Teaming up across sectors—healthcare, housing, social services, education—sounds like a smart, holistic way to improve population health. But does it actually work?
That’s the question Hugh Alderwick, Andrew Hutchings, Adam Briggs, and Nicholas Mays set out to answer. They sifted through a mountain of evidence to figure out not just if these collaborations are effective, but why some succeed and others don’t.
To bring order to this complex research, the team used NVivo to code and analyze findings from 36 systematic reviews. The qualitative data analysis software helped them surface patterns, track themes, and make sense of what really matters in these partnerships.
In this article, we’ll dig into this research and how NVivo helped analyze qualitative data to bring clear insights from complex data.
What the research team wanted to know
This wasn’t just another study—it was a review of reviews. The researchers combed through 36 previous reviews of collaborations between health care and other sectors at the local or regional level. These collaborations spanned a wide range of contexts including initiatives focused on improving service coordination, integrating care for specific populations, or addressing broader social determinants of health.
The research team didn’t just look at outcomes; they also focused on identifying mechanisms and factors that influence how these collaborations function. The goal was to better understand what helps partnerships succeed or fail—beyond whether they meet health-related targets.
To make sense of the large volume of qualitative and quantitative findings across reviews, the authors used NVivo to organize and code data on collaboration outcomes, mechanisms, and contextual influences.
What the review revealed
Overall, the evidence supporting the impact of cross-sector collaboration on health outcomes was weak. Some studies found modest improvements in health-related behaviors or service access, but others reported no change. Many of the reviewed interventions showed mixed results on outcomes like quality of life, mortality, or reductions in health inequality.
Service access and quality showed slightly more promising results, with some studies noting improved access to care—particularly for marginalized groups. But these findings weren’t consistent across settings, and some initiatives actually led to reductions in access or unintended consequences.
Cost and resource use were even harder to pin down. Some collaborations appeared to require additional resources without demonstrating cost savings. Others redistributed costs between agencies rather than reducing them overall.
What did emerge more clearly were examples of process-level impacts. Collaborations often generated increased organizational capacity, stronger working relationships between agencies, and broader engagement in local planning efforts. Still, evidence on whether these changes translated into health improvements was limited.
What makes or breaks collaboration?
Digging deeper with the help of NVivo qualitative data analysis, the review surfaced a handful of factors that consistently shaped whether collaborations between health and non-health care organizations function effectively.
Through NVivo, one of the most frequently mentioned elements that was identified in the coding of the data was the presence of a shared vision and clearly defined goals among partner organizations. Collaborations that lacked clarity or included conflicting agendas often struggled to make progress. Involving organizational staff in shaping the direction of the partnership helped build alignment and buy-in early on.
Trust and communication also played a central role. Strong interpersonal and interorganizational relationships helped partners coordinate tasks, share information, and resolve conflicts. On the other hand, partnerships that faced cultural or professional differences—such as conflicting norms between health and social care agencies—often experienced tension and misalignment. Role clarity was another common theme, with unclear responsibilities leading to duplicated efforts or disengagement.
Many partnerships encountered practical barriers related to limited resources. A lack of funding, staff capacity, or shared infrastructure constrained what could be achieved. While pooled budgets and shared tools sometimes helped, they also introduced challenges, such as disputes over cost allocation or access to data. Staff turnover was another barrier that disrupted continuity and institutional knowledge.
Leadership and governance arrangements influenced how decisions were made and how problems were addressed. Partnerships benefited when leadership was committed, inclusive, and engaged across multiple levels of the participating organizations. However, governance structures that excluded key stakeholders or failed to establish clear accountability often hindered progress.
Finally, broader political and policy contexts shaped the environment in which these collaborations operated. National mandates and funding initiatives could create opportunities for joint working, but frequent changes in policy or misalignment with local priorities sometimes undermined collaboration. Structural features, such as agency boundaries or institutional reorganizations, added another layer of complexity.
Rather than acting independently, these factors often influenced each other. For example, strong leadership could help overcome resource shortages, and better communication could improve trust and role clarity. The review emphasized the need to understand collaboration as a dynamic and context-dependent process shaped by multiple interacting influences.
How NVivo supported the analysis
To manage the complexity and diversity of findings across the 36 reviews, the research team used NVivo to support data extraction, coding, and synthesis. The qualitative data analysis software enabled systematic organization of both structured and unstructured data, including narrative accounts and formal outcome measures. Researchers developed coding templates to classify findings across domains such as outcomes, influencing factors, and contextual conditions. These templates were refined iteratively as new patterns emerged, allowing the team to build an evolving framework that captured key themes influencing collaboration.
NVivo’s ability to handle mixed methods data was particularly important in a review that combined qualitative insights with quantitative outcome assessments. The software supported comparisons across study types and helped the team trace how codes and categories developed over time. This process enhanced the transparency and credibility of the synthesis by making analytic decisions auditable.
Without NVivo, organizing such a large and heterogeneous evidence base—while maintaining a clear trail of analytic reasoning—would have been significantly more difficult.
Enabling researcher collaboration with NVivo Collaboration Cloud
Team research has its own challenges—especially when you’re dealing with qualitative data, complex codes, and people spread across locations. NVivo’s Collaboration Cloud is designed to support this kind of distributed teamwork by allowing researchers to work on the same project from different locations in a secure and organized environment.
With Collaboration Cloud, users can access shared NVivo projects online, work across Windows and Mac, contribute to coding and analysis in real time, and manage updates without the need to manually merge files. By offering a centralized, cloud-based workspace for qualitative research, Collaboration Cloud simplifies project management and supports more efficient, consistent collaboration throughout the research process.
Explore other ways NVivo has powered real-world breakthroughs in our Behind the Breakthroughs series:
Try NVivo for your next research project
NVivo makes it easier to work with complex qualitative data—whether you’re coding transcripts, reviewing policy documents, or analyzing healthcare research. And with NVivo Collaboration Cloud, it’s also easier to do that work as a team. Discover all the research benefits of NVivo today!