Article author and evidence
Dr. Kainat Amjad, MBBS
Dr. Kainat Amjad is a medical doctor and practice growth strategist who writes about healthcare operations, patient-facing workflows, administrative burden, and practical technology adoption. Her work focuses on helping providers improve the business side of care without weakening accountability, privacy, or the patient experience.
Hospital readmissions are one of the most stubborn problems in American healthcare. A patient gets discharged, seems stable, and within 30 days ends up back in a hospital bed. For providers, this raises a real question: can a virtual care coordinator actually prevent that cycle, or is this just another tech trend dressed up as a solution?
The short answer is yes, but the data is more specific than most marketing claims suggest. A 2025 study published in JMIR Medical Informatics gives one of the clearest pictures yet of what virtual care coordination actually does, and where it works best.
How Big Is the Hospital Readmission Problem?
Hospital readmissions carry a documented financial and clinical burden. According to the NIH StatPearls clinical reference on reducing hospital readmissions, approximately 20% of Medicare beneficiaries experience a readmission within 30 days of discharge. This is not a small operational inconvenience. In 2010, the Affordable Care Act formally tied readmission rates to CMS reimbursement, and by 2013 the Hospital Readmission Reduction Program (HRRP) was penalizing hospitals with high readmission rates across most facility types.
Not every readmission is preventable. Some reflect disease progression or unrelated new problems. But a systematic review of 34 studies found the median proportion of preventable readmissions was 27%, with individual studies ranging from 5% to 79%. A separate observational study of 1,000 patients readmitted within 30 days found a similar figure: 27% were considered potentially preventable, meaning there was more than a 50% chance the readmission could have been avoided with different care.
Source: NIH StatPearls, Reducing Hospital Readmissions (NBK606114).
That 27% is the target zone for virtual care coordination. It is not a claim that coordinators eliminate readmissions. It is a specific, addressable slice of the problem tied to communication gaps, medication issues, and follow-up failures.
Where Do Most Preventable Readmissions Actually Start?
The StatPearls review breaks down exactly where readmissions originate, and the pattern is consistent with what practice administrators already suspect.
During the hospital stay, the biggest issue is handoff quality. A meta-analysis found that only 12% to 34% of discharge summaries reached the aftercare provider by the time of the patient’s first follow-up appointment. When a primary care provider walks into a follow-up visit without the discharge summary, they are working blind. Missing or incomplete documentation on medication changes, pending test results, and follow-up plans directly increases readmission risk.
After discharge, the picture gets worse. Only about half of Medicare beneficiaries who were later readmitted within 30 days had actually completed a follow-up visit with any clinician before that readmission happened. Roughly 20% of patients experience an adverse event after discharge, and medication-related issues are the most common cause. About two-thirds of those adverse events were preventable or could have been mitigated with better oversight.
None of these are clinical failures in the traditional sense. They are coordination failures. The patient did not get worse because a physician made a bad call. They got worse because nobody was watching the gap between hospital and home.
What Did the UC San Diego Virtual Transitions of Care Study Find?
This is where the virtual care coordinator model gets tested directly. UC San Diego Health built a hospitalist-led virtual transitions of care (VToC) clinic and tracked outcomes over three years, from September 2021 to September 2024, across more than 25,000 patients.
The program worked like this. Patients discharged from the hospital medicine service line were automatically flagged for referral if their LACE+ index, a validated readmission risk score, fell between 50 and 75. Twelve hospitalists, two medical assistants, one pharmacist, and an on-demand interpreter service supported the clinic. Medical assistants scheduled video visits, patients got a reminder call the day before, and a pharmacist reviewed medications ahead of the visit to catch fill gaps, dosing problems, and interaction risks. After the visit, a standardized handoff went to the patient’s primary care provider and relevant specialists, and a separate note went back to the discharging hospitalist to flag any care lapses.
Source: JMIR Medical Informatics, Virtual Transition of Care Clinics and Associated Readmission Outcomes, UC San Diego Health (PMC12504893).
That is not a rounding difference. The effect was not uniform across all patients either. The researchers found the largest reduction in the moderate-risk group, exactly the LACE+ 50 to 75 band the program targeted. That detail matters for any practice thinking about building something similar. This is not a blanket intervention that helps everyone equally. It works best when it is targeted at the patients most likely to benefit from a structured follow-up, not the lowest-risk patients who would likely do fine anyway, and not always the highest-risk patients whose readmission drivers may be less coordination-dependent.
