Increased Utilization of Telemedical Emergency and Nonurgent Care Following Deployment of Virtual Triage and Care Referral in Australia

Abstract

Objective

To evaluate whether an artificial intelligence–based national virtual triage and care referral (VTCR) service in Australia improved care acuity level alignment, increased patient engagement of telemedicine services, and reduced emergency department demand by offering lower acuity, less costly options for urgent, virtual, or in-person care services.

Patients and Methods

Cross-sectional analyses examined changes in patient care intent following VTCR to determine whether it facilitated patient adoption of new emergency and nonurgent telemedicine and virtual care services.

Results

Virtual triage and care referral more than doubled the number of patients selecting appropriate, lower acuity nonurgent care from 330,279 (21.3%) to 820,800 (52.9%), an increase of 31.6 percentage points [PPs] (P<.01), and effectively eliminated uncertainty in patient care seeking from 670,502 to 2557 patients, a decrease of 99.6%. Intent for in-person emergency care fell significantly from 119,414 (36.7%) to 105,349 patients (24.6%) (–12.1 PP; P<.01), replaced by substantial growth in patient intent to use virtual emergency care (from 612 to 11,840 patients or +10.1 PP) and nonurgent virtual care use (from 20,467 to 26,289 patients or +2.9 PP) (P<.01). Victoria, a state within Australia, recorded the highest uptake. Extrapolated nationally, these shifts could prevent an estimated 2409 unnecessary in-person nonurgent visits and 19,286 unnecessary emergency department visits annually in Australia. Aboriginal and Indigenous patients showed similar benefits and engaged VTCR at higher rates than other patients.

Conclusion

Artificial intelligence–based VTCR improved alignment between patient perceived needs and recommended care pathways, not only driving greater use of appropriate, lower acuity, and telemedicine services but also reducing unnecessary in-person emergency visits. By eliminating uncertainty in care seeking and advancing adoption of new virtual emergency and nonurgent care options, VTCR offers a scalable, evidence-based solution for optimizing emergent and urgent care delivery and easing pressure on emergency departments across Australia.

Australia’s health care system, recognized for its performance, equity and administrative efficiency, faces increasing challenges in service demand, resource allocation, and access to care.1, 2, 3 Clinically inappropriate or unnecessary use of emergency department (ED) services is contributing to rising health care costs and delays. In 2021-2022, one-third of ED patient presentations were for lower urgency needs, which could have been managed in a primary care setting or by self-care.⁴ Moreover, among those needing primary care, inadequate access affects 28% of Australians who live in rural and remote areas, potentially resulting in delayed diagnosis, increased reliance on more costly hospital services, and suboptimal health outcomes.³ Consequently, Australians are turning increasingly to online sources for health information, with 73% of Australians using the internet to research a health issue before seeing a health care professional.⁵ However, unreliable online health content, often not based on scientific evidence, can lead to inappropriate self-diagnosis, causing either avoidable care delays or overutilization of high-acuity services. This underscores the need for accessible, evidence-based health information and guidance to better engage individuals with care needs.⁵

In response, health care delivery systems are integrating telemedicine, virtual care services and in-person urgent care to enhance and augment in-person care. As the variety of new services increases, health care systems are considering ways to influence patient care seeking behaviors as a means of realizing the opportunities and benefits of telemedicine and virtual care services for improved care delivery.⁶ Virtual triage and care referral (VTCR) is being used to assess patient acuity and refer patients to appropriate care delivery settings as an alternative to visiting EDs for medical concerns not requiring that highest level of care acuity.6, 7

Healthdirect Australia (HDA), a government-funded national health advice service, introduced a new artificial intelligence (AI)-based VTCR system to serve as a digital or virtual front door to care and to improve patient access to and engagement of health care information and services available 24/7/365.6, 8 VTCR offers symptom assessment and personalized referral recommendations to emergency services, urgent care, nonurgent general practice, pharmacy, mental health, midwifery care, or self-care, as clinically indicated.

By systematically evaluating patients’ clinical presentation and connecting them to the correct level of care, taking into account their circumstances and preferences, HDA aims to reduce patient care uncertainty, unnecessary health care usage, and care acuity misalignment between patient care intent and need, as well as avoidable carbon-equivalent emissions.9, 10, 11, 12 Although not a focus of the present study, HDA is also more broadly dedicated to improving the outcomes of patient care delivery through the innovative implementation of advanced health care and health information technologies.

