Monitoring ambulance calls for cardiovascular stress during epidemics

Outbreaks are often recognised after laboratory confirmation, but the earliest available clues may be changes in ordinary health-service demand. Ambulance records can reveal a rapid rise in calls for chest pain, an indirect marker of cardiovascular stress that may accompany severe respiratory infection, systemic inflammation or disruption to routine care. Used carefully, these records can shorten the interval between an emerging epidemic and a coordinated public-health response. Learn more about Case Studies.

A call for chest pain is not proof of infection, nor is every cardiac event caused by a contagious disease. Fever, dehydration, hypoxia, inflammatory and clotting responses, and delayed access to medicines or medical review can affect cardiovascular risk. Influenza, COVID-19 and other outbreaks may also coincide with more emergency presentations among people with existing heart disease. The signal therefore needs clinical context, seasonal adjustment and comparison with multiple data sources. Learn more about Data Use.

For Australia, the opportunity is to adapt Japan’s multi-channel syndicated surveillance approach rather than copy it wholesale. Ambulance dispatch data can be combined with emergency departments, hospitals, general practices, pharmacies, schools, aged-care facilities and laboratory reporting. This wider view helps distinguish a true change in illness from weather, pollution, media coverage, altered healthcare access or normal seasonal variation.

Why chest pain calls can signal epidemic strain

Ambulance dispatch data are valuable because they arrive close to symptom onset and cover events that may not reach emergency departments. Chest-pain complaints can be grouped with cardiac arrest, suspected heart attack, breathlessness, collapse and other cardiovascular symptoms to form a broad measure of acute distress. A wider definition may be more stable than a narrow myocardial infarction code, which is often assigned only after hospital assessment and is unavailable at the point of call.

During an infectious disease wave, several pressures can converge. More people may feel breathless, weak or dehydrated; existing cardiovascular disease can destabilise; and primary-care appointments, medication repeats or elective procedures may be interrupted. Busy waiting rooms can also delay recognition of warning signs. A surge in chest-pain calls may therefore represent both worsening disease and harder access to care, giving public-health teams a reason to examine respiratory illness and community transmission at the same time.

The pattern must be interpreted against alternatives. Hot weather in Melbourne, humid conditions in Sydney, smoke events, severe influenza seasons and changes in ambulance access can all alter cardiovascular demand. An apparent increase caused by a coding change, population growth or a new private transport provider would be misleading. Surveillance works when an alert describes an unusual departure from expected conditions and prompts investigation, not when any single rise is labelled an epidemic.

Building a reliable ambulance surveillance indicator

A workable system begins with a data dictionary that maps dispatch reasons consistently across participating services. Each event needs a timestamp, broad location, age band, response priority and a standardised symptom group. Free-text notes may improve classification, but extracting them should be limited to approved variables. Duplicate calls, transfers between providers and changes to dispatch terminology should be identified before an event enters the surveillance stream.

The analytical layer should compare the current period with a multi-year baseline using the same months and day of week. Heat warnings, influenza positivity, air-quality data and major public holidays can be included as contextual variables. An alert can be triggered by a sustained rise in call volume, a higher proportion of high-priority responses, or both. Statistical change is more meaningful when paired with clinical severity, ambulance workload and emergency-department pressure.

Japan shows how ambulance signals can sit within a broader syndicated system rather than operate as a stand-alone warning. Clinics, hospitals, pharmacies, schools, elderly-care facilities and laboratories can each provide a different view of community illness. Comparative case studies on syndromic surveillance illustrate how multi-source indicators can be combined without treating them as interchangeable. In Australia, the same principle applies: an ambulance rise should prompt checks of respiratory symptoms, medicine demand, school absence and test positivity, not a declaration that an outbreak has begun.

