Advancing quadruple helix theory for health systems innovation: evidence from Ageing@Coimbra | BMC Health Services Research

This study employs a single case-study design appropriate for investigating complex, context-dependent phenomena in emerging fields of inquiry [40,41,42,43]. We selected Ageing@Coimbra because it is the first European Innovation Partnership on Active and Healthy Ageing (EIP-AHA) Reference Site explicitly organised around a Quadruple Helix (QH) governance configuration and because its activities have generated peer-reviewed documentation of ageing- and cognitive-health–related innovations [7]. The initiative also exhibits a polycentric, citizen-anchored governance structure that aligns directly with our interest in how local ecosystems enable QH collaboration. Our methodological protocol followed established qualitative case-study standards, combining documentary review with semi-structured interviews in order to triangulate sources and capture diverse stakeholder perspectives. As researchers, we maintained an external but engaged role: none of the authors were formally affiliated with the Ageing@Coimbra consortium, yet we had prior exposure to its activities through European Reference Site networks. This position allowed for informed but independent observation, reducing risks of insider bias while ensuring familiarity with the institutional context.

Two of the authors were directly involved in the interview process. Access to Ageing@Coimbra was facilitated through a colleague in the European Reference Site Collaborative Network (RSCN), who introduced the research team to the consortium. This proximity supported contextual familiarity and facilitated participant recruitment, but also required careful attention to reflexivity. To mitigate potential bias, coding and analysis were undertaken collaboratively, with cross-checks between members more and less directly connected to the Reference Site.

The consortium operates under a memorandum of understanding endorsed by the Regional Health Authority, which grants it authority to pilot innovations within public primary-care units. Academia is represented by the University of Coimbra and the Pedro Nunes Institute, whose legal mandate allows direct knowledge-transfer contracts with SMEs. Industry partners—mostly digital-health start-ups clustered in the local incubator—benefit from fast-track procurement rules introduced by the 2019 Portuguese Innovation Decree, enabling them to test prototypes in public facilities for up to twelve months. Municipal governments contribute zoning powers and small-grant budgets earmarked for age-friendly infrastructure, while civil-society organisations (senior universities, patient associations) exercise agenda-setting influence through two seats on the consortium’s steering committee. Together, these levers create a polycentric but formalised governance environment that both empowers co-creation and ensures regulatory compliance.

Portugal’s institutional setting provides an especially fertile backdrop for analysing health-policy innovation. While the Serviço Nacional de Saúde (SNS) remains centrally financed, significant competencies for health promotion, primary care delivery, and ageing-related programmes have been delegated to regional health administrations and municipalities, creating opportunities for decentralised experimentation. Policy instruments such as the 2019 Innovation Decree and the National Programme for Healthy Ageing explicitly encourage pilot projects that involve universities, local governments, SMEs, and civil-society actors. This multi-level framework aligns with European Union priorities, notably the EIP-AHA Reference Site model, but Portugal has distinguished itself by embedding QH-inspired governance into statutory instruments rather than relying solely on soft coordination or project-based collaboration. Compared with more centralised systems (e.g., France) or those with stronger insurance-based pluralism (e.g., Germany, the Netherlands), Portugal represents a hybrid case where decentralised authority and targeted innovation policies converge. Against this backdrop, Ageing@Coimbra exemplifies how a territorially anchored, citizen-inclusive ecosystem can translate European policy ambitions into locally meaningful health-system innovations.

