Reducing hospital readmission rates
By Spencer H. Kubo, MD, FACC, FACP
From Minnesota Healthcare News
The high frequency of readmissions within 30 days of discharge from a heart failure related hospitalization is a hot topic among primary care physicians, hospitalists, and cardiologists. In 2015, the penalty for a hospital having rates above the expected reached its maximum, with up to 3 percent of all Medicare reimbursement withheld by the Centers for Medicare & Medicaid Services.
Clinicians are bombarded with countless journal articles and news releases that highlight specific interventions that have successfully reduced the rate of readmissions. Which intervention works best? Which one will work in your hospital? Our experience working with patients with heart failure at Allina’s United Hospital provides some information and insight that applies to a wide range of health care facilities.
Common themes emerge
Although the details of readmissions vary from hospital to hospital, two commonly accepted themes have emerged:
- One size does not fit all.
Many interventions have reduced readmissions, including medication reconciliation by pharmacists, implantable hemodynamic monitors, and referrals to cardiac rehabilitation. There is no single “magic” bullet that will work in every instance, so the best intervention for a particular hospital will vary, depending on geography, case mix, and available resources.
- Not all readmissions are preventable.
In fact, some readmissions indicate good medical care! This observation has been especially polarizing because the simple 30-day readmission statistic clearly includes some rehospitalizations that actually indicate excellent follow-up care. For example, patients with heart failure typically have multiple comorbid conditions that may require a hospitalization independent of the initial reason for admission to the hospital. Consider the patient who develops new-onset rapid atrial fibrillation or the patient who has a stroke while on therapeutic doses of Coumadin. Many experts have suggested a separation into preventable readmissions that occur within seven days of discharge and readmissions that occur after seven days, which frequently reflect a progression of the disease or other comorbid conditions and do not necessarily indicate inadequate medical follow up.
We assembled a small team that included a cardiologist, the director of the outpatient cardiology clinic, and an inpatient cardiovascular nurse specialist so that all the key patient settings were represented. Clinicians are trained to act quickly and there was a tendency to settle on a solution that “should work” and start implementation right away. However, we were very methodical about gathering the data to help us develop the most effective outcome.
First, we reviewed the medical record of every patient who had a 30-day readmission over a 12-month period. We found that the primary reasons for readmissions were varied and reflected medical complexity. Some were related to fluid overload or hypotension, but others were due to patient noncompliance, such as poly-substance abuse. Some were related to a scheduled procedure. This might include a patient who was allowed to “recover” at home prior to being readmitted for valve replacement surgery or cardioversion. Still others were related to comorbid conditions, such as new onset atrial fibrillation or an incarcerated hernia. From this analysis we understood that a simple intervention, such as a scale so that patients could weigh themselves at home, would not be uniformly effective. Further, since these patients had complex and interrelated medical problems, we quickly realized that we needed a clinician to help guide the patient through a successful transition from hospital to home.
A second observation was that it was very difficult to translate the issues and contingencies that occurred in the hospital to outpatient clinicians who were not involved in the hospitalization. Consider the patient with massive fluid overload and treated with an intensified diuretic regimen. At some point, when the patient becomes an outpatient, the diuretic regimen has to be decreased or the patient will develop hypotension or worsening renal insufficiency. But the exact timing of that transition is not easy to predict or communicate to the clinic staff. This problem is accentuated by our health care systems, which in an effort to maximize efficiency and effectiveness, inadvertently create silos and barriers to integration. Specifically, hospitalists and inpatient nurses stay with the patient only during hospitalization. Clinic nurses and primary care physicians only assume care once the patient visits the clinic. We have discharge orders in the electronic health record, but these documents can’t always reflect the nuances and changes in clinical conditions. Thus, it is not surprising that the plan derived during the hospitalization cannot always be effectively executed during the hand-off from hospital to home.
Coordination of care
Our solution was to introduce a heart failure care coordinator, something we modeled after a successful program at Mercy Hospital in Coon Rapids. We recruited nurses with cardiovascular experience in caring for patients with heart failure. We identified patients who had risk factors for rehospitalization based on an algorithm developed by Allina Health. The heart failure care coordinator established a relationship with the patient and closely followed him or her during hospitalization. This was important so the patient could contact a familiar person to address questions as an outpatient.
Another key is that the heart failure care coordinator, while not directly involved with inpatient care, had an intimate knowledge of the issues and the possible outpatient adjustments. The care coordinators made frequent phone calls (sometimes daily!) to the patient tracking weight, symptoms, and medications, which could be overwhelming, especially for an elderly patient. They connected with staff at skilled nursing facilities. Sometimes they had to prioritize and sequence follow up visits, such as those to dialysis, a primary care physician, or cardiologist. Many times, they reconciled medication doses and different pill strengths. They reviewed what the patient was eating and helped to interpret different symptoms. They alerted the clinic physicians with very specific follow-up information so that the physician could focus in on the most relevant issues at the clinic visit. This was the ultimate in care coordination because the coordinators helped to connect all the providers (primary care physicians, nurse practitioners, cardiologists, rehab personnel, home care nurses, and transitional care units) to improve patient outcomes.
Our program was launched at United Hospital in June 2012 with almost immediate results. Our average 30-day readmission rate was reduced from a peak of 22.4 percent in 2012 to 14.2 percent for the remainder of the year. Encouraged by these initial results, we expanded the program to Abbott Northwestern Hospital in August 2013, again with almost immediate results indicating a dramatic reduction in 30-day readmissions. We are now able to report results on a system-wide level. When combining the four cardiovascular hospitals for Allina Health (United [UTD], Mercy [MCY], Unity [UTY], and Abbott Northwestern [ANW]), we have seen a dramatic and progressive decline in the 30-day readmission rate. The readmission rate prior to implementation of the heart failure care coordinators was 19.7 percent, which has been reduced to 16.7 percent (p<0.001) (see Figure 1). Furthermore, the rates of follow-up visits that are completed within five days of discharge have increased from 39.8 percent to 58.2 percent (p<0.001) (see Figure 2).
We have now expanded our Heart Failure program to include representatives from all the major cardiovascular hospitals and all the various disciplines and patient touch points, such as pharmacy, rehab, and social work. This program provides education, training, and other resources for clinical staff at our other hospitals and clinics. Our experience points to the success and reward of taking a careful analytic approach to a problem and developing a focused intervention that improves the process of care by connecting and coordinating hospital, home, inpatient providers, and outpatient providers. We have clearly made an impact on 30-day readmission rates and are looking to extend our focus on care coordination beyond 30 days so we can continue to improve the care delivered to our patients with heart failure.
Spencer H. Kubo, MD, FACC, FACP, is a cardiologist with the United Heart and Vascular Clinic. He is board-certified in cardiovascular disease and has a special interest in patients with heart failure.