Prevalence and Accuracy of Medical Alert Information on Cellular Phones Among Emergency Department Patients

Immediate access to accurate medical information is critical in emergency care, particularly when patients are unresponsive or unable to communicate. Smartphone Medical Identification (ID) features offer a potential source of readily available health data, yet their prevalence and accuracy remain largely unknown. As the only tertiary care, Level 1 Trauma Center in our region serving a primarily rural population, determining the prevalence and accuracy of medical ID could aid in providing specified care sooner to patients who are unable to provide pertinent health information. This study was designed to determine the proportion of Emergency Department patients with Medical ID enabled, assess variation in access by phone type and year of birth and to evaluate the accuracy and agreement of Medical ID information compared to the electronic health record (EHR).

This cross-sectional study included adult patients presenting to a single Emergency Department by ambulance who had a smartphone between September 2018 and April 2025. Consented participants provided research staff with their smartphone to check for an active Medical ID. For those with Medical ID enabled, researchers documented phone type, birth year, and compared listed medical conditions, allergies, and medications with EHR data. Descriptive statistics, 95% confidence intervals, sensitivity, specificity, and overall accuracy were calculated using EHR as the gold standard.

Of 383 participants, 30.8% (95% CI: 26.2%–35.7%) had Medical ID enabled, predominantly on Apple devices (89%). Participants’ year of birth was centered around the median of 1967, with values ranging from 1933 to 2005. Overall accuracy was high (>90%) across most categories, driven by near-perfect specificity. Sensitivity was low for many items, including medical conditions (Asthma 56.3%, Diabetes 50.0%), medications (Beta-blockers 26.3%, Calcium Channel Blockers 5.9%), and allergies (Food 50.0%). Rare items showed perfect agreement (Kappa = 1.00), while common items had moderate agreement (≈0.60) and “None of the Above” categories were poor (≈0.25–0.48).

Medical ID adoption was limited, and accuracy varied. While specificity and overall accuracy were high, low sensitivity indicates frequent omissions of EHR-documented information. Medical ID should not be relied upon as the sole source of patient history in emergency care.