A Hospital Avoidance Pathway With Digitally Supported Follow‐Up for Low Back Pain: An Alternative to Emergency Department Care Authors: Joseph F. Orlando, Maria Zucco, Usha Ritchie, Anne L. J. Burke, Matthew Beard, Michelle Guerin, Saravana Kumar Published: 2026 ABSTRACT ABSTRACTBackgroundLow back pain (LBP) is a common reason for presentation to an emergency department (ED) but LBP‐care in this setting is frequently low‐value, costly and often fails to meet system and patient needs. Other service options are urgently required. This study evaluated a novel care pathway that diverted people with suspected non‐serious LBP seeking ED‐care to a community‐based, hospital avoidance service in a large South Australian health network.MethodsA retrospective quantitative analysis was conducted on data extracted from electronic medical records (demographics, service activity, subsequent presentations/admissions) for patients seen March–December 2024, and analyzed using descriptive statistics. Repeated measures analysis of selected EQ‐5D‐5L items (pain/discomfort, usual activities, anxiety/depression) was conducted using the Wilcoxon signed‐rank test. Semi‐structured interviews undertaken May‐Nov 2024 with consecutive eligible patients explored experiences of care via qualitative thematic analysis.ResultsThere were 143 patients (mean age = 52.1 years, SD = 19.0) managed via this pathway. All received multidisciplinary care and 101 (78%) registered for digitally supported follow‐up. Patients reported significant improvements in pain/discomfort and usual activities (p < 0.05) 2‐weeks following initial evaluation, but not anxiety/depression. Most patients (90%) did not present to a public hospital ED within 90 days of intervention. Interviews (n = 20) identified four key themes: faster access and reducing ED pressure, care that meets expectations, digitally supported care assists recovery, and recommendations for improvements in LBP‐care.ConclusionThis digitally supported hospital avoidance pathway provided an acceptable alternative to ED‐care for patients with LBP with potential to offer significant benefits to the health system. Introduction Low back pain (LBP) is a leading cause of disability [1] and for most people—where serious pathology is not present—clinical guidelines recommend community‐based management [2, 3]. In Australia, this care is often inaccessible or considered ineffective, especially by patients highly impacted by LBP, leading many to seek care from a public hospital emergency department (ED) [4]. In South Australian public EDs, non‐serious LBP is mostly triaged as needing non‐emergency care and patients wait on average five hours to be seen [5]. Perhaps unsurprisingly, given the urgent and emergent nature of this setting, care delivered in the ED has been shown to focus predominantly on traditional biomedical investigations, interventions and strong analgesics [6, 7]. This represents low‐value care and is delivered at high cost to the system [5]. There have been recent calls to improve public funding of, and access to, high‐value community healthcare for non‐serious LBP [8, 9, 10], but government uptake has been limited and demand on the public hospital system remains high. Australian public hospital services are grouped by designated geographical catchment areas and managed by state funded local health networks (or equivalent). In the absence of national reform, these networks are increasingly having to find innovative solutions to manage ED demands from patients with non‐urgent conditions, including non‐serious LBP. The Central Adelaide Local Health Network (CALHN) manages 130,000 patient presentations annually across its two public hospital EDs, of which approximately 1.8% are for non‐serious LBP conditions. LBP‐related ED representation rates have been reported as 7% within 30‐days and 17% in 12‐months [11]. In response to local demands on its EDs for LBP‐care [5, 11], the CALHN developed an evidence‐informed, digitally supported pathway for people with non‐serious LBP. Based on the ‘Optimal System of Care for LBP’ codesigned by the Commission on Excellence and Innovation in Health [12], the digitally supported pathway was created to enable people with non‐serious LBP who would otherwise seek care from an ED to be assessed, treated and briefly supported via a publicly funded, community‐based, hospital avoidance service, known locally as the Urgent Care Hub (UCH). This novel LBP‐pathway was designed to provide better care and outcomes for patients and the health system. It has potential to enable patients to avoid unnecessary hospital presentations whilst simultaneously improving the type of LBP‐care they receive, and this now warrants formal evaluation. This study aimed to evaluate the capacity of this care pathway to divert selected patients with non‐serious LBP seeking emergency care to receive care in a hospital avoidance service. Additionally, this study evaluated health system activity (including subsequent hospital presentations/admissions), patient reported outcomes (pain/discomfort, usual activities, anxiety/depression) and experiences of care (acceptability, alignment with expectations and needs). Methods Study Design This research study was conducted over two phases: a quantitative phase using descriptive analysis of service activity and patient outcomes; and a qualitative phase using qualitative descriptive methodology to summarize findings from