec and Sou h Aus alia [3,24,36]. Impo an ly, we also apply a ealis scien i ic app oach o ou case s udies, which is an on ology ha has ocused ou wo k on unde s anding he o -hidden gene a i e mechanisms ha explain implemen a ion ou comes (desc ibed u - he below) [5]. The amewo k below p esen s HiAP implemen a ion as a p ocess in ol ing in e ac ions be ween sys em com- ponen s and subsys ems in ou model ha leads o he eme gence o bo h a ou able and un a ou able imple- men a ion ou comes, as well as e ec s on he sys em componen s hemsel es (i.e. eedback). Figu e 1and Table 1ou lines ou sys ems amewo k o HiAP imple- men a ion in e ms o a se ies o componen s pe aining o h ee subsys ems, including he execu i e (heads o Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 3 o 10 s a e and hei appoin ed poli ical eli es), in e sec o al ( he milieu o policy-make s and expe s wo king wi h go e nance s uc u es ela ed o HiAP) and in asec o al (policy-make s wi hin policy sec o s). The implemen a ion o HiAP o en equi es policy co- o dina ion ac oss mul iple go e nmen le els o sys ems (e.g. na ional, p o incial/s a e, egional and local), as well as wi h o he sys ems ou side o go e nmen ha a ec heal h equi y ( ha do no necessa ily look like Fig. 1). The e o e, we also include a ole o ex a-go e nmen al sys ems, which can in luence he implemen a ion o HiAP in many ways (desc ibed u he below). In HARMONICS, sys ems heo y has been use ul o es ing hypo heses and explaining phenomena abou wo a ou able ou comes o he implemen a ion p ocess, namely he accep abili y (i.e. a e sec o s willing o collabo a e on heal h and equi y?) and easibili y (i.e. do sec o s ha e he capaci y o e ec i ely collabo a e on heal h and equi y?) o HiAP implemen a ion o pa ne s in di e se policy sec o s ela ed o he ongoing p ocess o implemen a ion [37]. Accep abili y and easibili y a e a ou able inasmuch as hey con ibu e o mo e sus ain- able implemen a ion o HiAP and hey may also lead o s onge ypes o HiAP. Fo example, accep abili y and easibili y may acili a e a b oade inclusion o policy sec o s ha pa icipa e in HiAP o e ime, e lec ing bo h a la ge numbe o sec o s and g ea e di e si y among hose sec o s, e.g. including social and economic-o ien ed sec o s. The e may also be s onge equi y in e en ions (i.e. in e en ions ha a e mo e ups eam), as e lec ed by he alues and ideas implici in he heo y o change o esul ing heal h equi y in e en ions. Case s udy: The implemen a ion o Heal h 2015 in Finland Below, we demons a e he alue o using he sys ems amewo k o HiAP implemen a ion (sys em componen s om Fig. 1and Table 1a e i alicised) o o ganise in o ma- ion abou a pa icula policy issue and ou come in he implemen a ion o Heal h 2015 in Finland. Following he case desc ip ion, we o e some e lec ions on gene a i e mechanisms and nex s eps o unde s anding he ou - comes o he case. Ou case s udy d aws e idence om a ealis explana o y case s udy o his p ojec [23], including he analysis o pee - e iewed and g ey li e a u e, go e nmen epo s and key in o man in e iews wi h 10 policy-make s in ol ed in he implemen a ion o Heal h 2015. Mo e de ails abou ou me hodology a e a ailable in Shanka dass e al. [5]. Backg ound o in e sec o al ac ion (ISA) o heal h equi y in Finland In he ea ly 1970s, i was ecognised by public o icials ha he heal h o Finland’s popula ion was lagging behind economic p og ess [38]. As a esul , he imp o e- men o popula ion heal h became a poli ical p io i y o he Finnish go e nmen (policy agenda). Since ha ime, policy sec o s ha e been g owing hei (p io ) expe ience wi h ISA. In 1972, he go e nmen adop ed a public heal h law ha in oduced a amewo k o in e sec o al collabo a ion be ween he heal h sec o and o he go e nmen sec o s, as well as non-go e nmen al o ganisa ions and he p i a e sec o , o educe ca dio as- cula disease in he popula ion. In 1986, a na ional Heal h Fo All s a egy was implemen ed o encou age b oade e o s owa ds ISA o heal h. Following an Fig. 1 A sys ems amewo k depic ing 14 componen s wi hin h ee go e nmen sub-sys ems in ol ed in HiAP implemen a ion Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 