This article is more than 5 years old. Information may no longer be current.

Mobile monitoring reduces preventable cirrhosis readmissions

WASHINGTON — A novel mobile device monitoring system reduced 30- and 90-day cirrhosis-related readmissions with no potentially preventable readmissions in this pilot study at 90 days, according to data presented at The Liver Meeting 2017.

“We are all acutely aware of the penalties for 30- and 90-day readmission rates,” Vandana Khungar, MD, MSc, director of inpatient hepatology at the Hospital of the University of Pennsylvania, said during her presentation. “We think of readmissions as value-based care becomes more and more popular in the U.S. We have seen many papers, not just related to all-cause readmissions, but also to disease-specific states. We should worry not just about reimbursements in value-based care, what is more important than reimbursement is the effect of readmissions on our patients and their caregivers. We know there are obvious benefits of improving patient well-being, delaying transplant and possibly even improving survival to transplant, [as well as] physician and health system benefits of improved quality metrics and decreased health care costs.”

Costly readmissions

Citing a 2012 study by Michael L. Volk, MD, MSc, and colleagues, Khungar said there are more than 150,000 hospitalizations and 40,000 deaths per year attributed to cirrhosis. Additionally, cirrhosis-related readmissions cost $4 billion per year.

Approximately 20% to 37% of patients with cirrhosis are readmitted each year within 30 days of discharge, and approximately 20% of those readmissions were considered potentially preventable, Khungar said. Ninety-day readmission rates among the 14-center North American Consortium for Study of End-stage Liver Disease – which the Hospital of the University of Pennsylvania is a part of – were 53%.

Khungar said 29% of patients with cirrhosis are readmitted within 30 days at an annual cost of $16.3 million at her institution. The potentially preventable readmission (PPR) rate was 23%, and more than half of the hospital’s cirrhosis service PPRs come back within 10 days of discharge, she said.

Early intervention

Khungar and colleagues developed a pilot study design with an intervention group compared with a cohort of patients admitted to the hospital’s cirrhosis service between May 2015 and July 2016. The aim was to evaluate a remote monitoring system with the ability to intervene early to determine the effect on 30- and 90-day readmissions in patients with cirrhosis, as well as potentially preventable readmissions.

Khungar said the belief was the hospital could reduce 30- and 90-day readmissions by remotely monitoring patients’ weight, mental function, medication adherence, temperature, blood pressure and pulse oximetry.

Patients received a 4G tablet, wireless blood pressure monitor, pulse oximeter and weighing scale following discharge from hospital. They kept each device for 90 days.

Using the digital platform, patients sent biometric parameters, medication administration and responses to symptom questionnaires to registered nurses at the hospital. And if necessary, a non-clinician liaison, nurse or doctor could intervene with the patient by phone or video conference.

The digital system analysis included 19 patients (median age, 58.3 years; range, 50.2-64; 52.6% women) and 143 patients (median age, 61 years; range, 54.2-66.9; 45.5% women; 65% white) received usual care.

Results of intervention

Median time to readmission was 71 days in the digital system cohort and 42 days in the usual care cohort.

Twenty-one percent of patients within the usual care cohort were readmitted within 30 days, whereas 15.8% of patients in the digital monitoring system were readmitted during the same time.

Khungar, however, acknowledged an “interesting” finding in the difference between cohorts and 90-day PPR rates. Researchers found that 34% of the usual care cohort’s 90-day readmissions were potentially preventable. However, there were no 90-day PPRs in the digital monitoring cohort (P = .03)

The researchers had a median of 18.5 digital encounters (range, 6-128) with patients in the digital cohort. And, as Khungar said, each patient received coaching on their health care as well as teaching on how to use their devices and the digital platform for reporting.

The mobile device system allowed for the researchers to intervene when necessary, Khungar said. Researchers changed the medications of six patients and another eight who were non-adherent to their medications received coaching to get back on their medications. Three patients received reminders to prevent them from missing follow-up appointments.

Khungar noted that if PPRs were avoided, the institution would save $1.1 million annually, and potentially more as more patients received training to monitor themselves and would make these behaviors habit.

Additionally, revenue could increase if another service, for example cardiac surgery, could use a bed that would be going to the PPRs.

“We want to implement proactive measures to prevent our hospitalizations, lead to lower health care costs and penalties for the hospital as readmission rates are becoming a national indicator,” she said. “Readmissions are becoming increasingly important, but we all have to remember our cirrhotic patients are very ill, and given their degree of illness, some readmissions are to be expected. So, we need to focus on the potentially preventable ones. We need to engage our patients and their caregivers, and with simple technology, this is completely feasible.”

Broad applicability

Khungar did acknowledge that some individuals required a fair amount of education on their disease process and device management, but that the system was easy to teach. Additionally, Khungar mentioned that the cost per patient per 90 days was $1,019 – which, when questioned about its broad applicability, she stated may be cost prohibitive in large populations.

PAGE BREAK

“That’s one of the key questions,” Khungar said. “What is the secret sauce? And can you deploy it to any other place? That is what we are trying to examine with [a larger] randomized control trial. Can we make it less high cost, less high touch so that everybody can benefit from this? Can we deploy it easily to many centers? Our hope is for any centers that are interested, that we are able to make it a package plug-and-play and we can deploy it to another center without a lot of staff on the backend.” – by Ryan McDonald

Reference:

Khungar V, et al. Abstract 170. Presented at: The Liver Meeting; Oct. 20-24, 2017; Washington, D.C.

Volk ML, et al. Am J Gastroenterol. 2012;doi:10.1038/ajg.2011.314.

Disclosures: Khungar reports no relevant financial disclosures.

Editor's note: This coverage has been updated to ensure clarity in reporting.