Patients And Methods

Study Objectives

This study evaluated whether AI-based VTCR technology can improve care acuity alignment and patient experience, and reduce unwarranted and avoidable care costs.

Study Design and Setting

A cross-sectional study compared patient care intent before and following VTCR to determine its impact on patient care intent to engage an appropriate level of care acuity to meet their needs. Funded by Australian governments, HDA is the national virtual public health information service, providing equitable and timely access to care alternatives and referral for all Australians wherever and whenever they need help, including rural and remote regions. It is accessible via multiple channels—helplines, video, websites, a service finder, and a mobile application. In conjunction with its health care service partners, HDA identified an unmet and growing need for scalable digital solutions not only to help manage increasing service volume demand on health systems, but also to reduce unnecessary or avoidable ED visits by connecting people to the right care at the first encounter. Emergency department presentations in the Australian health care delivery system have increased over the last 5 years from 8.2 million in 2019-2020 to 9.0 million in 2023-2024 (a mean annual increase of 2.3% per year).¹³ Between 2022-2024, hospitalizations in Australia increased 4.1%, from 12.1 to 12.6 million, with overnight hospitalizations increasing 3.9% (4.4-4.6 million) and hospitalizations in public acute hospitals increasing by 4.9% (7.1-7.5 million).^13,14 Further, HDA deployment of VTCR targets individuals experiencing health symptoms who are unsure about what care is needed or where to seek care, including patients needing immediate guidance beyond usual office hours, who may otherwise seek ED care for nonurgent concerns.

The study tracked implementation of new ambulatory and telemedicine care pathways introduced as alternatives to visiting an ED in person. A general practitioner (GP) helpline received referrals from a nurse triage helpline with acute presentations from callers who indicated they were unable to access a GP within the posttriage recommended timeframe to visit a GP (in 2 or in 24 hours). Urgent care centers offered walk-in treatment for acute, nonemergency conditions. In addition, virtual ED pathways provided video assessment by an emergency care clinician for non–life-threatening acute conditions, which could be resolved without in-person examination. Each service excluded high-risk cases such as chest pain, severe respiratory distress, major trauma, mental health crises, substance abuse, toxic exposure, pregnancy complications, or labor. Supplemental Figure 1 (available online at https://www.mcpdigitalhealth.org/) illustrates various care pathways in the HDA deployment of VTCR in Australia.

Description of VTCR Engine

The Infermedica VTCR engine, which is used by HDA as a key component of the virtual front door technology platform, conducts evidence-driven analyses considering 800 illnesses, 1500 symptoms, and 300 risk factors from any internet-connected device.¹¹ Artificial intelligence, including machine learning, and natural language processing enable evaluation of reported symptoms and medical history, and convey guidance to clinical acuity-level appropriate care. To minimize potential mistriage, Infermedica’s AI-driven VTCR engine may overtriage to higher acuity care rather than risk undertriaging a patient.¹¹ Clinical vignettes are used to evaluate VTCR validity, and prior research demonstrated that Infermedica’s VTCR engine provides safe recommendations in 97.8% of cases and is comparable in performance with rules-based triage protocols used by live nurses.^11,15,16 In the remaining 2.2% of cases, the result was a maximum of 1 level less conservative than expected, which is therefore not significantly different from expectations. The undertriage rate is lower than that previously reported in other studies, which found clinically relevant undertriage in 3.7% of cases for nurses using a clinical decision support system (7.3% for general practitioners).¹⁷ Infermedica’s VTCR corrects clinical acuity level misalignment of patient care intent with actual clinical need, and has demonstrated the potential to increase early detection and expedite care referral.^11,16, 17, 18, 19, 20

Data Captured and Analyses Completed

Data were obtained from 1,552,592 VTCR patient encounters across all Australian states and territories, except Queensland, from April 1, 2023, to March 31, 2025 where age, sex, and care seeking intent were reported. Analyses determined the level of clinical care acuity alignment or divergence between patient pretriage and posttriage care intentions, and whether VTCR was able to alter patient posttriage care seeking intent when it differed from VTCR determined clinical need. The care intent survey included 5 pre-VTCR care options: self-care; nonurgent in-person consultation; urgent in-person consultation within 24 hours; ED care; and patient did not know or was uncertain about what level of care to seek (unspecified). Posttriage intent included 3 other care pathways: nonurgent video or voice call consultation with a primary care clinician; urgent care center visit for immediate nonemergency care; and immediate virtual emergency care via a video call with an emergency care clinician. Analyses identified the extent and direction of intent change from in-person care to telemedicine.

