Temporary Inpatient Hemodialysis Access for Uninsured Patients Is Associated with Shorter Hospital Stays in a County Safety-Net System

Highlights

– Regulatory program flex approval permitting an inpatient hemodialysis (HD) unit to provide transitional outpatient HD to uninsured new initiators was associated with a marked reduction in hospital length of stay (LOS) at a Los Angeles County safety-net hospital.

– In the intervention hospital (LA General), mean LOS fell from 13.0 to 7.6 days after implementation (difference 5.4 days); control hospitals also showed LOS reductions, but difference-in-difference analysis did not reach statistical significance.

– Run-chart analyses suggested an immediate and sustained LOS reduction at the intervention site, while control hospitals showed more variable changes, raising concerns about the validity of difference-in-difference assumptions.

Background

Patients who lack insurance and who require initiation of maintenance hemodialysis face substantial structural barriers to timely transition from inpatient care to outpatient dialysis centers. In many systems, outpatient dialysis facilities require documentation of ongoing coverage or enrollment in programs before accepting patients, which delays discharge and prolongs hospital stays. Prolonged hospitalization for dialysis initiation increases costs, exposes patients to inpatient-related risks (eg, hospital-acquired infections), and disrupts timely transition to outpatient care and social supports.

Several health systems and policymakers have explored interim solutions — including emergency-department–based dialysis, localized outpatient slots reserved for uninsured patients, and regulatory flexibilities — to expedite safe discharge. The intervention studied here represents a novel approach: regulatory approval for an inpatient HD unit to provide temporary outpatient-style HD to uninsured patients newly requiring maintenance dialysis, allowing them to receive subsequent outpatient sessions outside the traditional inpatient encounter and thereby facilitating discharge from acute-care beds.

Study design

The report by Banerjee and colleagues (JAMA Network Open, 2025) describes a quasi-experimental, pre–post quality-improvement evaluation conducted within the Los Angeles County Department of Health Services safety-net system. The study period spanned January 1, 2016, through December 31, 2024. The intervention — implemented at Los Angeles General Medical Center (LA General) in February 2020 — consisted of regulatory program flex approval allowing the hospital’s inpatient hemodialysis unit to provide temporary, transitional outpatient dialysis to uninsured patients initiating maintenance HD. Two other safety-net hospitals in the system served as controls.

Population: Uninsured inpatients who newly initiated HD during the study period. The analysis included 951 patients across the three hospitals.

Comparison: Preintervention (before February 2020) versus postintervention periods at the intervention hospital, with concurrent data from two control hospitals. During the intervention period, control hospitals increased use of emergency departments to provide outpatient HD access.

Primary outcome: Hospital length of stay (LOS). Secondary outcomes specified in the study included all-cause 30-day readmission and mortality rates.

Key findings

The aggregated cohort included 951 uninsured patients who initiated hemodialysis in the included hospitals between 2016 and 2024.

Intervention hospital (LA General):

• Preintervention: 200 uninsured inpatients newly initiating HD (mean [SD] age, 52.0 [11.7] years; 130 men [65.0%]).
• Postintervention: 241 patients (mean [SD] age, 52.9 [11.1] years; 171 men [71.0%]).
• Mean LOS decreased from 13.0 days (SD 17.5) before the intervention to 7.6 days (SD 6.6) after the intervention (P < .001). This represents an absolute mean reduction of 5.4 days.

Control hospitals (two sites combined):

• Preintervention: 234 patients (mean [SD] age, 52.4 [13.6] years; 164 men [70.1%]).
• Postintervention: 276 patients (mean [SD] age, 52.7 [12.5] years; 209 men [75.7%]).
• Mean LOS decreased from 12.5 days (SD 15.3) before the intervention to 9.1 days (SD 9.4) after the intervention (P = .002), an absolute reduction of 3.4 days.

Difference-in-difference (DiD) analysis attempted to estimate the incremental effect attributable to the transitional outpatient HD model at LA General versus controls; the DiD estimate was a reduction of 2.0 days but did not reach statistical significance (P = .23).

Importantly, run-chart analyses demonstrated an immediate and sustained reduction in LOS at the intervention hospital after implementation, whereas control hospitals displayed more variable LOS patterns after the intervention began. The investigators note that the parallel-trends assumption necessary for valid DiD inference was likely violated, limiting the interpretability of the nonsignificant DiD result.

Secondary outcomes: The provided study summary identifies all-cause 30-day readmission and mortality as secondary endpoints. Detailed outcomes for these measures were not included in the excerpt provided here; readers should consult the full text for complete secondary-outcome data and subgroup analyses.