semi‐structured interviews of patients' perspectives of clinical care received [13]. Although the least theoretical of qualitative methodologies, it was ideal for exploring this novel topic by summarizing direct patient experiences [14]. Quantitative and qualitative phases were conducted in parallel but equally prioritized, with findings summarized to determine the impact of the LBP‐pathway on patients and suitability as an alternative to ED care. As this study used quantitative and qualitative methods, the manuscript was reported using the checklist of Mixed Methods Reporting in Rehabilitation and Health Services (MMR‐RHS) [15]. Ethics approval was granted by the Human Research Ethics Committees of the CALHN (#15482) and the University of South Australia (#205112). Setting Analysis of retrospective data occurred for patients seen within a 10‐month period from March to December 2024 across two metropolitan UCHs within the XXLHN; the largest local health network in South Australia. Semi‐structured interviews were undertaken between May and November 2024 and occurred via telephone call with the principal investigator (J.F.O.), 4 weeks after the patient's initial management in the UCH. Access to the hospital avoidance service was via referral from a CALHN ED for patients assessed by a triage nurse without formal evaluation by an ED physician; statewide ambulance service; or general practitioner (GP) based in primary care; self‐referrals were not accepted. Referrals to the UCH could be made seven days a week during operating hours of 8:30 a.m.–8:30 p.m. All patients had an initial clinical assessment by the referring clinician and were required to meet predefined referral eligibility criteria: adults (≥ 18 years of age); suspected non‐serious LBP with or without leg pain; Australasian Triage Scale (ATS) categories 3–5 (urgent, semi‐urgent, non‐urgent care) if referred from ED triage; not requiring immediate magnetic resonance imaging (MRI), which was unavailable at the UCH; and providing consent to receive care at the UCH. Patients who did not meet these criteria continued to receive care in the ED. This included patients with suspected serious pathology; ATS categories 1–2 (immediate, potentially life‐threating care); patients requiring immediate MRI; patients that declined UCH referral; and patients presenting outside of the UCH operating hours. Participants Referrals to the UCH were screened from daily clinic lists for eligibility in the study by one investigator (M.Z.). Inclusion in the study required that patients with suspected non‐serious LBP met the UCH referral criteria; patients underwent assessment and management in the UCH for a primary discharge diagnosis of non‐serious LBP with or without referred leg pain; and patients were discharged directly home from the UCH without requiring transfer to the ED for further care. Patients who presented to an ED following their UCH visit and were subsequently diagnosed with serious spinal pathology were retained in the analysis to capture missed pathologies and hospital re‐presentations. Patients were excluded from the study if they did not meet the UCH referral criteria for suspected non‐serious LBP; diagnosed with LBP disorders secondary to serious or specific spinal pathologies (i.e., spinal fracture, cauda equina syndrome, infection, neoplasm) or other primary diagnoses during the UCH visit; transferred to the ED for further care; discharged to a residential or other supported facility; or had spinal surgery in the previous 6‐months. All patients that met the study inclusion criteria were offered digitally supported follow‐up. Patients who consented to participate in this follow‐up care were required to own, and be able to use, a cell/mobile phone, and have sufficient English language, cognitive skills and hearing ability. Those who did not consent or were unable to participate with digitally supported follow‐up care were excluded from this part of the analysis. Selection and Recruitment All patients that met the selection criteria were included in the quantitative analysis. A waiver of consent to audit retrospective patient data was granted in accordance with the Australian National Statement on Ethical Conduct in Human Research [16]. Patients recruited for semi‐structured interviews were required to have engaged sufficiently with digitally supported follow‐up (defined as completing surveys on at least three timepoints) to provide informed feedback. The principal investigator (J.F.O.) who was not involved in patients' clinical care, telephoned consecutive eligible individuals 4 weeks after they were discharged from the UCH to formally recruit and gain their written consent for participation. Given most patients experienced improvements in pain and function, a subset of patients with poorer outcomes from the quantitative phase, including patients who subsequently re‐presented to an ED or required greater follow‐up care from the UCH on discharge, was identified for recruitment to specifically capture these experiences in the qualitative phase. The target sample size (n = 15–20) for the qualitative phase was based on recommendations from a recent systematic review [17]. Intervention Urgent Care Hub (UCH) Clinical assessments were undertaken by a medical officer (specialist physician or general practitioner), nurse and/or physiotherapist and comprised a patient history, screening for serious pathology, vital signs and physical