4 o 10 Table 1 De ini ion o sub-sys ems and sys em componen s Execu i e Subsys em: The p ocesses o go e nmen esponsible o he c ea ion and implemen a ion o legisla i e manda es ela ed o he implemen a ion o HiAP ini ia i es. Policy agenda: The ini e se o social and poli ical issues upon which go e nmen s ac on a a gi en poin in ime, which will be shaped by he pa y o ganisa ion(s) who con ol he go e nmen and in luenced by ex a-go e nmen al ac o s, and which ha e implica ions o he p io i y o heal h equi y ini ia i es like HiAP. Poli ical eli es: Ac o s who, by i ue o some o m o o mal au ho i y wi hin he go e nmen , exhibi a consis en and subs an ial le el o con ol o e he poli ical p ocess and i s ou comes, including as hey ela e o he implemen a ion o HiAP ini ia i es. Poli ical ideology: The clus e o ideas, belie s, alues and a i udes ha cons i u e he no ma i e lens h ough which poli ical eli es and pa y leade s in e p e and ac upon social and poli ical issues. Fo example, hey ha e poli ical in e es s in ela ion o hei opinions abou ‘wha ough o be’when i comes o de eloping o modi ying policies du ing he implemen a ion o HiAP ini ia i es. Those opinions may, in u n, be shaped by hei expe iences and unde s anding o he eal wo ld, and by hei mo e undamen al wo ld iews and ideological subsc ip ions, e.g. he ole o s a e o he p imacy o indi idual esponsibili y. In e sec o al Subsys em: The p ocesses o go e nmen ha acili a e he ho izon al and e ical coo dina ion o he HiAP policy agenda ac oss a ious sec o s o he go e nmen and wi h ex a-go e nmen al pa ne s. Expe ad iso s: Expe indi iduals (o en om ou side o go e nmen ) who a e o mally consul ed in planning and execu ing he implemen a ion o HiAP ini ia i es. Expe ad iso s a e a ype o policy eli e, i.e. hey ha e in luence o e he policy p ocess. HiAP inancial a angemen s: Financial a angemen s ha dic a e he magni ude, dis ibu ion and sou ces o unding a ailable o he implemen a ion o HiAP ini ia i es. HiAP managemen : The se o echnical p ocesses h ough which go e nmen s gene a e ins i u ional capaci y o implemen a ion o HiAP ini ia i es. HiAP manda e: O icial legisla ion and o mal s a egies con aining speci ic ins uc ions o he implemen a ion o HiAP ini ia i es (e.g. policy goals, di ision o esponsibili y, alloca ion o esou ces, p ocesses o moni o ing and en o cemen ), which may change o g ow in numbe o e he pe iod o implemen a ion (see F eile e al. 2013 o addi ional in o ma ion). In asec o al Subsys em: The p ocesses o go e nmen ha acili a e ac i i ies such as he pu sui o sec o al objec i es, which may be a ec ed by he implemen a ion o HiAP ini ia i es. High- anking ci il se an s: Bu eauc a s who may ha e au ho i y o e he policy p ocess delega ed o hem by poli ical eli es. High- anking ci il se an s a e a ype o policy eli e, i.e. hey ha e in luence o e he policy p ocess, and may be pa icula ly engaged in he echnical aspec s o implemen ing HiAP ini ia i es. P io expe ience wi h in e sec o al ac ion (ISA): A his o y o wo king in e sec o ally on sha ed policy objec i es, which may in luence how he implemen a ion o ela ed ini ia i es, such as HiAP, occu s. Sec o al ideology: The clus e o ideas, belie s, alues and a i udes ha cons i u e he no ma i e lens h ough which policy-make s wi hin a gi en sec o in e p e and ac upon social and poli ical issues such as heal h equi y, and which may a y gi en sec o al objec i es (e.g. heal hca e, popula ion heal h, economic g ow h, enginee ing), i.e. a wo ld iew. Sec o al objec i es: Goals and mo i a ions o policy sec o s, o en deli e ed h ough a o mal manda e om he execu i e, which may be a ec ed by a go e nmen ’s implemen a ion o HiAP ini ia i es. Sec o al powe : The o mal au ho i y alloca ed o policy sec o s in go e nmen manda es, indica ed in absolu e (e.g. di ec i es, budge size) and ela i e e ms ac oss sec o s (e.g. ela i e budge size). In e sec o al coo dina ion o policies in he p ocess o implemen ing HiAP ini ia i es may be shaped by he powe o he sec o s in ol ed. Wo k o ce capaci y o ISA: The ex en o expe ise among human esou ces wi h ools and p ocesses and wo k o