Post-discharge meal delivery was associated with fewer early readmissions

Design

Retrospective single-centre cohort study with 3-day landmark

Population

368 adults aged 60 or over at risk of food insecurity or malnutrition, discharged home

Primary outcome

Unplanned readmission at 14, 21 and 30 days

Effect

14-day aOR 0.31 (95% CI 0.14 to 0.69); 21-day 0.47 (0.23 to 0.95); 30-day 0.82 (0.44 to 1.53)

This single-centre retrospective cohort study followed 368 patients aged 60 or over, at risk of food insecurity or malnutrition, who were discharged home between September 2022 and April 2024. Some received two weeks of home-delivered meals through a state-funded Meals on Wheels discharge pilot; 62.9% received meals.

Meal recipients had lower odds of 14-day readmission (aOR 0.31, 95% CI 0.14 to 0.69) and 21-day readmission (aOR 0.47, 0.23 to 0.95). By 30 days the association had weakened (aOR 0.82, 0.44 to 1.53) and the interval crossed 1. Earlier delivery, within 9 days, was associated with lower 14-day readmission (aOR 0.23, 0.06 to 0.81), with wide intervals.

It is an association: patients who actually received meals may differ from those who did not, for reasons such as phone access, home stability and being reachable. The authors call for prospective evaluation.

• Screen older inpatients for food insecurity before discharge.
• If a meal-delivery programme exists locally, refer early, since earlier delivery was associated with lower readmission.
• Do not expect a lasting effect on 30-day readmission from two weeks of meals.
• Treat this as hypothesis-generating, not as proof that meals prevent readmission.

Why it matters

It suggests a cheap, non-clinical step at discharge may matter for older adults who are food insecure.

Don't overread it

This was a retrospective single-centre cohort; association with fewer readmissions does not show that meals caused them.

The statistics, in plain English

An adjusted odds ratio of 0.31 means the odds of readmission within 14 days were about two-thirds lower in recipients after statistical adjustment, but adjustment cannot remove all differences between people who got meals and those who did not. At 30 days the interval includes 1, so no association is shown.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.