- Design
- Retrospective cohort, quantile regression
- Population
- 20,986 adult inpatient encounters at a 714-bed US academic centre, 2023
- Primary outcome
- Excess hospital days beyond expected length of stay
- Effect
- Referral encounters 85.5% of excess days; +2.11 median excess days after adjustment
A retrospective cohort at a 714-bed US academic centre analysed 20,986 adult inpatient encounters in 2023. Nearly half (48.4%) involved a referral for post-acute care — rehabilitation, skilled nursing or home services.
Those encounters accounted for 85.5% of all excess hospital days — days beyond the expected stay for the diagnosis. After adjusting for age, sex, service and case-mix severity, a referral encounter had a 2.9-day longer median length of stay and 2.1 more excess days.
The authors are careful to say referral is a marker of discharge complexity, not a cause. The practical point for a ward team is that long stays are predictable early: the patient who will need somewhere to go after hospital is the one who will stay, so that planning should start on day one, not on the day of medical fitness.
- Identify on admission which patients are likely to need rehabilitation, nursing-home or home-care support.
- Start the referral process while the medical work-up is still under way, not after 'medically fit'.
- Record the expected discharge destination in the admission note and revisit it daily.
- Excess days cluster in a minority of patients — target discharge-planning effort there.
Why it matters
It moves length-of-stay effort from the whole ward to a predictable minority of patients.
Don't overread it
One US hospital with a US post-acute system; the share may differ where families, not institutions, provide post-discharge care.
The statistics, in plain English
The 2.1 excess days is a median difference after adjustment, meaning a typical referral patient stayed about two days longer than expected — not that every one did.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for internal medicine, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free