By JOSEPH ANTOS and YEVGENIY FEYMAN

New York has an opportunity to make real improvements to its Medicaid program; but it is leaving out key health care providers which might be best positioned to cut costs and improve value.

The state has undertaken a $6.4 billion dollar initiative, the Delivery System Reform Incentive Payment (DSRIP) program, to reduce avoidable hospital use by 25 percent over the next 5 years. DSRIP is testing ways to do everything from cutting unnecessary emergency room visits to reducing transmission of HIV/AIDS.

This could lead to healthier, more productive lives for New York’s 6 million Medicaid beneficiaries and could generate positive spillover effects for all New Yorkers—if hospitals, physicians, and other providers adopt better ways to manage Medicaid patients, they are likely to use those improvements to assist all their patients.

A team approach, coupled with preventive health care to avoid hospitalization and appropriate follow-up care, is an important part of the strategy. Better coordination between community physicians and the hospital, and better information for patients, can save money and improve health outcomes.

The problem? New York’s DSRIP program is mainly confined to “safety net” providers,  including hospitals, academic medical centers, nursing homes, individual physicians, and others with a high volume of Medicaid or uninsured patients. The state limits the participation of physician groups known as Independent Provider Associations (IPAs) because they are not considered Medicaid providers.

IPAs fall somewhere between health maintenance organizations and sole practitioners.  They are small networks of physicians that negotiate contracts with insurers on behalf of their members. IPAs cannot negotiate with Medicaid, which has a “take it or leave it” attitude when it comes to paying providers. That, and nothing inherent in the way IPAs provide care, is the reason they are left on the periphery of the DSRIP experiment.

Individual physician members of IPAs are allowed to participate in DSRIP, but have little reason to do so. DSRIP is focused on hospitals; without the scale of IPAs, individual practitioners would be unable to run the sorts of experiments DSRIP requires.

This misses an opportunity to engage physician groups better placed to cut unnecessary hospital use than the hospitals themselves. After all, physicians in the community are the ones who most often decide whether patients should go to the hospital. Their involvement with patients can prevent unnecessary ER trips. Keeping patients out of the hospital is the path to more effective care at a lower price.

An objective of DSRIP is to encourage greater integration and coordination among providers, with the goal of promoting value, not volume. Instead, DSRIP promotes consolidation among providers with hospitals in charge. Numerous studies show that increases in hospital market concentration lead to increases in the price of hospital care and higher spending. The FTC has already warned the state that incentives for providers to participate in DSRIP can encourage anti-competitive behavior, too. That is not what the designers of DSRIP had in mind, but it is the likely result.

Can physician-owned IPAs do a better job than hospital-centered organizations in responsibly lowering health care cost? An analysis of health care spending in California found that hospital-owned IPAs had 10.3 percent higher spending per patient than physician-owned IPAs. More evidence comes from Medicare’s Pioneer accountable care organization (ACO) program. The most recent results reveal no difference in savings from hospital-based ACOs and unaffiliated physician groups. Such studies indicate that financial integration with large hospitals is not necessary to achieve lower costs, and may, in fact, be counterproductive.

Can IPAs maintain the same level of quality as hospitals? The California study did not address quality, while many of the quality metrics used in the Pioneer ACO program track processes rather than outcomes. This means that DSRIP can be a useful tool in beginning to answer that question.

We don’t know the best ways to reduce costs and improve value in our health care system. By limiting IPA participation in DSRIP, we might just be taking the best option off the table.