Are 30-Day Readmissions Rates a Reliable Indicator for Poor Healthcare Delivery?  (Part 2 of 2)

A look at poor healthcare delivery through hospital readmission rates.

Obamacare (Patient Protection and Affordable Care Act) has provisions that require the Centers for Medicare and Medicaid Services (CMS) to financially penalize hospitals that have unacceptably high 30-day readmission rates for Medicare and Medicaid patients. Institutions with high 30-day readmission rates in just a handful of situations will suffer financial penalties for ALL Medicare and Medicaid charges during the following fiscal year – not just the handful that are monitored.  Specifically, the CMS tracks 30-day readmission rates for[1]:

  • Heart failure
  • Heart attack
  • Pneumonia
  • Chronic lung problems (emphysema and bronchitis)
  • Elective knee and hip replacements

The penalties can be as much as 3% of all the Obamacare charges for the coming fiscal year. In most organizations, this can easily amount to millions of dollars. In larger institutions, this can amount to tens of millions of dollars.

The rationale behind this policy is that high 30-day readmission rates are a reliable sign of poor healthcare delivery. The idea is that if the hospitals had done a good job in the first place, patients wouldn’t need to come back so soon.

Part 2 of this blog argues that 30-day readmission rates are not a good metric for assessing the quality of care.  I fully recognize that my position is incompatible with the current wisdom, but I’ll give several reasons to support this position.  I suspect there are many others who feel the same way but haven’t argued their position.

The most striking issue that occurs to me is that 25% of all hospitals will automatically be classified as “losers” regardless of the reasons for their high readmission rates. This automatic and simple-minded categorization is grossly unfair.

 

Hospitals Are Only One Component in a Complex Healthcare Web

Hospitals are highly visible nodes in a complex web of healthcare delivery.  Other components include general practitioners, medical and surgical specialists, independent laboratories, the social welfare system, and family support among others. Unfortunately, it is not unusual for the elderly to have no family support. Everyone knows healthcare is a highly fragmented and fragile system. Failure in any component of this web can lead to readmissions. Nevertheless, Medicare and Medicaid (and perhaps society at large) hold hospitals solely accountable for readmissions.

Given the complexity of the healthcare web, it is highly unfair to single out hospitals as culprits when many of the factors affecting readmissions are beyond the control of hospitals.

Readmissions are a function of hospital care and discharge planning.  That is true.  But it is not the full story.  Another factor that impacts readmissions is the severity of the illnesses treated; those with severe illnesses are more likely to be readmitted. Hospitals can lower their readmission rates by declining to treat patients with severe illnesses.  I know that this is gaming the system, but it makes the metrics look good.

In some communities, elderly patients are discharged into the care of loving, stable, supportive families. In other communities, elderly patients go back to a bleak room in solitude. When they need help – even a ride to see their GP – there is no one to turn to. In other cases, elderly patients may live with their children.  But their children often have jobs and lives of their own. Although they are available to give help sometimes, they are simply not available to help all the time.

At discharge, hospitals routinely advise patients to schedule follow up appointments with their GPs. Patients promise to do so – but often don’t. In some cases, they don’t have GPs to call.  In other cases, they simply forget to make the appointments.  Sometimes they try to schedule an appointment but cannot get one for a month or more. Then there are the patients who simply don’t have access to transportation to get to their appointments.

Discharge staff generally give extensive instructions to patients about their medications, diet, exercise, etc. But it is not unusual for patients to fail to understand these instructions. Or they understand but they don’t have the money to buy the medications. Or they have the money for their medicines but they forget to take them.

There are any number of points of failure and many of them are beyond the hospital control – but hospitals take the hit for readmissions.

 

Race and Minority Status Are Correlated with Readmission Rates[2]

Blacks and Hispanics have higher rates or readmission to hospitals than whites. Many of these readmissions are avoidable. This means that hospitals serving Black and Hispanic populations are doomed to look bad on their readmission stats. There is no justice in this.

Why is race and ethnic background so important in determining readmissions?  Well, for one thing, the research shows they are less likely to schedule follow-up visits with their GPs or ongoing care givers. They are also less likely to even have GPs and, therefore, are more likely to rely on their local hospitals. Many new immigrants don’t have adequate proficiency in the English language to understand their discharge instructions or read and understand the written materials their hospitals give them. Unlike whites, they have no experience in taking the initiative to look after their own health; they often take the position that whatever happens to them is beyond their control. Some don’t trust Western medicine and discount what they are told.

These demographics suffer more anxiety and depression than whites. These mental health issues contribute to the likelihood of readmissions.

These demographics often have co-morbidities. In other words, they often have several problems at the same time.  If patients don’t bring their other problems to the attention of hospital staff – or if hospital staff fail to stumble across them – those problems can pop up after discharge and trigger other, but unrelated readmissions.

