Single 30-day hospital readmission metric fails to reflect changing risk factors

Credit: Anne Lowe/public domain

A new study from researchers at Beth Israel Deaconess Medical Center (BIDMC) suggests that risk factors for readmission change significantly over the course of the 30 days following hospital discharge. Thirty-day hospital readmission rates have become a federal quality metric intended to reflect inpatient quality of care and unnecessary health care utilization.

Published today in the Annals of Internal Medicine, the research suggests that two distinct 8-day and 30-day would serve as better inpatient quality measurements and would better inform readmission prevention strategies.

A review of more than 13,000 discharges involving more than 8,000 patients in 2009 and 2010 found that early readmissions (0-7 days post discharge) were associated with markers of the acute illness managed during initial hospitalization. A patient's chronic illness burden was more important in predicting late readmissions (8-30 days post discharge).

"Our research found that for readmission evolved during the first 30 days following ," said lead author Kelly L. Graham, MD, MPH, a physician in the Division of General Medicine and Primary Care at BIDMC and an Instructor in Internal Medicine at Harvard Medical School (HMS). "Readmissions in the first week were more highly associated with factors related to the initial hospitalization than later readmissions. These findings suggest that the standard 30-day metric does not accurately reflect hospitals' accountability for readmissions."

The authors also note that other research has shown hospitals that strictly follow evidence-based care standards do not necessarily have the lowest readmission rates and that readmission rates do not serve as a benchmark to inpatient mortality. Under the Affordable Care Act, the Centers for Medicare and Medicaid Services (CMS) may reduce payments to acute care hospitals deemed to have excess readmissions within 30 days of discharge.

The study also found that discharge between 8 a.m. and 12:59 p.m. was associated with lower odds of an early readmission. The authors note that discharge in the first part of the day likely enables patients and their families more time to access community resources such as pharmacies and social supports, thus reducing the likelihood of readmission.

The authors also found that social determinants of health are closely tied to readmissions, as they impact how patients access care. They evaluated the impact of barriers to health literacy on readmissions and found that they were associated with both early and late readmissions. Insurance status was also relevant among patients readmitted in the late period; patients with unsupplemented Medicare or Medicaid were more likely to be readmitted eight or more days after discharge.

"The growing movement toward accountable care organizations and patient-centered medical homes may prove beneficial in preventing unnecessary readmissions," Graham added. "Patients discharged from the hospital need support from and teamwork among hospitalists, physicians, nurse practitioners, visiting nurses, pharmacists and others."

The authors stressed that both hospital and outpatient settings need systems of care that closely monitor patients as they transition their medical care from the hospital team back to the primary care team. Post-discharge monitoring would better enable a team to make sure adhere to the detailed care plan designed by the hospital team, such as taking medications correctly and keeping follow-up appointments.

Explore further

New model predicts readmission of congestive heart failure patients

Journal information: Annals of Internal Medicine

Citation: Single 30-day hospital readmission metric fails to reflect changing risk factors (2015, June 2) retrieved 4 December 2021 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors