Friday, April 1, 2016

Why have re-admission rates fallen?

The Affordable Care Act ("Obama care") began imposing penalties, and it seems as if hospitals have responded.  Mother Jones is skeptical about the results, as this would represent an almost perfect response to the ACA penalties:
the chart is almost too perfect. For four years the readmission rate is dead stable. Then, in a single month between December 2010 and January 2011 it suddenly drops by a full percentage point, and continues dropping for two years. This decline started about eight months after the passage of Obamacare, and it's hard to believe that hospitals could react that quickly. 
Then, the very instant that penalties begin for high readmission rates, everything stabilizes again. Apparently America's hospitals unanimously decided that once they'd hit a certain level, that was good enough and they wouldn't bother trying to improve even more.
I can think of several explanations, e.g., it may be that:
  • Hospitals are refusing admission to particularly sick patients, likely to get re-admitted; 
  • Patients are getting re-admitted to other hospitals; 
  • Hospitals are giving patients prophylactic antibiotics,  which reduces re-admits, but also creates new antibiotic-resistant superbugs.  
  • Hospitals are steering them to clinics and outpatient services, (the "good" outcome).
I would love to hear from former students (pls post in comment section) as to what they think is causing the change.  

9 comments:

  1. I can only speak from my experience; however, we created a triage clinic--essentially a "walk-in" clinic for our patient population to be seen with any urgent issue after discharge. This has cut down on both ED visits and re-admissions.

    ReplyDelete
  2. Here are some explanations that were e mailed to me:

    1. I think you see a change in the definition of readmission and a gaming of the system acounting for the decrease, e.g., hospitals could be reclassifying re-admits as a “23 hour observation”

    2. By offering better alternatives to hospital re-admission, e.g., Heather's "triage clinic" above.

    3. The Hawthorne effect--that people modify behavior only because they are being observed.

    ReplyDelete
  3. I particularly liked this comment on the government focusing on outcomes that are easy to measure, regardless of whether they are good metrics:

    CMS is not alone in preferring to focus on things that are highly visible and easy to measure. Readmission rates fill the bill. But as one who reviews and analyzes them daily, I think readmissions are a lousy quality measure and keep us from focusing on more important issues of care. But, penalties talk and hospitals are forced to listen so we’ll keep doing what we can to reduce readmissions until we have the incentive to devote that time to more important issues of quality care.

    ReplyDelete
  4. - Regarding the possibility that hospitals are refusing admission to particularly sick patients, who are likely to get re-admitted…EMTALA does discourage this behavior.
    The Emergency Medical and Treatment Labor Act (EMTALA) requires that:
    a. A hospital may not delay an appropriate medical screening examination, or further examination or treatment, to inquire about the patient’s payment method or insurance status.
    b. hospitals report the receipt of patients whom it has reason to believe may have been transferred inappropriately (under section 1867 of the act).
    (source: https://www.cms.gov/Regulations-and-Guidance/Legislation/EMTALA/)

    - Prophylactic anti-biotics is no longer an easy option…there are already so many resistant organisms. This decrease in readmission trend could not be attributed to over prescribing alone.

    ReplyDelete
  5. Early in my career, I was assigned to be the Commercial liaison on a Total Quality Management (TQM) team that was created to help improve the operations of the steel company I was working for at the time. At the time, I didn’t have much experience with the “scientific” side of business management but I entered into this assignment with hopefulness and an open mind. I took two important things away from that assignment that I still rely on today. The first thing I learned is that there is a tremendous amount of inter-dependency within an organization and the whole is really stronger than the sum of its parts. The most important thing I took away from this assignment came in the form of a book the TQM consultant gave me called, How to Lie with Statistics by written by Darrell Huff in 1954. This book taught me a very important lesson about analysis and reporting; whether or not the results support your position, you must question the results to make sure that they are not “slanted” toward a particular outcome.

    The Mother Jones article did a good job of questioning the timing of change in the re-admission data for Medicare patients. The last chapter of Huff’s book is titled, “How to Talk Back to Statistics”, instructs the reader on how to identify good, useable data, as well as identifying “murky” data (e.g., re-admissions data). The advice contained in the chapter is practical and supported by common sense. Some of the questions that should be used to “test” the results are:
    • How does the person know?
    • Is there anything missing?
    • Has the subject changed/
    • Do the results make sense
    These are the questions, I am sure, through the mind of the author of the Mother Jones story. This is important not only to “out” the truth but to understand the situation and what needs to be done to correct it. After thirty years of life in Corporate America, I’m still amazed that there are managers that do not want to understand the truth about their own organization so that improvements can be made to support long-term profitability and sustainability.

    One of my favorite examples of how important it is to identify the “lies” within the statistics is a Dilbert cartoon strip titled, “Our Products are Killing Our Customers”. In this cartoon strip, Dilbert’s dilemma is how to present this problem to Senior Management with a “positive spin”. After some consideration, Dilbert makes a presentation to Senior Management with an opening slide that states, “Good news! We have fewer dissatisfied customers!” This example reminds me not to be a “Dilbert” when presenting the data and not to fall into the trap of wanting to hear only good news.

    References
    Huff, D. and Geis, I (1954). How to Lie with Statistics. New York, NY. W.W. Norton and Company, Inc.

