Salmon says Cigna members prefer the open-access model for a variety of reasons, but chief among them is a belief that restricting where the patient can receive care doesn't force providers and hospitals and health systems to make themselves attractive to patients so that they want to stay in the same system whenever possible.
Participants in the Cigna program must have achieved patient-centered medical home standards, are committed to provide care set by the NCQA, and must commit to a dedicated embedded care coordinator. Collaborative accountable care debuted with Cigna in 2008, and by 2010, a study in the journal Health Affairs showed that its Arizona practice recorded per member, per month costs that were $27.04 more favorable than its comparison group. Quality measures were improved versus the peer groups, in Texas and New Hampshire, although cost differences were deemed insignificant.
"It does put a challenge on the groups. Certainly physicians would much prefer if most of the patients they were responsible for were in a more limited network," he says. "But this means they have to offer services the patients want so that patients voluntarily use them for their care."
He says in Cigna's experience, open access has encouraged physician groups to extend office hours, send reminders to patients, and generally make a greater effort to maintain oversight of that patient's care.
"It's just better customer service because of this structure. That drives a higher level of patient care," Salmon says.
He explains that in shared-savings programs like the ACOs Cigna operates regionally, the financial reward for provider participation is only one of several incentives. Another is increased patient volume; partners that show above-average ranks in quality and cost are prominently featured in tiered and network products to encourage patient growth. A third reward is increased professional satisfaction for physicians.
"We're striving to help the primary care physician who's been encouraged to see more patients more and more rapidly in order to make a living," Salmon says. "In these arrangements, how many patients you see a day is no longer a key metric; how well you take care of them is. Currency for services does not fully reward care coordination."
What also might be toxic is that many ACOs are still ramping up their management and technological capabilities to fully take advantage of their partners' ability to actually manage care, especially of populations that utilize the most expensive sites of care most often, such as those who use or who are more likely to use the emergency room or to be admitted as an inpatient. Aetna's Kennedy says the most important work the insurer is doing right now is adding to the number of programs with its partners to make them as financially successful as possible.
"There is a foundational set of care management programs and technologies, but over the next year, you'll see us add to those capabilities," Kennedy says. "By next year, we'll be able to say we have a new business model for healthcare with all the programs and technologies to make it work, including clinical and claims data integration, analytics based on clinical indicators, and process reengineering to reconfigure what they do and how they do it."
One reason Aetna is well positioned to help "enable" providers such as Banner is because its goal, where possible, is to offer what it calls "payer-neutral tools" to help organizations to demonstrate improved value. For instance, Aetna's tools, largely analytical and predictive in nature, help Banner improve the value proposition for its ACO populations, including Pioneer ACO, even though Medicare patients, not Aetna patients, are attributed to that structure.
Across all patient populations served by Banner ACOs, the structure recorded the following improvements between 2011 and 2012, the latest period for which comparison information is available:
The numbers represent a range of results based on the patient population being measured, and the results are not blended. For instance, the lowest medical cost savings during the period measured were 3.0% for one group, while the group on the high end of the range saved 5.5%. One direct correlation to those results is that the Aetna commercial products associated with the Banner Health Network ACO populations are priced anywhere from 8% to 15% below prevailing rates in the market, according to the insurer.
Another example of such results comes from Carilion Clinic in Roanoke, Virginia, which has 3,000 participants in a shared-savings ACO with Aetna, which was formed in 2011.
Compared with 2011, 2012–2013 results show that patients in the Aetna ACO saw 6% fewer inpatient days and had 23% fewer avoidable emergency department visits and 10% fewer high-tech imaging scans. The ACO reduced its readmission rate to 4.9% from 5.6%.
Banner's Decker says a big part of helping physicians integrate processes that help achieve this type of savings includes educating them on so-called big data principles.
"The bigger the data pool, there's a greater tolerance of small inaccuracies because the overall story is probably still accurate," he says. "The bigger the data gets, it almost becomes more philosophical, and an individual arguing that a small piece of data is inaccurate doesn't change the overall story."
Banner Medical Group had just less than three million discrete patient contacts in 2013. "We're getting physicians to understand the concepts of big data and maybe become less defensive about individual cases that are inaccurate," Decker says. "I'm a provider myself, but what relaxed me is that this is not aimed at individuals. If you look at overall performance for the year, there's a story to be told there."