PBRNs or RBPNs or Both?

In this issue, two reports by Kelly and colleagues describe the development of the Recovery-Oriented Care Collaborative, a practice-based research network (PBRN) focused on improving services for people with serious mental illness. In the Research & Services Partnerships column, these authors detail the stepwise process of developing infrastructure, clarifying priorities, and executing an initial research project. In a related letter to the editor, the authors report project findings regarding the impact of integrating mental health and general medical care.

As Kelly and colleagues describe, developing an effective PBRN requires attention to both infrastructure and culture. They note that the necessary infrastructure to support research includes administrative processes to facilitate communication, as well as technical tools or processes to collect and organize data. But establishing that infrastructure requires a cultural foundation of transparency and trust. Although the interests of clinical practice and research are often closely aligned, they are rarely identical. Serving those varying interests usually requires some compromise and sacrifice, which can be purchased only with trust. As Kelly and colleagues describe, developing transparency and trust is an iterative process. Trust facilitates successful collaboration, which further strengthens trust. Repeated successfully, this process expands beyond individual relationships to change the cultures of both clinical and research organizations. This process of mutual engagement characterizes community-based participatory research or what Van de Ven has called “engaged scholarship.”

This iterative process of collaboration enables another important cultural shift: the establishment of continuous measurement as a core function of effective health care. Although measurement-based care in mental health has lagged behind many areas of general medicine, use of standard depression outcome measures is being rapidly implemented in community practice. The spread of electronic medical records systems allows those measurement data to be organized in support of both care for individual patients and population-based research. As always, the most important barriers to this transformation are cultural rather than technical. Patients, providers, and health systems must have confidence that systematic measurement of care processes, care experience, and patient-centered outcomes will support improvement for all rather than punish those already disadvantaged.

A culture of continuous measurement can then enable a culture of continuous inquiry. As Kelly and colleagues describe, a learning health care system depends on engaging patients, clinicians, and health system leaders in a collaborative process of identifying and answering research questions. Behavioral scientists understand the power of immediate reinforcement. If the objective is to reinforce the behavior of asking questions, then a learning health care system should aim to provide timely answers to questions that arise from practice. Disagreements about best practices are healthy. In a culture of continuous inquiry, however, a disagreement leads to the questions: “What information would we need to actually know the answer? How can we find that information in our practice?”

It must be admitted, however, that a culture of continuous inquiry and rapid-cycle learning would create challenges for traditional researchers and traditional journals such as this one. Our current practice of disseminating research findings through journal publication may serve the career interests of researchers and the business interests of publishers. But delayed reporting of answers until publication does not always serve the interests of patients, clinicians, and leaders of health care systems. Notably, legislation establishing the Patient-Centered Outcomes Research Institute (PCORI) required that findings of PCORI-funded research be made public within 90 days. Significant changes to our research traditions may be in the works.

We can certainly lament the limited reach of research into community practice and the resulting delay in dissemination of evidence-based practices. But we can point to some advances on the national scene. PCORI’s National Patient-Centered Clinical Research Network (PCORnet) is developing infrastructure to “liberate” data from large health care systems and engage patients and patient-led organizations in the research process. NIH’s Health Care Systems Research Collaboratory is supporting a series of large pragmatic clinical trials based in real-world health care systems. The NIMH-funded Mental Health Research Network supports an infrastructure for clinical trials and observational research in 13 large health care systems across the United States.

The ultimate goal is the seamless integration of research and practice. In such a learning health care system, collaboration between patients, clinicians, health system leaders, and researchers would not be a special event. A culture of continuous measurement and continuous inquiry would extend to every place where people living with psychiatric disorders receive care, including rural practices and those serving traditionally disadvantaged groups. If this goal were realized, an observer might ask, “Is this a practice-based research network or a research-based practice network?” And the answer would be “Yes!”