examination. Onsite pathology and radiology services (x‐ray, computed tomography) were ordered as required. Management by the multidisciplinary team included education, advice on self‐management and physical activity, mobilization, and prescription of pain medication as needed. Patients were discharged from the UCH when deemed medically stable and able to independently and safely mobilize—even in the presence of ongoing pain. Discharge summaries and recommendations were communicated electronically to patients' GPs. Further community services, including temporary home support care, allied health or geriatrician review, were arranged through the UCH as required; services that otherwise may not have been available due to out‐of‐pocket costs to patients. Patients who needed hospital admission for further management or escalation of care due to newly identified or suspected serious pathology were transferred via ambulance to the local public hospital ED as the UCH was not an overnight admission facility. Digitally Supported Follow‐Up Care Prior to discharge, eligible patients were referred for physiotherapy‐led, digitally supported follow‐up care. This service provided patients with short‐term (2‐week) clinical monitoring and support. Patients were enrolled by the physiotherapists into the digital platform, Personify Care Pty. Ltd., which invited patients via text message to formally register. Patients who completed registration received automated text messages on days 1, 3, 6, 10 and 14 following their UCH discharge, prompting them to complete electronic surveys that assessed LBP‐related symptoms, functional restrictions and concerns for recovery. A combination of ordinal and free‐text responses was available. Patients could also indicate if they wished to receive a telephone call from a physiotherapist to discuss any concerns regarding their condition or recovery. Data were collated in an electronic dashboard, providing real time visual representation of patient responses, flagging individuals who scored poorly in their surveys and/or requested a telephone call. The dashboard was monitored on weekdays by the UCH physiotherapists and flagged patients were contacted via telephone the same day (or the following business day if surveys were responded to out‐of‐hours). In‐person clinic reviews were then scheduled for further assessment and management as required. The digital platform also provided patients with links to online resources that reinforced self‐management strategies and when to seek further care. The pathway formally concluded after 14 days, at which time it was anticipated that severe pain would have improved and/or that patients had sufficient time to link with their local healthcare providers if ongoing care was required. Patients were also provided an opportunity to exit the pathway at days 6 and 10 if they wished to do so, and open‐text feedback was able to be provided. Quantitative Data Collection and Analysis Data extraction was independently completed by two investigators (J.F.O., M.Z.) using customized data collection sheets. Data were de‐identified with unique study‐specific identification numbers and entered into a bespoke spreadsheet. Patient demographics (sex, age, residential postcode), patient flow (referral source, arrival mode, length of stay, subsequent ED presentations/admissions within 90‐days) and service activity (imaging, medications prescribed, treating disciplines, referrals to other services on discharge) data were extracted manually from electronic medical records. Data relating to patient engagement with digitally supported follow‐up care (acceptance rates, number of surveys completed, telephone calls with physiotherapists, opt‐out timeframes) were extracted from the digital platform. Three items from the electronic surveys that measured patient‐reported outcomes were extracted from the digital platform at baseline, 1‐ and 2‐weeks into follow‐up care. The items were derived from the EQ‐5D‐5L instrument, validated for patients with LBP [18], and covered pain/discomfort, usual activities and anxiety/depression. Each item was scored on a Likert scale using standardized descriptors: 1 = no problems, 2 = slight problems, 3 = moderate problems, 4 = severe problems, 5 = extreme problems. Self‐care and mobility items of the EQ‐5D‐5L were inadvertently administered on a four‐point Likert scale and so were excluded from the analysis. Descriptive statistics were used to analyze patient demographics, flow, service activity and engagement with digitally supported follow‐up. Categorical variables were described with frequencies (%) and continuous variables with means and standard deviations. Patient‐reported outcomes were presented separately for each item as the median, interquartile ratio and distribution (%) of scored levels. Individual differences between baseline, 1‐week and 2‐week timepoints were assessed using the Wilcoxon signed‐rank test, a non‐parametric test for ordinal data. Statistical analyses were performed using IBM SPSS Statistics version 31.0 with significance set at p < 0.05. Qualitative Data Collection and Analysis Mutually convenient times were determined to conduct the interviews via telephone with each lasting 20–30 min in duration. The interviews explored individual patient experiences of LBP care received at the UCH and on follow‐up with digitally supported care. An interview guide using open questions with prompts allowed