Data derive from two complementary sources: (i) documentary materials (policy reports, consortium records, and the published Frontiers in Medicine case description of Ageing@Coimbra) covering the period 2013–2025 [7]; and (ii) twelve semi-structured interviews conducted between March and April 2025 with stakeholders directly involved in the initiative. We used purposive, maximum-variation sampling to ensure that all four helices were represented. Interviews were completed with three participants from academia, three from government / public authorities, three from industry / technology or health-related enterprises, and three from civil-society organisations (community associations, patient / senior groups). The twelve interviewees represented a broad cross-section of the ecosystem. Academic participants included a senior gerontology researcher, a digital-health scientist, and a knowledge-transfer manager. Government/public authority representatives comprised two municipal health directors and a regional health-administration official. Industry voices came from two SME founders in the digital-health incubator and a product manager in a telemedicine start-up. Civil-society participants included leaders of senior universities, a patient-association chair, and a coordinator of community volunteers. This profile ensured that both strategic decision-makers and operational implementers were captured across all four helices. Eligibility required an active role in Ageing@Coimbra decision-making or programme implementation and the ability to comment on cross-sector collaboration. Recruitment proceeded until thematic saturation—defined a priori as the point at which two consecutive interviews no longer yielded new first-order codes relevant to our research questions—was observed after interview 10 and confirmed in interviews 11 and 12.

Interviews, lasting 45–75 min, were conducted in the participant’s preferred language (Portuguese or English) via secure video call or in person. A common guide covered seven domains: respondent role; organisational contribution to QH collaboration; perceived tangible and intangible benefits; collaboration challenges and inter-organisational dynamics; knowledge-sharing and governance practices; the role of the local ecosystem; and future perspectives for QH in healthcare innovation. With permission, all interviews were audio-recorded and professionally transcribed verbatim. Bilingual members of the research team checked each transcript against the audio for accuracy; where needed, Portuguese passages were translated into English and back-checked for meaning preservation. Participants were offered the opportunity to review and amend their transcript (member checking); minor clarifications were incorporated. The interview process was designed to facilitate open reflection. After presenting the information sheet and obtaining written consent, participants were encouraged to narrate both successes and tensions in their collaborations. Probing questions sought concrete examples of governance practices, co-creation challenges, and perceptions of ecosystem enablers. Field notes were taken to capture non-verbal cues and contextual observations, complementing the verbatim transcripts.

Analysis combined deductive sensitising concepts drawn from QH scholarship with inductive coding informed by constructivist grounded-theory principles [44,45,46,47,48,49,50]. First, two researchers independently read all transcripts and developed an initial open-coding frame anchored in the research questions (QH benefits; ecosystem role). Codes were iteratively compared, merged, and expanded to incorporate emergent categories (e.g., alignment challenges, distributed governance, empowerment). We then organised codes into seven higher-order thematic categories: (i) knowledge sharing, (ii) transformative collaboration, (iii) alignment challenges, (iv) patient-centred outcomes, (v) community engagement, (vi) distributed governance, and (vii) empowerment and inclusiveness. These categories were iteratively validated against both documentary evidence and the interview corpus, serving as the analytical scaffold for the Results section. By defining these clusters upfront, we make transparent how raw interview data were systematically connected to our conceptual claims. Documentary materials [7] were coded in the same NVivo project to support data triangulation and to trace temporal sequencing of initiatives referenced by interviewees. To enhance reliability, two coders double-coded four transcripts (~ 33% of the corpus); discrepancies were discussed until consensus and the codebook was revised accordingly. Given the small qualitative sample and our emphasis on interpretive depth, we did not calculate a formal inter-coder kappa; instead, agreement was established through negotiated resolution and full-team review of theme definitions.

Credibility was supported through triangulation of interview and documentary evidence, member-checking of transcripts, and maintenance of an audit trail documenting analytic decisions. Transferability is aided by providing thick contextual description in Sect. 4.1 and Supplementary Appendix A. Dependability and confirmability were strengthened by reflexive memos in which researchers recorded positionality, assumptions, and interactions with data throughout the study. The research involved professional stakeholders speaking in their institutional capacities and did not solicit personal health information or sensitive individual data; under the policies in effect at the authors’ institutions, such minimal-risk social-science interviewing was exempt from formal ethics-board review and therefore did not receive a protocol number. Nevertheless, we adhered to recognised ethical standards: all invitees received an information sheet describing the study purpose, voluntary participation, data anonymisation, and withdrawal rights; written informed consent was obtained prior to each interview; audio files and transcripts were pseudonymised and stored on encrypted drives in compliance with GDPR-aligned data-protection procedures.