The Invisible Crisis No One Wants to Quantify

There is a narrative about telehealth burnout that the industry has been repeating for the past three years. It goes like this: virtual care exploded during the pandemic, providers got overwhelmed by patient volume, and now they're exhausted. It's a tidy story. It's also incomplete — dangerously so.

The real driver of provider burnout in telehealth isn't the number of patients on the schedule. It's everything that happens after each patient leaves the virtual waiting room. The charting. The coding. The compliance paperwork. The after-visit notes that pile up like a second workday hidden inside the first one.

This pattern is now visible in published EHR log studies and health-system data. The trend is consistent: documentation and inbox work are consuming an outsized share of clinician time, especially as telemedicine has become a core mode of care delivery.

36.2 min Median EHR time per 30-minute primary care visit (JAMA Netw Open, 2023)
6.2 min Median after-hours EHR time per visit ("pajama time")
~2:1 EHR task time to face-to-face clinical time in primary care logs
43.2% U.S. physicians reporting at least one burnout symptom (AMA, 2024)

These are not abstract statistics. They represent clinicians who entered medicine for patient care, but now spend much of their day in documentation and digital message management. Telemedicine has expanded access and convenience, but without workflow redesign, it can also expand after-hours administrative load.


The Anatomy of a Documentation Burden

To understand why this problem is so persistent, you need to trace a single telehealth visit from start to finish — not the patient experience, but the provider's. Here is what that journey actually looks like:

Before the Visit: The Preparation Tax

Before a provider even joins the video call, they spend 4-7 minutes reviewing the patient's existing chart, recent lab results, prior visit notes, and any pre-visit questionnaire responses. In a well-integrated system, this information is consolidated. In most telehealth setups, it means opening three to five different tabs or applications and mentally synthesizing the information.

This preparation phase is invisible in scheduling metrics. It doesn't show up as "provider time" because the appointment hasn't technically started. But it's real cognitive work, and for a provider running 20+ visits a day, it adds up to over two hours of untracked labor.

During the Visit: The Dual-Task Trap

In a physical exam room, a provider can listen, observe, and dictate notes simultaneously — often with the support of a medical scribe. In a telehealth visit, the provider is alone. They're simultaneously managing the video connection, listening to the patient, taking mental notes, and often attempting to type abbreviated notes in real time.

Cognitive science research is clear on this: humans do not truly multitask; they task-switch. Frequent switches between patient conversation, documentation, coding prompts, and inbox tasks increase mental load and reduce available attention for clinical reasoning.

"I realized I was spending more time during the visit thinking about what I would write in the note than actually listening to my patient. That's when I knew something was fundamentally broken." — Internal provider survey, Q1 2026

After the Visit: Where the Hours Disappear

The visit ends. The patient disconnects. And now the real work begins.

A typical post-visit documentation sequence for a telehealth encounter includes:

  1. SOAP note creation — Capturing the subjective story, objective findings, assessment, and plan in a format suitable for continuity and billing.
  2. Diagnosis coding — Selecting ICD-10 codes that accurately reflect the encounter and support compliant reimbursement.
  3. Prescription management — Entering, verifying, and e-prescribing medications with interaction and formulary checks.
  4. Referral documentation — Preparing referrals with relevant history, urgency, and follow-up instructions.
  5. Patient instructions — Writing plain-language after-visit summaries, red-flag guidance, and next steps.
  6. Compliance documentation — Capturing consent, telehealth modality details, and jurisdiction-specific requirements where applicable.

Even when each subtask is short, the cumulative burden is substantial. Published studies show clinicians often spend equal or greater time in the EHR than in direct patient interaction, and this burden often spills into evenings and weekends.

The Pajama Time Epidemic

The industry term for after-hours EHR work is "pajama time." In a 2023 JAMA Network Open study of 307 primary care physicians, median pajama time was 6.2 minutes per visit, with wide variation across clinics. On high-volume clinic days, that quickly adds up and erodes recovery time.


Why Telehealth Amplifies the Problem

There's a pervasive assumption that telehealth should reduce administrative burden. Everything is digital. There's no room to clean, no physical charts to file, no waiting room to manage. The logic seems obvious — until you examine the actual workflow.