ce size dedica ed o implemen ing HiAP ini ia i es, enabling easibili y (see F eile e al. 2013 o addi ional in o ma ion). Wo k o ce HiAP awa eness: An unde s anding o he need and easons o an in e sec o al app oach o add ess heal h equi y, as pa o he p ocess o agenda se ing and, ul ima ely, buy-in o he implemen a ion o HiAP ini ia i es (see F eile e al. 2013 o addi ional in o ma ion). Ex a-go e nmen al sys ems: Sys ems ou side o go e nmen ha can in luence HiAP implemen a ion, including as o ganisa ions and indi iduals become pa ne ed o he implemen a ion o HiAP; o example, by pa icipa ing in planning o execu ing in e sec o al ac ion o in being he subjec o some a endan egula o y ac ion. The e a e also likely o be mo e indi ec in luences, such as policy en ep eneu s who ad oca e o lobby o in luence o e he implemen a ion p ocess, and c oss-na ional policy and agenda-se ing amewo ks. Finally, a he global and local le els, he e a e esea ch p og ammes and knowledge hubs p oducing in o ma ion o suppo implemen a ion. Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 5 o 10 e alua ion conduc ed by WHO expe s [39], a e ised s a egy was adop ed in 1993 o accommoda e a g ea e ocus on heal h equi y [40]. O e all, he e was a long his o y o p io expe ience wi h ISA be o e Heal h 2015 was adop ed, which would ha e c ea ed some ins i u- ional awa eness abou he p oblem o heal h equi y and he need o ISA o add ess his p oblem. A he u n o he millennium, as he e m o he Heal h Fo All s a egy was ending, Heal h 2015 was adop ed as a long- e m s a egy o add essing heal h equi y using ISA (HiAP manda e;see[41] o mo e de ails). As pa o he implemen a ion o he Heal h 2015, Finland’sMinis yo Social A ai s and Heal h de eloped guidelines o di ec o he sec o s o conside using Heal h Impac Assess- men s o con ibu e o he e alua ion o key policy decisions –a o mo HiAP managemen . This was a p ocess sugges ed by WHO and he EU (en i onmen al in luences –sup ana ional poli ical o ganisa ions). No laws we e in oduced alongside hese guidelines o make pa icipa ing in Heal h 2015 manda o y, as we e passed in he case o Quebec’s e ised Public Heal h Bill [41], ano he example o a HiAP ini ia i e. Ou case s udy ocuses on a single policy issue ha a ose du ing he implemen a ion o Heal h 2015 conce n- ing he go e nmen ’s decision-making abou whe he o e o m na ional alcohol policy (policy agenda) ollowing he in oduc ion o Es onia o he EU in 2004 (ex e nal in luences –sup ana ional poli ical o ganisa ion), in he con ex o a Heal h Impac Assessmen ha was con- duc ed on his issue. Since he 1970s, he cos o alcohol in Finland had been egula ed by a high ax a e ha discou aged he consump ion o alcohol on he one hand and p o ided a con enien sou ce o ax e enue on he o he [42]. F om he poin o iew o main aining high axes, he Finnish go e nmen conside ed his a angemen o be iable as long as Sweden and Denma k ac ed as bu e s be ween Finland and Ge many, wi h he la e being, a he ime, he nea es EU Membe S a e wi h low alcohol axes. Howe e , he easibili y o he go e nmen ’s exis - ing app oach o alcohol policy was unde mined when he EU (en i onmen al in luences –sup ana ional poli - ical o ganisa ion) app o ed he membe ship o Es onia, whose alcohol axes we e as low as hose in Ge many. On he one hand, economic assessmen s by he Minis y o Finance p edic ed ha , due o he ee mo emen o goods ac oss EU coun ies, a conside able inc ease in Finnish a elle s’impo o alcohol om Es onia would ake place. They p ojec ed dec easing p o i s o Finnish b ewe ies and o he businesses ha we e dependen on alcohol p oduc ion and sales ( ele an o he sec o al objec i es o he Minis y o Finance). In an a emp o a oid his scena io, he Minis y o Finance p oposed a educ ion o alcohol sales axes by one- hi d [43]. Howe e , a Heal h Impac Assessmen conduc ed by he Minis y o Social A ai s and Heal h on he p oposal p ojec ed ha lowe cos s o alcohol would lead o inc eased consump ion and ela ed nega i e heal h impac s ( ele an o he sec o al objec i es o he Minis y o Social A ai s and Heal h). The Heal h Impac Assessmen sugges ed ha a ax educ ion o 33% would