A New York Times article by Jane Gross highlights a costly healthcare problem: avoidable hospital readmissions, which affect one in five patients and account for $17.4 billion of the current $102.6 billion Medicare budget. (When people talk about readmissions, they're generally referring to an admission within 30 days of the previous admission for the same condition.)
When patients end up back in the hospital, it's often because their care didn't continue appropriately when they were first discharged. Patients leaving the hospital should receive detailed instructions about what they need to do to take care of themselves - ranging from what they should eat and how much they should move around to which medications they should take and what kind of medical care they should get.
There are many points where this can go wrong. Patients might not receive the appropriate instructions, or they might not understand them. They may not have caregivers who can help them follow the instructions - for instance, by bringing patients food while they remain in bed or helping them get around if their mobility is limited. If prescription drugs or doctor visits are recommended, patients' insurance status, financial resources, and location will influence their ability to comply. Once a patient arrives for the recommended follow-up appointment, the healthcare provider will need to have sufficient information about the hospitalization.
In her article, Gross focuses on one specific solution to the readmission problem - hospitalists:
By the time Djigui Keita left the hospital for home, his follow-up appointment had been scheduled. Emergency health insurance was arranged until he could apply for public assistance. He knew about changes in his medication -- his doctor had found less expensive brands at local pharmacy chains. And Mr. Keita, 35, who had passed out from dehydration, was cautioned to carry spare water bottles in the taxi he drove for a living.The hourlong briefing the home-bound patient received here at the Hospital of the University of Pennsylvania was orchestrated by a hospitalist, a member of America's fastest-growing medical specialty. Over a decade, this breed of physician-administrator has increasingly taken over the care of the hospitalized patient from overburdened family doctors with less and less time to make hospital rounds -- or, as in Mr. Keita's case, when there is no family doctor at all.
Because hospitalists are on top of everything that happens to a patient -- from entry through treatment and discharge -- they are largely credited with reducing the length of hospital stays by anywhere from 17 to 30 percent, and reducing costs by 13 to 20 percent, according to studies in The Journal of the American Medical Association. As their numbers have grown, from 800 in the 1990s to 30,000 today, medical experts have come to see hospitalists as potential leaders in the transition to the Obama administration's health care reforms, to be phased in by 2014.
... Dr. Subha Airan-Javia, Mr. Keita's hospitalist, splits her time between clinical care and designing computer programs to contain costs and manage staff workflow.
The Patient Protection and Affordable Care Act (the new healthcare law) will reduce Medicare payments to hospitals with excess hospital readmissions, so this potential loss of revenue might spur more hospitals to hire hospitalists.
The new law also includes Medicare demonstration programs, summarized by the Kaiser Family Foundation, that have the potential to start moving us toward a system that pays for quality of care, rather than just quantity of services:
- Establish a national Medicare pilot program to develop and evaluate paying a bundled payment for acute, inpatient hospital services, physician services, outpatient hospital services, and post-acute care services for an episode of care that begins three days prior to a hospitalization and spans 30 days following discharge. If the pilot program achieves stated goals of improving or not reducing quality and reducing spending, develop a plan for expanding the pilot program. (Establish pilot program by January 1, 2013; expand program, if appropriate, by January 1, 2016)
- Create the Independence at Home demonstration program to provide high-need Medicare beneficiaries with primary care services in their home and allow participating teams of health professionals to share in any savings if they reduce preventable hospitalizations, prevent hospital readmissions, improve health outcomes, improve the efficiency of care, reduce the cost of health care services, and achieve patient satisfaction. (Effective January 1, 2012)
- Establish a hospital value-based purchasing program in Medicare to pay hospitals based on performance on quality measures and extend the Medicare physician quality reporting initiative beyond 2010. (Effective October 1, 2012) Develop plans to implement value-based purchasing programs for skilled nursing facilities, home health agencies, and ambulatory surgical centers. (Reports to Congress due January 1, 2011)
If hospitals are rewarded for providing better-coordinated care, including high-quality discharge instructions, we may see fewer avoidable hospital readmissions. That'll be good for patients' health as well as Medicare finances.
