A Remote Nurse Triage System -
Designed to
Measurably Improve
Patient Outcomes
Explore a Remote Nurse Triage System Designed to “Make It Easier To Do the Right Thing”.(Institute of Medicine, 2000)
*
*
Explore a Remote Nurse Triage System Designed to “Make It Easier To Do the Right Thing”.(Institute of Medicine, 2000) * *
Nursing Informed & Designed
Nursing Process Driven
Transparent, Explainable, Consistent, Familiar, User Friendly
A Modern Remote Nurse Triage Guideline Design
TeleTriage Systems’ Design Advantage and the Role of AI
TeleTriage Systems’ Guideline Task Force conceptual model (Structure - Process - Outcome) is already human-centered rather than purely algorithmic. That matters because the strongest future triage AI systems are unlikely to function as rigid decision trees.
Instead, nurses will probably function as: expert contextual interpreters (Lephrohon & Patel, 1995) operating within structured safety boundaries — much closer to the Wheeler/Guideline Task Force approach than to older deterministic protocol engines.
TeleTriage Systems’ Guideline Task Force approach and design already contains concepts that resemble modern AI reasoning. An AI upgrade may not require replacing the framework, AI may become a support layer around the framework. For example, the 5-level, explicit acuity system, simplified for nurse, patient, researcher, and AI systems to understand, represents an important triage concept, functioning better than disposition schemes.
For remote clinical triage, the safest and most robust approach is to treat AI as one component of an integrated, evidence-informed, comprehensive clinical system, rather than as the system itself. Such a framework provides the structure that enables AI, clinicians, and clinical guidelines to work together to support consistent, transparent, and safer patient care. TeleTriage Systems is one such system, with a 40+ year track record to attest to its standing as a leader in the field, while working to achieve EBM status.
QUESTION TO ChatGPT with ChatGPT Response (Unedited)
Can AI-augmented remote clinical triage perform safely without a foundational and complete system -- clinical training, qualified experienced staff, evidence-based CDSS, reliable EMR, clinical standards?
In general, no. The available evidence and current regulatory thinking do not support deploying AI-augmented remote clinical triage as a safe, standalone capability. AI can be a valuable component, but it is not a substitute for a complete clinical triage system.
Remote triage is a high-risk sociotechnical system. Safety depends on the interaction of multiple components, not on the performance of the AI alone. Even a highly accurate AI model can produce unsafe outcomes if the surrounding clinical infrastructure is inadequate.
A comprehensive remote triage system typically requires:
Essential ComponentWhy it mattersClinical trainingNurses must recognize atypical presentations, identify red flags, detect cognitive biases, communicate effectively, and know when to override decision support.
Qualified, experienced staffClinical judgment is essential for interpreting incomplete, ambiguous, or conflicting information that AI may not resolve.
Evidence-based CDSSStructured, validated protocols reduce variability and ensure consistent risk assessment. AI should complement—not replace—these protocols.
Reliable EMR and documentationAccurate medical history, medications, allergies, and prior encounters are often essential to safe disposition decisions.
Clinical standards and governancePolicies define scope of practice, escalation pathways, supervision, documentation requirements, quality assurance, and continuous improvement.
Quality monitoringRegular auditing of calls, disposition accuracy, adverse events, and protocol compliance identifies safety issues before they become systemic.
Risk management and legal oversightDue care requires documentation, competency validation, incident review, and ongoing maintenance of the system.
Why AI alone is insufficient AI has important limitations in remote symptom triage:
Patients often provide incomplete or inaccurate histories.
Serious illnesses frequently present with nonspecific symptoms.
AI cannot directly observe the patient or perform a physical examination.
Clinical context—including comorbidities, medications, age, and social factors—often changes the appropriate disposition.
AI systems can hallucinate, misclassify rare presentations, or perform less well outside their training data.
AI cannot assume legal accountability for clinical decisions.
These limitations become especially important because remote triage is performed under uncertainty. Safe practice depends on managing that uncertainty through trained clinicians, standardized protocols, and organizational safeguards.
AI's appropriate role The strongest current use case is AI as decision support, not autonomous decision-making.
AI can assist by:
organizing symptom information,
suggesting differential diagnoses,
highlighting red-flag symptoms,
recommending evidence-based questions,
identifying possible protocol pathways,
summarizing documentation,
checking consistency,
flagging potential errors or omissions.
The clinician remains responsible for:
obtaining an adequate history,
interpreting context,
exercising professional judgment,
determining the final triage disposition,
documenting the rationale.
A systems engineering perspective
Patient safety in remote triage is often described using the "Swiss cheese" model of safety: multiple independent layers of defense compensate for the limitations of any single layer. AI should be viewed as one additional safety layer, not the entire system.
A simplified hierarchy might look like this:
Organizational governance
Standards and policies
Qualified clinicians
Clinical education and competency assessment
Evidence-based triage protocols (CDSS)
Reliable EMR and documentation
AI decision support
Continuous quality improvement and outcome monitoring
If any of these major layers are absent, the overall safety of the system decreases.
Implications for legal due care
From a legal and risk-management standpoint, demonstrating due care generally requires evidence that the organization maintained a safe clinical system, including:
qualified personnel,
competency-based training,
validated clinical protocols,
appropriate supervision,
documentation standards,
quality assurance,
governance and continuous monitoring.
