How Two Health Systems Made Cross-Organisation Partnership Work

Two primary health care partnerships in Canada set out to improve access for people the system usually reaches last. Neither had a single organisation in charge. A cross-case study followed both for close to two years, and what it found has less to do with funding or mandates than with the ordinary quality of how very different people worked with each other.

Two Partnerships With Nobody in Charge

Governments and funders often mandate partnership as a route to better health outcomes, on the assumption that gathering the right organisations around a table will produce better care. In practice many such partnerships underperform. A mandate can require attendance and a reporting line. It cannot require the thing that actually makes joint work productive.

The two partnerships examined by Loban and colleagues, in a 2021 cross-case analysis published in Health Science Reports, were each working to improve primary health care access for vulnerable populations, one in each of two Canadian provinces. Both brought together a genuinely mixed group: academic researchers and research coordinators, health system decision-makers and planners, family physicians and other clinicians, administrators from regional health authorities, staff from community organisations serving vulnerable groups, and, more prominently in one partnership than the other, patients and community members themselves.

Each of those actors held one piece of the picture and a different kind of authority. A clinician knows what happens inside a consultation. A health authority administrator knows what the system can be made to fund. A community worker knows who never reaches the front door in the first place. None of them could act on the whole problem alone, and none of them could instruct the others to move. That structural condition, not health care specifically, is what the study is really about.

The constraints were ordinary ones. Neither partnership had dedicated funding for the interventions it might want to run, and the people around the table answered to different organisations with different priorities, timelines and definitions of success. The study does not set out the individual workloads of participants, but in practice this kind of work almost always sits on top of a full-time job somewhere else. Any one of those pressures is enough to make a mandated partnership quietly stall.

Why the Researchers Looked at the Collaboration Rather Than the Outcomes

Long-term health outcomes take years to appear and are shaped by far more than any one partnership. Rather than waiting, the research team asked a nearer question: was the collaboration itself producing something beyond what any single member could have produced alone? The study calls that quality partnership synergy and treats it as an intermediate outcome, something observable while the work is still in progress.

The method was ordinary and careful. Over roughly two years the researchers reviewed internal documents such as meeting minutes and reports, sat in as non-participant observers across more than a dozen partnership meetings, and conducted around sixteen in-depth, semi-structured interviews with people directly involved in the work. They then used cross-case synthesis, comparing the two partnerships against each other to separate what held in both from what was specific to one.

It is worth being clear about what that design buys and what it does not. Observing meetings catches how decisions were actually made rather than how participants later described them. Interviews catch reasoning that never reaches the minutes. Documents catch what a partnership was prepared to commit to writing. Together they describe the working life of a collaboration in a way an outcome measure cannot.

What none of it measures is whether patients ended up better off, and the study does not claim that it does. That is a deliberate choice rather than an oversight. The question being asked is whether the collaboration was working as a collaboration, which is knowable long before any health effect could be attributed to it.

What Actually Held the Two Partnerships Together

The study reports synergy showing up in four consistent ways. Members pooled genuinely different resources, skills, perspectives and connections that no single organisation held on its own, and both partnerships found low-cost ways to act on what they had pooled even without dedicated intervention funding. The working atmosphere, described by participants as respectful, open and collaborative, sped up decision-making and let trust deepen as people became less guarded over time.

Members also reported benefits substantial enough to justify what participation cost them: a working understanding of what other sectors actually do, greater visibility for their own organisation, and professional learning. And both partnerships proved able to adapt when policy or context shifted around them, with one recovering from a significant disruption by drawing on trust and credibility it had already built up.

The researchers then traced where that synergy came from, and the sources are more mundane than the word suggests. Interpersonal habits: open communication, face-to-face relationship-building. Operational choices: how decisions were reached, how leadership was distributed. System-level conditions: notably, having decision-makers with real organisational authority actually present, rather than a delegate who has to take everything away and ask. The study frames synergy as dynamic rather than fixed. It builds over time, it can weaken under strain, and it needs deliberate attention rather than being assumed once a partnership formally exists.

What Two Cases Can and Cannot Tell Us

Two cases are two cases. Cross-case synthesis is a reasonable way to separate the general from the particular, but it cannot establish that these conditions caused the synergy, and it cannot say how often partnerships lacking them fail. Both partnerships operated in primary health care, in one country, under public funding arrangements, and both were functioning well enough to be worth studying for two years. This is evidence from partnerships that held together. It is not a sample that includes the ones that collapsed.

Part of the evidence is also self-reported. Participants describing an atmosphere as respectful and open are describing their own experience of a group they chose to keep attending. Observation of meetings offsets that usefully, and using both sources is a real strength of the design, but it does not remove the tendency of people to narrate a collaboration they are invested in generously. None of that makes the findings weak. It makes them findings about mechanism rather than proof of effect, which is what a study of this shape can honestly deliver.

The transferable lesson is about a structural situation rather than about medicine. Any outcome delivered by several organisations with no single owner faces the same problem: authority is distributed, so nothing can simply be instructed into place. A resettlement, a benefits claim crossing three agencies, a supply chain, a joint infrastructure programme. The study does not test its findings outside primary health care, but the conditions it identifies attach to that structural situation rather than to clinical work. Whether the people in the room can commit their organisations, whether decisions are made in a way everyone can see, whether members get enough back to justify the hours, none of those are health care questions.

Why This Matters Outside Health Care

Service-dominant logic, developed by Stephen L. Vargo and Robert F. Lusch, describes value as co-created through interaction among actors rather than delivered by a firm to a passive customer. A service ecosystem, in the definition published in the Journal of Service Management, is "a relatively self-contained, self-adjusting system of resource-integrating actors connected by shared institutional logics and mutual value creation." Primary health care is about as clear an instance of that as exists, and what this study adds is empirical texture: evidence from two real partnerships that the quality of relationships between actors is something observable and manageable rather than a soft precondition to be hoped for.

The practical version is unglamorous. If an organisation depends on other organisations to deliver something, the coordination between them is a thing that can be designed, attended to, and allowed to decay. The study does not say what to build. It says what to watch.