- Log in to post comments
This is pure garbage. You and Jane Gross have it dead wrong. Hospital readmissions are not the hospital's fault. It has nothing to do with the patient not receiving adequate care or adequate discharge instructions. For the past 2-3 years, I've witnessed readmission after readmission, almost exclusively due to patient non-compliance, and we're talking about gross non-compliance here. Do you have any idea how many congestive heart failure patients I've seen discharged only to come back two weeks later with their urine positive for cocaine? Do you have any idea how much time is already spent talking to the patients and writing down detailed instructions about diet, activity, and medications? The vast majority of patients simply just don't care enough to follow our directions. How far would you like to take it? Forcefeed them their meds everyday? Daily urine drug screens? This is America. The patient's are free to do as they please after they leave the hospital, and believe me, they exercise it. Inadequate care? Inadequate discharge instructions? Preventable? You're clueless.
There are definitely factors that are outside the hospitalâs control, and I donât doubt that some patients will fail to follow directions no matter how much information and advice they get. I wonder, though, whether providers giving discharge instructions have enough time to discuss things in a way that could improve compliance among the rest of the patients. In my own experience with a variety of doctors, Iâve had one who spent a few extra minutes talking to me about how I could fit particular health behaviors into my life, and it did help me change my habits. Iâve had others who were so clearly overscheduled and didnât have more than five minutes to spend with me, and so I didnât ask questions and left feeling not entirely certain about what I should do. (In those cases, I was able to get the info elsewhere, but not everyone can do that.)
There are also some performance measures where not all hospitals score well. In the CMS Hospital Compare data (I accessed it through Commonwealthâs Why Not the Best site), you can see a lot of variation. For instance, at the national level, hospitals give all the recommended care (on 24 process-of-care measures) 90% of the time â not the worst record, but one that could improve. I checked the individual scores for all the DC hospitals, and they ranged from 74% to 97%. When it comes to giving complete discharge instructions to heart failure patients (activity level, diet, discharge medications, follow-up appointment, weight monitoring, and what to do if symptoms worsen), the national average is 79%, and DC hospitalsâ scores range from 22% to 100%.
So, as you say, there are many factors influencing readmission that are outside of the hospitalsâ control â but there are also factors that are within their control, and many hospitals could do a better job with those.
The attitude of Medstudent is distressing. It goes back to the old idea that the doctor is a supreme being. Behind every recalcitrant patient you will find a recalcitrant health service.
I am in IT and acutely aware of the amount of jargon used in my profession. I have had conversations with doctors where most of my questions are "What does that word mean?" The technical jargon used by doctors is great for them in communicating with other health care professionals, but for others these words have no meaning and are not shortcuts to years of training and experience. Most of us feel dumb and do not want to admit that we do not understand what is being said to us. (The old Emperor's New Clothes story)
Put it in simple layman's terms. Give us examples of what you mean. Be graphic. Tell us "If you don't do these things, you will be back in here and even more miserable." Scare the crap out of us. And I am guessing that a phone call a week afterwards would be too much to ask.
Medstudent reminds me of software developers blaming problems on "user error" when really the software's user interface needs a complete overhaul or, perhaps, more than a few bugs need to be fixed.
Maybe if Medstudent doesn't like patients, Medstudent should choose another line of work.
When my partner was hospitalized a couple of years ago she was assigned a hospitalist Two issues made me think the practice has some problems.
The first issue was payment. She was admitted through the emergency room at 6am to the one hospital she knew was a participating provider in her health plan. When the hospitalist came by at about 8pm we simply assumed, in our drugged and exhausted state, that since she identified herself as a hospital employee she was also part of the plan. Not so. She was a contractor hired by an agency which had contracted with the hospital. She should have been on the participating provider list, but wasn't. So her agency billed us for the non-participating rate, for which we hadn't satisfied the deductible, and wanted a bunch of money. It took a complaint to the consumer protection division of the State Attorney General office to clear that up - we said that if the hospitalist had paperwork problems it was the hospitalist's problem and her agency should eat the charges, not us.
The second was the conflict of interest and possibly unnecessary and expensive tests. The hospitalist ordered some painful, intrusive, and expensive test which we never did understand. The surgeon - who we thought was handling the case - simply shrugged his shouulders and said something about hare brained ideas when we asked him. The only clear thing was that it looked like the hospital department that administered the test (something involving injection of radiochemicals, I dunno) looked like it wasn't getting much business. The hospitalist, as an (indirect) employee of the hospital, has something of a financial interest in making sure that the hospital's various facilities are used, and their services reimbursed. We wondered if it was the hospital's need for money, rather than my partner's need for another (in the end, unneccesary) diagnostic test that prompted the decision and order.
