Why Health Care Is Stuck — And How to Fix It
The pressures for fundamental change in health care have been building for decades, but meaningful change has been limited while the urgency of change only grows. The moment of discontinuity has arrived. Already unsustainable costs, an aging population, advances in medicine, and a growing proportion of patients in low reimbursement government programs have made the status quo unsustainable. Change is inevitable.
There is only one real solution, which is to dramatically increase the value of health care. Value is the outcomes achieved for patients relative to the money spent. Without major improvements in value, services will need to be restricted, the incomes of health care professionals will fall, and patients will be asked to pay even more.
In our October Harvard Business Review article “The Strategy That Will Fix Health Care” we describe the strategic agenda that is necessary to create a high value health care delivery system.We believe that there is no longer any doubt about how to increase the value of care. The question is whether providers can make the necessary changes.
Why has it been so hard for health care organizations to improve outcomes and efficiency, despite their best intentions? With so many good, smart people working so hard? With patients’ needs so obvious and so compelling? And with such deep societal concerns about health care spending? The answer is complex, but the result is clear: progress in health care has been all but paralyzed by self-reinforcing barriers to change.
We are confident that providers can overcome these barriers by starting with a shared understanding of why they are stuck. We think the most important barriers include the following:
Providers are organized and reimbursed around what they do, rather than what patients need.
Most health care delivery organizations are organized around physicians and specialties. Within a hospital, physicians are members of the departments and divisions and specialize in what they were trained to do. They treat a broad range of conditions relevant to their field—for example, a neurologist will see patients with headaches, provide stroke care, and treat multiple sclerosis and other conditions with a neurological component. Physicians are physically located in their specialty units, and patients are expected to find their way to them. In this structure, physicians generally work hard to help patients during each encounter – and their assumption is that if they do so, they are doing their job. In this context, their efforts to improve care have focused largely on raising the volume of the discrete services they provide, with “efficiency” gauged in terms of “throughput.”
This approach may have made more sense in the past, when there was so much less that medicine could do, and it was possible for specialists to know all there was to know in their field. Today, however, medical progress has made many previously untreatable diseases treatable, and some even curable. Many more types of clinicians must work together to deliver state-of-the-art care. A patient with diabetes, for example, might need care from physicians trained in endocrinology, nephrology, cardiology, vascular surgery, ophthalmology, podiatry, and primary care.
Yet the current siloed organization of care makes multidisciplinary, integrated care extremely difficult, even if clinicians are part of the same institution and utilize the same electronic medical records. At one well-known teaching hospital, a survey showed that 15% of staff physicians did not realize that they were all members of the same physician organization. This is an indication of how strong their main identification is with their specialty divisions.
In this legacy structure, effective teamwork is possible but it doesn’t happen naturally. Duplication and delay are built into the system. Patients are forced to coordinate their own care and make sure their various physicians are communicating.
Care fragmentation is reinforced by the fee-for-service model in which each doctor, specialist or otherwise, is paid separately, while the hospital receives its own payment. Physicians believe that they are compensated for what they do as individuals; basic teamwork functions that are critical to meeting the needs of patients, like meetings to review performance or even using the electronic medical record, are sometimes labeled “unfunded mandates.” Some high-value services are not reimbursed at all, such as follow-up telephone contact after hospitalizations or “virtual” or informal consultations that avert the need for an office visit. The result is that crucial work required for high value care does not get done.
Some new payment models provide reimbursement for care coordinators, but those coordinators are typically superimposed over the current fragmented care process, leaving the basic organizational structure intact. Although an overlay is less disruptive than restructuring the organization of care, it adds cost and essentially treats the symptom of faulty organization rather than the cause. And those added costs have led many to mistakenly conclude that excellent care is inherently more costly, which has hindered efforts to guide patients to the highest-value providers.
Free-agent physicians operate independently, rather than as part of an integrated team.
Not only is care siloed by specialty, but much specialty care in the U.S. is delivered by independent physicians in private practice. A study of Medicare patients, for example, showed that patients saw a median of seven different physicians in four different practices each year, with little or no integration among them.
In many ways, independent doctors have dominated medicine’s culture. “Free agent” physicians view themselves as equity-owners of autonomous businesses, and few have placed a high priority on integrating care with other clinicians as a means to improve value for patients. Instead, there are tensions among different specialists over what they are paid. And, there are inevitable conflicts with hospitals over compensation, staffing, desired facilities, and who should bear risk, not to mention threats to move business to other hospitals. All of this makes multidisciplinary, integrated care challenging – or impossible.
To address these issues, there are a growing number of joint venture models in which hospitals and independent physicians become partners, but the jury is still out on their impact and durability. Joint ventures can help boost integration and shift care to higher value models, but complexity remains about decision rights, responsibilities, and dividing revenues. Such joint ventures often fall short of true partnership, and prove fruitless.
The proportion of physicians that are employed is rising, an important enabler of high value care. When all the physicians involved in care are employed by a larger organization, teamwork can become part of the job description. However, the trust that is crucial to well-functioning teams takes time and work to develop especially when physicians have only recently become employed.
Patient volume is fragmented, making every patient a special case.
Health care systems in virtually every country, including the U.S., disperse rather than aggregate patients with similar needs. A century ago, hospitals sprang up in almost every small town or city, and served any patient walking in the door. This made sense when there was not much that medicine could do for many patients beyond relieving their symptoms. Because hospitalizations could easily last a month or more, close proximity was essential to allow visits by family members. The result was that most providers treated a relatively small number of most types of cases.
