Hospital Factors Drive Readmissions, Study Shows

Diana Swift

September 13, 2017

After disentangling patient-related factors, researchers find hospital factors significantly affect 30-day readmission rates, researchers report in an article published in the September 14 issue of the New England Journal of Medicine.

"There's been a lot of uncertainty about the degree to which readmission is about hospital environment, including policy and staff, and to what extent we're powerless as healthcare providers because in the end it's all about the patient," lead author Harlan M. Krumholz, MD, from the cardiovascular medicine section at Yale University School of Medicine in New Haven, Connecticut, told Medscape Medical News. "This study is an attempt to disentangle those factors."

The two-part study drew on the Centers for Medicare & Medicaid Services hospital-wide 30-day readmission measures cohort, consisting of Medicare fee-for-service recipients aged 65 years and older from July 2014 to June 2015, excluding cancer and primary psychiatric patients.

Dr Krumholz and colleagues divided the cohort into two random samples. The first sample consisted of 3,455,171 discharges and 2,741,289 patients at 4738 hospitals and was used to calculate hospital readmission performance.

The investigators calculated the risk-standardized rate for unplanned readmissions for all hospitals, classifying them into performance quartiles, with the lowest readmission rate (quartile 1) indicating the best performance.

Overall, the mean risk-standardized readmission rate was 15.6% (±0.6).

Across quartiles, the mean risk-standardized readmission rates from the best- to the worst-performing hospitals were 15.0%, 15.4%, 15.7%, and 16.3%.

In terms of hospital characteristics, such as size, location, and teaching status, the researchers found no concrete differences affecting readmission.

"It seems unlikely it's about the hospital's' structural characteristics. It seems more about hospital culture and the way the hospital operates," Dr Krumholz said, adding that his group will look at relevant hospital characteristics in a future study.

To minimize the effect of patient-specific factors in readmission rates, the researchers performed a second analysis using data from the remaining Centers for Medicare & Medicaid Services cohort. From those patients, the team identified 37,508 patients who had two or more admissions for similar diagnoses within a year and who were treated at more than one of the 4272 hospitals.

A significantly higher 30-day readmission rate emerged in patients admitted to the worst-performing quartile of hospitals compared with those admitted to hospitals in the best-performing quartile: 25.1% vs 23.1% (95% confidence interval, 0.4 - 3.5; P = .001). Other interquartile comparisons did not reach statistical significance.

"An absolute difference of 2 percentage points may seem to be small relative to the overall readmission risk, but it indicates that for every 50 patients who are admitted to a hospital in the lowest-performing quartile rather than in the highest-performing quartile, there is one additional readmission," Dr Krumholz and colleagues write.

"In the end we still have work to do to improve readmissions," Dr Krumholz said. "Even the best hospitals can do better, and we're just beginning to learn how to optimize people's recovery after discharge. And that will likely involve strategies to make sure people are in a stronger position to go home."

Allaying concerns about the possible negative effects of pushing for rehospitalization reduction, Dr Krumholz's group reported in July that the recent penalty-driven decline in readmissions is not associated with any increase in 30-day post-discharge mortality.

The authors believe the current findings validate use of readmission rates as a measure of hospital quality and performance, and a starting point for further reducing preventable readmissions.

The study was funded by the Yale-New Haven Hospital Center for Outcomes Research and Evaluation and other organizations. Numerous study authors report support from the Centers for Medicare & Medicaid Services, the US Food and Drug Administration, and the Agency for Healthcare Research and Quality. Dr Dhamarajan reports ties to Clover Health Insurance. Dr Krumholz reports ties to Johnson & Johnson (Janssen), United Health, Hugo (an online health data platform), IBM Watson, Element Science, and Aetna Inc. Dr Ross reports relationships with Medtronic, Johnson & Johnson, the Blue Cross-Blue Shield Association, and the Laura and John Arnold Foundation.

N Engl J Med. 2017;377:1055-1064. Abstract

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.