I'm a pediatric hospice and palliative care nurse. Here in IL we are fighting for a bill that allows both palliative care and home health to bill public aid (currently we use grant money to pay for our services and in this economy it's rough). We definitely keep kids out of the hospital and ER with symptom management in the home. The kids in our palliative care program get home nursing and occasional visits from a social worker and chaplain. The kids all have poor prognosis and some are hospice appropriate but don't qualify because they have many hours of home health that hospice reimbursement can't pay for. All of this costs far less than ER visits and hospitalizations. The bill is currently on Gov. Quinn's desk and it looks like he will sign it.
Medstudent - glad to see you confirming my view of medical doctors - assholes, assholes, and more assholes, stacks of them, far as the eye can see.
Managing medical care
Medicine had its last great bout of reform and reorganization between about 150 and 100 years ago. By any measure, that reform was wildly successful, transforming medicine from a practical status in most people's lives akin to astrology's, to a practical necessity. But one of the effects of that reorganization, the explosive growth in practical medical knowledge, led to a need for specialization that that last wave of reform did not address, because it was not yet a problem.
Managing that whole array of specialty care is a problem now. It is arguably the problem, for both quality and cost of medical care.
People grope towards some sort of solution, some way for patients to use that vast apparatus for their benefit, rather than have it use them. While the broad outline of a solution is fairly clear -- use one set of medical experts, the hospitalist and/or the primary care provider, to ride herd on the specialists -- how that gets the patient back in control has not been worked out.
The author of this article, after talking about the rather obvious benefits of having a hospitalist coordinate the inherently complex and multi-specialty care needed in hospital, ends up talking about finding a way to reward hospitals for coordinating care better. But if we are going to put generalists in charge of the specialists, that means we need to dethrone the old source of authority. The hospital is the creature of the specialists. If we are to put the generalist in charge, we don't need to figure out how to reward hospitals for better coordinated care, we need to get the hospitals out of the way so that our payment system can reward or punish the people we're putting in charge, the generalists. Hospitals need to become mere facilities, not agents, of patient care.
As illustrated by Michael Schneider's post, we also need to take the next step of getting the patient in charge of the generalists. For starters, these generaltists would have to work for the patient, and not the hospitals. We would still have the problem of how we get the layperson, the patient, to effectively control the expert, the generalist, and not have the relationship degenerate into the patient being just that, passive, and the expert in charge. But that's the irreducible nub of the problem, a problem that we laypersons have in relation to all expert knowledge, whether your auto mechanic's, your doctor's, or that of the pundit who claims he knows what to do in Afghanistan. We need to at least simplify the problem to that nub.
Once we have the problem down to that nub, once we've decided to have a system in which the generalists control the specialists, and the patients control the generalists, we can start on strategies to make the second part of that work. The patient needs to be given a means of control. The obvious such means would be to let the patient control a sizable premium for continuity of care. If you come to trust one generalist, because that person handles the many small and few medium problems you develope over the years, you reward that practitioner by continuing to use him or her for your care, primary care and the management of specialty care for you. By the time you develope big, hospitalizable problems, you wil have had a chance to develope a basis of judging generalist expertise by process validity. A provider who has worked well with you, had a good process for handling small and medium problems in a way that doesn't overtreat or overtest, but takes the time to let you make the best, informed, decisions about how aggressive to be in handling your hay fever, is someone you will know how to deal with when it comes to the big problems, and the big decisions later in life.
The idea of the continuity premium that the patient controls is not to make primary care/generalism more remunerative, the idea is to give the generalist the ability to back off on false standards of "time efficiency" so that he or she can work on mastering the control of the whole gamut of specialty care to the service of ther individual patient. This function, managing the care of specialists, as opposed to providing the direct medical services within the scope of practice of the generalist, is not directly compensated in our present system. It is very time-consuming, but not easy to directly compensate, because when it is done well, precisely to the extent that it is done well, with the most allotment of the generalist's time and attention, it results in no treatment, no tests, no interventions of any sort. In essence, these continuity premiums the patient would control, would be payments to the generalists to do as much as possible of as little as possible in the way of invasive and expensive intervention. You can't minimize care responsibly, you can't say that a given patient with chest pain doesn't need a raft of invasive tests, without a very careful history, physical exam, and maybe some minimally invasive tests. The patient is the only one who can be left controlling this minimalist approach without creating incentives to do too little.
