Background: Prior cross-sectional studies have confirmed the anecdotal experience of many hospitalists: that many older adults are getting “stuck” in the hospital after completing medical treatment. However, these studies have been limited to select hospitals using a single point-in-time estimate. Nationally representative data is needed to better understand this phenomenon and characterize potential underlying contributors. The VA system is uniquely able to characterize delayed discharges through its robust characterization of hospitalizations paired with robust daily utilization management reviews by more than 700 nurses using the same national framework.

Methods: We created a dataset of all Veterans age 65+ hospitalized in VA hospitals between January 1, 2018, and December 31, 2022, merging VA administrative data with utilization management reviews to create both a Veteran-day and a Veteran-stay (entire hospitalization) dataset. We characterized the proportion and number of Veterans who experienced a delayed discharge (defined as not meeting criteria to remain in the hospital – i.e., medical care is completed and there is another reason for their continued stay) as well as the proportion and number of hospital stays and days in which discharge was delayed. We evaluated changes over time in the number of hospital days in which discharge was delayed, and categorized reasons for delayed discharge recorded in utilization management reviews.

Results: Our final dataset included 776,955 Veterans, 3.75 million hospital stays, and 12.8 million hospital days, of which 7.8M days had a utilization management review. In total, 423,928 (54%) of Veterans experienced at least one delayed discharge, and delayed discharge days accounted for 2.5M (40%) of all hospital days with a utilization management review. The number of days of delayed discharge increased from 599,874 in 2018 to 678,369 in 2022, a 5.5% increase. Among stays with a delayed discharge, 63.9% of the days in that stay were delayed discharge days. The most common reasons for potentially preventable delayed discharge were related to post-acute care, including lack of skilled nursing facility beds.

Conclusions: More than half of Veterans experienced a delayed hospital discharge, accounting for 40% of all hospital days in VA hospitals, often due to lack of appropriate post-acute care options. Given the VA has more extensive supports for post-acute care (including its own nursing homes) than non-VA health care systems, these findings may underestimate the prevalence of this problem outside the VA. To our knowledge, our novel use of utilization management data (collected in the same way across all VA hospitals) represents the first nationally representative analysis of delayed hospital discharges – and is a necessary first step to understanding how to address them.