Ensuring timely patient discharges is central to managing a hospital's patient flow; however, discharges are dependent on the coordination of multiple care teams and thus are highly decentralized in nature. Many large hospitals have established capacity management centers to centrally direct and inform flow and support clinical teams across the hospital system, but they often lack transparency into what are the actionable, high-yield barriers to discharge that they need to focus on to be most effective. Moreover, these barriers are patient-specific and context-dependent, i.e., a patient's clinical-operational context determines what issues must be resolved and with which urgency. In this study, we leverage a machine learning model that predicts which patients are likely to be discharged in the next 24 hours together with a mixed-integer prescriptive optimization model to identify a subset of issues called minimal barriers that stand in the way of discharging a patient. Such barriers balance two aims: a high likelihood that the patient will be discharged from the hospital in the next 24 hours if these barriers are resolved; and a high likelihood that these barriers will indeed be resolved. We empirically demonstrate the efficacy of the proposed formulation and solution methodology in identifying a small number of minimal barriers using real data from a large academic medical center.

The authors have declared no competing interest.

This study did not receive any funding.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

Approved by Mass General Brigham Institutional Review Board, protocol 2011P001124

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.

Yes

No data is publicly available.