The degree of agreement between VTCR triage level and nurse-assigned triage level was used to determine the concordance of the recommended care acuity level. The nurse-assigned triage level was rendered by nurses based on their clinical judgment, following completion of VTCR. Concurrence was defined as an exact match between a nurse’s selected triage level and that recommended by VTCR (eg, nurse selects self-care and VTCR recommends self-care). The concurrence rate was calculated as the number of matching encounters divided by the total number of encounters.

Three periods were evaluated based on implementation stage of the new virtual emergency and urgent care pathways: period I, pre-implementation (April 1, 2023, to September 30, 2023); period II, early implementation (October 1, 2023, to March 30, 2024); and period III, mature implementation (April 1, 2024 to September 30, 2024). Encounters were also analyzed according to whether they occurred within or outside usual working hours. Pre-VTCR and post-VTCR patient care intent were compared and differences assessed for statistical significance with a Z-test using Google Sheets Online Spreadsheet Editor at a probability level of P=.01.

Ethics Statement

Patients provided consent before VTCR encounters for their data to be used for research purposes, with all analyses completed and all results reported in a de-identified, anonymous manner in the aggregate. Given the absence of risk of patient harm, ethical review board review was not engaged.

Results

Patient Demographics

Patients were predominantly women (934,678 or 60.2% female), and young, with 70.4% (1,092,416) under age 44 years (Table 1). Most were non-Indigenous, with 89,636 (5.8%) indicating Aboriginal or Torres Strait Islander race or ethnicity, who comprised a larger percentage of the sample than their contribution to the national population (3.2%). Aboriginal and Torres Strait Islander patients were younger than other patients, with 47,330 (40.9%) indigenous vs 496,822 (33.9%) non-indigenous patients under age 18 years, and 26,343 (22.9%) indigenous vs 252,184 (17.2%) non-indigenous patients age 18 to 29 years (P<.05), respectively.

Table 1.

Sex and Age Distribution of Patients Using Virtual Triage and Care Referral

Sex	Age (y)	No. of encounters (%)
Female	<18	258,551 (16.7)
	18-29	197,988 (12.8)
	30-44	195,627 (12.6)
	45-59	113,660 (7.3)
	60-79	124,374 (8.0)
	>80	44,478 (2.9)
Total female		934,678 (60.2)
Male	<18	276,168 (17.8)
	18-29	75,868 (4.9)
	30-44	87,625 (5.6)
	45-59	67,113 (4.3)
	60-79	82,205 (5.3)
	>80	28,243 (1.8)
Total male		617,222 (39.7)
Other	<18	126 (0.0)
	18-29	323 (0.0)
	30-44	140 (0.0)
	45-59	58 (0.0)
	60-79	37 (0.0)
	>80	8 (0.0)
Total other		692 (0.1)
Total		1,552,592 (100.0)

Patients were located mostly in New South Wales (36.7%) and Victoria (33.6%), higher than these states’ contributions of 31.3% and 25.6% to the national population (P<.01). VTCR patient use rate averaged 62 encounters per 1000 persons, with modest variation by state or territory: 72 encounters per 1000 in the Australian Capital Territory, 67 in New South Wales, 44 in the Northern Territory, 88 in South Australia, 85 in Tasmania, 74 in Victoria, and 65 in Western Australia. Queensland utilization was low with 1 encounter per 1000 likely due to an existing alternative service and geoblocking of HDA services. Aboriginal and Torres Strait Islander use was higher in Victoria (+12.8 percentage points [PPs]) and less common in New South Wales (–9.4 PP; P<.05).

Regarding the person engaging VTCR, in 848,057 encounters (54.6%) it was the patient themselves. In 415,916 encounters (26.8%), it was the mother of a minor age patient, and in 78,320 encounters (5.0%) it was the patient’s father. Other relationships included wife (43,378 encounters or 2.8%), daughter (35,181 encounters or 2.3%), husband (27,491 encounters or 1.8%), son (18,548 encounters or 1.2%), or another individual (85,701 encounters or 5.5%). A majority of encounters occurred outside usual health care business hours (1,067,489 or 68.8% of all encounters).

