Closing the Post-Discharge Follow-Up Gap
The hospital carries the 30-day return and the ACO carries the result, and the follow-up call that stops it is the one nobody has time to make..
Last updated: October 2026
The hospital carries the 30-day return and the ACO carries the result, and the follow-up call that stops it is the one nobody has time to make.
How do ACOs and health systems close the post-discharge follow-up gap?
They put a call to every discharged patient on the calendar and have an AI co-clinician make it, by phone or video, between the discharge and the next visit. The clinician sees a ranked worklist of who answered, who did not and who sounds different, and decides who gets a human call next.
The gap is not a knowledge problem. Every discharge planner knows which patients need a call in the next 72 hours. The gap is capacity. The call competes with everything else a nurse does that day, and it loses.
Closing it means taking the call off the nurse's list without taking the nurse out of the decision. That is the job of an AI co-clinician.
The patient is not the hard part either. Most discharged patients will answer a phone. They will say how they slept, whether the new medication is making them dizzy, and whether anyone has come by. The information is there for the asking. The asking is what does not happen.
What the gap costs
Hospitals are penalized for returns within 30 days under the CMS readmission penalties. Since January 2026 the TEAM model puts selected surgical episodes, including hip and femur fracture treatment, on the hospital's book for 30 days after discharge. A patient who falls at home in week two and comes back with a fracture is now an episode the hospital owns.
ACOs carry the same patients on a different ledger. Every avoidable return counts against the year's result. The cost in both cases is the same event: the visit that did not happen before the return did.
There is a third ledger. Star Ratings reward follow-up that happens after a discharge and after a hospitalization for mental illness. The measure does not ask whether the team meant to call. It asks whether the follow-up occurred in the window.
Why the call does not get made
A follow-up call takes longer than it looks. The nurse finds the number, dials, reaches voicemail, makes a note, and tries again tomorrow. Three attempts is a morning.
Then the attempts disappear. Most teams have no record of who was called, who answered and what was said. When the patient returns, nobody can show that the follow-up was tried.
The language problem makes it worse. A patient who answers in Spanish, Vietnamese or Tagalog needs an interpreter on the line. Scheduling one turns a ten-minute call into a two-day task. Those patients are called last, and often not at all.
What Gia does between discharge and the next visit
Gia places the call. Phone or video, any phone including a landline, in any of 44 languages. If the patient does not answer, Gia retries. No check-in silently disappears.
The call is a conversation. How she slept. Whether she took the new pill. Whether anyone has been by. Inside it runs a 40-second speech check-in that measures 2,500 speech biomarkers: pace, pauses, articulation and prosody. Results reach the clinician in under two minutes.
The clinician does not read transcripts. The team sees a ranked worklist with the reason beside each name: who did not answer after three tries, who reported a new symptom, whose speech has shifted since the last call.
Caregiver check-ins run alongside. The daughter who sees her father every evening gets a call too. What she has noticed becomes part of the same worklist, beside what the patient reported and how the patient sounded.
What the clinician gets, and what Gia never does
Each result arrives with Proactive Decision Actions. Options, never orders. The clinician picks one, adds a note and assigns the follow-up. A nurse call, a visit, a medication review, or nothing.
A check-in is a screening conversation. Gia is clinical decision support, not a diagnosis. The clinician decides what happens next. Every decision is recorded.
One click on Initiate Assessment sends Gia back to the patient to administer a structured assessment in conversation: the BIMS, a cognitive screening, or an ADL and IADL review. Results arrive in under two minutes. The clinician asked for more information and received it, without a visit and without a nurse's morning.
What the record shows at day 30
The follow-up record holds every attempt, every answer and every clinician decision, dated and attributed. Results, notes and follow-up write back to Epic, PointClickCare or your EHR.
When a patient does return, the chart shows what was tried and when. When a patient does not return, the chart shows why the team reached that one before the others. That is the record an ACO quality review and a hospital quality committee both ask for.
Where to start
Pick one discharge population and one 30-day window. Have Gia make the follow-up calls and route results to the clinicians already responsible for those patients. Count reach, follow-up completed and the decisions recorded.
See how ACOs and health systems deploy Gia, then book a working session with your workflow, your EHR and your population.
Sources & References
- CMS, Hospital Readmissions Reduction Program (HRRP): https://www.cms.gov/medicare/payment/prospective-payment-systems/acute-inpatient-pps/hospital-readmissions-reduction-program-hrrp
- CMS, Transforming Episode Accountability Model (TEAM): https://www.cms.gov/priorities/innovation/innovation-models/team-model
David Kaiser is the Founder and CEO of Scienza Health, the company behind Gia, an AI co-clinician that health plans, ACOs, health systems and senior-care operators use to check in with older adults between visits.
This content is intended for informational purposes and does not constitute medical advice. Citing research does not mean this page or the product has been independently peer reviewed.
