The Institute of Medicine (IOM) is an honorific organization established in 1970 by the National Academy of Sciences, with the purpose of advising the federal government on matters relating to medical care, education, and research. This advice has recently come through the Institute's Committee on the Quality of Health Care in America, which has issued two important reports. The Committee's first report, To Err Is Human: Building a Safer Health System,1 was released in 1999 and focused on a specific dimension of quality, viz. patient safety and medical errors. The key findings from this report were that errors are common in medical care, that such errors are caused by system or process (rather than human) failures, and that we must design safer health care systems in order to prevent medical errors. The Committee's second report, Crossing The Quality Chasm: A New Health System for the 21stCentury,2 was released in 2001. The "Chasm Report" took a broader "treetop" view of the American health care delivery system, concluded it was severely broken, and offered suggestions for how it can be re-engineered to "innovate and improve care." The Chasm Report has profound and far-reaching implications for the future of health care. In this article, I describe the implications of the report for neuropsychiatric care and suggest that the report provides a road map for dramatically improving such care.
The Chasm Report praises the unparalleled advances in medical science in this country, as well as the skill, dedication, and self-sacrifice of American health care workers, but it also indicts the health care delivery system for not translating those strengths into meaningfully better care for each and every patient. The report documents the wide variation in health care quality, noting that too many Americans fail to receive safe and effective care. The care is fragmented, the system is full of waste and inefficiency, and too many Americans lack health insurance and therefore are deprived of access to basic care. This gap between what should be possible in health care (given the incredible advances in science and technology) and what the patient actually experiences in the clinic or at the bedside is the "chasm" referred to in the report's title.
To remedy this state of affairs, the Chasm Report recommends that quality be made an explicit priority of the health care system in this country, and that all constituencies work together to improve the following six dimensions of health care:
Safety (the care should avoid harming patients),
Effectiveness (the care should be evidence-based, and should avoid overuse or under use),
Patient-centeredness (the care should be respectful of the individual's preferences, needs, and values),
Timeliness (the care should be available when the patient wants and needs it),
Efficiency (the care should be free of waste), and
Equity (everyone should get the best possible care, regardless of age, sex, race, financial status, or any other demographic variable).
These six dimensions of "perfect care" provide a revealing litmus test by which to assess the current quality of neuropsychiatric care. How much do we really know about the quality of neuropsychiatric care in the United States? I suppose the answer depends in part on whether one sees the proverbial glass as half full or half empty. In many ways, neuropsychiatry is a microcosm of our larger health care system in this country. The good news is that there is a growing scientific database within neuropsychiatry, as evidenced in part by the success of the Journal of Neuropsychiatry and Clinical Neurosciences and of organizations such as the American Neuropsychiatric Association (ANPA). The bad news is that there is a paucity of data on these dimensions of ideal neuropsychiatric care.
The treatment of depression in patients with neurological disorders provides a particularly revealing example of this "good news—bad news" situation. Depression is remarkably common in patients with neurological disorders,3 and has considerable impact on the patient and family. Important and innovative research is illuminating its neurobiology.4 Yet how much do we really know about the depression care in neuropsychiatry?
Safety—Antidepressant medications, and less commonly ECT, are typically recommended for treatment of depression in patients with neurological disorders (reviewed in5). Although a growing clinical experience suggests that these therapies are generally safe in many patients with brain disorders, there is also evidence that such patients may be at increased risk of certain adverse effects. For example, we and others have found that depressed patients with evidence of basal ganglia ischemia on magnetic resonance imaging are at increased risk of delirium from ECT or antidepressant medications (reviewed in6). Others report a higher frequency of cardiac and gastrointestinal side effects from antidepressant drug therapy in depressed patients with brain disorders.3 Despite the potential concerns raised by these reports, we lack firm data on the specific risk of morbidity of these treatments in our patients. And how do our patients themselves feel about the safety of such treatments—I am not aware of any published data on this issue. In a forthcoming report, my colleagues and I will describe a simple and effective tool for identifying adverse drug events in patients with neuropsychiatric disorders and for assessing the patient's satisfaction with the safety of such care.
Effectiveness—In a recent review published in this journal, Rummans and colleagues reported that there were few controlled data to support the use of antidepressant medications or ECT for the treatment of depression in patients with brain disorders.3 For example, only one randomized controlled trial has examined the efficacy of antidepressant medication (in this case nortriptyline) in patients with Parkinson's disease, and active drug was not clearly different from placebo. Of the five randomized controlled trials that have examined the efficacy of antidepressant medication in patients with Alzheimer's disease, only three reported efficacy superior to placebo. The evidence for efficacy is perhaps strongest in patients with post-stroke depression, where four of four studies report an advantage with drug vs. placebo, but even here one study finds an selective serotonin reuptake inhibitor (SSRI) (citalopram) effective where another does not (fluoxetine).
Patient-Centered Care—Ideal neuropsychiatric care would be care customized to the particular needs and values of each individual patient. The patient becomes the source of control for all decision making. While there are no systematic data on this issue, my personal experience indicates that much of the neuropsychiatric care in this country is not patient-centered. At a very fundamental level, if we lack meaningful data on the risks and benefits of a treatment, how can patients possibly make an informed decision about whether to undertake such treatment? We often resort to describing the results of studies of highly selected patients who are medically healthy, but the question remains just how applicable are those findings to this particular patient with this particular neuropsychiatric disorder? Even the information we do provide to patients is rarely designed with input from patients and families, which would no doubt improve its clarity and usefulness.
