185 - Ensuring Quality Transitions of Care from the Hospital to the Home Setting by Utilizing Virtual Visits

Publication Number: 4182.185

Lulu Xu, Boston Children's Hospital, Boston, MA, United States; Vandana L. Madhavan, Massachusetts General Hospital, Chestnut Hill, MA, United States; Emily Kung, Harvard Medical School, Boston, MA, United States; Chadi El Saleeby, MassGeneral Hospital for Children, Boston, MA, United States

Background: At least a quarter of hospitalized children experience post-discharge issues (PDIs). These include misunderstanding of discharge instructions and issues acquiring prescriptions or follow-up appointments. PDIs are exacerbated in patients who use a language other than English (LOE). Few studies have examined the utility of post-discharge telemedicine/virtual visits (VV) following discharge from a pediatric hospital medicine (PHM) service.

Objective: In this multi-phase study, we created and implemented a post-discharge VV program and assessed its acceptability, feasibility, and utility over several years.

Design/Methods: First, we conducted a needs assessment of families of PHM patients and referring primary care pediatricians (PCPs) to create patient selection criteria, educational materials, and guidelines for the program. Patient inclusion criteria included patients with pending labs at discharge, prolonged hospitalization, complex medical needs, or inability to follow up with their PCP. The program was started in 2018, halted in 2020 during the COVID-19 pandemic, and restarted in 2024. A retrospective review of visit notes was conducted for all 74 patients who completed a VV between June 2018 and May 2025. To assess utility, we evaluated the following variables: "Clinical Care Impact (CCI)" and "Patient Care Impact (PCI)". CCI was defined as either a new/changed prescription, new/change in follow-up appointments, or new orders for labs or imaging. PCI was defined as any patient who had either a CCI or additional care coordination.

Results: Acceptance of the program was high among responding PCPs (73.5%) and families (85.1%). 113 patients were approached, 100 accepted, and 74 (65.6%) completed the visit. Medical complexity and difficulty following up with PCP were the most frequent reasons for inclusion in the program. No statistical difference was seen in VV completion between LOE patients (70.6%) and patients who prefer English (74.5%). A CCI was made in 45% of visits and there was a PCI in 72% of visits. Families that use a LOE experienced slightly fewer CCIs (41.6%) compared to English-speaking families (45.2%), but slightly more PCI (75.0% vs 71.0%). None of these differences were statistically significant.

Conclusion(s): VVs reliably address post-discharge issues by impacting the plan of care and facilitating care coordination. Our program reached patients that use LOE in equal measure to English-speakers. Such programs are particularly beneficial for medically complex patients, patients unable to follow up with a PCP, and patients who use a LOE.

Table 1: Characteristics of the patient population
*Other respondents include American Indian/Alaskan native (n= 1) and multiple (n=1). Fourteen patients had no race listed.
†Languages other than English include Spanish (n=6), Arabic (n=2), Cape Verdean Creole (n=2), Bengali (n=1), Haitian-Creole (n=1).
‡Out-of-State residents include 12 within the Northeast and 7 from outside the Northeast.
§ These criteria are not mutually exclusive. A single patient can meet multiple inclusion criteria.
|| Determined at provider discretion
¶Defined as length of stay greater than 7 days
**Added as criteria during Phase 4
PCP: primary care physician

Figure 1: Impact of virtual visits from chart review
* Other care coordination = Any coordination efforts including reaching out to specialists or PCP or providing patient with contact information
† Clinical Care Impacted (CCI) = At least one of the following: New prescription(s) provided or changed, New follow-up appointment(s) made or changed, other new orders placed (labs/imaging)
‡ Patient care impacted = At least one of the following: CCI or other care coordination