Clinical Integration 2015: The Hospital Perspective
A recent article by John Morrissey lists four different stages of clinical integration and reports on three different organizational experiences in trying to achieve the goals. He frames the issue of clinical integration thusly:
1. Hospitals and physicians can’t thrive in the new health care environment alone.
2. The population health priority is a driver of clinical integration.
3. Clinical integration links physicians, acute care hospitals and service for chronic
care and post-acute care conditions.
4. Achieving integration can happen in different ways at different organizations.
He starts his discussion by pointing out that organizations must move forward legally. There is a wide assumption in the industry that the Affordable Care Act makes clinical integration legal, but as the recent case from Idaho makes clear, the Federal Trade Commission is still committed to enforcing anti-trust laws in the name of consumer protection. While the notion of effective competition in healthcare is problematic, the possibility of dealing with such legal actions is a major obstacle.
It is true that traditional hospitals “can’t thrive” in the new health care environment alone, but it is not as clear that physicians can’t. One problem we have encountered locally in thinking about clinical integration is the difference in motivations. The hospital sees clinical integration as a way to maintain its central place in the health care hierarchy and to maintain its financial integrity. Physicians may see integration as a way to get paid for work not directly related to patient care, but usually get committed only if they see it as a way to improve the care of their patients. Physicians in primary care, on the other hand, don’t expect financial reward, and in many places can opt out of any onerous payment arrangements. The growth in different payment structures for primary care reflect the growing divergence in the financial interests of primary care physicians from hospitals and physicians who use hospital services regularly.
The notion that population health drives integration is also problematic. Even in cases where physicians are committed to various structures such as the primary care medical home, which emphasize seamless care of individuals over time, the focus is on the care of individual groups of patients, not populations of patients. There is also a paucity of evidence that hospitals have any particular insight or expertise into improving or managing the health care needs of populations, as witnessed by the boom in organizations offering consulting services in this area. Further, there is, as yet, no defined mechanism for paying for the effort. While physicians may receive incentives for reducing hospital admissions in a medical home payment structure, hospitals are not being paid to reduce hospital admissions. And finally, as I have noted in other articles, there is little insight into how to motivate people to take steps to improve their health before they get sick.
So is clinical integration the latest fad that is destined to fail? I hope not, but I would suggest some alternate ways of thinking about it. Perhaps the biggest problem from the hospital perspective is to overcome the notion that hospitalization is an isolated event, rather than an “episode of care.” What does that mean? For healthy people, going to the hospital is likely to occur for having a baby or suffering a traumatic injury, as the days when healthy people who become acutely ill and are admitted to the hospital are largely past. The patient who was previously healthy who presents with his acute myocardial infarction has actually had his first episode of care, not, unless he dies acutely, his only episode of care.
I fear this seems like a semantic distinction without a difference, so let me take that example of the middle aged man with his first myocardial infarction. Presently, all of the pathways used by the hospital and its physicians, all of the quality metrics, and all of the computerized documentation are based on the assumption that this is an isolated event. Once the patient is discharged, the event is over. Yet it is clear that delaying (probably not preventing) the next admission for coronary artery disease means getting him to stop smoking, lose weight, take his statins, exercise, and keep his blood pressure under control. All of these fall under the rubric of the chronic care model, and are increasingly the business of the primary care doctor. Over time, the patient, and perhaps the physician, will forget the hospital admission and fall back into the old habits. In a sense, I am arguing for seeing the hospital admission from the medical view of chronic care of the patient, not the surgical “one and done” model that currently prevails.
In my hospital, a majority of patients on the medicine services have already been hospitalized at least once in the previous year for the same problem, and some, unfortunately, become “frequent flyers.” We, like most hospitals, have computerized care plans that can be activated by a single mouse click, but these plans are designed using the surgical model of admissions, so a patient can end up having the same series of tests done over and over to no clinical purpose, because it is easier to click once and move on than to unclick all the irrelevant items.
Recently I was talking to one of my hospitalist associates who is in a leadership position. He told me the story about “Mrs. Smith” who was admitted every few weeks because of shortness of breath. Since he knew her well, he said he would try to sneak a look at her in the ER without her seeing him. If she had her makeup on, he would not admit her; if she didn’t, he knew she was going to have to be admitted for a few days for a “tune up.” I suggested what was needed was a PATIENT-specific care plan, perhaps labeled Mrs. Smith’s dyspnea care plan, that was attached to her name. In fact, if he and his colleagues developed such care plans for a small number of patients, significant time and money savings could accrue. He was intrigued with the idea.
I approached the CMIO with the notion, and he instantly saw the value of it, but admitted he did not know if it could be done. However, he agreed with me it ought to be done, and is committed to trying to find a way to do it. Since we use one of the largest hospital EMR systems, the fact that it has not already been built into the system shows the constraining effect of seeing each admission in isolation. (It is a single billing event, but that does not mean we should see it that way clinically.)
A related problem is the difficulty of carrying forward from one admission to another relevant patient-specific information. One of my goals as chief of staff was to improve our organizational approach to end-of-life care. The attorneys objected to the notion that “DNR” orders should carry forward from one admission to the next, and it turned out the EMR did not have any way to do it anyway. The net effect was that patients for whom appropriate decision making had been done on the first admission were being subjected to unwanted care on the second, because the new hospitalist team did not know, or have the ability to find, the relevant documentation. With great effort, the assistant CMO was able to get the system to at least provide a prompt that patient was DNR during the previous admission. (But the lawyers still expect the busy hospitalist to reassert this with each admission, because, again, they see each admission in isolation.)
Given the potential legal obstacles, not to mention the competing financial and organizational interests involved in clinical integration, I suggest we focus more on the “transitions of care.” These do not require organizational structures of questionable legality, they only require organizations to focus on the needs of individual patients who come into and out of the boundaries of the organization’s care systems. The buzz phrase “population health” should not obscure the fact that we still have to care for the sick patients one at a time.
Physicians may also need to adjust their thinking toward a longitudinal perspective. For most of my years in practice, the organizing notion was the “Chief Complaint.” Getting away from that perspective is underway, but far from complete. I will consider this in a future article.
14 February 2015
 Morrissey J. Three Different Paths to Clinical Integration. Hospitals & Health Networks, January 2015, pp. 38-41.
 Saint Alphonsus Medical Center Nampa, et. al., v. St. Luke’s Health System. Case 1:12-cv-00560-BLW Document 463 Filed 01/24/14. Order located at http://www.ftc.gov/sites/default/files/documents/cases/140124stlukesmemodo.pdf
Challenges to Achieving the IOM Attributes of a High-Quality Healthcare System
Most people agree the Institute of Medicine's (IOM) description of the attributes of a high quality healthcare system are appropriate and worthy of trying to achieve. But it has turned out to be quite difficult to make progress.
The Practicing Physician and Medicare
What Hospitals Are Doing
How are hospitals and health systems responding to change? An AHA survey provides some insights, but suggests few are really working to improve the function and resiliency of their teams, and are thus likely to fail in attaining their strategic objectives.