inc ease he numbe o hea y d inke s by 200,000 (4% o he o al popula ion) and inc ease alcohol- ela ed dea hs by 600 a yea [43]. Finally, poli ical eli es in he go e nmen decided o implemen he ax educ ion. Gene a i e mechanisms in Finland: a sys ems iew Why did economic a gumen s ca y mo e weigh han public heal h and equi y conside a ions in his case, in spi e o p io ISA expe ience? A numbe o mechanisms in ol ing mul iple subsys ems and componen s can help o explain his ou come. Below, we a gue ha compe ing sec o al objec i es be ween he Minis y o Finance and he Minis y o Social A ai s and Heal h, i.e. economic and heal h objec i es e lec ing dis inc in e es s and ideas, a ec ed he go e nmen ’s decision-making abou alcohol policy in an icipa ion o Es onia joining he EU. In pa icula , c i ical ac o s included he in luence o a s ong alcohol lobby (en i onmen al in luences –p i a e sec o lobby o ganisa ion), an ideological commi men o neolibe alism since he ea ly 1990s (poli ical ideology) [44] and a concomi an ocus on p omo ing economic g ow h a he expense o o he aims, including social equi y (policy agenda) and a non-di ec i e implemen a- ion s yle o HiAP (HiAP managemen ). Joining he EU in 1995 led o libe alisa ion o he Finnish economy and inc eased ade, which augmen ed he ole o lobbyis s in he poli ical p ocess. By he ime Es onia’s membe ship was app o ed, a s ong lobby ad oca ing o he lowe ing o alcohol axes had es ab- lished i sel , e en amongs he coun y’s poli ical execu- i e (poli ical agenda)[45], in pa h ough elec ion campaign con ibu ions paid o by he alcohol and e ail indus ies (en i onmen al in luences –p i a e sec o lobby o ganisa ion). Mass media campaigns also shaped he poli ical agenda by a ousing popula conce n o e he economic consequences o main aining high alcohol axes (en i onmen al in luences –media). On he o he hand, he e was compa a i ely minimal lobbying agains he p oposal o lowe hem (en i onmen al in luences – lobby o ganisa ion)[45]. Poli ical eli es wi hin he go e nmen execu i e we e s ongly ocused on he sec o al objec i es o he Minis y o Finance, such as secu ing economic g ow h and es aining in la ion [46]. Ou key in o man s desc ibed how economic sec o s (such as hose conce ned wi h inance, ag icul u e and indus y) we e gene ally less esponsi e o he esul s o he Heal h Impac Assessmen , pa icula ly because he poli ical Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 6 o 10 ideology o he uling pa y a ou ed economic g ow h o e social objec i es such as heal h equi y. The Minis y o Finance, in pa icula , exhibi ed esis ance o e o s o ake in o conside a ion he po en ial heal h conse- quences o hei ax e o m p oposal [46,47]. Ou key in o man in e iews also sugges ha poli ical eli es in cen al go e nmen did no p o ide s ong leade ship in he poli ical p ocess (i.e. a guably a lack o HiAP managemen ) o achie e he sec o al objec i es o he Minis y, which hinde ed buy-in om sec o s o Heal h 2015. As Vilen [48] no es, widesp ead popula suppo o s ong ax policies in he 1970s, 1980s, and s ill in he 1990s had g own weake by he ime Heal h 2015 was being implemen ed, so i may ha e been poli ically unpopula o poli ical eli es o ad oca e o mo e ups eam in e en ions (conce n o sec o al powe ). Vilen [48] also desc ibes how he Minis e o Social A ai s and Heal h and his cabine we e occupied wi h he ongoing heal hca e e o m, lea ing less ime o p omo ing o he issues (compe ing sec o al objec i es). Finally, by using ou sys ems amewo k o help o gan- ise and in e p e ou da a, we migh a gue ha he ins i- u ional awa eness in he Finnish go e nmen o he need o ISA o p omo e popula ion heal h and equi y coupled wi h he Heal h 2015 S a egy we e less pe sua- si e on poli ical eli es in he Minis y o Finance and elsewhe e in he execu i e han he need o pu sue he economic in e es s o he alcohol lobby in Finland. This asse ion can se e as a ype o wha Me on e e ed o as “middle ange heo y”([49], p. 39), which can be explo ed and e ined ia a numbe o c i ical ques ions abou how and why ce ain subsys ems (and compo- nen s) enabled economic g ow h objec i es o be p