Some debates about health care policy represent black and white choices. But others are a more nuanced shade of grey: the new proposals could turn out well, or not so well, depending on how they are implemented.

One such reform getting broad-based attention is the creation of Medicare Accountable Care Organizations, or “ACOs,” which are scheduled to come online on a voluntary basis next January 1. The overall goal of an ACO — rewarding hospitals, physicians and others for providing higher quality, lower cost care, as well as better coordination and more effective management of chronic disease — is the right treatment to fix the ills of our nation’s health system. But depending on how the new program is implemented, ACOs could instead have unanticipated negative effects: a chilling of medical innovation and progress and a “cheapest is best” style of medicine that leaves some patients behind.

To make sure ACOs achieve the goals we all support without these negative consequences, we need to grapple with two key facts:

The real danger of stinting on care. By bringing ACOs into Medicare, the federal government is exposing elderly and disabled patients to a new shared financial savings model –- one that gives hospitals and physicians bonuses based on the strength of their performance in economizing and that, under the draft ACO regulation, also penalizes doctors and hospitals who do not achieve spending targets.

Recognizing the importance of quality care, the Centers for Medicare and Medicaid Services will make the amount of financial reward ACOs can achieve dependent on whether or not they meet a set of quality measures across five domains of care. But while meeting the quality measures may be a necessary condition for quality care, it is certainly not a sufficient condition. These quality measures do not generally measure outcomes and there are vast areas of clinical practice which are not addressed at all.

To cite just a few examples, there are no measures for treatment of cancer. There are no measures for treatment of severe arthritis. There are no measures for treatment of osteoporosis. There are no measures for treatment of chronic pain. CMS has stated its intention to move toward outcome measures as they become available, but such measures are not generally available now, and it will be challenging to create outcome measures that capture all key attributes of successful treatment.

The difficulty of effectively measuring quality for populations through standardized measures will only increase. As the mission of medical practitioners from the dawn of time has been to cure ills in the solo human being, so each new genetic revelation reminds us that science is hurtling toward a future in which care can be customized for each individual patient.

That we are each in our own way unique does not mean that we each need our own type of health care. But neither does it mean that quality measures designed for a whole population will ensure quality care for every individual. The answer lies in a marriage between services provided efficiently and protections to ensure appropriate care for individual patients.

One potential way to get to the desirable balance is to have independent monitors oversee the ACOs to ensure that arbitrary “stinting” does not keep patients from getting the treatments they truly need. Such monitors could provide a safeguard against subtle, but damaging practices that could deny individual patients the best care for their unique circumstances. The ACO draft regulations include provisions for monitoring by CMS, but an independent monitor could ferret out problems that CMS might not identify and would contribute to public confidence in the new entities. Public release of monitors’ findings would help ensure a degree of protection for Medicare beneficiaries.

An especially effective solution would be to assure that individual practitioners’ decisions on individual patients are not affected by financial incentives to stint on care. An option that the Office of the Inspector General raised in its portion of the ACO draft regulation would be particularly effective: reward individual physicians differentially for their quality scores, not the reduced costs they generate by their decisions on individual patients. Reward all physicians in the group equally for the progress the group as a whole makes in delivering care economically.

After all, ACOs are not intended to save money by limiting care for individual patients. The promise of the ACO is that it will reduce costs through better coordination of care, better management of chronic disease and more effective preventive health to keep patients healthier and reduce hospitalizations and expensive, avoidable interventions.

In fact, in order to get truly effective disease management, we may need to require services that are not currently covered or available, such as remote monitoring, home visits and patient education.

The danger of slowing the development of new treatment and cures. The health of the population has improved dramatically over the last few decades as the result of development of new treatments and cures. Between 1980 and 2000, life expectancy increased by more than three years. Death from heart attack plummeted by 50 percent. Death from stroke dropped 30 percent. Death from breast cancer dropped 20 percent. Disability among the elderly declined by one-quarter. The last thing anyone wants is to slow that progress — especially when, in this 21st century of the life sciences, the opportunities for progress are so great.

But without certain design elements, ACOs could be roadblocks in the path to improved patient care if they become blueprints favoring standardization and discourage physicians from adopting new forms of care. Avoiding this requires a handful of careful steps:

  • Quality measures must not freeze medical practice in place, and we should not discourage providers from using new treatments.  If a quality measure is based on whether a provider uses an existing treatment, ACOs and physicians should not be penalized by lower quality scores merely because they are early adopters of a new technology they may view as more effective for their patients. Furthermore, if the new treatment proves to be a better alternative, it should be incorporated into the quality measure.
  • ACO benchmarks should include temporary adjustments to encourage use of new technologies that can improve patient care. These adjustments could be similar to adjustments Medicare uses to pay for new treatments under the Diagnosis Related Group hospital payment system. For new treatments that meet CMS standards, ACOs and physicians within the ACO should not suffer financially for being first to adopt a treatment that promises to significantly improve patient care because it might raise costs in the short term. Many of these new treatments may not only improve patient care but may also reduce costs over the long term.
  • Patient outcome and cost measures need to take into account benefits that accrue over longer time periods than a year or even three years — for example, a replacement heart valve may be more expensive initially but it may last 25 years instead of 10 years.

Steps like these could help us achieve the essential goals of payment reform — improving quality care and reducing cost growth — while at the same time strengthening Medicare’s mission of guaranteeing each senior citizen the best care that American medicine can provide, today and in the future.