The place to end this line of thought is with the generalist. The next great reform and reorgnanization of medicine needs to be whatever will make the generalists ready to assume their role as the managers of medical care. In addition to the things external to medicine already mentioned -- stop using the payment scheme to keep the hospitals in charge and start paying generalists to take charge, under the control of the patients -- medicine itself has to change. Practitioners graduate from the residencies we have in place with all the skills needed to be inpatient generalists, but the longitudinal dimension, the outpatient care that lies in between, and hopefully prevents, inpatient care, is not really taught. It can't really be taught in the hospital-centric, inpatient-centric, training that is needed to get physicians-in-training their necessary level of exposure to serious pathology. We need to have a period of outpatient training added on after the current residency. These generalist fellowships would become the driving force in defining the ever-changing boundaries between primary care/hospitalism and the specialties, enabling the primary care provider to do what the patient gives him or her those hefty continuity bonuses to do right.
Medstudent has valid points. Bitter a little, maybe...but valid points nonetheless. It hurts to realize that one's fantasies about the medical field don't translate into reality. However.
The true reality is that our entire medical culture is focused on disease treatment, not illness prevention. Prevention doesn't make money for researchers, pharmaceutical companies, hospitals, insurance companies, ancillary services, etc...Disease does. And people are subconsciously encouraged to ignore their health because there is, after all, "a pill for that" that will treat their symptoms without actually curing anything, and ensure that they will inevitably need more and more treatment/drugs/hospitalizations/and so on in order to allow continued life. Not to mention the fact that many people, especially the retired folks, are under the impression that they earned all the health care they can access, and that they shouldn't have to pay for any of it. We've had fifty years of an entitlement culture, and the skyrocketing cost of health care in those fifty years is proof it doesn't work, at least how it is formed in the U.S.
It's a three fold problem: disease focused culture, lack of responsibility for one's own health, and an entitlement culture. Plus, of course, the simple fact that many many people fall through the cracks of the system due to financial constraints. There are no easy answers, and trying to blame hospitals or staff for re-admissions is just stupid. They have to follow Medicare guidelines, and they do the best they can under the constraints they have placed on them.
Believe me, I know first hand. I work in the medical field.
I agree with David Sencer. As a (pediatric) hospitalist myself, I know that there are times the patient doesn't comply. There are times it's because of depression or other infirmity, and other times it's because of inadequate prep (though these days the services are efficiently provided more times than not - medstudent is right about that.) And at times it's frustrating. But blaming the patient is the easy way out. And it certainly does nothing to fix the system.
But most times, things go well enough and the patient isn't readmitted. It certainly takes good communication and good continuity of care to make that happen.
The idea that these are 'new' issues are given short shrift in the article. I've been a hospitalist for 20 years, before the term was coined, and we do contribute to fewer readmissions and shorter lengths of stay. Many hospitals have squeezed what savings there is out of the system already. Still, it's an important concept and part of the outcomes-based medicine revolution we are starting to see unfold.
Um... didn't nurses used to do this stuff? In nursing school, we had the mantra 'patient education, patient advocate, patient education' drummed into us.
But I suspect that with nurses' increased workload - mostly in the form of increased patient acuity - it keeps them from doing the kind of education they should and that they'd like to do... if they had the time.
I also suspect that nurses' time is 'cheap' compared to a doc's time to do this kind of education. HOWEVER, you can't bill for a nurse's time the way you can for a doc's. This begs the question... is the growth of the hospitalist profession an outgrowth of hospitals looking for another way to bill for patient care and education at the higher MD rate?
Any nurses want to weigh in here?
Medstudent sees the cocaine users in his/her practice - it's unbelievably frustrating to attempt to give good health care to these patients. I had a similar patients when I was in residency. There were also mentally ill men and women who refused housing and meds, lived in the parks and returned through the ER in crisis. Later, as a family physician in a small town, there were and are the repeat admissions for alcoholic pancreatitis patients or the lady with heart failure who sneaks salt into her room.