The study authors were also transparent about limitations. About 13% of patients referred to the clinic could not be reached, and the researchers noted this could introduce bias since those patients might differ in ways that affect outcomes. It is an observational study, not a randomized trial, so selection effects are possible even after risk-matching the comparison group. That is a normal caveat for real-world operational data, and it does not erase the finding, but it is worth stating plainly rather than glossing over.
What Is LACE+, and Why Does the Risk Score Matter?
Any practice considering a similar program needs to understand LACE+ before trying to replicate it, because the targeting is what made the UCSD numbers work, not the video visit itself.
L: Length of stay
How many days the patient spent in the hospital during the index admission.
A: Acuity
Whether the admission was emergent or urgent rather than planned.
C: Comorbidities
A weighted count of conditions the patient carries into discharge.
E: ED visits
Number of emergency department visits in the six months before this admission.
Each component is scored and summed into a single number, typically running from 0 to somewhere in the high 90s depending on the version used. The reason 50 to 75 mattered in the UCSD program is that it captures patients who are neither low-risk nor so acutely unstable that a structured outpatient follow-up program cannot meaningfully change their trajectory. A patient scoring below that range is often stable enough that standard discharge instructions and a routine PCP visit are sufficient. A patient scoring well above it may need inpatient-level case management, home health, or a skilled nursing placement rather than a single virtual visit.
For a practice without an existing LACE+ calculation built into its EHR, the practical starting point is simpler than it sounds. Most EHR systems either have a LACE or LACE+ module available or can approximate it using length of stay, admission type, a comorbidity count already captured for risk adjustment purposes, and ED utilization history that is already tracked for quality reporting. The scoring itself is not the hard part. The hard part is building the referral trigger so that patients in the target band are automatically flagged rather than relying on a discharging clinician to remember to make a referral on a busy shift.
Why Does Virtual Coordination Close the Gaps That Cause Readmissions?
Line up the UCSD program’s design against the failure points identified in the StatPearls review, and the fit is direct rather than coincidental.
| Readmission risk factor | How virtual coordination addresses it |
|---|---|
| Only 12% to 34% of discharge summaries reach the aftercare provider in time | A standardized electronic handoff is sent to the PCP and specialists after every VToC visit, closing the documentation loop a busy hospital discharge process often leaves open |
| Medication issues are the leading cause of post-discharge adverse events | A pharmacist reviews fill history and dosing before the visit, catching problems before they become emergency department visits |
| Only half of readmitted Medicare patients completed a follow-up visit beforehand | Reminder calls and scheduled video visits remove transportation barriers, one of the most common reasons patients skip in-person follow-up |
| Universal follow-up spreads staff time thin without improving outcomes proportionally | The LACE+ referral trigger focuses resources on the moderate-risk population where intervention has the highest documented payoff |
Why Do Some Virtual Care Programs Backfire?
Not every virtual care initiative earns the trust of the clinicians working alongside it, and it is worth being honest about that rather than pretending every virtual program is automatically welcomed.
Frontline nursing staff at hospitals that have rolled out virtual nursing and tele-sitter programs have raised consistent concerns in professional forums: constant messaging that interrupts direct patient care, virtual staff monitoring more patients than they can meaningfully attend to, and, in some cases, a rollout that was framed as staffing-neutral but coincided with bedside staffing reductions. Similar skepticism shows up around fully remote case managers, where physicians report that case managers became harder to reach once they moved off-site, slowing down the exact coordination work they were supposed to speed up.
This is not an argument against virtual care coordination. It is a reminder that the model only works when it is built the way UCSD built it, not as a thinner version of the same headcount. The elements that made the difference were specific: hospitalist oversight rather than an unsupervised call center function, a pharmacist actually reviewing medications rather than a coordinator forwarding a pharmacy question later, and a standardized handoff that closed the loop rather than generating another message for an already busy PCP to triage. A virtual layer added on top of unchanged staffing, without those clinical components, is a different intervention than the one the data supports, and providers are right to be skeptical of the difference.
What Mistakes Do Practices Make When Building a Virtual Coordination Program?
