Action Research Helps Hospitals Build Care Pathways That Bend Without Breaking

Care pathways are among the most widely used tools in modern healthcare quality improvement. They map out the ideal sequence of care for a particular condition, coordinating the work of physicians, nurses, therapists and other professionals so that patients receive consistent, evidence-based treatment from admission to discharge. Yet the way these pathways are actually developed in real hospitals has long been a blind spot in the scientific literature. Most studies describe what a finished pathway looks like, but far fewer examine the messy, iterative process of building several pathways at once inside an organization that is itself constantly changing. A new study published in Health Research Policy and Systems addresses that gap directly, and its findings suggest that the research method known as action research may be the missing ingredient that allows care pathway development to survive contact with reality.

The study, led by Mildred Visser of the Erasmus School of Health Policy & Management at Erasmus University Rotterdam, followed a two-year care pathway development programme conducted in a Dutch specialized rehabilitation hospital with sixteen sites. Eleven multidisciplinary teams worked in parallel to design care pathways for different patient populations, guided by five quality improvement principles: collaborative improvement, stepwise and goal-oriented pathway development, hybrid development that combines design with learning, outcome monitoring, and shared ownership among participants. Rather than treating the programme as a fixed protocol to be evaluated at the end, the researchers embedded themselves in it, documenting how it unfolded, where it strained, and how it had to be reshaped in response to events on the ground.

The choice of action research as the guiding methodology is central to the paper’s argument. Action research differs from conventional observational research in that the researcher participates in the change process itself, generating scientific knowledge through cycles of planning, action, observation and reflection. In this programme, the research team collected reflexive data continuously: 169 journal reports, 44 recorded conversations with team members and senior managers, 25 research team meetings, and 149 programme documents. The data were analysed in four iterative phases, combining elements of process analysis and reflexive thematic analysis, so that insights from one cycle could inform adjustments in the next. This created a feedback loop between the programme and its own evaluation, which proved essential when the original design assumptions collided with practical constraints.

What emerged from the analysis were six distinct challenges that arose when multiple care pathways were developed in parallel. The first concerned alignment: teams differed in their pace of progress, their learning needs and their clinical scope, and the standardized programme structure did not always fit these differences. Teams working on complex conditions such as spinal cord injury rehabilitation faced different design questions than teams focused on other rehabilitation populations, and forcing everyone through identical steps at identical speeds created friction. The programme had to be refined repeatedly to allow teams to move at appropriate speeds while still benefiting from shared learning across the collaborative.

The second challenge was balancing innovation with manageable change. Teams were encouraged to redesign care in meaningful ways, but too much change at once threatened to overwhelm frontline staff and disrupt daily operations. The researchers found that the programme’s design had implicitly assumed a capacity for change that the organization did not always possess, and adjustments were needed to pace the innovations so that they remained implementable. Closely related was the third challenge: the organization’s limited readiness for hybrid care, in which treatment is delivered through a mix of in-person and remote modalities. The pandemic era accelerated interest in hybrid care models, but embedding them into formal care pathways required infrastructure, skills and cultural acceptance that could not be assumed in advance.

The fourth challenge involved outcome monitoring, a cornerstone of value-based care. Teams were expected to measure patient outcomes systematically, but the study revealed practical problems in doing so, including difficulties with data collection and a heavy reliance on expert support to make measurement meaningful. Without functioning outcome monitoring, the feedback that was supposed to drive improvement weakened. The fifth challenge concerned distributed leadership and capability development. Because eleven teams were working simultaneously, leadership could not be concentrated in a single group; the programme had to cultivate improvement capabilities across the organization, which took time and deliberate investment. The sixth challenge was reflexive in nature: the programme itself was hybrid, combining fixed design elements with ongoing learning and adaptation, and teams needed help navigating that ambiguity.

Each of these challenges, the authors argue, exposed assumptions embedded in the original programme design that were tested and often overturned in practice. Crucially, the action research approach allowed those assumptions to be surfaced and corrected in real time rather than discovered only in a post-hoc evaluation, when it would have been too late to help. Successive refinements to the programme were documented and analysed as part of the research itself, turning the inevitable turbulence of organizational change into a source of scientific insight. The researchers describe this as linking change, learning and knowledge generation in a single continuous process, which is precisely what conventional quality improvement collaboratives often fail to achieve.

The implications reach beyond this single Dutch hospital. Care pathway development is a global strategy in value-based healthcare, and organizations everywhere struggle with the same tension between standardization and local adaptation. The study’s conclusion is that action research offers a feasible methodology for implementing and adapting care pathway development methods through phased, reflective and context-sensitive learning. By embedding reflection and adaptation into the programme from the start, action research helped balance structure with flexibility, strengthened shared ownership among clinicians and managers, and integrated sustainability considerations early in the process rather than as an afterthought. In other words, the methodology did not just measure whether the programme worked; it actively helped the programme work.

The research also carries a message about patient involvement. The programme included patient representatives, including two from a post-discharge spinal cord injury rehabilitation network and one from multiple sclerosis rehabilitation, who participated as partners in the pathway design process and helped ground the designs in lived experience. This participatory element aligns with the collaborative improvement principle and reflects a broader shift in health services research toward co-design with the people who actually use the services. Combined with the study’s rigorous documentation of its own methods, including a published coding structure mapping data to the five quality improvement principles and adherence to recognized qualitative reporting standards, the work offers a template that other organizations can adapt.

Funded by the Dutch Organisation for Health Research and Development, ZonMw, the study arrives at a moment when health systems face accelerating pressure from aging populations, workforce shortages and technological change, all of which make rigid, one-size-fits-all improvement programmes increasingly untenable. The authors’ central finding is deceptively simple: care pathway methods describe what good pathways should contain, but only a methodology that embraces reflection, participation and continuous adaptation can bridge the gap between those methods and their practical application. Action research, demonstrated here across two years, eleven teams and sixteen sites, appears to be that bridge. For hospital leaders contemplating large-scale pathway programmes, the lesson is that the process of development matters as much as the product, and that building flexibility into the process from day one may be the difference between pathways that endure and pathways that quietly fade from use.

Subject of Research: Using action research to guide parallel, adaptive care pathway development in a dynamic hospital organization

Article Title: Action research as a methodology for adaptive care pathway development in a dynamic organizational context

Article References: Visser, M., de Mul, M., Ahaus, K., Osterthun, R., Pangalila, R., Grauwmeijer, E., Sluijter, A., van Pelt-Zoutendijk, M., & Weggelaar-Jansen, A. M. (2026). Action research as a methodology for adaptive care pathway development in a dynamic organizational context. Health Research Policy and Systems. https://doi.org/10.1186/s12961-026-01520-2

Image Credits: AI Generated

DOI: 10.1186/s12961-026-01520-2

Keywords: action research, care pathways, care pathway development, quality improvement, rehabilitation medicine, hybrid care, outcome monitoring, participatory methods, health services research, complex interventions, healthcare innovation, qualitative research