The Context-Switching Tax

A telehealth provider's desktop during a single visit typically includes the video platform, the EHR, e-prescribing tools, payer or prior-auth workflows, and sometimes separate diagnostic viewers. As these systems fragment, clinicians spend more effort navigating software boundaries rather than following a single clinical workflow.

Each switch carries a cognitive cost. Human-factors research consistently shows task-switching degrades efficiency and increases error risk, especially under time pressure.

The Compliance Multiplier

Telehealth visits that cross state or jurisdictional lines create exponentially more compliance overhead. Consider a provider licensed in three states seeing patients in all three on the same day. Each state may have different requirements for:

  • Informed consent documentation format
  • Prescribing restrictions (especially for controlled substances)
  • Visit documentation retention periods
  • Patient notification requirements for AI-assisted documentation
  • Telehealth-specific billing codes and modifiers

A single multi-state practice may need to maintain compliance with 5-10 different regulatory frameworks simultaneously. This isn't an edge case — it's the reality for a growing percentage of telehealth providers as cross-state licensure compacts expand.

The Missing Scribe

In traditional clinical settings, medical scribes absorb a significant portion of the documentation burden. They listen to the encounter, draft the clinical note in real time, and hand the provider a near-final document for review and signature. The provider's role shifts from "author" to "editor" — a fundamentally faster cognitive task.

In telehealth, scribes are rare. The economics don't support it for most virtual care platforms, and the logistics of a remote scribe listening to a telehealth call introduce their own privacy and workflow complications. The result: the provider becomes their own scribe, adding an entirely separate job function to every visit.


The Downstream Cost: Beyond Provider Wellbeing

It's tempting to frame documentation burnout as a "provider wellness" issue — important, but ultimately a workforce management problem. This framing vastly underestimates the downstream impact.

Patient Safety Erosion

Fatigued providers take shortcuts. One common shortcut is copy-forward (or "note cloning") — copying prior notes and editing them for the current encounter. Studies across EHR environments have shown substantial duplicated text in notes, which can undermine clarity and increase the risk of stale information.

The danger is subtle but serious. Cloned notes can propagate outdated information — a resolved allergy that still appears active, a discontinued medication still listed as current, a symptom marked as "new onset" that was actually reported six months ago. Each propagated error compounds over time, creating a clinical record that drifts further from the patient's actual health state.

Delayed and Degraded Care

When documentation piles up, everything downstream slows down. Referral letters are delayed. Prescription renewals sit in queue. Lab result reviews get pushed to the end of the day — or the next day. For the patient, this translates to delayed diagnosis, delayed treatment, and a growing sense that their provider doesn't have time for them.

This perception is not imaginary. High documentation load reduces cognitive bandwidth, making communication, synthesis, and follow-up decisions harder at the end of long clinic sessions.

The Financial Hemorrhage

The cost of replacing a single physician who leaves due to burnout is estimated at $500,000 to $1,000,000 — factoring in recruitment, credentialing, onboarding, temporary coverage, and lost patient relationships. For a telehealth startup operating with a lean provider network, losing even one physician can destabilize the entire operation.

But the financial impact extends beyond turnover. Rushed documentation increases coding inconsistency, denials, and rework, all of which raise administrative costs and delay revenue realization.


What Intelligent Automation Can — and Cannot — Fix

At this point in the conversation, someone always says: "Just use AI to write the notes." It's the obvious suggestion. It's also, in its naive form, the wrong one.

Unsupervised AI-generated clinical documentation introduces a different set of risks that are equally dangerous: hallucinated clinical details, incorrect drug dosages, fabricated patient history, and liability exposure from notes that no physician actually reviewed. The solution to documentation burnout cannot be removing the physician from the documentation process entirely. It has to be restructuring the process so that the physician's role is focused where it matters most.

The Right Model: From Author to Editor

The most effective approach — the one that the evidence supports and that we've implemented at LivoRx — is what we call the Doctor-in-the-Loop model. Here's how it works:

  1. Real-time structured capture: During the telehealth visit, the conversation is processed through a clinical NLP pipeline that extracts structured data — chief complaint, history of present illness, review of systems, assessment, and plan — in real time.
  2. Automated SOAP draft generation: Within seconds of the visit ending, a complete SOAP note draft is generated. This isn't a transcript — it's a structured, clinically formatted note with proper terminology, organized sections, and appropriate detail level.
  3. PII redaction at the pipeline level: Before any clinical text reaches an AI model, it passes through our automated PII redaction layer. Patient names, dates of birth, and identifiers are tokenized; only anonymized clinical content enters the processing pipeline.
  4. Physician review and attestation: The provider reviews the draft, makes corrections or additions, and signs the note. Their role has shifted from "author writing from scratch" to "editor reviewing a well-structured draft" — a task that takes 1-2 minutes instead of 8-12.

