Potentially avoidable transfers and emergency re-presentations among Australian rural and remote patients receiving low-intensity healthcare supported by a virtual advisory service: a cross-sectional study

Background Virtual health services improve rural-remote community access to emergency healthcare and are potentially resource-efficient by minimising interhospital transfers. Potentially avoidable transfers (PATs) and emergency department re-presentations among patients receiving low-intensity care may use resources unnecessarily, reducing the efficiency and sustainability of virtual health services. Our study aimed to quantify and identify factors associated with PATs and re-presentations among rural-remote low-intensity care (TISS ≤ 2) patients supported by a virtual critical care advisory service (vCare). Methods An observational cross-sectional study was conducted for vCare-supported patients receiving TISS ≤ 2 care in rural-remote public health facilities (Modified Monash (MM) remoteness scores 4–7) of western New South Wales, between 1 February and 31 March, 2021. Data were retrieved from 1600 electronic medical records. Transferred patients were received by MM3 facilities in western NSW. Re-presenting patients made unplanned return visits to their original facility within 72 hours of discharge for ongoing concerns related to the original condition. Pearson’s chi-square determined associations between transfer status, re-presentation status and other study variables. Alpha-levels were set at 0.05 for two-by-two comparisons and adjusted using Bonferroni’s correction for pairwise comparisons exceeding one. Odds ratios and Cramer’s V determined effect size. Results Of TISS ≤ 2 patients, 28% were transferred, with those among 0–4-year-olds significantly higher than other age groups (P < 0.0001). Also, 20% re-presented within 72 hours of discharge, with non-transfers more likely to re-present (24%) than transfers (10%) (OR = 2.906, 95CI:1.654, 5.104, P = 0.0001). Clinical conditions for which non-transfers were more likely to re-present than transfers included abdominal pain (OR = 9.44, 95CI: 2.55–34.94, P = 0.0001), chest pain (OR = 5.06, 95CI: 1.36–18.80, P = 0.0138) and injury (OR = 3.27, 95CI: 1.11–9.66, P = 0.0457). Conclusions Results challenge the notion that virtual health services are resource-efficient. High transfer rates among low-intensity care patients may reflect low risk tolerance among rural clinicians while high re-presentation rates among non-transferred patients may reflect limited access to follow-up care, limited understanding of condition trajectory and low confidence to self-manage their condition. Enhanced support for rural-remote clinicians may reduce avoidable transfers, while enhancing discharge procedures may reduce re-presentations.