As healthcare systems grapple with nursing workforce pressures while seeking to improve efficiency and patient care, virtual nursing is emerging as a technology-enabled approach to redesigning clinical workflows. The Medical University of South Carolina (MUSC) has expanded its virtual nursing programme from five inpatient units to more than 60 units across four regions, using a combination of centralised clinical capacity, data analytics and AI-enabled technology to support bedside teams.
At the centre of this evolution is the integration of virtual nursing into existing clinical workflows, with applications spanning admissions, discharge education, documentation, care coordination and quality surveillance. AI-enabled capabilities are also opening new possibilities for ambient documentation, predictive monitoring and intelligent workflow routing, potentially allowing virtual nurses to support larger patient populations while enabling bedside nurses to spend more time on direct patient care.
In this interview with Emily H. Warr, MSN, RN, Administrator, Center for Telehealth, Medical University of South Carolina, she discusses the factors behind MUSC's virtual nursing expansion, the metrics being used to assess its clinical, operational and financial impact, and how AI is reshaping collaboration between virtual and bedside nurses. She also highlights the challenges around clinician adoption, EHR integration and change management, and shares her perspective on how virtual nursing could evolve into a broader AI-enabled care delivery model.
MUSC’s virtual nursing program has expanded from five inpatient units to 60+ units across four regions. What were the key clinical, operational, and technological factors that enabled this rapid scale-up?
A major factor was that MUSC approached virtual nursing as a care delivery redesign, not simply as a staffing intervention or technology deployment. The program began in November 2023 on five medical-surgical units, but it was built from the start around centralized staffing, technology infrastructure, and analytics through MUSC’s Center for Telehealth. That created a foundation that could be extended into vastly different hospital environments – rural and urban -- spanning medical-surgical, pediatric, oncology, emergency, neonatal, and other care settings.
Clinically, MUSC focused the virtual nursing team on repeatable, high-value workflows where experienced nurses could meaningfully reduce bedside burden, including admissions, discharge education, documentation, quality surveillance, patient education, and care coordination. As the program matured, MUSC used performance data to identify admission support, discharge teaching, and documentation compliance as particularly high-impact use cases rather than continually adding tasks simply because they could be performed virtually.
Operationally, the phased --but rapid --rollout was important. MUSC expanded progressively, learned from early adopters, collected frontline feedback, developed dedicated analytic ownership, and maintained regular executive visibility into program performance while maintaining a scalable intervention relevant to the entire health system. This approach allowed the operating model and the evidence base to mature alongside the program.
Technologically, ThinkAndor is the AI-native intelligence layer supporting the program, orchestrating virtual clinical workflows directly within the EHR so virtual nursing is embedded in the care team's work rather than operating as a disconnected application. That combination of workflow integration, centralized clinical capacity, and analytics is a critical reason the model can scale across an enterprise.
The new report proposes measuring virtual nursing across workforce experience, patient outcomes, operational performance, and financial impact. Which metrics have emerged as the most meaningful indicators of a program’s success, and why?
There is no universal virtual nursing metric. One of the most important conclusions of the MUSC work is that metrics must follow the program's value hypothesis. A program designed primarily to improve workforce resilience should not be evaluated in the same way as one designed around throughput or clinical surveillance.
The strongest approach is therefore to combine leading indicators with lagging indicators. Leading measures, such as time returned to the bedside, virtual admission completion, documentation burden, and discharge efficiency, can show within weeks whether the workflow is changing. Lagging indicators, including turnover, HCAHPS performance, cost stabilization, quality outcomes, and retention, demonstrate whether those workflow improvements translate into sustainable enterprise value over six to twelve months or longer.