The LivoRx Impact: By the Numbers

In our internal pilot cohort using the automated SOAP pipeline, clinicians reported:

Post-visit documentation time reduced from 16 minutes to under 2 minutes (87% reduction)

Pajama time eliminated entirely — all documentation completed within the workday

Note quality scores improved by 14% as measured by clinical audit

Provider satisfaction scores increased 31 points on standardized burnout assessment

What Automation Cannot Replace

It's equally important to be honest about the limits. Automation cannot replace clinical judgment. It cannot determine whether a patient's vague complaint about fatigue warrants a thyroid panel or a mental health referral — that decision requires the physician's training, intuition, and the nuanced understanding that comes from the human interaction itself.

The goal of intelligent documentation automation is not to practice medicine. It's to ensure that the physician's limited cognitive bandwidth is spent on practicing medicine, not on translating what they already know into a format that the EHR, the billing system, and the compliance framework can consume.


The Systemic Failure We Need to Name

Here is the uncomfortable truth that the telehealth industry needs to confront: the documentation burden is not a technology problem. It's a design problem. We built telehealth platforms that digitized the video call but left the documentation workflow essentially unchanged from its paper-based origins.

Think about it. We have real-time video with adaptive bitrate streaming, sub-second appointment scheduling, AI-powered triage — and then we ask the provider to manually type a note into a text box. It's the equivalent of building a self-driving car but requiring the driver to fill out a paper log of every turn they made.

The documentation layer of telehealth was never thoughtfully designed. It was grafted on from legacy EHR workflows. And providers have been absorbing the cost of that design failure with their time, their health, and ultimately their career longevity.


The Path Forward: Three Principles for Sustainable Telehealth

If the telehealth industry is serious about addressing provider burnout — not with wellness apps and meditation breaks, but with structural solutions — three principles need to guide the next generation of platform design:

1. Documentation Must Be a Byproduct, Not a Task

The clinical note should be generated from the visit, not about it after the fact. When structured clinical data is captured during the encounter — through NLP, structured intake, and automated coding — the note becomes an output of the visit workflow, not a separate task appended to it.

2. Physician Oversight Must Be Preserved, Not Eliminated

The answer to documentation burnout is not to remove the physician from the loop. It's to move them from the "creation" step to the "review" step. This distinction matters enormously — both for clinical safety and for provider trust. A physician who reviews and signs an AI-drafted note maintains accountability and clinical ownership. A physician whose notes are generated and filed without their review has lost control of their own clinical record.

3. Privacy Cannot Be an Afterthought

Any system that processes clinical conversations through an AI pipeline must redact patient-identifiable information before that data reaches external models. This isn't a feature to add later — it's the foundation that everything else must be built on. (For a deeper exploration of this, see our companion piece: The Urgent Case for Automated PII Redaction.)

  • Arndt et al., Annals of Family Medicine (2017): primary care physicians spent nearly 2 hours on EHR tasks per hour of direct patient care.
  • Rotenstein et al., JAMA Network Open (2023): median 36.2 minutes total EHR time and 6.2 minutes of pajama time per visit in a large primary care sample.
  • Holmgren et al., JAMA Internal Medicine (2023): telemedicine expansion associated with increased physician EHR time and message volume.
  • American Medical Association (2025 report of 2024 data): 43.2% of physicians reported at least one symptom of burnout.

The Choice Ahead

The telehealth industry is at an inflection point. The providers who built virtual care during the pandemic are burning out. The next generation of physicians — digital natives who expect technology to amplify their work, not multiply their paperwork — are entering the workforce with different expectations.

The platforms that solve the documentation burden thoughtfully — with physician-supervised automation, privacy-first data handling, and workflows designed around clinical reality rather than billing convenience — will define the next decade of virtual care.

The ones that don't will lose their providers. And without providers, there is no platform.

We built LivoRx because we believe telehealth can be better than this. Not incrementally better — fundamentally better. And it starts with giving providers back the one resource they value most: their time.