Is TACo the future of Telehealth?
Mussaad Al-Razouki, Sophie Smith, Saba Alzabin
Researchers at the University of California San Francisco (UCSF) recently introduced a new integrated modality of specialty medical consultation: the targeted automatic electronic consultation (TACo), in a viewpoint published in June this year in the Journal of the American Medical Association (JAMA).
This new modality combines the best of both e-consults and automatic consults. E-consults are based on a doctor’s ability to respond to certain types of patient questions by reviewing the information in an electronic health record (EHR) and offering recommendations without spending the time to perform a full history and examination. This is becoming increasingly popular in large public healthcare systems where doctors are usually pressed for time, fitting as many patients as possible in their shift, which in turn has a large impact on the quality of care provided.
In the JAMA report, the authors cited a Veterans Affairs study of 554 e-consults and 938 traditional consults which concluded that that e-consults reduced response time from an average of 34.4 to 2.4 days across several medical specialties. Volume efficiencies exist, but what about value efficiency? Well, in fact both patients and physicians have reported high levels of satisfaction with e-consults. A systematic review by Vimalananda and colleagues in the Journal of Telemedicine Care found 70%-95% satisfaction rates amongst primary care physicians. Although e-consults expand the options for any doctor and patient, they fall short because, at present (in the USA at least), initiation of e-consultation lies in the hands of the primary care physician. This means that the timely provision of expert opinion and care is still an obstacle. This is where specialty first patient-driven portals, such as the one that Nabta Health is adapting, will elevate the healthcare experience ultimately leading to improved outcomes.
Automatic consultation (triggered by a diagnosis or lab results), take the e-consults a bit deeper. These types of consultations have been linked to improved outcomes for a range of healthcare challenges such as palliative care and acute infections. By having an automatic prompt trigger the consultation, automatic consultations are more likely to promote timely involvement by a specialist doctor. However, this trigger generally leads to traditional in-clinic consultation, so this particular modality may even worsen the lead time for a patient to seek the proper specialty care.
In the TACo model, the EHR identifies and preselects patients who meet defined criteria for automatic consultation and presents a customised view of any pertinent information to a designated specialist, who then reviews the case virtually. The consultant can choose to provide personalised advice, suggest a formal consultation, or neither. Similar to e-consults, this model allows the specialist to quickly and efficiently review and respond to the pertinent information the EHR presents. However, unlike regular e-consults, these targeted automatic consults are AI-triggered by patients’ EHR data rather than a dependency on a consultation request. Unlike automatic formal consults, the TACo model enables efficient virtual consultation since the bulk of the interaction is performed online (as well as in real time in most cases).
Today, the UCSF Diabetes Service has started implementing the TACo approach. Each day, the UCSF EHR screens inpatients for any single patient meeting any one of four criteria defined by the system. Then, an experienced specialist, in this case a diabetologist, reviews each flagged case after reviewing a custom-built EHR portal (showing glucose trends, insulin hypoglycemic doses, medications, nutritional information and lab test results). If the specialist has a disease management suggestion, he or she writes a brief consult note with the recommendations. The process takes 2 to 5 minutes per case. Published evidence on the 1132 admissions that included such virtual consults showed a significant improvement in diabetes management, with a 39% and 36% reduction in hyperglycemic and hypoglycemic events, respectively, during the study period compared to events before the study commenced. Based on these outcomes, the model has received sustained financial support for consultant time by the UCSF health system. Despite initial skepticism from the inpatient clinicians, there appears to be increased broad enthusiasm.
Nabta Health is also trialling the TACo approach as a part of its first Hybrid Healthcare pathway for the diagnosis of infertility due to Polycystic Ovary Syndrome (PCOS). Users of the Nabta app who exhibit symptoms of PCOS are prompted by the integrated ML-powered health assistant, Aya.ai, to explore one or more virtual care components designed to move them efficiently along a personalised care pathway to diagnosis and treatment. For example, women with irregular cycles who are trying to conceive would be directed to use the OvuSense realtime fertility monitor, a market leading medical device capable of detecting ovulation with up to 99% accuracy in-cycle. Those requiring medical assistance have instant access to clinicians via TACos, courtesy of a complementary medical encounters app (Nabta Care).
One issue facing the widespread adoption of TACo is the way traditional healthcare consultations are paid for. Nabta has taken the TACo approach one step further by introducing a Pay-As-You-Go (PAYG) option for women who are short of time or money. Instead of paying for a fixed 15 or 30 minute consultation, users are triaged upfront by Aya.ai, and the symptom assessment report is then passed to the selected clinician before the TACo is initiated. This means that users pay only for the time they use “in conversation” with the clinician, on a per minute basis, and while the clinician issues any ePrescriptions.
Another challenge is with reimbursement for e-consults and automatic consultations. Because the TACo model would most likely increase the total number of consults, any investment in a doctor’s time would have to be offset by improvements in care and costs. If this approach results in improved quality, time and cost, the business case for support could be strong, particularly when payments to the system are bundled or triaged, based on diagnosis group or type. Just like Telemedicine eventually became eligible for payers reimbursement, so will TACos if proven to be beneficial.