At MUSC, some compelling indicators have been nurse workload and time reallocation, nurse satisfaction, turnover, labor capacity, and throughput. For example, 87 per cent of surveyed bedside nurses reported a reduction in overall workload; 76 per cent reported less time documenting admissions and discharges; 68% reported more time for patient rounding; and bedside nurse satisfaction with virtual nursing as a service averaged 4.41 out of 5. On the financial side, turnover across early-adopter units declined from 47 per cent to 30 per cent year over year, with MUSC estimating $2.4 million in associated turnover-related savings.
We are seeing that same connection between workforce impact and enterprise ROI across other health systems. At Sentara Health, virtual nursing has scaled across 118 units at 12 hospitals, giving more than 41,000 hours to care teams. Sentara has also seen a 7 per cent increase in discharges before 1 p.m., a 3 per cent improvement in patient experience when patients receive a virtual nursing touchpoint, and RN turnover below 2.5 per cent. These results reinforce why successful programs need to measure more than utilization. The real value is in capacity returned, workforce stability, throughput, patient experience, and ultimately the financial impact of those improvements.
What makes those measures meaningful is that they connect virtual nursing to outcomes executives already care about, including workforce stability, capacity, quality, patient experience, and cost, rather than simply measuring how many virtual encounters occurred.
How has AI-enabled technology, such as ThinkAndor, changed the role of virtual nurses and the way they collaborate with bedside teams?
The most important change is that technology allows the virtual nurse to become an integrated member of the clinical team rather than a remote task resource.
Within MUSC, ThinkAndor orchestrates virtual clinical workflows within the EHR and enables experienced virtual nurses to support admissions, discharges, education, documentation, quality surveillance, and care coordination. That gives the bedside nurse another experienced clinician who can take ownership of appropriate work while also serving as an additional set of clinical eyes. It’s truly an advancement in nursing care to a scalable team-based model.
AI takes that model further. Rather than requiring the virtual nurse to manually search through the chart, identify every task, and determine which patient needs attention next, AI can increasingly support documentation, surface relevant information, identify potential risks, and intelligently route work. The MUSC report specifically identifies ambient documentation, predictive deterioration monitoring, and intelligent workflow routing as emerging AI capabilities that can allow virtual nurses to oversee larger populations without sacrificing quality.
The result is a different division of labor. AI absorbs more of the information processing and workflow orchestration, while virtual nurses apply clinical judgment, and bedside nurses regain capacity for hands-on care. Nurses stay central to the model, and technology gives every member of the team more time at the top of license.
What measurable improvements has MUSC observed in areas such as nurse workload, documentation, patient education, care coordination, patient experience, or clinical outcomes since implementing virtual nursing?
The workforce results are among the clearest. In MUSC’s survey:
87 per cent of bedside nurses reported a slight or significant reduction in overall workload.
76 per cent of nurses reported spending less time on EHR documentation associated with admissions and discharges.
68 per cent of nurses said they had more time available for patient rounding.
Bedside nurse satisfaction with virtual nursing averaged 4.41 out of 5.
100 per cent of virtual nurses who participated reported being satisfied or very satisfied with the program.
Operationally, 62 per cent of bedside nurses reported improved admission and discharge throughput with virtual nursing support. The average discharge time declined from approximately 2.8 hours in July 2025 to 2.7 hours in April 2026 even as the program expanded. A 2024 analysis found that 23,516 virtual nursing encounters, averaging 36.8 minutes each, represented nursing capacity equivalent to approximately 10.2 full-time nursing positions.
Also, there are encouraging impacts on workforce economics. Overall turnover among early-adopter units declined from 47 per cent in FY24 to 30 per cent in FY25, with estimated turnover-related savings approaching $2.4 million.
MUSC provides an instructive example regarding patient education and experience. After identifying an opportunity in discharge-related HCAHPS performance at one regional unit, virtual nurses assumed responsibility for standardized discharge education and were reaching approximately 72 per cent of discharges, or 248 of 346, by July 2025, without adding bedside workload. The report says virtual nursing cohorts have remained steady on several nursing-sensitive HCAHPS measures amid broader system fluctuations, with modest improvement in "Willingness to Recommend," although continued evaluation is needed to isolate the program's contribution.