Concordance of VTCR and Nurse Triage by Care Acuity Level

Triage nurses concurred with the VTCR acuity level recommendation in 83.3% of encounters (Table 2). Nurses increased the clinical acuity of VTCR recommended care more frequently than decreasing it (P<.01). No meaningful difference in nurse triage concordance by Aboriginal or Indigenous identity was observed.

Table 2.

Concordance of Recommended Care Acuity Level Comparing Virtual Versus Nurse Triage

		Nurse triage level, n (%)			Total
		Self-care	Nonurgent consultation	Emergency care
Virtual triage level	Self-care	91,524 (5.9)a	98,077 (6.3)b	14,296 (0.9)b	203,897 (13.1)
	Nonurgent consultation	1021 (0.1)c	683,469 (44.0)a	107,857 (6.9)b	792,346 (51.0)
	Emergency care	101 (0.0)c	38,184 (2.5)c	518,063 (33.4)a	556,348 (35.8)
	Total	92,646 (6.0)	819,729 (52.8)	640,216 (41.2)	1,552,592 (100.0)

^aNurse concurred with virtual triage and care referral acuity level.

^bNurse up-triaged to a higher level of care acuity.

^cNurse down-triaged to a lower level of care acuity.

Impact of VTCR on Initial Patient Care Intent

The largest post-VTCR changes in care intent were an increase in nonurgent care from 330,279 (21.3%) to 820,800 (52.9%) patient encounters (+31.6 PP; P<.01), and a 99.6% reduction in the number of patient encounters with unspecified or uncertain care seeking intent from 670,502 (43.2%) to 2557 (0.2%) (–43.0 PP; P<.01) (Table 3). Patient encounters with self-care intent decreased modestly, and those with emergency care intent increased moderately. No meaningful differences were observed by race or ethnicity.

Table 3.

Change in Patient Care Intent Following Virtual Triage and Care Referral

Care level	Pre-triage patient-user care seeking intent, n (%)	Post-triage patient care seeking intent, n (%)	Direction and absolute (and relative) magnitude of change in care seeking intent, % (PP)	Statistical significance (P)
Self-care	129,064 (8.3)	107,044 (6.9)	−17.1 (−1.4 PP)	<.01
Nonurgent care	330,279 (21.3)	820,800 (52.9)	148.5 (+31.6 PP)	<.01
Emergency carea	422,747 (27.2)	622,191 (40.1)	47.2 (+ 12.8 PP)	<.01
Unspecifiedb	670,502 (43.2)	2557 (0.2)	−99.6 (−43.0 PP)	<.01
Total	1,552,592 (100.0)	1,552,592 (100.0)

PP, percentage points.

^aSelf-transport or transport by ambulance.

^bUnspecified includes: call this service, did not know, general practitioner not available, and unknown.

Change in Post-VTCR Care Intent by Implementation Period

A step-wise reduction of patient intent to use in-person emergency care occurred as implementation of virtual emergency care services progressed (from 119,414 to 105,349 patient encounters, or –32.9% and –12.0 PP) (P<.01) (Supplemental Figure 1). In-person emergency care intent decreased from 119,414 (36.7%) to 105,349 encounters (24.6%) (P<.01). Intent to use virtual emergency care increased significantly from 1839 (0.6%) to 37,642 (8.8%) of all encounters (P<.01), which coincided with increasing availability of these pathways across different regions. Intent to use telemedicine increased from 85,481 (26.2%) to 129,881 (30.3%) of encounters (+15.6% or +4.1 PP; P<.01), as did intent to use urgent care (increase of 35,803 encounters or +16,923.5% and +8.8 PP from 0.1% to 8.9% of encounters; P<.01). No differences by race or ethnicity were observed.