The factors listed here are not due to unsubstantiated biases but to solid research funded by the Centres for Medicare and Medicaid Services and conducted by the The Disparities Solutions Center, Mongan Institute for Health Policy, Massachusetts General Hospital. Yet, even with this solid research, well-known in the healthcare community, hospitals serving these disadvantaged populations are held responsible for readmission rates beyond their control.

 

Readmission Rates Vary by Geography and No One Knows Why[3]

In Part 1 of this blog, I showed a map of the readmission rates across the country. Now there are two interesting points about those maps.  The first is that the maps remain unchanged year after year. This means that the geographic-based dynamics are consistent year over year.

 

The other interesting point is that the underlying health profile across these geographic regions is essentially the same.  In other words, the factors that drive readmission rates are not tied to differences in the health of the general population on a regional basis.  There are other drivers, but those drivers are not well understood.

 

We Think We Know the Answers; Not Sure We Do

The experts are in general agreement about how to reduce readmission rates. Surprisingly, only very few of the hospitals that adopt the recommended practices actually see reductions in readmission rates!  This is counterintuitive.

The four generally recognized ways to reduce readmissions are:

  • Improve discharge management with follow-up
  • Patient coaching
  • Disease/health management
  • Telehealth services

Unfortunately, the evidence shows that these common-sense techniques do NOT generally lead to lower readmissions.  The research is consistent on this finding in both community hospitals as well as teaching and research hospitals. What the data for a study CMS conducted looking at changes in readmission rates during 2008 to 2010 showed is that reductions in readmission rates are slow and inconsistent.

 

Do You Like to Play Whack-A-Mole?

As a boy, I remember going to the country fairs in August and playing Whack-A-Mole. Some of you may know the game.  The game has a board with about a dozen holes cut into it. “Moles” would pop out of the woodwork at random times; I never knew when and where the next one would pop out. My job was to hit the mole on the head with a mallet.  I often missed.

In some respects, taking steps to reduce 30-day readmission rates reminds me of playing Whack-A-Mole – although it shouldn’t. It seems that even though we know what we should do to reduce readmission rates, doing the “right thing” rarely leads to the desired outcome. To the extent this is true, it suggests that we don’t understand the underlying problem or that we don’t know how to address the problem.

 

Here Are the Best Ways to Reduce Readmission Rates

The best way to reduce readmission rates is to only accept patients who are not very sick in the first place. These folks can be patched up fairly quickly and put back on the street with a much lower chance of being readmitted.

Another technique is to reduce the overall intensity of healthcare delivery.  One would think that intensive levels of healthcare would lead to healthier populations. That, in turn, would lead to lower rates of readmission. Not true.

A third technique is to change the regional practices of hospital site care.  In some areas, patients are more likely to go to a hospital for initial care rather than a local clinic or a GP. In those cases, readmission rates are higher. If we could discourage patients from using hospitals as their primary source of healthcare, we could reduce readmission rates.

We also need to change the financial incentives. Hospitals that are given the choice between leaving a bed empty and losing the revenue or readmitting a patient and increasing its readmission counts will rarely pass up the opportunity to earn a dollar today.

Experience also shows that taking steps to reduce readmissions in only one area (e.g., better discharge planning) has little impact. But if steps are taken in a number of mutually reinforcing areas, the hospital will see better results.

 

So, What Does It All Mean?

So, what’s the “take away” from all this?  Well, the first thing that occurs to me is that this is a very complex problem that we don’t seem to understand well in spite of the focus it has received.

Second, we should not hold hospitals accountable for outcomes they cannot control.  We need system-wide changes, not simply improved hospital procedures.

Third, even teaching and research hospitals – where we presumably find the best-of-the-best in healthcare – have not shown significant improvements in spite of their efforts.

Fourth, readmission rates vary geographically but change very little over time for any given geography.  That means there are forces at play we have not yet identified.

Fifth, racial and ethnic minorities have higher rates of hospital readmissions. These demographics have lower levels of trust in the “system,” take less personal responsibility for their health, have lower levels of health literacy, and suffer from higher rates of mental illness.

Sixth, 30 days is an arbitrary time frame.  It’s even possible that hospitals that focus on reducing 30-day readmissions will create unexpected negative consequences in other parts of the delivery system – although no research has substantiated this fear.

 

Read Part 1 HERE

 

[1] A Guide to Medicare’s Readmissions Penalties and Data, https://khn.org/news/a-guide-to-medicare-readmissions-penalties-and-data/

[2] Guide to Preventing Readmissions Among Racially and Ethnically Diverse Medicare Beneficiaries,

https://www.cms.gov/About-CMS/Agency-Information/OMH/Downloads/OMH_Readmissions_Guide.pdf

[3] The Revolving Door: A Report on U.S. Hospital Readmissions,

https://www.rwjf.org/content/dam/farm/reports/reports/2013/rwjf404178