    ReplyDelete
  6. Early in my career, I was assigned to be the Commercial liaison on a Total Quality Management (TQM) team that was created to help improve the operations of the steel company I was working for at the time. At the time, I didn’t have much experience with the “scientific” side of business management but I entered into this assignment with hopefulness and an open mind. I took two important things away from that assignment that I still rely on today. The first thing I learned is that there is a tremendous amount of inter-dependency within an organization and the whole is really stronger than the sum of its parts. The most important thing I took away from this assignment came in the form of a book the TQM consultant gave me called, How to Lie with Statistics by written by Darrell Huff in 1954. This book taught me a very important lesson about analysis and reporting; whether or not the results support your position, you must question the results to make sure that they are not “slanted” toward a particular outcome.

    The Mother Jones article did a good job of questioning the timing of change in the re-admission data for Medicare patients. The last chapter of Huff’s book is titled, “How to Talk Back to Statistics”, instructs the reader on how to identify good, useable data, as well as identifying “murky” data (e.g., re-admissions data). The advice contained in the chapter is practical and supported by common sense. Some of the questions that should be used to “test” the results are:
    • How does the person know?
    • Is there anything missing?
    • Has the subject changed/
    • Do the results make sense
    These are the questions, I am sure, through the mind of the author of the Mother Jones story. This is important not only to “out” the truth but to understand the situation and what needs to be done to correct it. After thirty years of life in Corporate America, I’m still amazed that there are managers that do not want to understand the truth about their own organization so that improvements can be made to support long-term profitability and sustainability.

    One of my favorite examples of how important it is to identify the “lies” within the statistics is a Dilbert cartoon strip titled, “Our Products are Killing Our Customers”. In this cartoon strip, Dilbert’s dilemma is how to present this problem to Senior Management with a “positive spin”. After some consideration, Dilbert makes a presentation to Senior Management with an opening slide that states, “Good news! We have fewer dissatisfied customers!” This example reminds me not to be a “Dilbert” when presenting the data and not to fall into the trap of wanting to hear only good news.

    References
    Huff, D. and Geis, I (1954). How to Lie with Statistics. New York, NY. W.W. Norton and Company, Inc.

    ReplyDelete
  7. If patients are admitted to other hospitals, penalties still apply.

    If a patient is being readmitted, they are often placed in observation status with hopes of correcting problems quickly. This does not count towards readmission rates.

    I do think that hospitals are contacting patients more frequently following discharge. Our hospitalist group has nurse in the office in charge of followup with patients. We have also established a goal of 5-7 day followup with providers. This has always been stated but now outpatient providers are able to bill post-discharge services and this has encouraged much more timely followup.

    ReplyDelete
    Replies
    1. Chris,
      You touched on a topic that came to my mind when reading this. Observation status is a good alternative rather than readmission. The ultimate goal should be to make patients healthy in their first visit so they don't need to come back. This will help with patients well being and costs.

      In the event that a patient might come back, they can be under observation which they can receive similar care at a lower cost. There are also cons to observation because if a medicare patient needs to move to a skilled nursing facility, they may need to have been in the hospital as an inpatient status for three days, none of which can be in observation status. Some insurances question whether patients need to be inpatient, or can they be in observation status? This may create problems when it comes to billing.

      While the readmission rates are declining, I believe the necessary follow up doctors visits, follow up calls, and out patient visits attribute to this. Better quality care, and follow up on the patients well-being makes for quicker recoveries, healthier people and less money spent on inpatient stays and readmission.

      Delete
  8. I think the change is due to a requirement that holds physicians and hospitals to a higher standard of care in order to assume the responsibility of patient outcomes due to the high costs / reimbursements of hospital billing in the past. The new quality metrics was formulated to save governmental agencies monies that can be better used to incentivise physicians and hospitals by rewarding high performing hospitals and doctors with monies saved from the avoidance of hosptial readmissions for the same illness.
    This can be achieved by identifying high risk patients and preparing patients at discharge by making follow up appointments with their PCP and providing a quality hone care service if warranted. Improving the medication reconciliation process so that patients are receiving the correct medications are the first step to ensure quality outcomes.
    According to AMED news "Experts say strategies for reducing readmissions amount to two deceptively simple objectives: Make sure patients understand how to care for themselves when they leave the hospital, and make sure they get the follow-up medical attention they need to keep their conditions under control."
    Experts also agree that Preventing readmissions -- even among patients at the highest risk of rehospitalization -- is relatively straightforward.
    Patients most likely to return to hospitals are Cardiac patients who were not appropriately assessed in order to provide a plan of self managed care techniques and or were not given an appointment for follow up care with their PCP within 48 hours of discharge.
    Then there is the small percentage of patients who do not follow up with their physicians and follow the plan of care. However, as long as hospitals and physicians are correctly documenting their records with these performed actions (which are viewed by Medicare and insurance companies via EHR /EMR/ Paper files) they cannot be held responsible for the outcomes of non participating patients who are not following the provisions.

    Reference: Reducing readmissions: How 3 hospitals found success. amednews.com Retrieved from: http://www.amednews.com/article/20110207/profession/302079939/4/

    ReplyDelete