AI Telemedicine Support for Primary Care Physicians in El Salvador

• DoctorSV AI Assistant (Device): An AI tool integrated into the telemedicine platform, built on Google's Gemini Large Language Models (LLMs). The system operates via two modules: (1) a clinical history assistant that supports structured documentation of patient information in real-time and (2) a pre-diagnosis tool that analyzes documented clinical data to generate differential diagnosis suggestions for the physician's consideration. The model uses contextual prompting to ensure suggestions are culturally and clinically appropriate for El Salvador.
• Standard Telemedicine Workflow (No AI) (Other): Standard primary care consultation via videocall without the assistance of artificial intelligence tools. Physicians rely solely on their own clinical judgment and manual documentation without automated summaries or diagnostic prompts.

Primary outcomes

• Diagnostic Accuracy [Through study completion, ~ 12-16 weeks]: The proportion of consultations where the primary diagnosis recorded by the participating physician matches the "gold standard" reference diagnosis. The reference diagnosis is established by a panel of three independent, blinded expert evaluators reviewing the anonymized clinical notes. A diagnosis is considered "correct" (value = 1) if it matches the reference diagnosis within the same clinically equivalent diagnostic group; otherwise, it is considered "incorrect" (value = 0). The analysis will compare the proportion of correct diagnoses between the AI-enabled and AI-disabled arms.

Secondary outcomes

• Diagnostic Concordance [Through study completion, ~12-16 weeks]: The level of agreement between the physician's diagnosis and the expert reference diagnosis, measured using Cohen's Kappa coefficient. This measure evaluates the reliability of the diagnoses beyond simple percentage agreement, accounting for agreement occurring by chance.
• Diagnostic Accuracy Stratified by Physician Experience Level [Through study completion, ~12-16 weeks]: Evaluation of diagnostic accuracy (proportion of correct diagnoses) compared between subgroups of physicians with "High Experience" (≥1 year in the program or ≥20 consultations) versus "Low Experience" (<1 year in the program or <20 consultations).
• Diagnostic Accuracy Stratified by Clinical System [Through study completion, ~12-16 weeks.]: The proportion of correct diagnoses stratified by the physiological system of the pathology: Respiratory, Digestive, Urinary, or Ophthalmic. This outcome assesses if the AI's performance or utility varies depending on the specific type of clinical condition.

Sex: All, ages: 18 Years to any

Inclusion Criteria:

* Must be a physician employed by the DoctorSV telemedicine program
* Must provide written informed consent to participate in the study
* Consultations must be for acute pathologies of the digestive, respiratory, or urinary systems, or acute ophthalmic infections manageable in primary care
* Consultation must be the first medical contact (first visit) for the current acute episode
* The condition must correspond to specific ICD-11 codes defined in the study protocol

Exclusion Criteria:

* Physicians who are inactive on the platform for more than 4 consecutive weeks
* Physicians who transition to work modalities other than telemedicine or reduce their working hours to less than 20 hours per week
* Physicians whose employment contract ends (resignation, dismissal, or contract completion) during the data collection period
* Consultations classified by the physician as requiring immediate in-person attention
* Consultations requiring referral to another level of care or specialty for definitive management
* Consultations interrupted or incomplete due to connectivity or system failures
* Consultations solely for administrative purposes (e.g., certificates, repeat prescriptions without clinical evaluation)
* Consultations that are follow-up visits or controls for a previously evaluated episode

• Manuel Bello, MD, Principal investigator, Hospital Nacional El Salvador