Science to Practice: Making What We Know What We Actually Do

Abstract

The perspective of the medical director of a large public mental health agency is provided regarding how to close the gap between what we know and what we do in mental health care. Tools for change, actions required, and key actors are identified. The author believes the moment is propitious for improving care systemically.

Changing the world of medical care is not for the faint hearted.

Many have tried, and stories of failure, in general medical care as well as mental health (and addictions) care, fill the archives of medical and social science history for those who can bear reading it. For a short survey of how limited our success has been to this day, despite the valiant and persistent efforts of many, take a look at the 3 articles in this issue by Motjabai and colleagues, Kreyenbuhl and colleagues, and Drake and colleagues. Some highlights, or are they lowlights?

• Forty percent of people meeting criteria for a diagnosis of schizophrenia report not receiving any treatment in the past 6–12 months, and only a fraction of those in treatment receive treatment consistent with the evidence base—with psychosocial treatments faring far worse than medications on both access and fidelity to evidence-based practices (EBPs).

• Managed care, loathed by many clinicians, has substantially driven out psychosocial care for people with psychotic disorders and in some instances contributed to discontinuities in care.

• Engaging clients in services remains an elusive goal, especially for young men from minority cultures and people who are economically disadvantaged. Having co-occurring mental and substance use disorders has a profound negative impact on engagement, as does living with the early phase of psychotic illness. Notably, high-need times like after an emergency department (ED) visit or a hospitalization are when consumers are most at risk to disappear from care, and we must do more to use these crises as opportunities to engage people in needed services.

• The methods we have relied on for dissemination and adoption of EBPs, such as guidelines and promotion of quality practices, do not sufficiently do the job; quality improvement and information technology efforts have shown modest and uneven effects.

• Still, we have treatments that work, methods that can make them work, and systems of care (like state or Veterans Administration [VA]-driven initiatives) to drive their dissemination and adoption. Maybe we are approaching what Malcolm Gladwell called “the tipping point”?

We cannot throw in the towel. We cannot accept a “quality chasm” that rivals the Grand Canyon. We cannot and need not. First of all, we have had some success, just not enough to responsibly meet our mandate to improve the lives of people, and their families, who turn to us (clinicians, systems of care, and government agencies) to have a life like everyone else. Second, in the graveyard of failed or poorly realized efforts, we can discover not only what does not work but also what can work, as well as what is needed to make our efforts work a lot better. Third, we have tools today, like electronic medical records (EMRs), learning communities, web-based training, and shared decision making, which will be able to go to scale in the years ahead in ways that we could not be achieved by our predecessors (see review by Drake, Bond, and Essock, this issue). Last but not at all least is the emergence of recovery as a framework for mental health policy and practices; not only does recovery emphasize hope and expectations (for clinicians as well as consumers and their families) but also it establishes “nothing about us without us” thereby enabling consumers to actually drive the process of improving services. In the past, passive diffusion of EBPs was coupled with a passive role for consumers and families—which did not work. Imagine the forces we can mobilize by turning passive to active on both fronts. It is a time for cautious optimism, for determination, and for leadership. More of the same has no future. It is time for the kind of change that only upheaval allows.

I have written this commentary from the perspective of the medical director of a large public mental health system to help make the case for identifying the crucial tools, actions, and actors needed to depart from the performance problems of the past and deliver quality and accessible care to people who suffer from a serious mental illness (including any co-occurring mental and medical disorder). What follows is an itemization of those tools, action, and actors that can close the science to practice gap. Any success will depend on dissemination, adoption, and problem solving of these tools and ideas by those directly providing services or implementing policy, though the perspective and valency of this commentary is that of a state government agency medical director charged with closing the quality gap.

Tools

A variety of tools now exist. Common sense and some evidence suggest that they complement and enhance one another.

• Decision trees or algorithms: These are prescriptive or suggestive pathways by which clinicians, administrators, regulators, and payers are instructed on what to do or what to consider and when. The Schizophrenia Patient Outcomes Research Team is a fine example of this type of work.^1,2 Clinical pathways are more readily employed in practice than are lengthy guideline documents (though guidelines often inform the development of the pathways).

