Readmission prevention starting at admission

Patients were screened up front to determine their likely readmission risk.


Background

Reducing readmissions is every hospital's goal, but “There is a component of readmissions that has very little to do with hospital care,” said Amy Deutschendorf, MS, RN, vice president for care coordination and clinical resource management for the Johns Hopkins Health System.

She groups the common causes of readmissions into five categories: 1) poor care coordination, 2) problems with the quality of care during the first admission, 3) disease progression, 4) patients' socioeconomic and sociodemographic issues, and 5) access to primary care. “Of those five reasons, the two that we could really do something about have to do with care coordination and ensuring that our patients are getting the appropriate care during the first admission,” said Ms. Deutschendorf.

With a grant from the Center for Medicare & Medicaid Innovation (CMMI), Johns Hopkins developed a bundle of readmission-prevention strategies that spans the entire hospitalization and extends after discharge. Two parts, Transition Guides (TG) and the Patient Access Line (PAL), were modeled after existing examples in the literature.

How it works

Under the new strategy, all patients were screened by nurses at admission to determine whether they would need complex discharge planning. “This comes from a premise that every patient needs their care coordinated, but not everybody needs a high-touch intervention,” said Ms. Deutschendorf.

If a patient was deemed to be at high risk for readmission, he or she received a TG, a registered nurse who remained in contact for 30 days after discharge to facilitate care and, when necessary, visited the patient at home to help with self-care management. All other patients received the PAL intervention, in which registered nurses called within 48 hours of discharge to comprehensively review discharge instructions and identify any issues. These phone calls typically lasted 20 to 30 minutes.

At the start of the interventions, the two hospitals hired 14 TG nurses and six PAL nurses, said Ms. Deutschendorf.

Results

Between Jan. 1, 2013, and Oct. 31, 2015, 8,635 patients were referred to TG services, and between July 1, 2013, and Oct. 31, 2015, 16,993 patients were referred to the PAL intervention. Overall, 1,973 (22.8%) patients in the TG group and 2,409 (14.2%) patients in the PAL group were readmitted within 30 days of discharge, according to results published in November 2017 by the Journal of General Internal Medicine. The proportion of patients who actually received the interventions was similar between groups: 60% in the TG group and 64% in the PAL group.

Patients who were referred to the TG program but did not participate were at significantly increased risk of readmission (adjusted odds ratio, 1.83; 95% CI, 1.60 to 2.10) compared to those who did. Similar results were found among patients referred to the PAL (adjusted odds ratio for readmission with nonparticipation, 1.27; 95% CI, 1.12 to 1.44). To prevent one readmission, the number of patients needed to treat with the TG and PAL interventions was 11.2 and 50, respectively.

Patient characteristics associated with not receiving either intervention were younger age, male sex, black race, Medicaid insurance, more comorbidities, and discharge from a medicine unit. These factors were also linked to higher readmission rates. “What we can show is that if you actually do receive the intervention, you're going to do very, very well, but our challenge then becomes to make sure that we engage people who, for whatever reason, do not,” said Daniel J. Brotman, MD, FACP, director of the hospitalist program at Johns Hopkins Hospital and a professor of medicine at the Johns Hopkins University School of Medicine.

Challenges

Initially, uptake of the TGs was a meager 25%. “A lot of patients don't want it. They don't want the Transition Guides coming into their homes,” said Ms. Deutschendorf. To address this issue, the hospitals brought the nurses onto the units to introduce themselves to patients before discharge, she said.

Another challenge was getting all clinicians on the same page, added Dr. Brotman. “Everyone needed to understand what the services do (and don't do) and make sure they were integrated with existing care processes so there was no duplication of efforts,” he said.

Dr. Brotman added that for others wanting to replicate this project, a potential challenge is the cost. “We were fortunate in the state of Maryland to have an expectation that some of our health care dollars will be used for infrastructure for care coordination,” he said. “That allows sustainability, but the initial seed money was provided through CMMI.”

Next steps

Johns Hopkins has since expanded the two strategies to another community hospital and is piloting them in pediatrics, said Ms. Deutschendorf. “We hope and expect to be able to spread it to our fourth community hospital in the Maryland area and eventually keep it going,” she said. “It's a very important thing that we do.”