However, while I hate the Friday and Saturday discharges (and marvel at our past history of admissions for work ups and even respite care) and see some problems with obtaining follow up appointments, I don't believe that the majority of re-admissions are due to failure of discharge instructions or inadequate follow up care. Mostly, it's the patients who have crisis after crisis due to long term, chronic illnesses. Some are near the end of life and some are not.
The fact that there are hospitalists is actually an argument against the myth of the physician doing more in the hospital in order to make more money: they exist because Family docs, general internists and pediatricians don't earn enough money with the hospital practice to justify taking extra hours out of their office practice to make hospital rounds. Overhead in the office goes on, whether the doc is there or not, and other patients can't make appointments if the doc is over at the hospital.
Back when I went into private practice (nearly 20 years ago), we were paid to be first assistant on surgeries and at least invited to witness or notified of the time for other procedures. We explained technicalities - and understood them better because we maintained our own knowledge by assisting - to the patients and their families before and after the surgery or procedure. Then, Medicare and other insurances decided we weren't medically necessary, bundled the surgeons' fees and made it more profitable for him to hire a nurse practitioner, RN, or Physician Assistant to serve as his first assistant in the OR or lab. Primary docs lost contact with the patient and his care. For a while, some primary or attending docs were paid for care of medical problems that surgeon didn't cover and then some admitted the patient and the surgeon, cardiologist or other proceduralist was the consultant. It seemed to me that just about the time we figured out the rules and regulations enough to get paid, the rules and regulations changed.
We primary docs were told that the hospitalist was more efficient for coordinating care and encouraged to stay in our offices - and we could get out of some of the tedious hospital responsibilities like committees and early morning staff meetings. I've never seen any quid pro quo for procedures in the small towns I've worked in. Everyone's up to their capacity and increased consults and tests are increased hours at work for what is essentially a flat fee (bundled fees in the case of the hospital and the proceduralist who are paid per diagnosis) or the allowed single daily charge. For instance, if I see the patient in the office or nursing home and then admit them to the hospital, or if the hospitalist or I see them more than once a day in the hospital or take the time to meet with the family, I only get paid once for each day.
Medstudent--
As a possible future patient of yours (since I don't expect to die before you finish your training), I would like to something out. The first is that many noncompliant patients are not drug abusers. In some cases, there's just no communication. I was talking to someone a couple of days ago who described an incident in her medical training. Shadowing a resident, she walked into a patient's room. The resident looked at the patient, looked at the chart, and basically shrugged. The med student went over and introduced herself to the student. The resident said "there's no point, he only speaks Russian." Because she could, she started over by introducing herself in Russian and asking the patient how he was doing.
What sort of "compliance" are you likely to get from a patient who has just spent several days seeing that his doctors cannot or will not talk to him. If he's lucky, they will send him home with some written instructions in Russian, but he's probably not going to get a chance to ask a doctor or nurse questions. Sure, the hospital has someone who translates, once in a while, but the default state, the thing the patient mostly sees, is that his doctors can't talk to him and think this is normal and not their problem.
"Compliance" is difficult, and not just for communication reasons. It's easy to say "they'd rather buy cocaine than statins," but what about the person who is choosing between dinner and statins, or baby formula or the rent and her own meds? Yelling at her isn't going to help. What about the person who won't take blood pressure medication because when she tried it, she lost 14 hours of her life, with no recollection of what happened in that time? You might get her to try a different medication. You might get her to eat less salt, take up jogging, or give up red meat. But blaming her won't achieve anything.
If a patient is using cocaine, then clearly they need medical treatment for the problem instead of a bunch of finger-wagging and blame. Shouldn't a med student realize that addiction is real and it's nearly impossible for an addict to just stop using cocaine without help? It seems to me that the bigger problem is lack of access to effective rehab programs.
And as for people who don't medications, maybe it's not that they just "don't care"; maybe they can't afford it or there is some other issue that makes it too difficult to get what they need. It's really easy to blame the patient, but much harder to understand that there are many factors involved in their non-compliance.
Written instructions are far better than oral if the patient and family are literate. I think doctors forget how much stress effects short term memory. A patient can hear every instruction and be so relieved on the way home they forget which scrip is for which symptom, etc.
It's a good idea for as patient to take a notepad to a doctor's office, but even better if they give you printouts with the labwork they want done, where to go for it, what medicines you need to take when, and any recommendations. I don't always know what I will need to refer back to later, even if I'm taking notes.