io i- ised o e popula ion heal h and heal h equi y objec i es [50]. Fo example, wha ole did he sec o al powe o he Minis y o Finance and he Minis y o Social A - ai s and Heal h play in his ou come, pa icula ly in he in e ac ion wi h he execu i e subsys em (i.e. a cen e- le wing coali ion uling in he e a o neolibe alism wi h s ong indus y lobby)? Would we expec a di e en ou - come wi h di e en poli ical eli es in powe ? Gi en ha he execu i e subsys em ( ia he ac ions o poli ical eli es) was mo e powe ul han he in e sec o al subsys- em ( ia he HiAP manda e o Heal h 2015), would a s onge manda e ha e led o a di e en ou come (e.g. a law s a ing ha a chie medical o ice o heal h mus e- iew Heal h Impac Assessmen s and p o ide o e sigh o ela ed policy ecommenda ions)? Discussion Ou use o sys ems heo y yields se e al bene i s o he implemen a ion o HiAP app oaches, including (1) an explici acknowledgemen o he in e ela ionships be ween go e nmen subsys ems and hei componen s in ol ed in HiAP implemen a ion, (2) a amewo k o con ex ualise he complex and eme gen p ocesses o implemen a ion o help explain how and why sys em componen s wo k oge he –and unde wha ci cum- s ances – o enable HiAP o succeed o ail, and (3) con- side a ion o how he sys ems o ex e nal in luences impac go e nmen sys ems in HiAP implemen a ion. We iew his amewo k as a ool o policy-make s cha ged wi h managing implemen a ion p ocesses, and o esea che s in e es ed in heo ising abou HiAP implemen a ion and explaining obse a ions. I will also help hose in e es ed in o he pa s o he HiAP policy p ocess (including design, adop ion and e alua ion) o unde s and he complexi ies o implemen a ion. O e ime, i obse a ions wi hin and ac oss case se ings e eal pa e ns in how sys em componen s in e ac and impac implemen a ion ou comes, gene al p inciples o his sys em may become appa en . Fo example, we no ed ha he a gumen s o he p i a e sec o lobby o ganisa ions in Finland ( o educe alcohol axes) ul ima ely had a nega i e impac on he imple- men a ion o Heal h 2015 because he uling pa y had a poli ical ideology ha a ou ed economic g ow h o e heal h equi y and, hus, op ed o no heed he indings o he Heal h Impac Assessmen ha was conduc ed. I his se o ela ions p oduces a simila ou come in Finland o e ime, his may cons i u e a a ionale o in oducing go e nance s uc u es and s a egies o managing ela ionships wi h p i a e sec o s akeholde s in he implemen a ion o HiAP. I such ou comes eme ge ac oss a a ie y o se ings whe e HiAP is being implemen ed, hen sup ana ional heal h o ganisa ions, such as WHO, may ake he lead in ecommending such ac ions globally. As local sys ems a e be e unde s ood o e ime, policy-make s can d aw on hei unde s anding o in e - ela ionships amongs componen s o acili a e wha No man e e s o as sys ems-based “change s a egies” ([18], p. 870) o heal h p omo ion. Fo example, p io o adop ing Heal h 2015 in Finland, a g oup o WHO expe s iden i ied he eliance on al eady o e s e ched ci il se an s o moni o and e alua e he p og ess o implemen a ion, in addi ion o hei no mal du ies, as a weakness o implemen ing he Finnish Heal h Fo All s a egy. The idea a ose o an Ad iso y Boa d o Public Heal h o unc ion as a collabo a i e body o in e sec- o al coope a ion [43], and his Boa d con inued o se e as a coo dina ing mechanism h ough he implemen a- ion o Heal h 2015. Using he sys ems amewo k, he ini ial challenge can be unde s ood as a p oblem o wo k o ce capaci y o ISA in he in asec o al subsys- em, and he change s a egy as esponding o ha need by in oducing a new body o HiAP managemen in he in e sec o al subsys em. Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 7 o 10 As one example o how esea che s can u ilise his amewo k, in HARMONICS, we ha e d awn on i using a ealis science app oach o s udy he social mechanisms o HiAP implemen a ion. Realis science d aws on c i ical ealism, which is an on ology ha posi s he exis ence o a wo ld ou side he obse e ha emains la gely hidden om he obse e , like he gea s o a clock. Roy Bashka (a key p oponen o c i ical ealism) inco po a ed sys ems hinking in his la e wo k [51], and Minge s ecen ly discussed se e al concep s ha belong o sys ems heo y [27], which we ha e ac ualised in HARMONICS. Fo example, he ‘s uc u e’o a sys em can be hough o as he sum o ela ions wi hin a sys em, which we ha e ep esen ed as a se ies o componen s wi hin subsys ems in Fig. 1.