The integration of primary care and behavioral health care services requires system-, clinic-, and patient-level changes that are not easily made unless proper attention is paid to choosing relevant measures. The old adage "That which is measured improves" speaks to the fact that when organizations focus on measuring integration, integration begins to happen.

A review of the literature reveals a wide variety of integration measures from which to choose. Unfortunately, few resources actually explain which integration measure is the best to choose and why. From the perspective of a clinic administrator or clinician, a good measure is broadly applicable and adds value in understanding program performance while mediating the significant resource allocation required to execute it. One approach to controlling the cost related to developing and implementing a new measure is to review the measures your agency already collects for state and accreditation reporting. An often overlooked set of measures can be found in current and past short-term quality improvement projects and strategic planning documents. These measures can be leveraged to describe your integration efforts with minimal changes to your data collection infrastructure.

Because health care services integration requires significant system-level collaboration (which includes the integration of two or more agencies within the context of a specific community), identifying system-level measures of change is paramount. Recent findings show that focusing only on patient-level changes without measuring system-level changes can inhibit progress toward integration. A good place to start when beginning integration is to choose a model for integration that best describes your efforts as a framework for your strategic planning documents that you create with your partner(s). Progress toward completing the integrated health care strategic plan objectives and goals is a good measure of your integration efforts. Examples of system-level measures include measures of collaboration, cost sharing, and the number of cross-agency partnership agreements.

Another level of integration occurs at the clinic level. Clinic-level measures roll up into the system-level measures and often depend on the system-level measures being in place (e.g. to successfully refer a patient to a partner clinic, both agencies should have a clear collaboration agreement on the most efficient and effective protocol). Measures detailing the number of patients referred to a community partner agency for care, rehospitalization rates, numbers of patients seen, and cost of care are all examples of clinic-level measures. At the level of clinical care provision, measures describing the specific kind of care provided include those specific to screening and disease management. Patient-level measures include the perception of care provided. Wellness self-management achievements and patients’ perception of their health are also important measures to capture.

A final measure worth considering for inclusion in your integrated health effort is one Paul A. Nutting describes as "adaptive reserve." Adaptive reserve includes measures of your organization’s adoption of learning organization or participatory leadership approaches to system change. Regardless of the measures you choose to monitor and describe your integrated health efforts, it is always helpful to reach out to others who have already started the journey toward integration.