Also: doctors need to listen, maybe even prompt patients who may be nervous to explain their noncompliance. I'm on an albuterol syrup for my asthma, although all my doctors have wanted me to be on an inhaler for emergencies. But there's a reaction where the lungs close up tighter in response to the medicine (paradoxical bronchiospasm). If I hadn't done the research, I wouldn't have been able to advocate for a medicine that didn't make me sicker; and I might have meekly accepted an inhaler I knew I couldn't take because it would make me feel worse. Rare conditions happen; rare side effects happen-- and excessive respect for a doctor who denies the likelihood doesn't make the patient comply, it just makes them not argue.
Readmission is tough. There are certainly problems with adherence, but there are a lot of problems that are systems problems.
I've seen ER and floor discharge documents that are 8 pages long, with a brief sentence buried on page 5 on follow up. Most of the rest of the pages are filled with mediolegal mumbojumbo and JACHO-required admonishments.
IT departments and hosp admins can be incredibly un-creative in solving these issues, often tacking on more forms and creating discharge process that is JACHO-compliant and patient-unfriendly.
Locally, our hospitalists are reserved for the rehab and geriatric patients, the psych patients, and patients unattached to a family doc with admission privileges. I think one of our main issues with readmission is too early discharge.
It's not the fault of the hospitalists, it's the fault of the system: we have a 7-floor hospital with two floors not used. We don't have enough nurses and beds. There are always overcapacity notices begging doctors to discharge ASAP. And frail patients suffer for this.
The patients I follow up a few days after discharge often look as terrible as they did on admission. And we have *spectacular* communication with the local hospital in terms of discharge instructions and follow-up. But patients continue to bounce.
I see the frustration "medstudent" feels- before you call him/her an unmitigated asshole, just remember that this is a phase, wherein a new doctor learns that not all illness is valiantly fought by brave patients doing all they can. This too shall pass, and I suspect "medstudent," with the scales of idealism falling from his/her eyes will be a better physician for it.
Care includes recognising one's own limitations in terms of what you can and cannot accomplish. So much of my job is to be a cheerleader in convincing people that working toward better health is doable.
Let's see if Medstudent et. al. can explain my aunt's readmission. She was admitted with across the board low blood counts while under radiation therapy for cancer (I can get the numbers if you wish, but suffice to say they were well below normal). She was given a number of transfusions and released when her marrow recovered and she was able to walk out of the hospital. She looked extremely pale and her radiation doctor insisted she restart treatment ASAP. The other doctors on the case did not contest this, even though her CT scans showed no cancer spread beyond the original area. She did go back to radiation treatment within 2 weeks and landed in the hospital again within a week, with blood levels so low the staff was amazed she was still alive. She received numerous transfusions and pulled through again after two extra weeks of intensive care in the hospital. She, BTW, is doing well now, but it took a year for her to regain any strength whatsoever (I visit her often).
So arrogant genius(es), how do we fix that problem? There was a hospitalist on her case both times, BTW. Fat lot of good it did.
My view is that it's something of an intransigent problem since one-size-fits-all treatment just doesn't work unless you want to shrug your shoulders when a patient ends up getting dead due to inappropriate treatment just to make things cheap.
Reducing hospital readmission's would be a great start and would save a load of money because the doctor's already know the conditions of their patients. They don't have to go through all of the test's again most of the time. However it is still both of their faults for coming back in.
Obviously MedStudent has not been in practice enough to understand that readmission is multifactorial and not exclusively on patient non compliance. But I do sense his/her frustration. We all explain to patients what is best for them; they go home with you expecting them to follow your advice only the return doing exactly opposite of what you have told them. Readmissions vary in all shapes and sizes. Think of it as a house of cards, with each card representing different aspects for patient care (Physician, Nursing, Pharmacist, Social worker...etc and of course the Patient). If any of these "cards" fail, then the house would collapse and a possible readmission may result. Patient non-compliance is one factor and often the first one we blame, but we really need to look into why they were non compliant and how we as a healthcare team can help. Every patient is different and and we need to understand that. Sometimes I feel that the patients are so bombarded with information at the point of discharge it all goes to short term memory and when they get home they just plain forget. Patients' are getting so much paperwork on "instructions" that things get lost in the pile when they go home. There is no simple answer to readmissions or else the problem would have been solved a long time ago.