• Decision support: A corollary of algorithms is the provision of specific information to the clinical decision maker in real time, namely, when a decision is being made. The gold standard for decision support is an EMR that has this capability built in and prompts the clinician as he or she writes in the chart or when diagnostic and therapeutic orders are made, including both medications and psychosocial services.

• Care management electronic tools, such as the Psychiatric Services and Clinical Knowledge Enhancement System (PSYCKES): The PSYCKES (http://www.omh.state.ny.us/omhweb/psyckes/demo/full_system/index.htm) is a New York State (NYS) Office of Mental Health (OMH) clinical support tool that provides clinicians, administrators, and quality improvement teams with current information on a patient, a service site (like a hospital ward or outpatient clinic), or a population of interest. Currently, PSYCKES is being used throughout NYS to identify and improve medication-prescribing practices for polypharmacy and for medications known to worsen a consumer's health (eg, those medications that increase risk for hyperglycemia, hyperlipidemia, or hypertension).

• Drug utilization review (DUR): Often criticized, appropriately, for denying needed care, DUR can also be conceived and administered to achieve quality goals, such as improving safety, reducing polypharmacy, drug-drug interactions, or helping to ensure that the right drug is being prescribed for a given condition. When quality goals are achieved, unwanted outcomes are avoided, and this can also have the (secondary but welcome) effect of reducing cost.

• Standards of care: NYS issued mental health clinic standards of care in 2008 for all outpatient clinics in the state (over 650 programs), based on the quality concerns identified after a series of violent episodes in New York City (NYC) involving people with serious mental illness (as victims as well as perpetrators) (http://www.omh.state.ny.us/omhweb/justice_panel_report/). These clinic standards stress what is well known, namely, that the best way to reduce risk of violence (and other adverse outcomes) is to engage people in care, provide thorough evaluations, collaborate with other caregivers and families, supervise and support inexperienced staff, and step up care (not step it down) when the going gets rough or people do not show up for care. In NYS, these standards of care will become the basis of licensing surveys throughout the state by the fall of 2009 (see below under actions).

• Performance monitoring: Clinicians and administrators are remarkably attuned and responsive to contractual and public reviews of their performance, especially when measures are developed collaboratively and are meaningful and feasible. This phenomenon is well known as “what gets measured gets managed.” A good example is the core indicators that psychiatric services in hospitals collect for hospital accreditation with the Joint Commission. Public posting of service provider performance (eg, on a state or county mental health authority Web site) is another powerful tool in reshaping and driving improvement in provider performance because no one wants to be listed among the poor performers.

• Learning communities: These are an element of the process of quality improvement and are important in adoption and dissemination of competencies. Learning communities typically meet face to face and by phone on a regular basis (eg, monthly) while transformation is underway and thereafter to maintain or improve upon success. They generally require expert leadership and staff dedicated to the process of quality improvement. Recently, more focused and less costly learning communities called Practice Improvement Networks (time-limited quality improvement networks among providers that rely on technology-based communications like telephone, Webinars, list serves, and web-based consultation rather than face-to-face learning) have evolved in response to the need to do more with less.

• Distance learning: Universities are perfecting this method whereby courses and training are provided by the Internet in interactive and engaging ways. This is how you too can get your MBA from a school thousands of miles away. NYS is developing a comprehensive set of web-based training modules for clinicians to gain skills in the diagnosis, engagement, and treatment of people with co-occurring mental and substance use disorders with the support of the NYS Psychiatric Institute/Columbia Department of Psychiatry and the Dartmouth Evidence-Based Research Center. Clinicians will be able to train and take examinations in variety of subjects that can both demonstrate their competence and that of the agency for which they work. Distance learning is crucial to mental health and chemical dependency agencies that are known to have annual turnover rates in excess of 30% as well as limited resources to support needed staff training.

• Training and train the trainer: This traditional method is the “retail” approach to increasing staff competencies. Either line staff are trained directly or a cohort of staff are trained with the expectation that they will train others. Service systems with large geographic distances, high staff turnover, and limited financial resources are finding this method increasingly difficult to employ.

