COVINGTON, KY. – A healthy 2-year-old is admitted for incision and drainage of a MRSA thigh abscess and is given intravenous clindamycin during a 1-day hospital stay before being discharged home with a prescription for oral clindamycin.
The child refuses to take the clindamycin at home, and is readmitted 3 days later with a new abscess on her arm that requires a 2-day hospital stay.
Was this readmission preventable?
Some attendees at the Pediatric Hospital Medicine 2012 meeting thought it was out of the hospital’s hands, while others suggested the hospital was at fault because clindamycin is such an unpalatable medication that an oral dose should have been given at the hospital, along with instructions for how to make it more palatable at home.
The scenario is part of the ongoing Vanderbilt Readmissions Project, which seeks to identify patient and hospitalization characteristics of 15-day readmissions in children, create a 5-point "preventability scale" for early pediatric readmissions that can be applied by multiple reviewers, and institute measures to decrease potentially preventable readmissions.
Dr. James C. Gay
What the investigators have found so far is that, even after reviewing the same clinical information for 200 pediatric readmissions, a panel of four knowledgeable pediatricians gave exactly the same ratings in 37.5% of cases, Dr. James C. Gay said. There was 94% agreement on planned readmissions (47/50 cases), but only 19% agreement on unplanned readmissions (28/150).
"Further studies are needed to develop concrete rules for assessing preventability that can be applied reproducibly by multiple reviewers in multiple types of readmissions," he said.
The Vanderbilt findings have financial implications for hospitals, as the federal government has already taken to heart the issue of preventable readmissions following the sentinel article reporting that 19.6% of Medicare beneficiaries were readmitted within 30 days at a cost of $17.4 billion in 2004 (N. Engl. J. Med. 2009;360:1418-28). Many of these readmissions were thought to be avoidable by improvements in care and the discharge planning processes during the initial hospitalization.
The Affordable Care Act has also taken up the issue, and beginning Oct. 1, 2012, prospective payment system (PPS) hospitals will experience decreased Medicare payments for three index admissions – myocardial infarction, heart failure, and pneumonia – with "higher than expected" 30-day readmissions, coauthor Dr. Paul Hain explained at the meeting, which was sponsored by the Society of Hospital Medicine, the American Academy of Pediatrics, and the Academic Pediatric Association.
The Centers for Medicare and Medicaid Services (CMS) will calculate hospitals’ actual readmissions, excluding planned readmissions and readmissions unrelated to the index admission, and then compare these to hospitals’ expected readmission rates. Hospitals with "higher than expected" rates will be required to pay back the payments they’ve received for readmissions deemed to be excessive.
Beginning in 2015, the CMS may expand the list of conditions to include chronic obstructive pulmonary disease and several cardiac and vascular surgical procedures, he said. Similar penalties are likely to befall children’s hospitals, as rules that start in Medicare trickle down to Medicaid in 3-5 years.
Part of the problem is that researchers have yet to identify what can reliably drive down adult or pediatric readmissions or even determine whether the readmission interval should be 3, 7, 15, 60, 90, 365, or 30 days, as the CMS uses.
Dr. Paul Hain
"We looked at 30 days, and the noise that comes in is incredible," said Dr. Hain, now with Children’s Medical Center, Dallas. Ultimately, the researchers chose 15 days for their analyses because of the intuitively greater relationship to events in the index hospitalization.
Based on 4-year data from Monroe Carell Jr. Children’s Hospital at Vanderbilt in Nashville, Tenn., the 15-day readmission rate among non-newborns was 9.6% in 2007, 9.6% in 2008, 8.8% in 2009, and 8.9% in 2010; and among newborns, the rates were 2.5%, 3.2%, 2.5%, and 3.0%, respectively.
The first real global study of pediatric readmissions reported that 16.7% of patients between 2 and 18 years old at 38 U.S. children’s hospitals were readmitted within 365 days, and that readmissions were strongly associated with any complex chronic condition, female gender, older age, black race, public insurance coverage, longer length of stay during the initial admission, and number of previous admissions (Pediatrics 2009;123:286-93).
One year later, the same group reported that the likelihood of readmission among children aged 2-18 years actually increased as a states’ health system performance ranking improved (J. Pediatr. 2010;157:98-102.e1), observed Dr. Gay of the Children’s Hospital at Vanderbilt.
Similarly vexing results have been observed among adults. The Mayo Clinic in Rochester, Minn., recently reported that general medicine patients with a documented follow-up appointment were slightly more likely to have a hospital readmission, make an emergency department visit, or die within 180 days after discharge than those without an appointment (Arch. Intern. Med. 2010;170:955-60).