The research, conducted by Tianran Wang and Zhongjin Yao of the School of Public Health and Management at Guangzhou University of Chinese Medicine, focuses on county-level medical alliances, the networks that link county hospitals with township health centers in an effort to redistribute medical expertise downward. Vertical collaboration in these alliances refers to cooperation across the hierarchical tiers of the system: specialists and administrators at the county level working with frontline clinicians at the township level. On paper, such alliances are meant to harmonize referrals, training, quality control and resource sharing. In practice, the researchers argue, collaboration often remains shallow, sustained more by administrative mandate than by the intrinsic motivation of the medical personnel involved.

To understand why, the authors grounded their work in the organizational collaboration theory developed by Denise D’Amour and colleagues, a framework widely used in health services research to decompose collaboration into its underlying dimensions. From that theoretical base, Wang and Yao constructed what they call the CLFI vertical collaboration theoretical framework. Within this framework, they identified ten dimensions as antecedent conditions, meaning factors that plausibly set the stage for strong or weak collaboration. These dimensions include shared goals and a patient-centered orientation, alongside other elements spanning the structural, cognitive and relational layers of how county and township personnel interact.

With the framework in place, the team designed a questionnaire and administered it through random sampling within a single county-level medical alliance in Fujian Province, in southeastern China. The survey achieved a valid response rate of 96.95 percent, an unusually high figure that suggests the instrument reached a large share of the target population. Rather than treating each respondent’s answers as independent data points to be averaged, the researchers used fuzzy-set qualitative comparative analysis, or fsQCA, a method that treats combinations of conditions rather than individual variables as the objects of analysis.

FsQCA deserves a technical explanation because it is central to the study’s contribution. Rooted in the mathematics of set theory, fsQCA evaluates how far each case belongs to particular sets, for example the set of respondents reporting high trust or the set reporting high collaboration capacity, assigning fuzzy membership scores that can range from fully out to fully in. The method then searches for configurations, which are specific combinations of conditions that are consistently associated with the outcome of interest. This makes it well suited to questions of the kind Wang and Yao posed, because collaboration is almost certainly equifinal: there may be several distinct recipes for success rather than a single dominant cause.

The analysis delivered exactly that kind of plural answer. From the data, the researchers extracted eight different configuration combinations of causal conditions, which they then synthesized into five broader capability enhancement paths. They named these the policy-driven path, the functional synergy path, the service-oriented path, the innovation-driven path and the relationship-building path. The consistency of the solutions across these five combinations of conditions reached 82.2 percent, a measure indicating how frequently, in the fuzzy-set sense, the identified configurations align with high vertical collaboration capacity.

Each of the five paths carries its own logic. The policy-driven path emphasizes coordinated rules, incentives and institutional arrangements as the engine that pulls county and township personnel into alignment. The functional synergy path highlights the professional division of labor and mutual capacity support, so that each tier of the alliance contributes complementary functions rather than duplicating or competing with the other. The service-oriented path embeds patient outcomes throughout the processes of collaborative goal setting, process design and performance evaluation, making the benefits to patients the organizing principle of joint work. The innovation-driven path calls for a tolerance and assessment mechanism that provides pilot support and dynamic monitoring for collaborative innovation, allowing personnel to experiment without fear that a failed experiment ends the partnership. The relationship-building path, finally, foregrounds the consolidation of mutual trust among county and township medical personnel.

The practical recommendations that follow from the findings are correspondingly multi-pronged. The authors argue that strategic policy coordination should be strengthened so that directives aimed at county hospitals and those aimed at township centers do not pull in different directions. Cognitive linkages between the two tiers should be enhanced, so that personnel understand and internalize each other’s roles and constraints. Patient outcomes, in their view, should not be an afterthought but a thread running through goal setting, process design and evaluation. Innovation needs institutional protection through tolerance and assessment mechanisms, and trust needs deliberate cultivation alongside a clear professional division of labor. Taken together, these moves are intended, in the study’s framing, to promote the medical community’s transition from spontaneous cooperation to conscious coordination, with the ultimate aim of improving the quality and efficiency of primary healthcare services.

Why does this matter beyond one county in Fujian? China’s health reforms have invested heavily in county-level medical alliances, also discussed in the literature under the umbrella of integrated delivery networks, precisely because grassroots institutions often lose patients to large urban hospitals, weakening both primary care and system sustainability. If vertical collaboration between county and township personnel is the mechanism through which expertise and standards flow downward, then understanding the configurational causes of strong collaboration speaks directly to whether those reforms deliver. The study’s message is that single-lever interventions, such as a new referral protocol or a training program, may fail not because the idea is wrong but because the supporting configuration of conditions is incomplete. A policy push without trust, or a shared mission without functional complementarity, may simply not generate the collaboration capacity reformers expect.

Methodologically, the study also illustrates why fsQCA is gaining ground in health services research. Traditional regression-based approaches estimate the independent effect of each variable, which can obscure the way conditions interact or substitute for one another. FsQCA, by contrast, accepts that the same outcome can arise through different routes and analyzes cases as wholes. The authors’ use of the method within a D’Amour-based theoretical framework offers a template for studying collaboration in other settings, from hospital mergers to multidisciplinary tumor boards. The study was supported by the 2024 Guangdong Province Graduate Education Innovation Program Project of China, and the authors report no competing interests. For the doctors, nurses and managers who populate China’s county alliances, the research suggests a hopeful conclusion: better vertical collaboration is achievable, but only when policy, cognition, service goals, innovation support and trust are assembled together, in whichever combination a given alliance can realistically build.

Subject of Research: Pathways to improve vertical collaboration capacity of medical personnel in Chinese county-level medical alliances using the CLFI model and fsQCA

Article Title: Study on pathways to improve the capacity for vertical collaboration of medical personnel in county-level medical alliances: an empirical exploration based on the CLFI collaboration model

Article References: Wang, T., & Yao, Z. (2026). Study on pathways to improve the capacity for vertical collaboration of medical personnel in county-level medical alliances: an empirical exploration based on the CLFI collaboration model. BMC Health Services Research. https://doi.org/10.1186/s12913-026-15813-3

Image Credits: AI Generated

DOI: 10.1186/s12913-026-15813-3

Keywords: county-level medical alliance, vertical collaboration, medical personnel, CLFI collaboration model, fuzzy-set qualitative comparative analysis, D'Amour collaboration theory, primary healthcare, health services research, fsQCA, China health reform, Fujian Province, patient-centered care