Getting it right on readmissions
Getting it right on readmissions
What didn't work is helping discover what does
If you say it out loud, people will agree intuitively: You can learn more from your failures than from your successes. But that doesn't mean people want to trumpet what doesn't work. That makes what's happening at Henry Ford Health System in Detroit so special: They are actively looking at what is still wrong as they try to get a handle on unplanned readmissions so that they can figure out what's right.
Long before payers started saying they'd stop paying for unplanned 30-day readmissions, the leadership at Henry Ford Health System decided to look at the topic, says Beth Anctil, RN, MSN, Director for Care Coordination for the system. "Our CEO thought we should tighten up our processes for discharging patients," she says, noting that over four years, the rate hasn't changed — or at least not enough, or not for long enough. "It will be down for three or four months, and then it pops back up again."
Anctil says she gets that it is complex — part is culture and getting people to change the way they do things. Part of it is that every patient who bounces back is different and figuring out which variable led to the readmission is hard. "We study it and it seems like everything we do is more complex than we thought initially."
Currently, there is a bundle that is considered standard of care, based on Institute of Healthcare Improvement suggestions. This includes doing a risk assessment for readmission and flagging the charts of patients considered at risk; providing education for patient and care provider; doing a medication reconciliation and consultation; ensuring a follow-up appointment within a specified time period; and providing discharge instructions and a summary to the patient and next provider.
"It's not rocket science, but it is different than what we did before, and there are barriers to it at every turn," she says, noting that she is trying to implement this in five hospitals, including two community hospitals and one medical-group based hospital. "I don't have the same control at the community hospitals I do with the staff model hospital."
What has been particularly difficult to get right is ensuring that patients have a timely follow-up appointment with their community care provider. "There are four drivers to that," says Anctil. "The hospital processes to facilitate the appointment and information flow; physician issues; patient/family issues; and financial issues."
In the first instance, there was no place in the record to capture who the primary care provider is, nor was there an obvious place to flag patients who were deemed by assessment to be at higher risk for readmission, she says. That meant that high-risk patients either weren't having appointments made because no one knew who to call, or because someone didn't realize they were at high risk. There was also an issue of ensuring a physician was available to see the patient in three to five days, she says, and if an appointment is made, how do you ensure that all the documentation — which in the past might have taken a week or more for a physician to complete — is done within the time period and delivered to the primary care physician? "In the regulatory world, they have 30 days to proof and sign their notes. Now we are telling them to do that on the day of discharge, and to make sure that it includes information on pending test results and how the attending physician thinks the primary care doctor should proceed."
What happens if the appointment isn't kept? Anctil asks. Who is supposed to follow up — the physician or the case manager? "We told them that until the handoff to the primary care physician is successful, they still have ownership of the patient." Not all the physicians agree. "Most would agree it's a good idea, but putting systems in place is hard. What do we expect them to do — drag the patient in?"
Further, each of these items is variable depending on the physician, the patient, the patient's support network, and insurance. In the latter category, Anctil notes that some may provide an outpatient case manager, and some may not. "You have to get buy in, you have to know if they have transportation, and whether they have money to pay the copay for their doctor," she says.
All of these areas of conflict and trouble have plagued Anctil's efforts to reduce readmission rates. But they keep plugging away.
Currently, they are asking every patient who is readmitted within 30 days of discharge why they think they were readmitted and what they might have done differently to avoid it. Often, the reason is medication-related.
"Patients won't take them or they have a generic at home and a name brand here and when they get home they take both, which is a double dose of something. We have to spend a lot more time educating patients now. We ask them what their goals are. If they are on 15 medications, they might say they want to only take two, because that's all they can afford." While the patient decision is based on money, that may not be the best decision for their health. If they will only take two medications, it's better that a provider determine the two most important medications to take.
Another issue they have struggled with at Henry Ford is how much attention to give to those at the far end of the bell curve. "If the tail end of the bell curve is 75% of the readmissions, though, you have to address it. The issue is, though, that you need a completely different strategy to deal with the patients who are far outside the norm than with the patients in the middle. There is a group of people in the inner city who come back into the hospital as a respite. I think you have to have separate programs for them than for the other readmissions."