‘Eme gence’deno es p ope ies o a sys em ha a e no held by any componen o he sys em. This includes he ou comes o HiAP implemen a ion ha we ocus on in HARMONICS such as sus ainabili y. Finally, ealis science gene ally aims o explain phenomena in ways ha allow o complex gene a i e mechanisms. Fo example, he explana o y case s udy me hodology de el- oped o HARMONICS uses “con ex -mechanism-ou - comes pa e n con igu a ions”([5], p. 464) o ocus ou wo k on unco e ing hese gene a i e mechanisms in he cou se o analysis. O he ealis science me hodologies ha e also been use ul o enac ing a sys ems heo y app oach in HAR- MONICS. We ha e d awn on ecen wo k o apply ealis science o heal h sys em ans o ma ion (e.g. [52, 53]), which ecommends s a ing wi h some “ini ial p og amme heo y”([53], p. 7) abou how an in e en- ion wo ks (including concep ual amewo ks and hypo heses), and hen using his o guide he a icula ion o gene a i e mechanisms and o e en ually p oduce a e ined heo y. Ra he han wo king owa d a g and heo y ha explains all causal ins ances, he end p oduc he e is wha sociologis s desc ibe as middle ange heo y, which e lec s he causal mechanisms in speci ic ci cums ances (i.e. as e lec ed in a pa icula da abase) [49,50]. Because in o ma ion abou disc e e mechanisms can be ha d o summa ise in a simple manne –and causal a ibu ion di icul o pinpoin –when s udying he implemen a ion o HiAP, his sys ems amewo k has se ed as a use ul heu is ic o de eloping and es ing hypo heses ac oss cases in HARMONICS by e ealing componen s and guiding heo y abou in e - ela ionships. This sys ems amewo k is no a panacea o unde - s anding he policy-making p ocess. In pa icula , we ocus mainly on he policy-making p ocess wi hin go e nmen a he mac o/s uc u al le el, wi h less a - en ion o p ocesses a he mic o le el, al hough hese a e also occu ing wi hin he sys em. While policy- make s can endea ou o plan ce ain aspec s o HiAP implemen a ion, some ou comes will ma e ialise ou side he con ol o indi idual policy-make s and o he social echnologis s [54]; his is a key quali y o complex sys ems. We ha e de eloped his amewo k h ough ou expe ience analysing a se ies o explana o y case s udies o HiAP implemen a ion (e.g. [24]), and heo izing on he opic (e.g. [55,56]). This amewo k will be u he desc ibed and s eng hened as we and o he s con inue es ing i s explana o y powe in empi ical s udies o his opic. Al hough Table 1desc ibes some o he in luences o ex a-go e nmen al sys ems, i is wo h no ing ha hese a e likely o be nume ous and a ied ac oss se - ings, and his aspec clea ly can be u he elabo a ed on. Fo example, while he e a e heal h esea ch sys ems ha c ea e, ansla e and dissemina e knowledge ha in- o ms implemen a ion [57,58], he impac o his know- ledge may be o limi ed u ili y o policy-make s engaged in HiAP implemen a ion since scien is s o en ocus on one aspec o mul i- ace ed poli ical p oblems [59,60]. O he public, p i a e and ‘ hi d sec o ’sys ems (exis ing “be ween he ma ke and he s a e”[61], p. 5) may indi - ec ly in luence implemen a ion because hei manda es and in e es s include public heal h, heal h equi y o so- cial wel a e. They may also pu sue o he objec i es ha con lic wi h hese goals, like iscal conse a ism o p o i . Conclusions He ein, we desc ibed a sys ems amewo k o he implemen a ion o HiAP ha a icula es a se ies o ele- an subsys ems and componen s, and hen demon- s a ed he explana o y alue o his amewo k in a case s udy o he Heal h 2015 S a egy in Finland. Al hough policy implemen a ion has been desc ibed as a complex p ocess, his sys ems heo y o e s an o ganised iew o his p ocess by ocusing on ce ain cha ac e is ics o a go e nmen and i s ex e nal en i onmen while unco e - ing unde lying ela ionships [62]. This con ibu ion is imely as he e has been e y li le applica ion o Sys ems Theo y o heal