free‐flowing conversations (Table S1). Mock interviews had been piloted with patients from the UCH not enrolled in the study prior to collecting data, with feedback provided by the research team. The interviews were audio‐recorded, transcribed verbatim and de‐identified. Field notes were recorded in a journal. Data collection ceased when minimal new codes emerged across three consecutive interviews and no additional insights were generated. Data were managed using qualitative analysis software, NVivo version 14 (QSR International Pty Ltd., Melbourne, Australia). A six phase process for thematic analysis involved: initial reading and familiarization of data, generation of initial codes, organizing codes into themes, reviewing themes for appropriateness against the coded extracts, naming and analyzing how themes are defined, and producing the report [19]. Two investigators (J.F.O., S.K.) coded the transcripts, and these were discussed with the broader research team who then developed a thematic map of analysis to describe the dataset. All patients were invited to review a summary of the study's results, but none chose to do so. Research rigor strategies were used throughout the data analysis process, including data triangulation with multiple investigators (J.F.O., M.Z., S.K.), using an audit trail of research decision‐making processes, documenting contextual information of patients' experiences, recording the interviewer's subjective responses through the research process, and reflexive discussions with the research team to critically examine emerging interpretations, analytic decisions and potential researcher assumptions [20, 21]. Reflecting on the research team's assumptions and experiences is important in how these may have influenced the research process. The principal investigator (J.F.O.) is a physiotherapist with clinical experience in managing people with LBP in hospital settings, and he is undertaking a higher degree by research on this topic. His interest in improving the delivery of care for LBP was openly disclosed to patients. He attended workshops run by the University of South Australia on qualitative research, and received expert mentoring by an experienced qualitative researcher (S.K.) in the team. Results Patient Flow and Characteristics Patient flow through the study is summarized in Figure 1. There were 143 patients seen in the UCH who met the study selection criteria. Four patients referred and evaluated for LBP in the UCH were excluded from the study after being diagnosed with serious spinal pathology (n = 2 vertebral fracture), non‐spinal pathology (n = 1 kidney infection), and required transfer to the ED and subsequently admitted for ongoing pain management (n = 1). Characteristics of patients included in the study are summarized in Table 1. All patients (N = 143) were referred for digitally supported follow‐up care; however, 14 patients did not meet the inclusion criteria and 28 did not register with the digital platform, leaving a total of 101 (71%) patients who participated in follow‐up care. Patient flowchart. TABLE 1Characteristics of patients undergoing low back pain care in the Urgent Care Hub (N = 143). n %Age, mean ± SD (range) = 52.1 years ±19.0, (20–91)SexFemale7049Male7351EthnicityIndigenous Australian32Non‐indigenous14098ResidenceMetropolitan13695Regional75Residential postcode, IRSAD decilesDisadvantaged (1–5)6243Advantaged (6–10)8157Referral sourceAmbulance service6948Emergency department triage6848General practitioner64 Note: Data presented as number of patients (%).Abbreviations: IRSAD, index of relative socioeconomic advantage disadvantage; SD, standard deviation. Characteristics of patients undergoing low back pain care in the Urgent Care Hub (N = 143). Note: Data presented as number of patients (%). Abbreviations: IRSAD, index of relative socioeconomic advantage disadvantage; SD, standard deviation. Urgent Care Hub Service Activity This care pathway demonstrated capacity to divert selected patients with non‐serious LBP seeking emergency care to receive care in a hospital avoidance service. All patients (N = 143) received multidisciplinary care. The primary treating clinician was mostly a medical officer (n = 130, 91%). Provision of physiotherapy varied according to the time of day and week. Of patients who presented during weekday hours, 91% (n = 104/114) received physiotherapy, compared with 34% (n = 10/29) of those who presented on weekends or after hours, when physiotherapy staffing was reduced. Of the total sample (N = 143), 135 (94%) patients were prescribed pain medications, of which, simple analgesics (72%) and opioids (54%) were the most common. There were 34 (24%) patients who received imaging, of which, CT lumbar spine (20%) was the most common. The overall mean length of stay in the UCH for this patient cohort was 213 min (SD 91.3). UCH service activity is summarized in Table 2. TABLE 2Urgent care hub service activity (N = 143). n %Occasions of service within clinic hoursWeekday (8:30 a.m.