• Peer supports and shared decision making: More than ever peers are needed to engage and retain consumers in services as well as to improve upon the services they receive. A remarkable example of consumer-led work is the shared decision making method developed by Dr Pat Deegan. An example of her work is when peers aid other consumers in completing a short summary of their response to medication treatment and their goals for meeting with a prescribing psychiatrist right before that meeting. Peer supports can and need to be far more comprehensive and delivered more globally, perhaps through web-enabled technologies.

Actions

Tools require that someone pick them up and use them. Left to their own, as history has demonstrated, they are passive elements that have as little effect in building a house as a saw, hammer, and nails that sit unused on a stack of boards.

• Regulatory relief: Is it correct to say that the experience of providers and advocates with government policy makers and auditors is that you cannot do this or that because of some regulation? The first action needed is to investigate whether that statement (belief) is wrong. Countless myths and misinformation accrue over time about government rules and regulations. So first find out if something indeed is the case; when not the case, government agencies should issue clarifications and questions and answers that inform providers and others how much liberty they actually have. For an example, the NYS OMH's Web site includes a section to help clinicians understand that they can disclose protected health information without authorization when such a disclosure is part of “… the provision, coordination, or management of health care and related services for an individual, including consultation between providers and referral of an individual to another provider for health care” (http://www.omh.state.ny.us/omhweb/hipaa/phi_protection.html). In other instances, regulations exist that contribute little more than burden and impediments to effective work. Those need to be excised from the rule books, however much time that may take.

• Regulation: Many think that regulation is a potent means of achieving results. Often, that is not the case. Regulation (usually) requires a lengthy public and political process making the product likely a reflection of compromise and clout rather than of quality. At times, regulation is required to change existing statutory provisions or mandates, in which case it needs to be done. At other times, regulation is needed to permit or support novel and necessary funding. Find the shortest and best way to drive change and rely on regulation judiciously in light of the time and the gauntlet it requires.

• Financing: “Money makes the world go round” said the barker in cabaret. Reward works better than punishment, especially over time. Simple financing systems that are clinically meaningful and readily adopted are highly desirable—and often hard to achieve. It is amazing how difficult it can be to avoid complicating payment (and other administrative) systems. But when it can be done right, the force of change is remarkable.

• Licensing: The power of licensing is huge. Without a license, or with license restrictions or conditions, programs cannot operate effectively if at all. Licensing is a hammer that can build or destroy. NYS OMH is using its licensing authority in over 650 outpatient clinics to assess adherence with the Clinic Standards of Care noted above; OMH is adopting what is called the tracer methodology, implemented many years ago by the Joint Commission. OMH surveyors will go to a clinic with a list of 6–8 active and former clients whose care needs would be very telling about the quality of the clinic and its adherence to the published Clinic Standards of Care. Each client's care would be “traced” from intake through various processes of care to determine, eg, if a co-occurring substance use of medical problem existed and what was done, if ongoing collaboration was achieved with other caregivers or family, or what the clinic did when there were engagement or retention problems. Licensing would then reflect the degree to which the clinic achieved quality standards.

• Contracting: It is remarkable how little contracting is used as a tool for quality. Mostly contracting is for how many services rather than how good the services were in meeting their goals. Pay for performance has yet to become a standard means by which payers achieve what consumers and families want from their caregivers. One example of performance-based payment is the homeless outreach contracting that we redone in NYC several years ago that paid not for profit agencies for housing the street homeless; in the past 3 years, that has resulted in a 50% reduction of chronic street homelessness throughout the city (http://www.huffingtonpost.com/lloyd-i-sederer-md/real-progress-on-homeless_b_178217.html).

• Provider and system accountability: In the summer of 2009, a joint NYC-NYS project will commence in NYC that will identify the most vulnerable clients in the public sector (principally using Medicaid data). These clients will be tracked, using the database, and alerts issued that indicate breaks in service (no outpatient or assertive community treatment visits or no picking up medication) and sudden escalations in acute services (like ED visits or hospital admissions). A government-run care-monitoring service receiving the alerts will then contact the provider(s) last identified for a given client and check to ensure that all that can be done is being done. This project should reveal both provider and system accountability and support improvements in both areas.

