Tuesday, July 15, 2014

Admissions about readmissions

The ever-thoughtful Brad Flansbaum presents a cogent summary of what we have learned and not learned about the causes of readmissions and the cost and benefits of reducing them.  I'll summarize briefly but then turn to the ramifications for a new generation of companies that have been created to capitalize on the current readmissions fad. First, here are some excerpts from Brad's post:

Think of the drunk looking for his lost car keys under the only light post in a parking lot.  As unlikely as he might find the keys, he does use logic in his approach despite his low odds of success.

We use readmits for the same reason.  We have mediocre tools to assess outcomes, but nothing better exists.  We use readmissions because we do not have much else to employ.

If one could equate a hospital stay to a restaurant experience, readmits as part of our composite grade would rank somewhere between arranging the table china and the linen choice.

Yes, as a country, we have improved—readmits have dropped 1-2% in absolute terms the last few years.  However, we do not know precisely why, nor have confidence in what interventions we have implemented to make the strides.  

Preventable readmits do not factor much into our large health care price tag.  It is also hard to claim any positive externalities impacting other aspects of care from whatever practice changes we have adopted, given we cannot reliably identify them.

The readmit metric has a long way to go—both on the knowledge discovery front and how we as practitioners should seek wisdom in a very crude measurement.  I am still amazed at how many non-clinicians find meaning in a measure so few clinicians view as high impact.

Brad also includes a list of what we actually know about readmissions.  I include a few:

1.  Not one intervention has magic bullet status.  Most studies indicate success involves the interplay of many items, often involving people outside the hospital in the community.
2.  Medication reconciliation, patient education, family involvement, and follow-up appointments matter—but not to everyone and we do not know if we should apply them to each patient or use them selectively.
5.  Preventable readmissions remain a mystery as a percent of total readmissions.  Based on my read, the low could be 5% and the high could be as great as 20%.
6.  Most readmissions arise secondary to diagnoses unrelated to the initial one.
9.  Administrative claims data, as used by the readmission measure, correlate poorly with clinical data derived directly from medical records.

Now, let's turn to the world of commerce. I'm going to focus on one company, Dovetail, as an example.  It's not the only one, but it appears to me to be represent the attempt of investors to take advantage of the current policy with regard to financial penalties for above-average readmission rates.  The concept, at heart, is to employ pharmacists to make home visits to patients and monitor medication adherence and adjust drug usage in light of the patient's changing condition. (I'm sure the company would say that other parts of the regime are of importance, too, but my sense is that much is based on the home visit model.)

Is there anything wrong with this?  No, not clinically.  It certainly won't do any harm.  And no, not as a short term business strategy, i.e., marketing to hospitals who seek arrows in their readmissions quiver.  Especially when the hospitals can enter into contracts for such services rather than increasing their own staff count.

It is on this latter point that I suspect the business model is unsustainable.  If we review Brad's conclusions, we can see that there is no proven relationship between medication reconciliation and readmissions reductions.  As I understand it, Dovetail and others offer their service on a fee-for-service basis, i.e., so much per patient or interaction or whatever.   In the absence of rigorous direct correlations between interventions and results, hospitals will eventually make a financial judgment that the cost is not warranted compared to their other alternatives.

So, then we turn to the underlying business model of such companies.  Is their goal to have a long-term sustained business, or to exit by sale to other private equity investors?  If it is the former, the company will need to move to a fee based on risk-sharing, at which point their revenues will be more uncertain.  If it is the latter, the multiple used for the purchase price will be heavily discounted given the uncertainties set forth by Brad.  Why?  If, as Brad suggests, effectiveness cannot be demonstrated, hospitals will stop using the service altogether or will demand that prices be reduced to an unprofitable commodity level.

In short, this is a business model without a future. It offers a short-term fit for a psychological niche while hospital administrators flail for answers to the government's financial penalty program. As a sustainable part of the health care system, it is a nullity.  If it is proven to be an effective approach, hospitals will figure out that employing their own staff is more cost-effective than paying a price that includes an equity return to private businesses. 

Ironically, I'm told that many of the investors in these kinds of firms are actually doctors in hospitals.  Well, if so, that is a fact that speaks for itself.


Toni Brayer said...