Treating it as a call center function
The UCSD model combined clinical oversight, medication review, and administrative scheduling into one coordinated workflow. Stripping out the clinical and pharmacy components and leaving only scheduling will not replicate the result.
Skipping risk stratification
Following up with every discharged patient at the same intensity burns staff capacity without improving outcomes proportionally. LACE+ or a comparable tool prioritizes outreach toward patients most likely to benefit.
Underestimating technical barriers
Fewer than 5% of UCSD visits converted from video to phone, but only because the workflow planned for a phone fallback rather than assuming every patient can complete a video visit.
Ignoring the handoff step
The visit is only half the intervention. Without a standardized handoff back to the PCP and discharging provider, the documentation gap that drives readmissions persists.
Not staffing pharmacy review
Medication reconciliation is not a nice-to-have. Given that medication issues drive the largest share of post-discharge adverse events, a program without pharmacist involvement is missing the highest-risk component.
Presenting it as staffing-neutral when it isn’t
A program layered on top of existing responsibilities without dedicated staffing tends to generate the same frustration frontline clinicians describe with poorly implemented virtual nursing programs: more messages, no more capacity.
What Does It Take to Run This Operationally?
Setting up virtual care coordination touches several administrative functions at once: scheduling, medical assisting, pharmacy review, and documentation routing back into the EHR. For most practices, the barrier is not clinical willingness. It is staffing bandwidth. Finding and training medical assistants who can manage discharge-based scheduling on top of existing patient panels, along with pharmacist time for medication review, is a real operational lift, particularly for practices already running lean.
The day-to-day mechanics matter more than they might seem. A pharmacist review that happens the morning of the visit, using an updated fill history pulled from the pharmacy benefit or state prescription monitoring database, catches far more than a review done from stale chart notes. Flagging a gap between what was prescribed at discharge and what was actually filled gives the coordinator something concrete to raise with the patient before the visit even starts, rather than discovering it live on camera.
The handoff routing works the same way. A standardized note that goes to the PCP and to the discharging hospitalist, built as a structured template rather than free text, gets read faster and acted on faster than a narrative summary buried in a fax queue. Practices that skip building this template tend to see the same handoff failure the intervention was supposed to fix, just shifted one step later in the process.
What Compliance Controls Apply to Virtual Care Coordination?
Any virtual care coordination program handling patient scheduling, medication history, or discharge documentation needs to operate under HIPAA-compliant controls, not a claimed certification, since HIPAA compliance is a program of administrative, technical, and physical safeguards rather than a certification a vendor or practice receives. Documentation routing between the coordination team, the EHR, and outside specialists should follow the same access controls and audit logging a practice already applies to other protected health information workflows.
What Are Healthcare Providers Asking About Virtual Care Coordination?
Practices evaluating virtual care coordination are not just weighing whether the model works. They are describing past programs that added messages without adding staff, remote case managers who became harder to reach, and a fear that this becomes another call center function with better branding. The pairs below connect those concerns to what the data actually shows.
We tried a virtual program before and it just added more alerts without adding staff. Why would virtual care coordination be any different?
It depends entirely on whether the program adds dedicated staff or just adds messages to an unchanged team. The UC San Diego program worked because it added twelve hospitalists, two medical assistants, and a pharmacist to the workflow, not because it layered alerts on top of existing headcount. Programs that generate frustration are almost always the alerts-without-staffing version. If a proposed program cannot name the additional staffing it comes with, that is the detail to press on before rollout.
Source: JMIR Medical Informatics, UC San Diego Health VToC studyOur case managers went remote and became harder to reach. Won’t a virtual care coordinator run into the same problem?
That risk is real when follow-up depends on the patient successfully reaching someone. The UCSD model avoided it by scheduling the visit proactively, with a reminder call the day before, rather than waiting for the patient to initiate contact, and by requiring a standardized handoff back to the PCP regardless of how the visit went. Only about half of readmitted Medicare patients had completed any follow-up visit beforehand, which is the access gap this structure is built to close. The structure is what matters, not whether the role is remote.
Source: NIH StatPearls, Reducing Hospital ReadmissionsIs this just a call center function with better branding, not something that actually changes readmissions?