Interpretation and mechanistic considerations

How might a transitional outpatient HD service embedded in an inpatient dialysis unit shorten hospital LOS? Several plausible mechanisms exist:

• Operationally, the model decouples provision of follow-up dialysis from the inpatient admission, freeing acute-care beds once the patient is medically stable and has initiated dialysis.
• It streamlines care coordination because the treating health system retains responsibility for initial dialysis sessions and can schedule near-term outpatient sessions without awaiting external facility acceptance or completion of payer eligibility processes.
• Providing dialysis in a familiar hospital-affiliated environment may reduce administrative friction (eg, credentialing, vascular access arrangements) that otherwise slow discharge.

The authors’ run-chart data—an approach well suited to quality-improvement assessment—support a temporal association between implementation and rapid LOS shortening at LA General. The concurrent LOS reductions at control hospitals likely reflect broader secular changes (for example, system-wide policy shifts, process improvements, or adaptations during the COVID-19 pandemic) and highlight the complexity of attributing causality in uncontrolled environments.

Expert commentary and critical appraisal

Strengths of the report include use of a real-world safety-net population, a pragmatic implementation strategy, and multiple analytic approaches (pre–post, DiD, and run charts) to explore the signal of effect. The large sample size for a single-system study and consistency of directionality (LOS fell at both intervention and control sites) suggest contemporaneous operational changes that merit careful interpretation.

Limitations and cautions:

• Quasi-experimental design: Without randomization, unmeasured confounding and secular trends can bias effect estimates. The likely violation of the parallel-trends assumption undermines the robustness of the DiD result.
• Potential influence of the COVID-19 pandemic: The intervention began in February 2020, coinciding with the pandemic onset in the United States. Pandemic-era changes in admission thresholds, bed management, and discharge practices could have contributed to LOS changes.
• Contamination and concurrent innovations: Control hospitals increased use of emergency-department–based outpatient HD during the study period, which could have reduced the observable contrast between intervention and controls.
• Generalizability: Los Angeles County’s safety-net environment, local regulatory flexibility, and available inpatient dialysis capacity may not exist in other regions. Resource intensity (staffing, infection control, and billing adaptation) required to implement this model could limit scalability.
• Incomplete reporting in the supplied summary: Key secondary outcomes (30-day readmission and mortality), cost data, patient-reported outcomes, and longer-term access to outpatient dialysis were not reported in the excerpt and are essential to judge net benefit and safety.

Policy, operational, and ethical considerations

This model raises important policy questions. Granting regulatory flexibilities that permit hospitals to provide temporary outpatient dialysis may offer a pragmatic, patient-centered approach to prevent prolonged hospitalization and to reduce system costs. However, responsible implementation requires safeguards: clear clinical eligibility criteria for transitional outpatient dialysis; defined time limits or criteria for transition to established outpatient centers; mechanisms to ensure continuity of care, including vascular access surveillance and infection prevention; transparent billing practices; and attention to equity so that solutions do not inadvertently create two-tiered standards of care.

Conclusions and next steps

Banerjee et al. report that regulatory approval enabling an inpatient HD unit to provide transitional outpatient dialysis to uninsured patients initiating HD was associated with a substantial, immediate, and sustained reduction in hospital LOS at a large county safety-net center. Although concurrent LOS decreases in control hospitals and limitations inherent to the quasi-experimental design temper causal inference, the findings identify a promising operational strategy to address a pervasive discharge barrier for uninsured patients needing dialysis.

Important next steps include multicenter evaluations to assess replicability, randomized or stepped-wedge designs if feasible, measurement of patient-centered outcomes (quality of life, dialysis continuity, vascular access complications), economic analyses, and assessment of equity implications. Policymakers and health system leaders should weigh the benefits of reduced inpatient bed use against the operational requirements and oversight needed to ensure safe, equitable care transitions.

Funding and clinicaltrials.gov

Funding and clinical trial registration details were not provided in the excerpt furnished for this summary. Consult the full JAMA Network Open article for complete disclosures and acknowledgments.

References

1. Banerjee J, Gordon H, Walsh VE, Espiritu JS, Canamar C, Tangprahaphorn S, Oh HH, Nguyen JP, Yun T, Seo YS, Song Y, Redulla M, Alvarez M, Ghaffari A, Hutcheon D, Shoenberger J, Varnal M, Blake N, Coffey CE Jr, Spellberg B. Provision of Temporary Access to Inpatient Hemodialysis to Uninsured Patients Initiating Hemodialysis. JAMA Netw Open. 2025 Nov 3;8(11):e2544295. doi: 10.1001/jamanetworkopen.2025.44295. PMID: 41252168.

Author note

This article synthesizes the methods and main results reported in the cited JAMA Network Open paper with interpretation intended for clinicians and health system leaders. For full methodological detail, secondary outcomes, and authors’ disclosures, please consult the original publication.