Case managers beat Houston Methodist's discharge AI on the last day
JAMA Network Open published a quality improvement study on September 3rd, 2026. It compared discharge dates from a commercial AI tool built into Houston Methodist Hospital's electronic health record against dates estimated by the hospital's own case managers. The study covered inpatient encounters discharged between August 1st, 2023 and February 28th, 2024. At admission the two were close. Nearer to discharge, which is when the answer is used to plan a bed, the case managers were substantially more accurate. It reports estimation accuracy rather than any effect on a patient, and says so.
How to take part
The study is published open access and can be read in full, with its methods and its table of results. A hospital's choice of a clinical software tool is not a public proceeding and there is no room a resident can attend.
Where
1 county. In Houston-Pasadena-The Woodlands.
Timeline
- filed
Published in JAMA Network Open
- Today
How this decision moved
One dated line per check, oldest first. A line that says nothing changed means somebody looked and it had not.
- 2026-09-09
Admitted on the published study. The hospital has said nothing about whether the tool is still in front of its case managers, and the study names no vendor.
- 2026-09-12
Checked and unchanged. The decision still stands as decided.
- 2026-09-13
Checked and unchanged. The decision still stands as decided.
- 2026-09-16
The Houston Methodist result still reads the way it was published, with the case managers ahead of the tool where the date matters most. No correction has been issued.
- 2026-09-19
The finding stands as published, and the hospital has said nothing further about what it did with the tool afterwards.
- 2026-09-23
The finding still stands that the case managers were closer than the model on the day that mattered most.
- 2026-09-26
The study's finding still stands. Case managers were closer than the model on the last day.
- 2026-09-29
The published result holds. On the final day the case managers' calls were closer than the discharge model's.
- 2026-10-02
The published comparison still stands, with the case managers closer than the model on the last day.
- 2026-10-05
Houston Methodist's own account still has its case managers beating the discharge AI on the last day.
- 2026-10-08
Houston Methodist's own account still has its case managers outperforming the discharge AI on the final day.
- 2026-10-11
Houston Methodist's account still has case managers beating the discharge AI on the final day.
- 2026-10-15
Houston Methodist's account still has its case managers beating the discharge AI on the last day.
- 2026-10-18
Houston Methodist's account still has its case managers outperforming the discharge AI on the last day.
- 2026-10-21
Checked and unchanged. The decision still stands as decided.
- 2026-10-24
The case managers' result against Houston Methodist's discharge AI is unconfirmed today, since the journal page answered with a browser check instead of the article.
The evidence
Every fact above rests on one of these. The words are the source's own.
At 24 hours, 79.5% of case manager estimations fell within 1 day vs 37.9% for AI estimationsJAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
This quality improvement study included inpatient encounters at Houston Methodist Hospital with discharge dates between August 1, 2023, and February 28, 2024.JAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
Artificial intelligence–estimated discharge dates were generated by a commercial clinician-assisted tool integrated with the EHR.JAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
This quality improvement study found that AI-estimated discharge was comparable to usual care at admission but outperformed by case managers as discharge approached (when accurate estimations are most actionable for bed management).JAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
The strong admission correlation did not persist nearer discharge and should not be interpreted as evidence that the tool captured the same information clinicians use throughout the stay.JAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
Case manager estimates were part of usual care and visible to care teams, so they may be partly self-fulfilling and, thus, advantaged nearer discharge; AI estimations were accessible in the EHR but not enabled automatically.JAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
Whether the AI’s incremental accuracy in selected subgroups justifies the complexity of implementation warrants prospective, outcome-based evaluation.JAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
Nearer to discharge, case managers vs AI were more accurate (48 hours: MAE, 1.29 [95% CI, 1.26-1.32] vs 1.59 [95% CI 1.57-1.61] days; 24 hours: MAE, 0.98 [95% CI, 0.96-1.01] vs 1.93 [95% CI, 1.91-1.95] days) (both P < .001).JAMA Network Open, Artificial Intelligence vs Case Manager Estimation of Discharge Dates for Hospitalized Patients Primary source, official · pmc.ncbi.nlm.nih.gov
Questions about this decision
Answered from the record itself. Every answer is assembled from stored fields, so an answer the record has no basis for is left out rather than guessed.
What is this decision?
JAMA Network Open published a quality improvement study on September 3rd, 2026. It compared discharge dates from a commercial AI tool built into Houston Methodist Hospital's electronic health record against dates estimated by the hospital's own case managers. The study covered inpatient encounters discharged between August 1st, 2023 and February 28th, 2024. At admission the two were close. Nearer to discharge, which is when the answer is used to plan a bed, the case managers were substantially more accurate. It reports estimation accuracy rather than any effect on a patient, and says so.
Who decides it?
Houston Methodist Hospital decides. The record names the deciding body for every entry it carries.
Can the public take part?
The study is published open access and can be read in full, with its methods and its table of results. A hospital's choice of a clinical software tool is not a public proceeding and there is no room a resident can attend. No dated public window is on the record. The deciding body is named and reachable.
Where in Texas does it apply?
It covers Harris County in the Houston-Pasadena-The Woodlands, TX area.
Has it been decided?
It has been decided. The dates on the item page carry when.
What happens next?
No future date is on the record. The last dated step on it was filed on September 3rd.
When did it start?
The earliest date on its record is September 3rd, 2026.
What kind of decision is it?
It is filed under health and education.
What sources back it?
One source backs it. It is primary.
Is it on the ERCOT grid?
Yes. It sits inside the ERCOT interconnection.
When was it last checked?
Every fact on it was last verified against its source on October 21st, 2026.
Cite this
Texas AI Docket, Case managers beat Houston Methodist's discharge AI on the last day. Tracked since September 3rd, 2026. Last verified October 21st, 2026. https://texasaidocket.com/item/tx-2026-0137/. Reuse permitted under CC BY 4.0 with attribution. The same entry is in the docket JSON as item tx-2026-0137.
Beat
Filed under Health and education, with every other decision on that beat.