h equi y in e en ions, including o ms o in e sec o al ac ion such as HiAP. Abb e ia ions HARMONICS: HiAP Analysis using Realis Me hods On In e na ional Case S udies; HiAP: Heal h in all Policies Acknowledgemen s The au ho s would like o hank Deb Finn o he con ibu ion o he esea ch. Funding This wo k was suppo ed by Canadian Ins i u e o Heal h Resea ch g an numbe s 111608 and 96566, he On a io Minis y o Heal h and Long-Te m Ca e and Wil id Lau ie Uni e si y. A ailabili y o da a and ma e ials The da ase s gene a ed and/o summa y ables om ou analysis du ing he cu en s udy a e no publicly a ailable due he sensi i i y o con iden ial Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 8 o 10 in e iews wi h Finnish go e nmen employees, bu a e a ailable om he co esponding au ho in a de-iden i ied manne on easonable eques . Au ho s’con ibu ions All au ho s con ibu ed o he concep ualisa ion and w i ing o he manusc ip . E hics app o al and consen o pa icipa e Resea ch E hics Boa d, S . Michaels Hospi al, #10–162. Consen o publica ion No applicable. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Publishe ’sNo e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. Au ho de ails 1 Cen e o U ban Heal h Solu ions, Li Ka Shing Knowledge Ins i u e, 209 Vic o ia S ee , To on o, ON M5B 1T8, Canada. 2 Dalla Lana School o Public Heal h, Uni e si y o To on o, 155 College S ee , To on o, ON M5T 3M7, Canada. 3 Depa men o Heal h Sciences, Wil id Lau ie Uni e si y, 75 Uni e si y A e W, Wa e loo, ON N2L 3C5, Canada. 4 Bloombe g School o Nu sing, Uni e si y o To on o, 155 College S ee , To on o, ON M5T 3M7, Canada. 5 Finnish Ins i u e o Occupa ional Heal h, PL 40, 00251 Helsinki, Finland. 6 Depa men o Medicine and Ins i u e o Heal h Policy, Managemen and E alua ion, Uni e si y o To on o, 155 College S ee , To on o, ON M5T 3M7, Canada. Recei ed: 22 Ma ch 2017 Accep ed: 8 Feb ua y 2018 Re e ences 1. Sola O, I win A. A Concep ual F amewo k o Ac ion on he Social De e minan s o Heal h: Discussion pape o he Commission on Social De e minan s o Heal h. Gene a: Wo ld Heal h O ganiza ion; 2007. 2. Fa a d P. Heal h in All Mee s Ho izon al Go e nmen . G enoble: Fi s In e na ional Con e ence on Public Policy; 2013. 3. Shanka dass K, Sola O, Mu phy K, F eile A, Bobbili S, Bayoumi A, O’Campo P. Heal h in All Policies: Resul s o a Realis -In o med Scoping Re iew o he Li e a u e. In: Ge ing S a ed Wi h Heal h in All Policies: A Repo o he On a io Minis y o Heal h and Long Te m Ca e. To on o: Cen e o Resea ch on Inne Ci y Heal h; 2011. h p://www.s michaelshospi al.com/ c ich/wp-con en /uploads/Heal h-in-All-Policies-A-Snapsho - o -On a io_ FINAL.pd . Accessed 24 Feb 2018. 4. Heal h in All Policies (HiAP) amewo k o coun y ac ion. Heal h P omo In . 2014;29:i19–28. 5. Shanka dass K, Renahy E, Mun ane C, O'Campo P. S eng hening he implemen a ion o Heal h in All Policies: a me hodology o ealis explana o y case s udies. Heal h Policy Plan. 2015;30(4):462–73. 6. Wo ld Heal h O ganiza ion, Go e nmen o Sou h Aus alia. Adelaide S a emen on Heal h in All Policies. Adelaide: Go e nmen o Sou h Aus alia; 2010. 7. Wo ld Heal h O ganiza ion. Fi s D a o a F amewo k o Coun y Ac ion Ac oss Sec o s o Heal h and Heal h Equi y. Gene a: Wo ld Heal h O ganiza ion; 2014. 8. Wo ld Heal h O ganiza ion, Minis y o Social A ai s and Heal h (Finland). The Helsinki S a emen on Heal h in All Policies. The 8 h Global Con e ence on Heal h P omo ion. 2013. h p://www.who.in /heal hp omo ion/ con e ences/8gchp/8gchp_helsinki_s a emen .pd . Accessed 20 Feb 2018. 9. Ca michael L, Ba on H, G ay S, Lease H. Heal h-in eg a ed planning a he local le el in England: Impedimen s and oppo uni ies. Land Use Policy. 2013;31:259–66. 10. Shanka dass K, Sola O, Mu phy K, G ea es L, O'Campo P. A scoping e iew o in e sec o al ac ion o heal h equi y in ol ing go e nmen s. In J Public Heal h. 2012;57(1):25–33. 11. Bha a G. In e na ional Dic iona y o Public Managemen and Go e nance. London: Rou ledge; 2005. 12. Cook S, Leppo K, Ollila E, Pena S, Wisma M. Heal h in All Policies: Seizing Oppo uni ies, Implemen ing Policies. Helsinki: Minis y o Social A ai s and Heal h o Finland; 2013. 