From Facebook:

Excellent article and observation. Lots of brain power in a room yesterday talking about readmissions and the variables involved. You are right. It is so many things.

Anonymous said...

Brad’s article is interesting, but the base statement about struggles with attribution of results to specific interventions permeates all of healthcare. You can almost always find data to support an alternative result/theory.

Focusing on readmissions is a metric that really forces hospitals to think about and reconsider continuity of care and the handoffs from acute to ambulatory care facilities. Did the hospital solve the problem or just pass it on. It is one way to measure efficacy.

Private, venture/equity backed companies can play a huge role in the pace of innovation in healthcare. The lure of financial riches drives people/companies to develop ideas and then companies that offer new/different services. An efficient market weeds out those products/services based on performance (ideally). The best ideas then become best practice and whether they continue to be delivered via the private company or are adopted by the health system is a function of business model and its execution by the management team. Regardless, there should be some reward for taking the risk to introduce new products/capabilities into the market.

Paul Levy said...

Well put, Anon. Our only disagreement might be in the wording of the last sentence, i.e., "there should be some reward." I'd say there should be some reward if the task is executed well and if there is a proven value to the purchaser.

Barry Carol said...

As I understand it, we can’t even define what a preventable readmission is with any precision. I think hospitals should focus primarily on reducing hospital acquired infections and do more thorough discharge planning including using nurse case managers to go to the patient’s home if necessary to assess needs and arrange for appropriate support. I also don’t know to what extent nursing homes send patients back to the hospital or to the hospital in the first place unnecessarily mainly because of litigation fears.

Instead of using a blunt instrument like readmissions to penalize hospitals financially if they fall below a target metric, I think we would be better served to put hospitals at financial risk by moving more aggressively toward bundled payments for surgical procedures, capitation where appropriate and shared risk / shared savings contracts with insurers.

Tim Huntley said...


I recently joined a company that is participating in this space, and while reducing readmissions is certainly part of our value proposition, it is only a small part. In our model, we actually partner with the hospital to help them extend care to the home with hospital-employeed NP's or PA's. Our role is to help them deliver this care in a way that is efficient and cost effective (i.e. a sustainable business model).

While I agree that readmissions is a blunt instrument, we made a major impact at our first hospital last year by reducing their rate from 16% to under 8% for complex CHF patients.

Paul Levy said...

Thanks, Tim, but if it works well, why will they need you after a while?

nonlocal MD said...

I like Barry's idea. The problem with cherry picking items to incentivize is that it promotes studying to the test rather than comprehensive overhaul of care. I thought we learned that with process-based incentives like antibiotics within x hours of admission,etc.
That said, I think looking at readmissions forced hospitals to realize they do have some accountability for what goes on after the patient leaves the hospital. Too many want to say 'that's not my job.'

Tim Huntley said...

"if it works well, why will they need you after a while?"

Along with our service, we provide quite a bit of technology. And without this technology, it is unlikely that the hospital will see the same results. In fact, the second hospital we are working with had attempted to do provider-level home visits with almost no impact.

Paul Levy said...


Interesting question as to whether it is their job. That's a very hospital-centric view of the world.

For example, what if patient is at a skilled nursing or rehab facility with no common electronic health record to that of the hospital? Does the hospital have an obligation to oversee the care given by another institution? Should it be held responsible for the actions of that institution?

nonlocal MD said...

No Paul; I was not aspiring to be that specific. For now, I meant it in more general terms such as thinking ahead to what the post-hospital situation will be like for the patient and shaping discharge instructions, medication prescribing, etc. accordingly. Physicians in particular often do not think about that, leaving family members to try to fill the gaps as best they can.
But yes I do believe that eventually someone will have to be assigned responsibility to fill those gaps, even the situations you describe. The care should be seamless. Right now it is perfect for the individual silos to point fingers at each other when something goes wrong. The buck needs to stop at one place, and everyone should recognize it. Anything less is not really patient centered, but puts the patient and family in the position of serving the institutions by being the default coordinator.

nonlocal MD said...

And ps, I couldn't care less about the excuse that the EHR's are not interoperable. If hospitals really wanted to they could put incredible pressure on the IT industry to make it so. They don't want to.