Robert Chang writes…
Stephen Jencks, Mark Williams and Eric Coleman recently published what some rightfully consider a landmark article in NEJM 2009 that discusses the readmission rates in the Medicare population. Most of us have some awareness that the readmission rate is high although I doubt most would have empirically said 1 in 5.
Some significant proportion of these readmissions is preventable. The Society of Hospital Medicine has collected a large number of resources in Project BOOST and hospitalists in our institution are eagerly participating. We all can recollect patients that “if only”…perhaps they wouldn’t have had to return to the hospital. Medication reconciliation, patient literacy and a host of other factors that fill our daily work lives…the story of preventable readmissions syncs with our experience in daily care.
But there are portions of the readmission story that do not sync with a lump categorization of “one readmission is like another”. We sometimes walk away from a day of work knowing that someone is coming back, no matter what we do, for medical, social, and logistical reasons that are sometimes (even often) beyond our control.
Here are some patients I worry about falling into the lump categorization:
1. “We can’t fix you” – We are far too enamored with our ability to cure. The reason COPD, CHF and psychosis have the highest readmission rates of identical cause as the index admission is that we cannot yet fix some of these medical problems or the fix is to control their environment in a way that they cannot or will not. Even an expertly managed failing heart still leads to a basal readmission rate. By definition, hospitalization cannot eliminate readmission. It was never intended to.
2. “Don’t fix me” – corollary to “we can’t fix you” is this is a miniscule but demoralizing group of people that come in who have no desire to be fixed. The attention, the medications and the environment are too good. “Good care” isn’t the problem for these patients – that’s what attracts them. We have no societal answer for this group of patients. Medicare fraud applies to physicians (top 20 Google hits all speak of physician fraud), not patients.
3. The “good catch” – we want people to come back if they are sick. The very reason we schedule a follow-up visit in this “less than 100% follow-up visit for all discharged patient” state is that we are worried about someone. To be readmitted is a good thing in this case.
4. The “sojourner” – we all know patients that move hospital to hospital (but we probably just don’t know that they do) for mental disability, illicit or abused substances like alcohol, or secondary gain. An article on habitually wandering patients conduced in 1994 in the VA system is a good example, found here. They find a mix of the people we can’t fix and people who don’t want to be fixed.
I’m sure you can come up with more. The point I bring to the table is not that the cause of reducing readmissions is hopeless or unworthy of our attention. We must participate in this venture and we must find ways to address reversible issues. My point is that our local leadership is incredibly important and one of the key measures we need to promote as leaders is appropriate expectations. Jencks et al does not state that the readmission rate should be zero but the quoted sum of $17.6 billion dollars spent on readmissions is derived from the elimination of all unplanned readmissions. In the “we can’t fix you” patients, that is an impossible expectation to meet and in the “good catch”, it is not an expectation that we would want to meet.
As our hospitals and hospitalists start viewing this issue with a keener and keener eye, we have to balance the splintering of an issue into fragments that no longer mean anything and lumping things together into an undifferentiated blob that doesn’t actually inform. But then again, that’s what hospitalists do best.
This article really hits the nail on the head. It is beautifully written by the authors. I sometimes feel guilty when patients that I worked extremely hard to care for and stabilized are back to the hospital sooner than later, sometimes for the same reason and sometimes for something totally different.
For those who are genuinely sick there is nothing we can do about them but for those who go from one hospital to the other, there should be a genuine patient fraud set up to check these abuses. I think this will reduce our readmission rate to an extent.
Benedict,
Thanks for your comment. I do think that we can and should do something about the majority of them but I also worry that we’ve set ourselves up for failure if we don’t think carefully about what we are trying to achieve. I think this relates to your first paragraph.
An underlying point which I think also struck a chord with you is that we are also affected individually and personally by these re-admissions. I think there are few physicians that could care less that someone they took care of comes back – we want to know why and learn from it! I am not practicing in a system built in such a way that I can learn from those cases without receiving some sort of penalty – be it self-imposed, financial or social. Do other readers have such a system in place?
Third and last, the fraud cases live in the realm of social policy and expectations. If someone visits the hospital 100 times for chest pain, is society willing to pay for that to catch the MI on 101st visit? What if the patient did not take their medications regularly? What if they were not paying for their medications to allow their children to eat? What if someone stole their car and they were working three stressful jobs to get another car to then get to the pharmacy and their doctor visits? And so on. Is our society able to hold a reasonable discourse on this issue? Have we prepared them to?
Ultimately, the physician and hospital staff bear the brunt of the responsibility in these cases, not the patient. “Due diligence” rules the day. There is no clear penalty for overuse, especially if you as a patient are bankrupt. There is a fairly severe penalty if you as a physician miss a diagnosis even within the context of someone crying wolf. It’s a question of how we promote social responsibility and whether we are solely responsible.
This post was really intended to set the background. The next post I make will flesh out a bit more of the responsibility of approaching these types of issues and that in leading in these areas, both careful consideration and appropriate expectations are key. I’ll include another story about, of all things…fighting over computer access related to expectations.
These are essential problems that need solutions. A basic tenet of human nature is that anything obtained for free has no value. People must be invested in some way for that value to be achieved. A simple solution is a co-pay at the ER door for all visits, for children, for immigrants, for everyone. Why not? Everyone has money for the cell phone, the cigarettes, the beer, you name it. Of course this will happen along with tort reform and the return of clinical medicine without every patient having advanced scanning and troponins.
Dr. Chang,
Thank you for an outstanding article about a central issue and challenge we face. I have learned not to view “bounce-backs” as personal failures but rather as learning opportunities, or sometimes as lessons in humility.
I look forward to your future posts on this and other topics.
Floyd and Lenny,
For some reason, I was just notified of your posts yesterday. Thanks for your comments and sorry for the delay in replying!
I think there is a lot to be said for “anything obtained for free has no value”. When working overseas, that’s a basic tenant. Free services do not really give patients a sense of value about the care they receive – it becomes an expectation rather than something they contribute to, participate in and appreciate.
Whether up front payment for all services is the mechanism to implement this is a serious consideration, but complicated within our third-party payment. Current laws also prevent turning away patients from a medical evaluation when they present to the ED. I think that’s an absolutely beautiful thing about our country that we should preserve, but the cost is a patient who comes to the ED 47 times in 2 months at my hospital (among other patients).
Lenny, thanks for your feedback and perspective. How we translate “systems errors” on a policy level down to us, the individual physician, is one of the very tangible issues this paper brings forth.