–5 p.m.)11480Weekday (after 5 p.m.) or weekend2920Medications administeredParacetamol4934NSAID8761Any simple analgesic10372Opioid weak2820Opioid strong5236Any opioid7754Benzodiazepine96Gabapentinoid64Tricyclic antidepressant32Investigations orderedX‐ray spine75Computed tomography spine2820Any spinal radiology3424Other (X‐ray pelvis/hip/knee)64Treating cliniciansMedical13091Nursing143100Physiotherapy11480Length of stay, mean ± SD (range) = 213 min ±91.3, (60–643)Subsequent public hospital presentations within 90‐daysEmergency department1410Hospital inpatient admission64 Note: Data presented as number of patients (%).Abbreviations: NSAID, non‐steroidal anti‐inflammatory drug; SD, standard deviation. Urgent care hub service activity (N = 143). Note: Data presented as number of patients (%). Abbreviations: NSAID, non‐steroidal anti‐inflammatory drug; SD, standard deviation. Digitally Supported Follow‐Up Care Activity There were 101 (71%) patients who registered with digitally supported follow‐up care, and data on this service activity is summarized in Table 3. Self‐assessment surveys were completed by most patients (n = 65, 64%) by the end of the first week. This reduced to 52 completions (51%) at the end of 2 weeks; 12 patients formally opted out of the digital supported care after completing the 6‐day survey with all cases indicating improved outcomes. Physiotherapists made telephone calls to most patients (n = 84, 83%) at some point during the 2‐week pathway; 37 patients requested a telephone call and the remaining patients were called based on their flagged survey responses. There was an average of 1.9 telephone calls (SD 1.5) made per patient. In‐person physiotherapy appointments at the UCH were arranged for seven patients who reported high levels of pain, and three of these patients subsequently re‐presented to hospital. In‐home or telephone appointments with other allied health or multidisciplinary geriatric services were also arranged for nine patients; three of whom subsequently re‐presented to hospital. TABLE 3Digitally supported follow‐up care activity (N = 101). n %Self‐evaluations completedBaseline9089Week 16564Week 25251Physiotherapy telephone calls017171262623232≥ 32626Physiotherapy clinic appointments09493177Other service referralsPharmacy telephone call44Social work telephone call22Occupational therapy home assessment22Geriatrics multidisciplinary home service11 Note: Data presented as number of patients (%) who registered for digitally supported follow‐up care. Digitally supported follow‐up care activity (N = 101). Note: Data presented as number of patients (%) who registered for digitally supported follow‐up care. Patient‐Reported Outcomes Patient‐reported outcomes are presented as the median and interquartile ratios at each timepoint in Table 4 and the distribution in the responses in Table S2. Individual differences in outcomes between timepoints are presented in Table 5. Results indicated significant improvements with large effect sizes for pain/discomfort and usual activities following initial UCH evaluation, with greatest differences indicated between baseline and 2‐weeks (p < 0.001, r = −0.88; p < 0.001, r = −0.76 respectively). In contrast, anxiety/depression was relatively stable over time, with median scores remaining at a moderate level. TABLE 4Patient reported outcomes during digitally supported follow‐up care.ItemsBaseline1‐week2‐weeks n = 90 n = 65 n = 52Pain/discomfort4 (3–4)2 (1–3)1 (1–3)Usual activities2 (1–4)1 (1–3)1 (1–2)Anxiety/depression3 (2–3)3 (1–3)2.5 (1–3) Note: Items were derived from the EQ‐5D‐5L instrument and scored on a Likert scale using standardized descriptors: 1 = no problems, 2 = slight problems, 3 = moderate problems, 4 = severe problems, 5 = extreme problems. Data presented as median (interquartile ratio). Patient reported outcomes during digitally supported follow‐up care. Note: Items were derived from the EQ‐5D‐5L instrument and scored on a Likert scale using standardized descriptors: 1 = no problems, 2 = slight problems, 3 = moderate problems, 4 = severe problems, 5 = extreme problems. Data presented as median (interquartile ratio). TABLE 5Differences in individual patient reported outcomes during digitally supported follow‐up care.ItemsTime comparisonTotal pairs (N)Non‐zero differences (n) W Z p Effect size (r)Pain/discomfortBaseline versus 1‐week653713.5−5.15< 0.001−0.85Baseline versus 2‐weeks52373.5−5.35< 0.001−0.881‐week versus 2‐weeks462333.0−3.280.001−0.68Usual activitiesBaseline versus 1‐week653799.5−3.86< 0.001−0.58Baseline versus 2‐weeks523441.5−4.43< 0.001−0.761‐week versus 2‐weeks461816.5−3.090.002−0.73Anxiety/depressionBaseline versus 1‐week6530139.5−1.960.050−0.36Baseline versus 2‐weeks5226110.5−1.720.085−0.341‐week versus 2‐weeks461344.0−0.110.913−0.03 Note: Non‐parametric testing using the Wilcoxon signed‐rank test for paired samples with statistical significance set at p < 0.05. Differences in individual patient reported outcomes during digitally supported follow‐up care. Note: Non‐parametric testing using the Wilcoxon signed‐rank test for paired samples with statistical significance set at p < 0.05. Hospital Presentations Of the 143 included patients seen in the UCH, most patients (n = 129, 90%) did not present for further care at a public hospital ED within 3‐months (90‐days) of being seen in the UCH. Of those who did (n = 14, 10%), seven attended for difficulty managing pain; five with missed spinal pathologies (fragility vertebral fractures = 3, herpes zoster = 1, spinal metastases = 1); and two proceeded for elective surgery for claudicant neurogenic leg pain secondary to spinal canal stenosis. Six of these 14 patients were admitted to hospital, all of whom were aged > 65 years. Moreover, of those who subsequently attended an ED, nine had participated in the digitally supported follow‐up pathway and three of the admitted patients