Change in Posttriage Intent Among Patients Intending to Access Emergency Care

Among patients who had a pre-VTCR intent to access emergency care, there were reductions of intent to seek in-person nonurgent care (14,699 to 10,615 encounters or −36.0% and −5.4 PP), and in-person emergency care (44,913 to 37,406 encounters or −26.2% and –12.0 PP; P<.01) as new care pathway implementation progressed (Table 4). Intent to use emergency care with ambulance transport decreased (13,508 to 12,465 encounters or −18.3% and −2.5 PP; P<.01). Intent to access nonurgent telemedicine increased (20,467 to 26,289 encounters or +13.8% and +2.9 PP; P<.01), as did intent to use virtual emergency care (612 to 11,840 encounters or +1612.7% and +10.1 PP) and in-person urgent care (34 to 8947 encounters or +23,209.7% and +8.0 PP; P<.01). There were no meaningful differences by race or ethnicity.

Table 4.

Change in Post-Triage Care Intent Among Patients With Pre-Triage Intent to Access Emergency Care

Post-triage intent	No. of encounters (%): pre-virtual emergency carea	No. of encounters (%): mature stage of virtual emergency care	Direction and absolute (and relative) magnitude of change, % (PP)	Statistical significance (P)
Self-care	3725 (3.8)	3247 (2.9)	−22.8 (−0.9 PP)	<.01
Nonurgent care, teleconsultation	20,467 (20.9)	26,289 (23.7)	13.8 (+2.9 PP)	<.01
Nonurgent care, in-person	14,699 (15.0)	10,615 (9.6)	−36.0 (−5.4 PP)	<.01
Urgent care center visit	34 (0.0)	8947 (8.1)	23,209.7 (+8.0 PP)	<.01
Virtual emergency care	612 (0.6)	11,840 (10.7)	1613.7 (+10.1 PP)	<.01
Emergency care, in-person	44,913 (45.8)	37,406 (33.8)	−26.2 (−12.0 PP)	<.01
Emergency care with ambulance transport	13,508 (13.8)	12,465 (11.2)	−18.3 (−2.5 PP)	<.01
Unspecified	197 (0.2)	0 (0.0)	−100.0 (−0.2 PP)	<.01
Total	98,155 (100.0)	110,809 (100.0)

PP, percentage point.

^aPeriod I was prior to the introduction of virtual emergency care (April 1, 2023, through September 30, 2023); period II was the early phase implementation of virtual emergency care (October 1, 2023, to March 30, 2024); and period III was when the implementation and operation of virtual emergency care reached maturity (April 1, 2024, to September 30, 2024).

Change in Post-VTCR Intent Among Patients With Initial Intent to Access Emergency Care in Victoria State by Implementation Period

Table 5 compares patients with initial intent to access emergency care across the 3 study periods in the state of Victoria, which had the highest adoption of new care pathways (Supplemental Figure 2 conveys the HDA care pathways implemented in the state of Victoria, available online at https://www.mcpdigitalhealth.org/). A reduction of intent to seek in-person nonurgent care occurred (4441 to 2945 encounters or –44.2% and –7.1 PP), as did intent to use in-person emergency care (14,361 to 8931 encounters or −47.7% and −24.6 PP; P<.01) and emergency care with ambulance transport (1828 to 1917 encounters or –11.8%, and –0.8 PP). Increases were observed in intent to access nonurgent telemedicine (5867 to 7584 encounters or +8.7% and +1.8 PP; P<.01), urgent care services (0 to 2353 encounters or +7.1 PP; P<.01) and virtual emergency care (228 to 8424 encounters or +3007.4% and +24.7 PP; P<.01).

Table 5.

Change in Posttriage Intent Among Patients With Initial Intent to Access Emergency Care in Victoria State by Virtual Emergency Care Implementation Period

Posttriage intent	No. of encounters before virtual emergency care (%)	No. of encounters mature stage of virtual emergency care (%)	Direction and absolute (and relative) magnitude of change, % (PP)	Statistical significance (P)
Self-care	1038 (3.7)	912 (2.8)	−26.1 (−1.0 PP)	<.01
Teleconsultation nonurgent care	5867 (21.1)	7584 (22.9)	8.7 (+1.8 PP)	<.01
In-person nonurgent care	4441 (16.0)	2945 (8.9)	−44.2 (−7.1 PP)	<.01
Urgent care center	0 (0.0)	2353 (7.1)	— (+7.1 PP)	<.01
Virtual emergency care	228 (0.8)	8424 (25.5)	3007.4 (+24.7 PP)	<.01
In-person emergency care	14,361 (51.6)	8931 (27.0)	−47.7 (−24.6 PP)	<.01
Emergency care with ambulance	1828 (6.6)	1917 (5.8)	−11.8 (−0.8 PP)	<.01
Unspecified	47 (0.2)	0 (0.0)	−100.0 (−0.2 PP)	<.01
Total	27,810 (100.0)	33,066 (100.0)

PP, percentage point.

