BEST PRACTICES DO NOT SCALE EQUALLY: IMPLEMENTATION PRECONDITIONS IN CARDIOVASCULAR CARE ACROSS 15 EUROPEAN SYSTEMS
Author(s)
DR MARCOS Gallego Llorente, PhD1, Liselotte Tjon, MS2, Ruben van Zelm, PhD3, Alice Masini, PhD4, Alexander Carter, PhD5, Areej Malik, MS6, Lian Y Rekker, MS7, Luuk C Kieviet, MS7, Daniel Breucker, MS8, Moritz von Scheidt, MD8, Gerard Pasterkamp, MD7, Marish Oerlemans, MD7, Hugo Bastiaan Amesz, MSc2, Pim van der Harst, MD9.
1Manager, Utrecht, Netherlands, 2Vintura BV, Utrecht, Netherlands, 3HU University of Applied Science Utrecht, Utrecht, Netherlands, 4University of Eastern Piedmont, Novara, Italy, 5London School of Economics, London, United Kingdom, 6London School of Economics and Political Sciences, London, United Kingdom, 7UMC Utrecht, Utrecht, Netherlands, 8TUM University Hospital German Heart Center, TUM School of Medicine and Health, Technical University, Munich, Germany, 9Universitair Medisch Centrum Groningen, 9713 GZ, Netherlands.
1Manager, Utrecht, Netherlands, 2Vintura BV, Utrecht, Netherlands, 3HU University of Applied Science Utrecht, Utrecht, Netherlands, 4University of Eastern Piedmont, Novara, Italy, 5London School of Economics, London, United Kingdom, 6London School of Economics and Political Sciences, London, United Kingdom, 7UMC Utrecht, Utrecht, Netherlands, 8TUM University Hospital German Heart Center, TUM School of Medicine and Health, Technical University, Munich, Germany, 9Universitair Medisch Centrum Groningen, 9713 GZ, Netherlands.
OBJECTIVES: A best practice is only beneficial as long as it can be translated into systems where it is not yet implemented. However, best practice transferability is often assumed rather than measured. We examined 184 cardiovascular best practice candidates across four pathways and 15 European systems, and how transferable those practices actually are.
METHODS: From the EuroHeartPath best‑practice library (best practice candidates across 60 country‑condition patient pathway pairs spanning coronary artery disease, heart failure, cardiac amyloidosis, and out-of-hospital cardiac arrest, each reconciling survey evidence, guidelines, and desk-research) we classified every candidate on two axes: practice domain (national/regional pathway, specialised-centre network, data/registry infrastructure, early-response ecosystem) and a three-level transferability class: protocol-level (a process or SOP a single site can adopt), network-level (requires reorganising referral relationships), or infrastructure-level (requires a built registry, or complex infrastructure assets). We report proportions with Wilson 95% confidence intervals (CIs), and the intervention-class-by-transferability association (chi-square; Cramér's V).
RESULTS: Most best practices were protocol-level (i.e. adoptable by a new site without new infrastructure requirements) (104/184; 56.5%, 95% CI 49.3-63.5), while 80/184 (43.5%) required network or infrastructure reorganisation. Distribution was highly condition-specific: most out-of-hospital cardiac arrest best practices were infrastructure-bound (27/47); amyloidosis best practices leaned on referral-network reorganisation (12 hub-and-spoke models). Practice domain was strongly associated with transferability class (Cramér's V=0.75; p<0.001): every data/registry (35/35) and early-response (12/12) candidate was infrastructure-dependent, whereas national/regional practices were overwhelmingly protocol-level (73/81).
CONCLUSIONS: Infrastructure- and network-dependent practices such as registries, hub-and-spoke networks and citizen-responder ecosystems, which in literature tend to constitute the highest-leverage cardiovascular interventions, were the least transferable, because they presuppose national infrastructure and governance. For HTA bodies and payers, judging an organisational best practice means assessing what it requires to be implemented elsewhere, and not only the potential impact of the fully-implemented practice.
METHODS: From the EuroHeartPath best‑practice library (best practice candidates across 60 country‑condition patient pathway pairs spanning coronary artery disease, heart failure, cardiac amyloidosis, and out-of-hospital cardiac arrest, each reconciling survey evidence, guidelines, and desk-research) we classified every candidate on two axes: practice domain (national/regional pathway, specialised-centre network, data/registry infrastructure, early-response ecosystem) and a three-level transferability class: protocol-level (a process or SOP a single site can adopt), network-level (requires reorganising referral relationships), or infrastructure-level (requires a built registry, or complex infrastructure assets). We report proportions with Wilson 95% confidence intervals (CIs), and the intervention-class-by-transferability association (chi-square; Cramér's V).
RESULTS: Most best practices were protocol-level (i.e. adoptable by a new site without new infrastructure requirements) (104/184; 56.5%, 95% CI 49.3-63.5), while 80/184 (43.5%) required network or infrastructure reorganisation. Distribution was highly condition-specific: most out-of-hospital cardiac arrest best practices were infrastructure-bound (27/47); amyloidosis best practices leaned on referral-network reorganisation (12 hub-and-spoke models). Practice domain was strongly associated with transferability class (Cramér's V=0.75; p<0.001): every data/registry (35/35) and early-response (12/12) candidate was infrastructure-dependent, whereas national/regional practices were overwhelmingly protocol-level (73/81).
CONCLUSIONS: Infrastructure- and network-dependent practices such as registries, hub-and-spoke networks and citizen-responder ecosystems, which in literature tend to constitute the highest-leverage cardiovascular interventions, were the least transferable, because they presuppose national infrastructure and governance. For HTA bodies and payers, judging an organisational best practice means assessing what it requires to be implemented elsewhere, and not only the potential impact of the fully-implemented practice.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
HSD98
Topic
Health Service Delivery & Process of Care
Disease
No Additional Disease & Conditions/Specialized Treatment Areas