That smaller cultural changes must accompany delivery system innovation is a truth already starting to take root among the largest institutions in health care. “All health care reform is local,” is a common saying among health care experts, and one of the key tenets in the mission statement for the Center for Medicare and Medicaid Innovation. In a speech on health system improvement given in early May, Kathleen Sebelius stated, “this transformation will happen one hospital and one community at a time. We can provide support and establish incentives, but you are the ones who will have to do the hard work of putting better systems into practice.”
In this way, the Partnership for Patients and the CMMI appear to be elegant solutions. The programs aim to encourage efficient, integrated care through a series of pilot programs and local interventions. Far from mandating top-down solutions, HHS is encouraging collaborations between hundreds of provider groups, hospitals, consumer groups, employers, and state and local agencies. The specific mechanisms of those solutions have not been articulated, but $1 billion has been allotted to test interventions in hospitals that can demonstrate both the organizational capacity and the will to implement quality efforts. In many ways, the programs accelerate the national conversation on quality, but success still rightfully hinges on the prerogative of local systems to improve themselves.
“I’ve been working with patient safety for the better part of twenty years, and I haven’t seen anything like this before,” Donald Berwick, head of the Centers for Medicare and Medicaid services, said. “It’s thrilling.”
Yet despite this initial enthusiasm, the goals are plainly unrealistic. In the next two years, the Partnership aims to reduce preventable injuries and complications by 40 percent, and reduce all hospital readmissions by 20 percent. President Obama’s proposed FY2012 budget is attributing to both programs $50 billion in savings over the next decade. If our previous forays into quality culture are any indication, system-wide changes involve a long and arduous process on every level of organization.
This is where laudable large-scale visions may run into hard politics: In a few years time, when those outcomes and savings fail to materialize, the agency may be forced to rely on policy levers that will inadvertently (and necessarily) undo those cultural changes. PPACA gives the HHS Secretary the legal authority to expand the scope of demonstration projects in Medicare, Medicaid, and CHIP through formal rule-making processes. In such a case, all hospitals and provider groups who accept those beneficiaries will be subject to newly mandated initiatives and data collection requirements, regardless of how that program fits into the local culture of care. Expanding the scope and scale of demonstrations is an easy solution, but policy by itself can do little to change the social norms within a hospital.
The much more real concern, though, is that the notion of “a culture of quality” itself may become codified and abstracted in ways that suit the data-driven needs of policymakers and not caregivers or patients.
By necessity, policymakers and providers live in two different worlds. And in the same way doctors use their social environment to understand health care, policymakers rely on what is most immediate and concrete for them -- data. Serious reviews of high-performing systems in medical journals and government reports invariably center on technocratic details: Medicare fee schedules, HMO case manager scripts, NCQA accreditation, HEDIS scorecards – all the things that can be tracked, measured, standardized, and then replicated. In the policy world, data is the social capital, and its language is a series of acronyms, outcome measures, and glossary terms.
Observing system-wide trends are, of course, vitally necessary for public and private sector researchers. But in the search for replicable interventions, data will always tend to discount what it can’t see: idiosyncratic interactions with patients, complicated webs of friendships between doctors, unique cultural norms within hospitals, and all the (literally) immeasurable ways communities and people learn how to deliver care better. This incongruity will always exist: There can never be a finely-tuned implementation strategy for, “everybody has to be on the bandwagon of beta-blockers”; and fee schedules can no more incentivize doctors to talk and influence one another than they can mandate genuine friendship. In health care delivery, the most important innovations take place in the spaces between policy.
In a 2006 commentary on the Agency for Healthcare Quality’s website, SUNY-Albany professor Timothy J. Hoff cautioned what happens when, in the surging era of patient safety, the language of policy increasingly fills those spaces. He lamented that “all sorts of safety culture measurements and assessments now exist. We treat the concept as if it has universality in both form and function across health care settings. We treat it holistically and distinctly, something to be developed in its entirety. Accreditation and awarding organizations come up with neat little checklists to quantify it for their reviews. For these reasons, the concept of safety culture is in danger of becoming mechanistic and rhetorical, and, as that happens, it loses some of its power to promote safe health.”
Hoff’s criticism raises a point that, in the efforts to promote quality care, should be an ongoing question: to what extent does policy have the power to promote cultural change? If we are to understand innovation as a social process, then the data-driven tools of policy offer few solutions. If the integrated care demonstrations reduced readmissions by a significant amount, was that a triumph of national policy or local culture? Even we begin to unravel those social influences, can they be superimposed onto different networks of people? We know there is no silver bullet – is there a bullet at all?
As the policy world inevitably involves itself in the socialized world of local care, it is not hard to envision this future: a system of improvement that is precise, accurate, and wrong.
And perhaps that is the unsatisfying conclusion to be reached: policy can only implement policy; it cannot implement behavior. When, on a national level, we discuss innovations for highly local and personal problems, we are really discussing something much more complex and profound: strengthening the ties between a community of professionals and their patients. And for that to work, the only policy solution is to cede governance to forces we cannot easily see, measure, or understand: people.