In Victoria, the magnitude of the increase in virtual emergency care intent from period I to period III was larger than that observed in other states (mean of 11,228 encounters and +10.1 PP for all other states vs 8196 encounters and +24.7 PP for Victoria; P<.01). Accordingly, the magnitude of the reduction of in-person emergency care intent from period I to period II in Victoria was larger than the mean reduction across all other states (–12.0 PP for all states and –24.6 PP for Victoria state; P<.01).

If the magnitude of change in acuity level of care intent achieved in Victoria is applied to encounters occurring across all Australian states annually and the volume of use remains constant, 2409 clinically unnecessary in-person nonurgent care visits would be avoided per annum, and 19,286 unnecessary in-person emergency care visits would be avoided yearly once VTCR implementation reaches maturity in all states and territories.

Discussion

Key Study Findings

Virtual triage and care referral elicited a substantial increase in nonurgent care intent among patients and effectively eliminated (−99.6%) patients indicating that they remained uncertain of, or did not know, what kind of care to obtain. Following VTCR, there was a modest decrease in self-care intent, and an increase in emergency care intent and demand for subacute care previously not met due to these options not being available or unknown to patients. These shifts aligned with HDA clinical care guidance, and may reflect VTCR’s ability to identify patients who underestimated the severity of their condition and required urgent care. Patient post-triage intent to use in-person emergency care declined as the implementation of virtual emergency care progressed. This trend persisted among patients who intended to access in-person emergency care before VTCR, with a substantial reduction of intent to seek such care following virtual triage and diversion of these patients (as reflected by care intent) to use telemedical emergency, nonurgent care, and in-person urgent care services. Thus, a shift to lower care acuity intent was observed both for the entire patient population and those in particular who intended pre-triage to visit an ED. The finding of greater intent diversion from in-person emergency care services in Victoria–with the most advanced implementation of the new telemedicine pathways–bodes very well for the future national trend once VTCR integration with telemedical services delivery becomes more widely known, accepted, and trusted by the Australian public. These outcomes were also remarkable for their persistence among Aboriginal and Indigenous patients when compared with those of other patients, a population segment facing longstanding health care inequities and access barriers.

Although nurses concurred with the care acuity level recommendation of VTCR in 83.3% of encounters, the fact that nurses mostly increased to a higher level of clinical acuity than recommended may be due to additional clinical information available to them, such as patient distress, not captured by VTCR. It is also possible that this difference may reflect VTCR error. This warrants further investigation, given that VTCR is configured to overtriage to minimize patient risk. The demographic skew of the sample toward younger females is similar to that observed in other studies of VTCR.^11,19 A majority of patients (68.8%) engaged VTCR outside usual medical office business hours, confirming the value of increased around-the-clock accessibility to care guidance the technology enables.

Study Implications for Practice and Ongoing Program Evaluation

The HDA objective of deploying VTCR to safely divert patient care intent from EDs to virtual or telemedical care in a clinically appropriate manner is being achieved. A simultaneous increase of patients intending to visit an urgent care center suggests that a segment of patients still felt a need to seek in-person rather than virtual care. These patients may require additional support to engage the full lifecycle of virtual triage and telemedical care delivery, or may have symptoms better suited to or requiring in-person care (such as changing a wound dressing). It is critical in ongoing and future evaluation of the virtual front door to confirm that these changes in patient-reported care intentions translate into actual care engaged.

The implementation of VTCR by Healthdirect Australia is notable for success in effectively engaging Aboriginal and Indigenous patients. Given the health care inequities among racial and ethnic minorities in many nations, the Australian experience should be examined to determine the best practices that have produced this positive outcome. Further, VTCR has the potential to ameliorate longstanding race-related care access and service delivery inequities, and may enable heightened focus on a population segment with hyperendemic incidence of preventable chronic diseases and other major public health problems.