Connecting ambulance data with Australia's health system

Australia’s geography and service structure make local implementation important. Ambulance operations are divided among bodies such as NSW Ambulance, Ambulance Victoria, Queensland Ambulance Service and SA Ambulance Service, with different dispatch systems and coverage patterns. Metropolitan services, rural and remote operations, and private non-emergency transport can therefore produce signals that are not directly comparable. A national framework needs common definitions while preserving the detail needed for local response.

Everyday conditions also shape the data. Melbourne’s repeated heatwaves, Sydney’s humid summers, crowded peak-hour public transport and interrupted exercise routines can influence cardiovascular stress. Long travel distances and thin service coverage outside major cities can affect when people seek help and whether ambulance crews are the first point of clinical contact. These realities call for baselines built by region and season, with community input where remoteness, disadvantage or cultural barriers affect access and trust.

Coordination should use existing national and state structures. The Australian Centre for Disease Control and state and territory communicable-disease units can interpret signals alongside notifiable-disease notifications, general-practice reports, hospital admissions, pharmacy sales, school absenteeism, aged-care alerts and laboratory results. The local health market is not uniform: publicly funded ambulance services, interstate contracts and private patient-transport operators may all contribute records. Major international events also create predictable surges in visitors and emergency demand, so enhanced monitoring should establish event-specific baselines before they begin.

Privacy, equity and governance

Ambulance records remain health information even when names are removed. Australia’s Privacy Act 1988 and applicable state and territory health-records legislation set important duties, while professional confidentiality rules and contractual obligations may also apply. Surveillance teams should use the minimum data needed, restrict access by role, log queries and publish transparent methods. Small geographic or demographic cells need special controls because they can identify people in small communities.

Governance should also cover the secondary use of information. A framework for responsible data use can specify lawful purposes, de-identification standards, retention periods, data-sharing agreements and independent oversight. Research or operational analysis should not be repurposed for commercial targeting, policing or individual performance management without a separate ethical and legal basis. Public transparency statements can explain what is collected, why it is needed and how alerts differ from individual clinical decisions.

Equity checks are part of data quality, not an optional addition. A national average can hide delayed access or increased risk in regional, remote, Aboriginal and Torres Strait Islander, or lower-income communities. Communities affected by surveillance should be involved in deciding how information is interpreted and governed. False alarms carry costs too: unnecessary public anxiety, disrupted clinical services and diverted resources can damage trust, so every alert needs a defined review and escalation pathway.

Operational recommendations for outbreak response

A practical Australian programme would use ambulance data as an early contextual indicator of cardiovascular strain and service pressure. Its purpose would be to trigger faster, cross-sector investigation during a possible infectious disease wave while protecting routine clinical care and patient privacy. Four measures make that purpose operational and testable.

• Adopt a national minimum dataset and common syndrome groups, including chest pain, dyspnoea, syncope and cardiac arrest, with consistent coding, deduplication and regional coverage rules.
• Establish seasonally adjusted baselines and tiered alerts that combine call volume, response priority, symptom severity and ambulance workload, with heat, pollution and influenza indicators considered.
• Pair every alert with a rapid review of emergency-department presentations, pharmacy sales, school absenteeism, aged-care symptoms and laboratory positivity so investigators can triangulate likely causes.
• Pre-agree privacy, community governance, escalation, communication and evaluation procedures, including clear thresholds for standing down an alert and auditing whether responses reduced harm.

Implementation can begin with retrospective, de-identified ambulance data to test definitions and establish local baselines. A pilot could compare a metropolitan service with a regional service, then incorporate emergency-department and pharmacy indicators before expanding nationally. Historical data must be used to determine whether the proposed alert thresholds would have identified known influenza or COVID-19 periods, heatwaves and major events without generating excessive false positives.

The immediate action is for health departments, ambulance services and public-health units to convene a cross-jurisdictional working group, agree the minimum dataset, nominate pilot sites and publish a response protocol within the next planning cycle. That collaboration could turn an indirect cardiovascular warning into early, evidence-led investigation while preserving patient privacy and routine care.