Developing the Business Case and Supporting Practice Change for Collaborative Care

For those who are uncertain about the prevalence and impact of mental health conditions, worldwide epidemiological survey data suggest that in a given year, mental health conditions affect almost one-fifth (17.6%) of individuals, with estimates of lifetime prevalence rates of 29.2% (1). Mental disorders and substance misuse are among the top five conditions leading to disability worldwide (2) and have been linked to dying 10 years earlier than the general population (3). In the Unites States in 2013, mental disorders were first on the list of costly conditions (4), in the range of $201 billion. The majority of spending on patients with mental health conditions is for comorbid general medical disorders; therefore, organizations accountable for improving medical metrics will likely need to better address mental disorders (5).

The deficiencies in our current system of care delivery are significant enough that in 12-month reviews, almost 60% (i.e., 58.9%) of patients with psychiatric diagnoses received no treatment (6). Moreover, general medical settings are seeing the largest growth in care seeking by patients with mental health conditions (7). For those aware of the shortages in mental health resources, this is not surprising. In a study in which researchers posed as patients in Boston, Houston, and Chicago, two rounds of calling for appointments led to actual appointments in only 26% of cases (8).

In 2009, the mental health shortages by county in the United States revealed that nearly one-fifth of counties (18%) had unmet needs for nonprescribing mental health professionals and 96% of counties had unmet needs for prescribing mental health professonals (9). When estimating the mental health resource needs for community health centers in the United States, researchers found that a more than fourfold increase in staff would be needed to meet the demand (10). The U.S. Department of Health and Human Services (data available on its website) estimated a need for 2,772 additional psychiatrists as of September 2016 (on the basis of an overall ratio of 1/30,000 and a ratio of 1/20,000 people for high-need patient populations) to remove the health professional shortage designation for mental health. Given that the population of U.S. practicing psychiatrists is declining (11), new models of practice are needed to address poor access to mental health care.

A meta-analysis of more than 79 randomized controlled trials demonstrated consistently positive results for collaborative care for depression and anxiety (12). One can make the argument that there are very few health care delivery processes already in practice with that degree of research support. In addition, the primary care interface provides an opportunity for improved outcomes for the all-too-common challenge of management of depressed patients with other chronic illnesses, such as diabetes and cardiovascular disease (13). These models have been tested with multiple patient groups and settings, with improved clinical and patient-level outcomes. Costs from the perspective of the insurance company have been reduced in models of care coordination for depression (14) and depression with diabetes (15) and in settings in which primary medical care is delivered collaboratively to patients with serious mental illness (16). When looking over the impact of this model in an entire system, researchers noted improvements in both quality process measures (e.g., screening for depression) and utilization measures (e.g., emergency and hospital visits) for those in integrated practices, as compared with others (17).

Some of the expected improvements when collaborative care is implemented effectively should include improved patient and provider satisfaction (18), improved access (19), improved required process measures for the Healthcare Effectiveness Data and Information Set (HEDIS) and other groups, improved patient-level outcomes (e.g., depression 6-month remission rates), improved ability to work for patients, and improved effective use of health care resources. In addition, the variety of roles for the psychiatrist and the backup of the primary care team may reduce burnout and improve job satisfaction.

For specific audiences, one might choose to highlight different aspects of collaborative care. Administrators might be most interested in improved access, improved quality process measures, improved patient outcomes, leveraged psychiatric time, lowered costs, and new payment opportunities with Medicare collaborative care codes. Insurance companies might be swayed by improved quality measures and either no additional cost or cost savings. Employers might be drawn by the potential for improved return to work (decreased absenteeism) and decreased impact on productivity (presenteeism).

If one’s audience is psychiatrists, one might highlight that collaborative care offers a new practice opportunity with reduced isolation (now with a payment mechanism with Medicare collaborative care codes), improved ability to measure outcomes, improved patient-level outcomes, and possible reduced burnout. In addition, there is no evidence for increased liability. Other mental health professionals might enjoy the reduced isolation, increased support, opportunity to work on a team, and targeted referrals, and nursing professionals might like the new role in care coordination and increased support that collaborative care offers.