had received significant physiotherapy input with digitally supported follow‐up prior to admission. Qualitative Findings Twenty‐two patients were invited and 20 consented to participate in semi‐structured interviews. The mean age of this subset of patients was 49.0 years (SD 11.7), of which just over half (n = 11, 55%) were male, and most (n = 17, 85%) had participated in digitally supported follow‐up care for the entire 2 weeks of the pathway. From this group, one patient presented to ED within the first week of being discharged home from the UCH due to difficulty managing pain (participant 1) but remained on the pathway till 2 weeks, and one person received multidisciplinary geriatric services (participant 19). A thematic analysis map was developed with codes grouped into four themes. Supporting patient quotations are provided in Table S3. Theme 1: Faster Access, Reducing ED Pressure Patients attributed considerable importance to accessing timely care to manage their condition, especially when experiencing high levels of pain. Most interviewees expressed a preference to have their LBP managed in a UCH rather than an ED as they reported experiencing shorter wait times for care in this setting compared with what they expected to encounter in an ED. Many also said that they believed it was appropriate to receive care in an UCH as they recognized that their condition was not “life‐threatening” [P15] and “not that big of an emergency” [P3]. They described benefits of being managed in this service, with some expressing the opinion that they had assisted the health system by alleviating pressures in EDs rather than “block the hospital up” [P10]. Theme 2: Care That Meets Expectations Patients reported high levels of satisfaction with the care they received in the UCH. They characterized it as being “thorough”, that it “worked through everything” [P5] and provided a range of management options “not just [being] medicated” [P6]. During a period of reported uncertainty and significant distress, patients also valued being treated by a clinician that “communicates with you and reassures you” [P18] and provides “really good information on how to manage… to put my mind at ease” [P5]. While some patients reported leaving the UCH with ongoing concerns regarding pain and restricted function, on reflection they acknowledged that implementing the advised management strategies on return home had been appropriate as they “helped a lot in the end” [P12]. Theme 3: Digitally Supported Care Assists Recovery Patients valued the support provided via digital care on discharge from the UCH. They described it as having given them “confidence” [P8] and “peace of mind” [P7] to recover at home, rather than feeling “abandoned” [P18] by a healthcare service. One patient asserted that digital care provided “tools and some more knowledge” [P4] while another stated that it “kept me on track in managing myself” [P17]. Some patients who reported having experienced ongoing concerns regarding their symptoms wondered if the digital pathway had potentially helped them to avoid “having to go back and see (the GP) … or go to ED” [P8] because of the prompt and ongoing support provided. One patient also recounted how he had been able to access further community healthcare services “all based on the inputs that the team (at the UCH) gave me” [P15]; bridging the gap between emergency and primary care for patients with LBP. Theme 4: Recommendations for Improvements in LBP‐Care Minor barriers were identified by patients who had received care for LBP at the UCH, the most common of which was around service availability. One patient stated, “The only thing I hated about it was it shut” [P10], asserting the belief that her care had been rushed to enable the facility to adhere with its scheduled closure time. Other patients expressed frustration with the fact that they had not received input from a physiotherapist, which was “the whole reason they had referred us there” [P8]. This appeared to align most commonly with times when physiotherapy staffing would have been reduced (e.g., weekends and evenings) and although not provided at the first point of contact, all of these patients said they had been subsequently offered an UCH physiotherapy appointment during business hours either via telephone or in‐person. Another patient reflected on the UCH requirement for an approved referral pathway into the service and suggested that direct access (e.g., via self‐referral) would have ensured rapid care “instead of waiting for two hours for the ambulance” [P20]. Discussion This study investigated a novel, digitally supported community‐based, hospital avoidance pathway for patients with suspected non‐serious LBP who were seeking emergency care. Quantitative and qualitative findings aligned to demonstrate a range of positive patient impacts. Improvements in pain/discomfort and function (usual activities) were observed 2‐weeks following initial evaluation in the UCH in keeping with the natural history of acute LBP [22], while mood (anxiety/depression) was unchanged. The service was largely perceived by patients as meeting their needs and providing timely, comprehensive and supportive care that enabled ongoing management in the community. This study also indicated that the pathway provided a range of care improvements compared to what is currently provided for LBP in South Australia's public EDs [5, 11]. These included reduced treatment