In the UCSD data, it worked specifically because it was not a call center function. Hospitalists retained clinical oversight, a pharmacist reviewed medications before every visit, and referrals were targeted using the LACE+ risk score rather than offered to everyone equally. Strip out the clinical and pharmacy components and keep only scheduling, and the result is a different intervention than the one that produced the 14.9% versus 20.1% readmission gap. The branding is not what closed the gap. The staffing mix was.
Source: JMIR Medical Informatics, UC San Diego Health VToC studyDoes this only work for large hospital systems with the staff to run a dedicated clinic?
The clinical structure was hospitalist-led because it matched UC San Diego’s own setting, but the underlying components, risk-based referral, pharmacist medication review, scheduled follow-up, and standardized handoffs, scale down to a practice’s own discharge volume. A smaller practice needs proportionally less staffing, not a fundamentally different model, as long as it targets the same moderate-risk LACE+ band instead of trying to follow up with every discharged patient equally.
Source: JMIR Medical Informatics, UC San Diego Health VToC studyWhat Are Providers Asking About Virtual Care Coordination and Readmissions?
Does a virtual care coordinator replace in-person follow-up entirely?
No. In the UC San Diego study, virtual visits were converted to telephone encounters in fewer than 5% of cases due to technical barriers, and the model was designed to supplement, not replace, in-person care when clinically indicated.
Which patients benefit most from virtual care coordination?
The UC San Diego data showed the strongest effect in patients with a LACE+ index between 50 and 75, a moderate readmission risk band. Very low-risk patients may not need the intervention, and the benefit for the highest-risk patients was less pronounced in this study.
Is this only relevant to hospital systems, or can outpatient practices use a similar model?
The clinical structure was hospitalist-led because it was tracking hospital discharges directly, but the underlying components, risk-based referral, pharmacist medication review, scheduled follow-up, and standardized handoffs, can be adapted by any practice managing post-discharge or post-procedure follow-up, scaled to the practice’s patient volume and risk profile.
What is the biggest operational risk in building this kind of program?
Based on the failure points identified in the broader readmissions literature, and echoed by frontline clinicians in other virtual care rollouts, the two most common gaps are incomplete discharge handoffs and missed medication reconciliation. A program that addresses scheduling but skips these two elements is unlikely to see a comparable reduction in readmissions.
What is the LACE+ index?
LACE+ is a readmission risk score built from length of stay, acuity of admission, comorbidity burden, and emergency department visits in the prior six months. It is used to identify which discharged patients are most likely to benefit from structured follow-up.
How is virtual care coordination different from a telesitter or virtual nursing program?
Telesitters and in-hospital virtual nursing monitor patients during an inpatient stay. Virtual care coordination, as tested in the UC San Diego study, is a post-discharge outpatient function focused on medication review, scheduling, and handoff documentation for patients who have already left the hospital.
What Are Healthcare Leaders Asking About Reducing Readmissions?
Healthcare leaders want to reduce readmissions without adding a program that generates more noise than value. These answers focus on staffing, targeting, and the operational details that separate a working program from a stalled one.
How much staffing does a virtual care coordination program actually need?
The UC San Diego program supported over 25,000 patients across three years with twelve hospitalists, two medical assistants, and one pharmacist, working alongside their existing clinical duties. Smaller practices should scale proportionally to their discharge volume within the target LACE+ band, not to their total patient panel.
Do I need LACE+ built into my EHR to start?
No. Many EHRs can approximate the score using length of stay, admission type, an existing comorbidity count, and ED utilization history already tracked for quality reporting. The harder part is building an automatic referral trigger rather than relying on a discharging clinician to remember.
Can this program work without a dedicated pharmacist?
The data suggests it loses much of its effect without one. Medication-related issues are the leading cause of preventable post-discharge adverse events, and the UCSD model’s pharmacist review caught fill gaps and dosing problems before they reached the patient.
How do we avoid the alert fatigue frontline staff report with other virtual programs?
Scope the coordinator’s role narrowly around scheduling, medication review, and handoff documentation rather than general monitoring or messaging. Programs that generate frustration tend to be layered on top of unchanged staffing rather than given a defined, adequately resourced function.
Which Sources Support This Readmission Reduction Guide?
Ready to Build Post-Discharge Follow-Up That Actually Reduces Readmissions?
Map the discharge volume, the risk band you want to target, the staffing it requires, and the handoff documentation that needs to close the loop back to your PCPs.