On clinical quality, virtual nurses have also identified real-time safety gaps involving pressure-injury orders, Foley catheter documentation, and CLABSI bundle adherence and enabled those gaps to be addressed before they became potential harm events. The report presents these findings as evidence of improved surveillance and risk identification. However, it does not claim that virtual nursing alone caused statistically significant reductions in falls, infections, or other clinical outcomes.
What are the biggest challenges health systems should anticipate when integrating virtual nursing into existing clinical workflows and electronic medical records, particularly around clinician adoption, interoperability, and change management?
The biggest mistake is thinking implementation is primarily a technology project. It is fundamentally a clinical operating model and change management project enabled by technology.
The first step is the workflow design. Health systems must be very explicit about what the virtual nurse owns, what remains with the bedside nurse, how handoffs occur, and when virtual support is activated. Without that clarity, organizations can create duplicated work rather than eliminating work.
The second step is clinician adoption. Frontline nurses need to see virtual nursing as an additional clinical capacity, not another system or another layer of oversight. MUSC's experience reinforces the importance of nurse leadership, frontline feedback mechanisms, and continually using data to focus the program on workflows that demonstrably reduce burden.
The third step is interoperability and data architecture. The report notes that virtual nursing data may span EHRs, HR systems, staffing platforms, operational databases, finance systems, and patient and employee surveys, often with inconsistent unit definitions and data structures. MUSC itself initially used supplemental REDCap data capture because full EHR workflow builds take time, and health system IT resources are constrained. As the program scaled, leadership alignment was needed to secure informatics resources and build the workflows more deeply into the EHR.
That is why deep EHR integration and intelligent orchestration matter. The virtual nurse should not have to operate across a collection of disconnected point solutions. The technology should bring the workflow, context, and appropriate information to the clinician while fitting naturally into the systems the care team already uses.
Finally, change management must continue after go-live. Roles, staffing models, technology configurations, and workflow scope need to evolve based on measured results. MUSC explicitly cautions against allowing virtual nursing to accumulate low-value tasks over time just because a task can be handled remotely.
Looking ahead, how do you see AI and virtual nursing evolving together, and what will be required for health systems to move from virtual nursing as a workforce-support tool to a broader care redesign strategy?
The next stage is moving from virtualizing nursing tasks to creating an AI-native model of care delivery.
Today, many virtual nursing programs begin by moving activities such as admissions, discharge education, documentation, or surveillance away from the bedside. While this creates value, it is only the reallocation of human work. The bigger opportunity is for AI to continuously reason across clinical information, identify what needs attention, prioritize work, automate appropriate administrative activity, and bring the virtual or bedside clinician into the workflow when human clinical judgment is required.
The MUSC report points directly toward this future, identifying ambient documentation, predictive deterioration monitoring, and intelligent workflow routing as capabilities that can allow virtual nursing teams to manage larger populations while maintaining quality. It argues that health systems implementing virtual nursing today need technology infrastructure and governance capable of supporting those capabilities as they mature.
We are already seeing evidence of that evolution at scale. Sentara has expanded virtual nursing across 118 units at 12 hospitals, giving more than 41,000 hours to care teams while improving early discharges, patient experience, and nurse retention.
For health systems, this advancement requires several shifts: moving away from point solutions toward an integrated clinical orchestration layer; embedding AI and virtual care directly into the EHR and existing workflows; establishing governance that allows AI to operate safely with clinician oversight; designing the model around measurable outcomes rather than technology utilization; and measuring performance across the full enterprise, including workforce, patient experience, quality, capacity, and economics.
That is why the MUSC experience matters. It shows that virtual nursing can evolve beyond a response to the nursing shortage. When the clinical workforce, AI, workflows, data, and measurement strategy are designed together, virtual nursing becomes part of a broader AI-native clinical services model that can redesign how care is delivered across the health system.
By Shraddha Warde|