Anctil has also taken a page from an article she read last year in the New Yorker that talked about hot spots — how a group of people who were falling and being readmitted to a hospital in Camden, NJ, all came from a single apartment building (http://www.newyorker.com/reporting/2011/01/24/110124fa_fact_gawande). Thinking of the readmissions as some sort of cohesive group — hot spotters rather than frequent fliers — is allowing Anctil to better profile the group. She knows, for instance, that a third of them have end-stage renal disease; two thirds have some sort of addiction issue or mental health problem. "That's only 92 patients last year who had six or more readmissions within 30 days — and probably more besides that. Those patients were responsible for over 2,000 admissions last year. And for all I know, they are at another hospital if they aren't here."
Even among that group of patients, only 4% are uninsured. "Our efforts with them will focus on end-stage renal disease, which should hit a third of the patients." One thing research has shown is that some of those patients aren't getting their full dialysis — they leave early or come late, dependent on someone else to drive them and at the mercy of that driver's imperative to be somewhere else. "We are actually working on a checklist to capture the reasons for under-dialysis."
They have also found a problem with how well they communicate dry weight changes to the dialysis center, Anctil adds.
"We have a cross functional team now that is looking at solutions," she says. On the table is mentoring and starting a primary care clinic in the dialysis center so that comorbidities can be addressed while the audience is captive, as it were.
The initial bundle includes a home care visit paid for by Henry Ford if the patient doesn't have insurance to cover it. "The reality is we have to be frugal, so I don't know that we will need a social worker at the dialysis center, and a mentor, and a home care visit. But I don't know which will work. We'll be able to evaluate some of it by looking at compliance and measures of other health status," says Anctil.
Shortly, Anctil says they will begin following a group of patients from home to dialysis — some who are compliant and some who are not. Patients on home dialysis will be monitored, too, so that they can see if those patients have something to teach about end-stage renal patients who can't stay out of the hospital.
As for that bump in the middle? Anctil knows something isn't working there, either. Part of the problem is that not all aspects of that five-pronged bundle have been implemented consistently, she says. "We want to flag those high-risk patients, but we couldn't find a good place to do it. On the chart? Where on the chart? On the board? There are pockets of providers who do this well and always have, but others don't. The only thing we can do now is track it and publicly show who is doing well and who isn't. We'll make it competitive, hospital to hospital. We'll tell Henry Ford Hospital that someone else is doing it better and they'll figure out a way to get it done." For a list of items that lead to flagging, see box below.
Diagnosis: Medications: Health history: Living situation: Source: Henry Ford Health System |
They are also doing additional disease education in the hospital, and then doing it again post-discharge, she says. "It is the same education in all settings." A pharmacy consultation and home health referral are also provided, and if the home care visit isn't covered, Henry Ford will pay for at least one visit. "It costs less than a readmission," says Anctil."
They are looking at how they perform on all these processes, and then looking to see if there is movement in the readmission rates. "If we end up doing this all and still getting a high rate, then something isn't working. But we aren't there yet. Three of the five things are in place — we still have to work on the follow up appointment and pharmacy consultation. There just isn't an easy way to determine if those things happened yet. But I think if we do all these things, we will have good results."
Anctil says that without a system readmission team headed up by CEO Nancy Schlichtin and chief quality officer Bill Conway, MD, she wouldn't hear about all of the various pilots and tests. "We all learn from each other and call the question about what should be standard work processes for the system." She also says that the system case manager council group, who work through the specific improvement efforts and reach a consensus on definitions of process measures are helpful.
Each of the five hospitals has its own team working on readmissions, too, and data are sent out monthly to the system and hospital leadership. It's on everyone's radar. One thing she thinks would probably be as good a predictor as anything else would be to ask the care team, "Do you think this patient will be readmitted in the next 30 days? You'll find the team generally knows the answer."
For more information about this story contact Beth Anctil, RN, MSN, Director for Care Coordination, Henry Ford Health System, Detroit, MI. Telephone: (313) 874-2490.
If you say it out loud, people will agree intuitively: You can learn more from your failures than from your successes. But that doesn't mean people want to trumpet what doesn't work.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.