Electronic Consultations in a Community Neurology Practice: A Retrospective Study Informing Best Practice

Objective

To describe our practice of electronic consultations (e-consults) and assess safety and risk factors for subsequent face-to-face consultations.

Patients and Methods

A retrospective cohort study of all e-consults completed in a community neurology practice between May 5, 2018, and June 31, 2019, was completed. Clinical and demographic variables were compared between the successful and unsuccessful (defined by presence of subsequent face-to-face consultation) cohorts. Hazard ratios (HR) were calculated using Cox regression model. Kaplan-Meier probability analysis (with 95% CIs) of subsequent face-to-face consultation was performed. Case examples highlighting potential harm were summarized.

Results

In total, 302 e-consults were reviewed. The most frequent referrals were for headache (n=125, 41.4%), dysesthesia (n=40, 13.2%), and abnormal imaging finding (n=27, 8.9%). The most common e-consult questions were for treatment (57.6%) and diagnostic evaluation (48.0%) recommendations. Moreover, 24.8% (n=75) of e-consults were followed by face-to-face consultations, with primary risk factors including female sex (HR, 1.9), referral for headache (HR, 1.7), and final diagnosis of migraine (HR, 2.0) or long-term migraine (HR, 5.0). Potential harm related to delayed diagnosis/treatment was identified in 6 (2.0%) patients with migraine and 4 (1.3%) without migraine presenting to emergency department.

Conclusion

Utilization of e-consults may safely improve access to neurologic expertise and prevent the need for some visits, which may have required a face-to-face visit. In patients with chronic migraine, e-consults should be considered short-term and followed by face-to-face consultation as soon as access allows. Neurologists performing e-consults should be able to triage patients to face-to-face consultation, particularly when diagnosis is uncertain or the neurologic examination may help guide appropriate testing.