It will be important to quantify the potential health system resource savings resulting from diversion of patients from in-person to virtual nonurgent and emergency care services, including the extent of clinical care capacity recovery and the clinical and financial value of accelerating appropriate treatment of higher acuity patients. Moreover, VTCR must yet prove unequivocally that it improves actual clinical outcomes and organizational financial performance. Once confirmed, programmatic innovation and evaluative research should focus on what factors impede some patients from following VTCR guidance and using telemedical nonurgent and emergency care services. This will provide an understanding of how to increase the number of patients engaging the full spectrum of VTCR-integrated telemedical care delivery.

With advances in generative and conversational AI, including the advent of large language models, VTCR will soon offer voice automation including voice-to-voice interaction with translation, which may encourage patients thus far hesitant to act upon its guidance to do so, produce further cost reductions in live and automated call centers, and divert more routine cases from overloaded clinicians so they can focus on the highest acuity, most urgent patients.

Study Limitations and Future Research

This study has several limitations. First, these analyses relied on patient self-reported pre-triage and post-triage care intent, which we used as a proxy for actual care seeking behavior, rather than confirmed clinical service use. We have no a priori reason to suspect patients were not candid in stating their pre-VTCR and post-VTCR care intentions. However, care seeking behavior must be verified by validating clinical care obtained by patients in order to accurately quantify avoided care utilization and associated reductions in care costs. Future studies should collect both diagnostic and care service utilization data from patient records.

The level of general population awareness of new VTCR and telemedical care options was not evaluated in this study. Public awareness will increase over time, as the service was renamed to 1800MEDICARE, a national toll free call number, on January 1, 2026 in order to improve recognition. In addition, a national promotional campaign is planned for mid-2026. Thus, the results reported in this study may represent their impact conservatively and quite possibly underestimate the potential national value of VTCR implementation. Further, patients vary in their access and ability to use self-care resources, access to and comfort with internet-based VTCR and telemedical services, and in their technology and health literacy.21, 22, 23, 24, 25 These patients may be underrepresented in this study population. Future research should endeavor to systematically evaluate these issues.

No detailed examination of more complex demographic, comorbid, care access, health insurance coverage, cultural, or socioeconomic factors was possible in this evaluation, and these influences should be a focus of future research on VTCR and telemedicine utilization in Australia and elsewhere. This study focused on a single, publicly funded health care system that provides universal nurse access and advice, and the patient care intent changes observed in this setting require validation in other care delivery systems. Future studies should consider not only prospective patient follow-up to confirm clinical diagnosis and service utilization, but should also include a randomized design to evaluate more nuanced patient-centered outcomes, including clinical course with and without VTCR use, and influencing factors which induce positive patient change from incorrect pre-triage care intent. Finally, the potential value and impact of VTCR within a broader national virtual front door strategy in Australia have been well described, and future research should be guided by an agenda that aligns with this national vision.⁶

Conclusion

Virtual triage and care referral in Australia is accomplishing the objectives of Healthdirect Australia by better aligning the acuity level of patient care intent with that recommended by evidence-based guidance, and is effectively facilitating the transition of patient care from in-person to telemedicine for nonurgent and emergency care services. Simultaneously, the number of patients that remain uncertain of or do not know what kind of care to obtain has fallen dramatically. Patient intent to use in-person emergency care declined as the implementation of virtual emergency care progressed, and a shift to lower acuity level care intent was observed for the entire patient cohort and particularly those who intended pre-triage to visit an ED. If the magnitude of changes in care intent acuity level achieved in the state of Victoria is applied to all Australian states, 2409 clinically unnecessary in-person nonurgent care visits and 19,286 unnecessary in-person emergency care visits would be avoided annually, once alternative care pathway implementation reaches full maturity nationwide. Given the present early stage of public awareness and potential utilization of VTCR within a national virtual front door strategy,⁶ the magnitude of positive outcomes and impact reported in this study may be conservative.

Potential Competing Interests

Dr Gellert, Mr Price, and Dr Marecka are advisers to Infermedica. Ms Kabat-Karabon is an employee of Infermedica. Ms McMahon, Dr Liu, Mr Burger, Dr Ma, and Dr Luckraj are employees of Healthdirect Australia.

Ethics Statement

Patients provided consent before VTCR encounters for their data to be used for research purposes, with all analyses completed and all results reported in a de-identified, anonymous manner in the aggregate. Given the absence of risk of patient harm, ethical review board review was not engaged.

Supplemental Online Material

Supplemental Figure

1