times and a more comprehensive multidisciplinary approach, which incorporated allied health and digitally supported community‐based follow‐up. Some aspects of care warrant further review. For instance, while spinal imaging requests and medication prescriptions were comparable to those associated with ED‐based LBP care in Australia [5, 11, 23, 24], they were higher than clinical guidelines might recommend. As such, the UCH may benefit from targeted interventions such as clinician and patient education to enhance provision of guideline‐consistent care [25]. The study's 10% rate of ED presentations following the UCH visit was not unexpected, and not too dissimilar to locally reported ED re‐presentation rates for LBP [11]. A comparison of ED re‐presentation rates for LBP at consistent timepoints between traditional ED and the new UCH pathway will form the basis of a subsequent study. Importantly, unlike traditional ED care, the UCH offered patient monitoring and follow‐up services on discharge through a hybrid digital pathway allowing UCH clinicians to escalate patient care when required. Among the re‐presentations, some cases were reviewed and discharged without changes to management, while two cases proceeded to spinal surgery after a considerable trial of appropriate conservative management. Although five cases of missed diagnoses were identified, the overall rate was low and consistent with reported rates (5.7%) of diagnostic error following ED presentations [26]. Evaluation of patients with LBP in the UCH was completed by a multidisciplinary team of emergency care clinicians, making it unlikely that UCH care disadvantaged patients compared to standard ED. Rather, the UCH pathway enabled timely assessment by a multidisciplinary team in a low‐cost setting, with digitally supported follow‐up that facilitated escalation of care when required. Whilst the study did not directly compare outcomes between patients receiving and not receiving multidisciplinary care, patients did express high levels of satisfaction when treated by a multidisciplinary team. This is consistent with findings from a recent qualitative study of patients discharged home following hospital admission for LBP where patient expectations of care were met when supported by a multidisciplinary team, despite experiencing high pain impact [27]. Multidisciplinary care may also have an important role in supporting patients with LBP at risk of poor outcomes, such as older people [28, 29]. The subgroup of people that experienced higher incidences of hospital presentations and admissions in this study were older individuals, a finding that is consistently reported across the literature [30]. There may be a role to expand multidisciplinary geriatric services for some older people with LBP. This warrants further investigation, in addition to the most effective ways to support this patient cohort manage their condition in the community, and where possible, avoid hospital presentation. Digitally supported follow‐up care was also implemented in this study to briefly support patients to manage their LBP out of hospital on discharge from the UCH. A hybrid model of digital care was utilized to monitor patient outcomes, empower patients to self‐manage their condition, and provide brief clinician support as required, mostly via telephone. This allowed clinicians to target patients who required further supports where there were concerns for recovery. Our study contributes to the literature that patient recovery from LBP can be briefly supported using a hybrid digital care model with low use of clinical resources. The benefit and acceptability of digital care for LBP following an ED presentation have also been reported by a qualitative study set in a large Australian hospital [31]. The digital care pathway was delivered with low burden on health service and workforce resources, and given the positive patient experiences and engagement, reflected a feasible follow‐up intervention. Whether digitally supported follow‐up care can identify patients at risk of poor outcomes, including hospital presentation and admission, needs further investigation. Combined, these findings suggest that the UCH‐care pathway was an acceptable alternative to ED‐care for patients with LBP and as such, has significant potential to benefit the health system. The next step in understanding the impact on the health system is to undertake an economic evaluation. The UCH's low‐cost setting is likely to be cost‐effective compared to the ED. Allied health‐led care is also delivered at lower costs compared to traditional medical care, arguing for broader investment in multidisciplinary, community‐based care for LBP. This will form the basis for a subsequent study. Strengths and Limitations This study is strengthened by its use of quantitative and qualitative research paradigms, the collection of data from frontline care, and grounding in the real‐world context of healthcare delivery. A rigorous approach to patient recruitment was undertaken to ensure appropriate selection of patients to have their care transferred from an ED to the UCH with only a small proportion excluded from the study after being diagnosed with serious or non‐spinal pathology in the UCH and/or transferred to the ED for further care. There are several limitations including that some eligible patients may have been missed. Identifying patients using discharge diagnostic codes was not available, as