Timeliness—In the absence of systematic data, my personal experience suggests that there are many sources of delay in neuropsychiatric treatment. Perhaps the largest problem is that of diagnosis—too often the behavioral disturbances are ignored, misinterpreted, or mistakenly dismissed as “expected psychological reactions” to the neurological disorder. Once the need for a neuropsychiatrist or other similar specialist is recognized, then there is the delay associated with trying to find one. And once one of us is found, how long must a patient wait before we can fit him/her into our already overcrowded schedules (assuming we can get authorization from the payor)? And let's not forget how long it takes to get the brain imaging study. …
Efficiency—Our current health care delivery system contains too much waste, and in my experience the same is true for neuropsychiatric care. In addition to the waste resulting from ineffective or untimely care (discussed above), there is waste resulting from complicated processes and from unnecessary paperwork and bureaucracy. There is also the waste associated with obtaining unnecessary laboratory studies, particularly brain imaging studies, when such studies do not rationally impact therapeutic decision making. This waste contributes to higher costs of neuropsychiatric care and in some cases even exposes the patient to unnecessary risk (e.g., radiation exposure from X-rays).
Equity—The quality of our neuropsychiatric care should not vary simply because of a patient's age, sex, ethnicity, socioeconomic status, geographic location, or any other demographic factor. We lack a body of systematic data that addresses this issue, but the limited information available suggests that these factors may be predictors of neuropsychiatric care (reviewed in7). For example, neuropsychiatric disorders abound in prisons and in state mental health hospitals and systems, yet such locations have a relative paucity of clinicians with neuropsychiatric expertise. Similarly, a review of the membership directory of the ANPA indicates a concentration of members in the northeast, in large urban regions, and in private and academic health care settings. These observations suggest that those patients in different locales, in rural areas, or who lack health insurance will have relatively less access to skilled ANPA clinicians.
In summary, while we lack definitive information on many of these dimensions of perfect care, the available data and our experience suggest a system of neuropsychiatric care far from ideal.
There is a quality chasm in neuropsychiatric care, and I suggest that our profession should set as a goal the closure of this chasm by the end of the decade.
So how do we accomplish this ambitious goal? I believe that for each of the most common interventions in neuropsychiatry (as for any medical procedure or intervention), there should be a defined range of acceptable outcomes. I encourage the ANPA and the Journal of Neuropsychiatry and Clinical Neurosciences, to work together to establish this database, and in so doing, define "quality" for the field of neuropsychiatry. Both entities should establish a joint strategic plan that incorporates the Chasm Report's six dimensions of perfect care as a framework to guide future activity and resource investment. The main strategic objectives in this plan would be to:
Obtain baseline data on the status of the field's performance on each of the six dimensions of neuropsychiatric care,
Define "ideal" care for the most common conditions encountered in neuropsychiatry,
Set measurable goals and develop tactics for achieving this "ideal" care, and
Assist individual neuropsychiatric practitioners in measuring and improving their specific care.
As an example of this work, let us return to the issue of depression care in patients with neurological disorders. A "pursuing perfection" team from the ANPA and the Journal would review the extant literature on the care of persons with depression secondary to a neurological disorder; identify additional data needs (e.g., What are the indications for brain imaging? Is there a role for psychotherapy or for nontraditional treatments? What constitutes evidence of effectiveness? etc.); make recommendations for care within the limitations of the current dataset (e.g., establish care guidelines or pathways that are published in the Journal, and ensure that the voice of the patient is incorporated into the design of such guidelines via Consumer Advisory Groups); set measurable goals (e.g., We will reduce the number of unnecessary laboratory studies by 50% over the next year. We will double the percentage of patients with stroke who are screened for depression, etc.); and then help the practitioner implement these "ideal" practices (e.g., ANPA would sponsor workshops on implementing advanced office design and clinic access, as well as develop model patient education materials (paper, video, CD ROM, Internet, "Ask your neuropsychiatrist", etc.) for depression screening, diagnostic procedures, and treatment interventions, and the Journal would help distribute these.).
Such a plan is perfectly in keeping with the ANPA's mission of "…applying neuroscience for the benefit of people".8 Further, the timing for this initiative is particularly propitious for ANPA, which is due to update its rolling 3-year strategic plan in 2003.8
I also challenge each and every neuropsychiatric practitioner to begin a process of systematic data collection in order to provide outcome data for their individual practices (in a future article I will describe a simple yet highly effective tool that we have developed at Henry Ford Behavioral Health for obtaining such data and improving service quality). Each of us should make such "clinic-specific" data available to patients when they are referred for our neuropsychiatric care, and we should encourage our patients to compare these data to those of other providers and to the range of acceptable outcomes established by our field.
Of course, we face many challenges in our efforts to implement this bold plan. Much more research is required to define the database, and neuropsychiatric practitioners need tools and skills to establish their own outcome metrics and pursue their improvement work. Such work is by its very nature integrative and collaborative across specialties, and as such it may be met with resistance (as has the field of neuropsychiatry itself) from the disparate medical specialties of neurology and psychiatry.9 But perhaps the greatest challenge will be whether we can summon the will to change. I don't see that we have a choice—we owe it to our patients to provide the best possible neuropsychiatric care while at the same time respecting their individual wants and needs. Given the incredible determination and commitment to quality that characterizes so many neuropsychiatrists, I am sure we will succeed.
I am indebted to John Campbell, M.D., a fellow neuropsychiatrist who "pursues perfection" and who provided helpful comments on an earlier draft of this manuscript.