TELEHEALTH AND SYSTEM ERROR

From 1995-2018,  Ms. Wheeler served as an expert witness on 35 malpractice cases involving telephone triage - a turning point in her career.  Recurrent error was often involved.   For 40+ years Ms Wheeler has worked diligently on solutions to avoid recurrent error.  TeleTriage Systems represents the legacy of Ms Wheeler and other experts to make telephone triage safer. Many recurrent errors are still unaddressed. Clinician error may be related to corporate (system) error.

FAILURE TO ASSESS or TO ELICIT ADEQUATE INFORMATION: Incurious, passive clinicians, placing the burden to provide all essential information on a lay-person -- patient or family.

FAILURE TO DOCUMENT & TO COMMUNICATE with patients, family, healthcare providers, Failure to obtain Patient Informed Consent

IIGNORING RED FLAGS: In addition to extremes of age and chronic illness,  Failure to identify high risk symptoms  frequent calls over hours or days are signs of patient anxiety Failure to factor in recent environmental events  -- Heat/Cold extremes, occupation or

GUIDELINE ERROR: Failure to use any guideline, selecting the wrong guideline, using a guideline incorrectly.

COGNITIVE ERROR: Stereotyping symptoms or patients, jumping to conclusions, confirmation bias, attribution error and others

RED HERRINGS (Distractors): Relying on the previous misdiagnoses and/or ED visit outcome,  or relying on patient misperceptions of symptoms

System or Corporate Error: Failure to provide adequate After Hours patient access to care (alternative venues for further evaluation).  Inadequate follow -up Error Feedback to clinicians about their triage decisions - appropriate or inappropriate. Excessive Call volume or Inadequate Staffing. Inappropriate Staffing  non-clinicians. Failure to provide Standards or Clinical Training,  Inadequate Guidelines, Technology, EMR and CDSS. 

TWO HISTORIC SYSTEM ERRORS: CLINICIANS WORKING IN A VACUUM & WORSENING LACK OF PATIENT ACCESS AFTER HOURS RESULTING IN DELAY IN CARE (see below)

RESEARCH: AI & Telephone Triage

NEW relevant, Innovative research: Machine learning models predict under triage in telephone triage, Inokuchi et al., Annals of Medicine.

To Err is Human, to delay is deadly New research describes the role of AI in reducing under triage (and delay in care) in pre-hospital telephone triage by nurses. Unlike any current studies - typically focused on in-person triage of emergent symptoms — this research focuses on how AI might help reduce undertriage of ostensibly non-urgent symptoms in remote triage — a more difficult task. Questions and suggestions follow;

·      The amount of symptom data collected (and relied upon) is not clear and four risk categories seemed sparce (patient age, sex, chief complaint category, comorbidities).

·      Were “Fire and Disaster Management Agency Protocols” medically developed or EBM?  A sample protocol and previous protocol outcome results would be helpful for comparison. 

·      How are nurses trained to use these protocols -- as decision-making or decision support tools?  If used for decision support, then what clinical training for telephone triage did they have?  Will nurses be required to use AI-supported protocols as decision-making tools?

·      In telephone triage, proficiency is linked to safe outcomes, not speed (Perrin & Goodman, NEJM,1978)

 Telephone Triage -- remote clinical risk assessment and decisionmaking “under conditions of uncertainty and urgency” (Lephrohon & Patel, 1995) – requires pattern recognition (with or without AI). Pattern recognition requires adequate detail. If telephone triage is plausibly analogous to a virtual history and physical, then the duty of due care (the standard), requires that nurse elicit information and identify red flags (risk factors) to rule out urgent symptoms.

Generally speaking, eliciting symptom history, functional status, and medical history are all essential. Is it a high risk patient, with high risk symptoms? Both? Or none?  In addition to the Patient Medical History checklist described,   rule out questions might include: immunodeficiency, immunosuppressive disease, Implant, Invasive events: Post-Op; Post-Partum; Post-Procedure; Trauma; Large Burn, Recent injury or  International Travel.

Protocols, EMR, clinical training, experienced clinicians and professional standards are all part of a complete telephone triage system.  Hopefully, as Computerized Decision Support Systems improve, AI will serve as a Clinical Co-pilot – suggesting reasonable, prudent triage levels, with the nurse as the final decision maker.

Perrin EC, Goodman HC. Telephone management of acute pediatric illnesses. The New England Journal of Medicine 1978;298(3):130-5.

Leprohon J, Patel V. Decision-making Strategies for Telephone Triage in Emergency Medical Services. Medical Decision Making. 1995;15:240-53.

RESEARCH: AI & ED TRIAGE

A recent JAMA Commentary provides an analysis of ED triage research and the application of AI and Large Language Models (LLM) to ED Triage on-site. Friedmans’s analysis suggests it may be too early to know AI’s effect on ED triage. He suggests a human-machine hybrid — AI as a clinical-co-pilot to clinical decision makers. Williams asks “Is just being able to do something the bar for using AI, or is it being able to do something well, for all types of patients?”

Since the 1980s, ED overcrowding has continued to increase. Are patients (perhaps low-income groups, lacking adequate access) increasingly self-triaging to the ED, thereby compounding overcrowding? Recent studies point out negative consequences: reduced patient safety, increased staff stress level, increased error, and delays in care. 

One wonders if increasing research studies on how AI can reduce ED overcrowding is on the right track? Increasingly crowded EDs are a major systemic problem that likely needs to be solved first.  Is overcrowding solved by enlisting AI to support ED staff to triage new arriving patients more rapidly? Given ESI triage mismatch, how can the ESI tool meaningfully apply to self-triaged patients, inappropriately now on-site?

ED overcrowding is likely a long-standing symptom of inadequate patient access to less emergent venues – open for expanded hours, not just office hours. Inadequate access also can also sabotage the effectiveness of telephone triage, its original purpose — to reduce inappropriate ED visits. It was also intended to make health systems overall more cost-effective. Without expanded access and improved pre-hospital triage systems, ED overcrowding will likely continue to worsen.