Today, however, medicine is far more advanced and specialized and lengths of stay are much shorter. Treating a high volume of patients with a particular medical condition is critical to value, to build experience in highly sophisticated and technical diagnostics and procedures, work more effectively in multidisciplinary teams, and better measure how patients are doing. In the existing fragmented system, providers have limited experience in each type of case, leading to less efficiency, more dropped balls, and worse patient outcomes. Fragmentation also means that most providers are unable to integrate support personnel with specialized skills in a disease area directly into the team —such as nutritionists or behavioral health specialists. There is overwhelming evidence that having a high volume of patients with a particular condition is important to value, and, conversely, that care by local providers with small populations can lead to poor outcomes.
Despite the clear benefit of focusing on the areas with adequate patient volume, providers tend to cling to every service line and duplicate services across health systems. Boards of directors are loath to close services in any facility, in part because politicians mistakenly equate local service with better care. Antitrust regulators are also remain wary of consolidation, mistakenly seeing the relevant market as highly localized and believing that the more providers of a given service in the region the better.
Massive cross-subsidies in reimbursement for individual services have distorted care and stalled care integration.
Under the prevailing fee-for-service payment system, there is a loose relationship at best between the fees paid and the actual costs of performing that service. Flawed reimbursement methodologies have made some services lucrative (for example, radiology and chemotherapy), while others are reimbursed below actual costs (mental health and palliative care, for instance). Organizations use high-margin services to cross-subsidize the money-losing areas, with severe, perverse consequences. Virtually every provider organization is motivated to invest in profitable services like bariatric and vascular surgery in a desperate grab for enough lucrative business to stay alive. The result is excess capacity and overprovision of these services, yet insufficient volume for most providers to deliver excellent or efficient care.
Cross-subsidization across services, even those needed in caring for the same condition such as less reimbursed cardiology drug therapy and highly reimbursed interventional procedures, works against making the highest value care choices while creating tensions among providers, undermining team-based, integrated care. Providers involved in patients’ care fight over responsibility and compensation rather work together. High value but poorly reimbursed services, such as palliative care, are underprovided.
No participant in the system has good information about patient outcomes and the cost of care.
Flying blind is dangerous. When there are no data on how you are doing, and whether new interventions or practices actually improve outcomes or lower costs, initiatives to improve performance can end up doing more harm than good.
The shocking truth in health care is that there are few data on the actual outcomes that matter to patients with specific conditions. Instead of recognizing that quality is determined by outcomes, providers tend to define quality on the basis of compliance with guidelines (for example, reliability in ordering certain tests or “door-to-balloon” time for patients with myocardial infarction) and patient status as measured by a limited number of clinical indicators (such as LDL cholesterol levels and hemoglobin A1c) which are incomplete predictors of outcomes but not actual outcomes themselves.
There is also a near complete absence of data on the true costs of care for a patient with a particular condition over the full care cycle, crippling efforts to improve value. The lack of cost information starts with widespread confusion about the difference between costs and charges. Most clinicians also have no way of knowing what things actually cost or how much time care processes take. Without the ability to understand the costs of the care for specific conditions, or how costs compare to outcomes, efforts at cost reduction revert to power struggles and arbitrary cuts. Efforts to improve performance become mired in turf wars, personal opinions, and clashes of ego. Resources tend to flow to services that seem to be the most profitable or whose advocates are most skilled in internal politics.
Information technology has often made care integration and value improvement harder, rather than enabling it.
Most clinical information systems have been designed around specialties, procedures, or care sites, and focused on scheduling and fee-for-service billing. Few systems were designed to keep track of individual patients over a full care cycle, and provide all the caregivers involved with comprehensive patient information. Few if any clinicians involved in the care of a patient have complete information. Information systems can also make it almost impossible to collect information on outcomes that matter. Highly relevant data (for example, incontinence or falls) are not captured in EMRs at all, and much outcome information is buried in “free text” fields within clinician notes, which makes it hard to extract or act upon.
The information systems used by health insurance plans have been no better, and maybe worse. They are designed to adjudicate and pay bills for individual services not measure the overall care and value for patients. Most insurers cannot even capture whether a patient is dead or alive. Good luck trying to piece together the overall charges for one episode of care if it spans the end of the calendar year. Faulty information systems make it all too easy to give up and continue with business as usual. For example, insurers throw up their hands about bundled payments because legacy systems are coded for fee-for-service payments.
BREAKING DOWN THE BARRIERS
As the figure below shows, the barriers to change in the legacy system are interdependent and mutually reinforcing.
Fee-for-service payments for specialties as well as private practice physicians reinforce the siloed organization of care. Fragmented IT solutions work against multidisciplinary care models, rather than enabling them. Misunderstandings about profitability because of inaccurate costing leads to overly broad service lines, a problem exacerbated by the fact that providers attempt to serve all the needs in their service area. Low patient volumes in many conditions, due to serving only the immediate geographic area and duplicating services across locations, reinforce the siloed structure of care delivery because providers cannot afford to have dedicated teams. And so on.
Because these barriers to change reinforce each other, incremental fixes do not work. As a result, progress in truly restructuring health care delivery has been stymied. The legacy structure almost guarantees low or uneven value for patients, yet it is extremely resistant to change.
These barriers to change make it clear that to move from the legacy system to a value-based system needs to be a true strategic transformation, not just a series of isolated steps. In our article, we describe the six components of this transformation, from organizing into integrated practice units, and measuring outcomes and costs, to expanding excellent service geographically and building a new kind of IT platform, as shown in the figure below.
This transformation will not happen overnight and each component will take time to roll out. But a true solution to our health care problem is within our grasp.