An audience of primary care providers might be drawn to collaborative care by the improved satisfaction rate and improved access to specialty, case-based learning. Patient groups will be drawn to data showing improved satisfaction, better health outcomes, improved access, decreased stigma, improved ability to remain employed (and thus insured), and improved communication among their providers.

Successful arguments for collaborative care over time in a given setting depend on that site’s ability to measure need and change in its own setting. Those making the case for collaborative care will find they are more effective when they take part in understanding and strengthening that site’s capacity to measure patient-level outcomes, meaningful clinical processes, and changes in utilization. Research on the impact of measurement-based care has shown that patient outcomes improve when measurement is done correctly (20). Measures must be seen as meaningful in the care of patients and actionable by patients and providers.

In some settings, the first step in making the case for collaborative care may be to simply measure the level of depression among the population treated in a given primary care setting. This step shifts the focus away from an individual patient’s depression toward a broader population-based view that can show a clinic that its own data can help its professionals to realize the extent of the need. Ideally, several patient-level tools need to be integrated, and a registry is needed to track patient progress, identify those who not improving even if they are not in the clinic, and compare outcomes across sites to develop best practices. A strength of the collaborative care model is the designation of a behavioral health care manager to own this task for the team.

Using this type of approach allows psychiatrists to stratify the population of patients and focus efforts on the patients who need the most attention. This routine use of a measurement-based approach is thought to be an important driver of the improved outcomes seen in collaborative care. An argument can be made that one cannot have evidence-based practice unless the practice can generate evidence.

The best advice in making the case for collaborative care is to start where you are and build. Much of this work is based on relationships among providers from different backgrounds. A first step may be to set up noon conferences in primary care to review cases and get to know each other. This may be followed by a search for common ground for further collaboration.

Common areas of shared interest include a need to improve access to mental health services; concerns about quality measures that might include depression; an interest in reducing emergency department and hospital rates; and an awareness of the impact of mental health patients’ needs on the workload of the primary care providers, health behaviors (especially for chronic diseases), and health utilization. Primary providers are more likely to be willing to put up with practice changes and to add measures to their work when the providers making the suggestions are in the trenches as well. When building consensus for collaborative care, the psychiatrist often has to take on the new role of champion for behavioral health and work to engage all team members to deliver effective care.

Consider these approaches when exploring a new opportunity to partner with a group interested in collaborative care. First, conduct an environmental scan. Learn about the needs and current initiatives of potential partners. Is the organization part of an ACO or other value-based initiative? Is the organization working toward meeting national standards, such as HEDIS?

Second, define value broadly, and look to improve the patient’s experience with more access and less stigma. Aim for improved patient outcomes. Improved medical outcomes are often measured, but these are linked to better mental health outcomes, and both lead to increased quality of life. These models also lead to improvement in the provider experience, both for psychiatrists and for medical providers. Finally, in the world of cost-conscious spending, look for ways to leverage scarce psychiatric resources efficiently and take advantage of new revenue for primary care, such as new Medicare codes for providing collaborative care. In addition, point out ways these models may reduce medical spending in organizations that are involved in ACO models.

Third, build on past successes and motivation for a better model. Who are natural partners for a first effort? It is often advantageous to start with a group or practice that has been successful in accomplishing other types of practice change. Make sure there are common areas of interest, a clear vision for collaboration, realistic goals, and a strategy to measure progress.

Fourth, start small, and grow capacity over time. Start with measuring your baseline. Pick one measure, one group or practice, or one target population to start. Commit to continuous quality improvement. Celebrate any progress toward goals, and build on that for the next initiative.

Fifth, get support. Information and resources on this type of practice are provided in Box 1.