13. MQ DV, Wisma M, Lin V, Jones CM, Da ies M, edi o s. In e sec o al Go e nance o Heal h in All Policies: S uc u es, Ac ions and Expe iences. Copenhagen: WHO Regional O ice o Eu ope; 2012. 14. S . Pie e L. Go e nance ools and amewo k o Heal h in All Policies. Quebec Ci y: Na ional Collabo a ing Cen e o Heal hy Public Policy, In e na ional Union o Heal h P omo ion and Educa ion and Eu opean Obse a o y on Heal h Sys ems and Policies; 2009. 15. S åhl T, Wisma M, Ollila E, Lah inen E, Leppo K, edi o s. Heal h in All Policies: P ospec s and po en ials. Finland: Minis y o Social A ai s and Heal h; 2006. 16. Thow AM, Ques ed C, Ju en in L, Kun R, Khan AN, Swinbu n B. Taxing so d inks in he Paci ic: implemen a ion lessons o imp o ing heal h. Heal h P omo In . 2011;26(1):55–64. 17. Cai ney P. Unde s anding Public Policy: Theo ies and Issues. London: Palg a e Macmillan; 2011. 18. No man CD. Heal h p omo ion as a sys ems science and p ac ice. J E al Clin P ac . 2009;15(5):868–72. 19. S oh DP. Sys ems Thinking Fo Social Change: A P ac ical Guide o Sol ing Complex P oblems, A oiding Unin ended Consequences, and Achie ing Las ing Resul s. Whi e Ri e Junc ion: Chelsea G een Publishing; 2015. 20. Plsek PE, Wilson T. Complexi y, leade ship, and managemen in heal hca e o ganisa ions. BMJ. 2001;323(7315):746–9. 21. Leykum LK, Pugh J, Law ence V, Pa chman M, Noel PH, Co nell J, McDaniel RR J . O ganiza ional in e en ions employing p inciples o complexi y science ha e imp o ed ou comes o pa ien s wi h Type II diabe es. Implemen Sci. 2007;2:28. 22. Baum A, Lawless A, Williams C. Heal h in All Policies om in e na ional ideas o local implemen a ion: policies, sys ems, and o ganiza ions. In: Cla ie C, De Leeuw E, edi o s. Heal h P omo ion and he Policy P ocess. Ox o d: Ox o d Uni e si y P ess; 2013. 23. KokkinenL,Mun ane C,O’CampoP,F eile A,OnekaG,Shanka dassK. Implemen a ion o Heal h 2015 public heal h p og am in Finland: a wel a e s a e in ansi ion. Heal h P omo In . 2017; h ps://doi.o g/10. 1093/heap o/dax081. 24. Molna A, Renahy E, O'Campo P, Mun ane C, F eile A, Shanka dass K. Using win-win s a egies o implemen Heal h in All Policies: a c oss-case analysis. PLoS One. 2016;11(2):e0147003. 25. Boulding KE. Gene al sys ems heo y- he skele on o science. Manag Sci. 1956;2(3):197–208. 26. Von Be alan y L. An ou line o gene al sys em heo y. B J Philos Sci. 1950; I(2):134–65. 27. Minge s J. Sys ems Thinking, C i ical Realism and Philosophy: A Con luence o Ideas. London: Rou ledge; 2014. 28. Mun ane C, Lynch J. Income inequali y, social cohesion, and class ela ions: a c i ique o Wilkinson's neo-Du kheimian esea ch p og am. In J Heal h Se . 1999;29(1):59–81. 29. Golds ein J. Eme gence as a cons uc : his o y and issues. Eme gence. 1999; 1(1):49–72. 30. Cai ney P. Complexi y heo y in poli ical science and public policy. Poli S ud Re . 2012;10(3):346–58. 31. Mo çöl G. U ban sp awl and public policy: a complexi y heo y pe spec i e. Eme gence. 2012;14(4):1. 32. de Sa igny D, Adam T, edi o s. Sys ems Thinking o Heal h Sys ems S eng hening. Gene a: Wo ld Heal h O ganiza ion; 2009. 33. B own WB. Sys ems, bounda ies, and in o ma ion low. Acad Manag J. 1966;9(4):318–27. 34. Lawless A, Williams C, Hu ley C, Wildgoose D, Saw o d A, Kickbusch I. Heal h in All Policies: e alua ing he Sou h Aus alian app oach o in e sec o al ac ion o heal h. Can J Public Heal h. 2012;103(Suppl 1):S15–9. 35. Hummelb unne R. Sys ems hinking and e alua ion. E alua ion. 2011;17:4. 36. Pin o AD, Molna A, Shanka dass K, O'Campo PJ, Bayoumi AM. Economic conside a ions and heal h in all policies ini ia i es: e idence om in e iews wi h key in o man s in Sweden, Quebec and Sou h Aus alia. BMC Public Heal h. 2015;15:171. 37. Mo es in F, Gau in F-P, Hogue M-C, Benoi F. Me hod o Syn hesizing Knowledge Abou Public Policies. Quebec: Na ional Collabo a ing Cen e o Heal hy Public Policy; 2010. 38. Economic Council Finland. Repo o he Wo king G oup Explo ing he Goals o Heal h. Helsinki: Economic Council. [In Finnish]; 1972. Shanka dass e al. Heal h Resea ch Policy and Sys ems (2018) 16:26 Page 9 o 10