the UCH operated using a mix of inpatient and outpatient models, the latter not coded. A longer recruitment timeframe may have increased sample representation and improved robustness of the dataset. Not all missed diagnoses may have been captured in this study, as identification was restricted to hospital medical records of re‐presentations and did not include primary care records, where subsequent diagnoses might have been identified. Regarding qualitative data capture, selection bias was inherent as interviews were conducted with patients who had high engagement with digitally supported follow‐up care. This was intended to capture the experiences of patients engaged with digital follow‐up; however, including patients who did not register or who had low participation may have yielded different insights. Some patients were excluded from digitally supported follow‐up care if they were unable to engage with digital technology. This may limit the generalizability of the findings for certain cohorts, particularly older people and those from culturally and linguistically diverse backgrounds. As previously noted, errors in administration of the EQ‐5D‐5L instrument meant that some items could not be analyzed. Implementing the instrument in full and with longer follow‐up timeframes would have strengthened the rigor and impact of the study. Lastly, the study's setting is unique to the local health network limiting the generalizability of findings. Provision, access and funding of hospital avoidance services vary across Australian states, and these may not be available in regional areas. The care pathway's generalizability may also be challenging in international healthcare systems, including the United States of America. However, components of this pathway may be transferrable, including collaborations between hospital and community healthcare services, optimizing allied health—especially physiotherapy—in the care of patients with LBP in emergency settings, and the use of digital technology to support follow‐up with a relatively low burden on health service and workforce resources. Implications Future studies using more rigorous research designs and larger patient recruitment are needed to investigate the full impact of this LBP‐pathway on patient outcomes, health systems and society at broad. Digitally supported care could be adapted for other low acuity conditions that can be followed up in the community, such as musculoskeletal disorders, minor injuries and wounds, and minor infections. Evidence for these models of care would inform policy makers and hospital administrators to appropriately fund and implement these pathways providing clinicians and patients with alternatives to an ED where appropriate. Further research is needed to determine the most effective ways to deliver evidence‐based care in the most appropriate setting for patients seeking emergency care for LBP. Studies may build on the multidisciplinary approach to care, as well as offering community‐based services delivered either in the home, clinic or via digital technology providing patients with continuity of care and support as they recover from LBP. This study provides preliminary evidence that a hybrid model of digital care to support people following discharge is acceptable, effective for pain and function, and delivered with low service demand. Future research could explore ideal digital follow‐up timeframes for LBP and the ability for technology to identify those patients at risk of poor outcomes. This could lead to the development of validated tools that predict hospital presentations or admissions for LBP to enable clinicians to best support this patient cohort. Conclusion This hospital avoidance service provided an acceptable alternative community‐based pathway for patients with non‐serious LBP seeking emergency care. It adopted a multidisciplinary model with digitally supported follow‐up, enabling ongoing recovery in the community while identifying patients with poor outcomes or persistent concerns who required further care. It benefited patients by delivering rapid comprehensive care that aligned with patient needs and expectations as patients experienced improvements. This pathway has potential to provide significant benefits to the broader health system and warrants formal evaluation. Author Contributions This study was designed by J.F.O., M.Z., U.R., and S.K. Data was acquired by J.F.O. and M.Z., and analyzed and interpreted by J.F.O., M.Z., U.R., A.L.J.B., and S.K. The manuscript was first drafted by J.F.O., and critically revised by M.Z., U.R., A.L.J.B., M.B., M.G., and S.K. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. Funding The principal investigator (J.F.O.) received a seed funding grant through the Allied and Scientific Health Office (ASHO), Department of Health and Wellbeing, SA Health, and was supported by an Australian Government Research Training Program Scholarship that funds his Higher Degree by Research. Conflicts of Interest The authors have declared that no competing interests exist. The provider of the digital platform, Personify Care Pty. Ltd., had no involvement with the study's conception, design, analysis, interpretation of results or report writing. Supporting information Table S1: Interview guide. Table S2: Distribution of patient reported outcomes during digitally supported follow‐up care. Table S3: Supporting patient quotations from semi‐structured interviews.