Simply showing that an AI system was used would generally not, by itself, demonstrate fulfillment of the duty of due care. Courts and regulators are more likely to evaluate whether the healthcare organization implemented and maintained an integrated system reasonably designed to produce safe patient outcomes.
Bottom line: AI can substantially enhance remote clinical triage, but current evidence supports its use within a comprehensive clinical system—not as a replacement for clinical training, experienced staff, evidence-based protocols, reliable health records, governance, and quality oversight. The safety of remote triage is an emergent property of the entire system rather than any single technology.
-
About Sheila Quilter Wheeler, RN, MS
It’s Been A Journey. Sheila Quilter Wheeler had no intention of pioneering a new nursing subspecialty. While managing advice calls as an ED nurse, she suggested guidelines for the process, her new mission — find solutions by developing this remote nurse triage clinical practice.
Initially, she suggested two books by physician-authors Vickery, Fries and Pantell to management, for the lay audience. Still in print, these resources contain instructions for assessing, triaging, first aid and home treatment for common adult and pediatric symptoms.
In the early 1990’s, telephone triage was an embryonic field, fraught with uncertainty. Many nurses were worried about their licenses -- anxious to know how to practice remote triage safely.
As a clinical call center advice nurse, Wheeler developed and taught an in-house telephone triage class. The training syllabus gradually evolved into a book proposal. The proposal was subsequently accepted by Delmar Publishers after nine publisher rejections.
Wheeler and medical writer, Judith Windt completed the training manual, including real-life case-study audiotapes in 1993,
Due to the ongoing scarcity of pre-existing research, Wheeler extrapolated, applied and relied upon research from foundational clinical subspecialties.
From 1993-1995, Ms. Wheeler served as Editor-in-Chief, directing a 23-member expert Nurse Task Force developing the first and only three volume, specific age-based, five-level triage guidelines. Ms. Wheeler served as founder of the first national Telephone Triage conference.
From 1995-2003 she served as an expert witness on 35 malpractice cases, gradually noticing patterns of recurrent error (Atul Gawande, MD). She began devising solutions —safety prompts — integrated into clinical training, the guidelines and new rules of thumb.
-
A Nurse-Driven Remote Triage System Evolves
It Took a Village Ms. Wheeler, colleagues, nurse-consultants, and a team of of 23 + Nurse-experts and three physician-reviewers developed a complete remote nurse triage system.
Two nurse legal experts —Barbara Siebelt, RN, MSN, and Laura Mahlmeister, RN, PhD. emphasized the importance of having a "paper trail" should malpractice arise — written evidence of a complete system to bring to court — representing the effort to fulfill the “duty of due care”.
In “From Novice to Expert” Pat Benner, RN, PhD developed the concept of nursing expertise. Ms. Wheeler adapted the concept to the emerging practice. Carolyn Smith Marker, RN, MS, nursing standards expert, recommended specialized standards.
Robert Smith, JD, advised that remote triage system components be integrated, cautioning against having “just bits and pieces". He added that each component served as a “layer of protection” — “an overcoat for safety”, working together.
Jeff Clawson, MD, Pioneer of 911 - Emergency Medical Dispatch, served as Ms. Wheeler’s mentor, paving a path with his training manual, rules of thumb and devotion to patient safety. Ms. Wheeler followed suit with training manual, training, standards and triage tools
Remote Triage Research
Researchers discovered that nurses relied on symptom pattern recognition, patient context, heuristics as clinical decision-making strategies for triage, (Lephrohon, Patel, 1995). They theorized that medical diagnoses are unnecessary in remote nurse triage.
Nurse-Driven Guidelines
For two years, Ms. Wheeler. and a team of 23+ expert nurses developed the first 5-level, 3-volume, Age-based Guidelines. They also developed the first Universal Guideline — serving as both a Guideline template and then evolving into a semi-working prototype suitable as an AI-augmented co-pilot.
-
Remote Nursing Care
The Wave of the Future. Rapidly evolving and expanding, remote nurse care is unique and ubiquitous. Nurses practice in range of ambulatory and acute care settings (office, clinic, ED, post-operative care).
Ambulatory care is predicted to expand rapidly in the future; bringing with it remote nurse triage and the technologies rquired to perform telephone triage, nurse televisits, and virtual chronic care monitoring.
Various technologies: telephone-only , video, biometric devices, and patient wearables will enhance nurses’ ability to perform this challenging task.
Remote encounters will include patients calling about worrisome symptoms to pre-scheduled virtual visits for non-acute consultations.
Grounded in the nursing process, care will include triage— referrals to ED, Urgent Care or Office visit — or virtually prescribing, treating, referring and monitoring patients, medications or treatments. A secondary goal is to enhance cost-effectiveness by reducing inappropriate ED, Urgent Care and Office visits for institutions offering this.
All forms of telehealth nursing will require the process of triage, if only for the limitations of access, remote encounters, incomplete information that will require avoiding potential delays in care or diagnosis that must always be considered in remote clinical encounters.
A high risk task and work environment, nurses experience stressors: sensory deprivation, rapid clinical decision-making, remote enounters.
Like other nursing specialties, remote nurse care is based on the nursing process, and requires a support system — specialized components: clinical training, nursing standards, clinical decision support systems (CDSS), and electronic health records (EHRs).
Finally, nurses must have a substantial amount of clinical experience. Patient safety depends on system integrity. All components must be proven safe, valid, and reliable, including CDSS and EHRs.