Actors

We are never alone in life as in policy and practice. Identifying which actors have influence and authority is essential to changing the landscape of care. Doing so with as many actors aligned as possible is the art of politics.

• State government: Dr Drake in his article in this issue declares the states “decisive.” We agree. Perhaps, the greatest role that states play is that they administer Medicaid that has become the primary payer of mental health services—more so than state mental health authorities. Because financing drives health (including mental health) care, the states are positioned to use Medicaid to support what works and to make doing more of what does not work financially unsupportable (see Federal government's role reguidance below). State mental health agencies can employ their authority over licensing to foster science to practice by creating quality standards that are the basis of licensing surveys.

• County or municipal governments: In many states, the counties and cities are seeking a new identity as public mental health payment is primarily from Medicaid, and licensing and quality review is often held by state government. “Local government units” (LGUs) know their constituents and politics best and can be instrumental in service planning, selection of local providers, and gaining stakeholder support. In many counties, as well, LGUs are providers themselves and thus can both pitch and catch when it comes to policy and practice. Yet, there is enormous regional and municipal practice and contract variation that unless improved will continue to contribute to less than optimal consumer outcomes.

• Federal government: The President's New Freedom Commission on Mental Health (http://www.mentalhealthcommission.gov/) recommended that the Centers for Medicare and Medicaid Services (CMS) issue guidance on how Medicaid could be used to support EBPs. Federal-level recommendations would be important in advancing a uniform set of recommendations and thus avoiding state-by-state variation. The Federal government also funds the VA system of care, increasingly a centerpiece in the care of veterans whose mental health needs are growing exponentially. The Federal government, through the Department of Housing and Urban Development (HUD), is also critically important in housing policy and the provision of affordable housing, especially through the Section 8 program. Substance Abuse and Mental Health Services Agency (SAMHSA) is the Federal agency charged with mental health services policy to complement the research agendas of National Institute of Mental Health, National Institute on Alcohol Abuse and Alcoholism, and National Institute on Drug Abuse. Interagency work is essential in the years ahead to bridge policy formation and funding allocations across CMS, HUD, SAMHSA, the Federal NIH agencies (noted above), and the VA as well as the White House Interagency Council on Homelessness to ensure that Federal policies, priorities, and funding meet the complex needs of people with mental and substance use disorders. One point made by Dr Drake bears emphasizing, namely, that Federal services and research agencies need to focus on what he has called the “95% solution”—where research aims to improve dissemination and adoption of effective practices because we know that near to half of recipients receive no care and over 40% receive poor care. The greatest improvements in public mental health in the next 5–10 years will come from increasing access, engagement, and retention of consumers, not by more refinements of EBPs that only 5% or recipients are now getting.

• Providers: The provision of clinical care is the most challenging domain of all. Clinicians and administrators have the direct responsibility of delivering science to practice. Providers will have to change the oil while driving the car: They will have to do transformative work while keeping day-to-day operations going.³ Those that meet the challenges should be preferentially given contracts and performance payments in order to systematically improve the quality curve.

• Consumers and families: The greatest source of support for the conversion of science to practice is today the one that is least used, namely, consumers and their families. Recovery is possible, and consumers and families understand and are more and more demanding it. PSYCKES, noted above, is developing a consumer portal so that recipients can go into their appointments knowing their medical information and assisted in becoming informed consumers of care. Wellness Recovery Action Plans developed by Mary Ellen Copeland are another fine example of how recipients can drive their own care and accelerate closing the science to practice gap (http://www.mentalhealthrecovery.com/).

Conclusions

Remember Bob Dylan's lyric “you don't need a weatherman to know which way the wind blows.” When science tells us what needs to be done and it is not part of routine practice, a gap exists that demands closing. We need to do what we know needs doing.

The moment for change could not be more propitious. Doing more of the same is unsustainable. We have developed a set of tools and a variety of methods, or actions, for their implementation. Now is time for the actors to go to work.