One of the themes of health care reform is that providers take on more risk and responsibility for reducing the cost and increasing the care provided to patients. Figuring out the best way to do so is also a risk taken on by providers when they participate in Centers for Medicare and Medicaid Services demonstration programs.
Selecting the right demonstrations to participate isn't a decision taken lightly by providers. Dr. Mitchell Rubenstein, vice president of medical affairs at The Valley Health System, said in addition to any financial risks hospitals face when participating in an unsuccessful demonstration, there are credibility issues to be considered, as well.
Some demonstrations only provide positive incentives for savings, while other demonstrations combine incentives with possible penalties for low quality or overspending. Another factor is whether a demonstration can be canceled by the provider, or if CMS is the only party that can terminate the project.
Rubenstein said that the hospital would be hesitant to move forward with future demonstration projects if a gain-sharing project Valley participated in starting in 2008 had been terminated early. The gain-sharing project allowed hospitals to financially incentivize physicians to implement best practices and reduce their overall cost of care for Medicare patients.
"One of the questions we had to consider before making a decision … was what happened with the demonstration," Rubenstein said, referring to a new CMS demonstration started in late 2011 that was built from the gain-sharing demonstration. "If it wasn't successful, from our standpoint, what makes us think that model one can be successful?"
When demonstration projects were first created by the Medicare Modernization Act, the New Jersey Hospital Association decided it was better to help the state's hospitals navigate the risks of the innovative projects than sit on the sidelines. The NJHA worked with 12 hospitals, including Valley, to apply for and establish the gain-sharing demonstration in the state.
"We were maybe a little out in front of health care reform," Sean Hopkins, senior vice president of health economics for NJHA, said.
"Our thought process was … with all the pressures on health care spending … the only opportunity a hospital has to remain financially viable is to try and figure out a way to have better control on the cost side," he said.
"You jumped in, you tested it, you saw if it worked, you learned and figured out ways to improve on the original structure, and ultimately, CMS was taking that information and looking at it, reviewing it, and learning from it themselves," Hopkins added.
Having an organization like the NJHA out front on the demonstration allowed natural competitors a chance to collaborate on best practices. Hopkins said finding a demonstration project that focused on provider alignment was a "foundational element" of success.
Hopkins said none of the 12 hospitals participating opted out of the demonstration, despite the annual opportunity during the three-year project. Since then, the NJHA has embarked on other demonstrations, including the Partnership for Patients project that looks to reduce readmissions and hospital-acquired infections.
"If this didn't have some parameter of success, it would be less likely to go into a similar program," Rubenstein said. "But keep in mind, what Hospital A considers a parameter of success might be different from what Hospital B considers a parameter of success."
Currently, CMS is running 16 demonstrations around the country. New Jersey hospitals have participated in the accountable-care organization demonstration, the shared savings — now called bundled payment — demonstration and the Partnership for Patients. Several hospital-based primary-care practices also are involved in the comprehensive primary-care demonstration.
When a CMS demonstration does not reduce costs or improve care, it just ends. There is no chance for modification once the federal government decides the model being tested does not work, because of legal issues CMS waives through the programs, and any risks taken by participants are fully absorbed without assistance.
Following a program ending, any modified payment plans are returned to their original status, and any additional expectations placed on physicians through the project are removed. Months of research, implementation, data analysis and work to get the project running can no longer be used as hospitals return to their original models.
E-mail to: firstname.lastname@example.org
On Twitter: @mcaliendo33