Clinician’s capsule

What is known about the topic?

During the Covid-19 pandemic, ambulance calls have decreased in many jurisdictions.

What did this study ask?

In British Columbia, what are the yearly changes in ambulance calls from spring 2020 compared with spring 2019?

What did this study find?

We found a 15% overall reduction in calls extending to every health region. All chief complaints were decreased except dyspnea and anxiety. Critical illness calls decreased 9%.

Why does this study matter to clinicians?

There should be planning for fewer overall ambulance transports and ED visits. Delayed presentations of critical illness could result in atypical presentations.

Introduction

Recent reports have described significant changes in patients accessing emergency health services (EHS) during the COVID-19 pandemic, [1,2,3,4,5] although results range from stable volume for ST-elevation myocardial infarction and stroke [2, 3], to an increase in out-of-hospital cardiac arrest [4, 5]. These observations have been generally recorded in areas with a high incidence of COVID-19. Since EHS service is optimized if the response matches patient need, EHS and emergency department (ED) planners can plan such responses around changing encounter patterns. We examined all-cause EHS utilization in British Columbia (BC), Canada, a jurisdiction with a low COVID-19 incidence [6], to see if encounter patterns also decreased.

Methods

Study setting and design

This observational study used aggregate provincial data collected and maintained by BC Emergency Health Services, (BCEHS) the sole prehospital provider covering nearly one million square kilometers. We examined all out-of-hospital EHS encounters within BC (population 5,031,893 and 5,110,971 in the first quarters of 2019 and 2020, respectively). The province is divided into five health authorities: Fraser Health Authority (FHA; approximate population 1.9 million persons), Interior Health Authority (IHA; 745 000), Northern Health Authority (NHA; 300 000), Vancouver Coastal Health Authority (VCH; 1.25 million), and Vancouver Island Health Authority (VIHA; 870 000).

The first BC case of COVID-19 was identified on January 26, 2020, and 100th on March 14, 2020 [6]. On March 15, the provincial government closed schools and non-essential businesses and encouraged people to remain at home. On June 4, the provincial social contact rate was estimated at 40% of normal [7]. The institutional ethical review boards of Providence Health Care and the University of British Columbia approved this study. We adhered to STROBE criteria [8].

Data collection

BCEHS uses Siren (Medusa Medical Technologies, Halifax, Nova Scotia) to create electronic records for all encounters. These include transfers from home to hospital (99.8% of total) and hospital to non-hospital location (0.2% of total) but not-interhospital transfers. (Sung Lee, personal communication September 29, 2020) Paramedics select a mandatory diagnostic code from a standardized list of complaints, such as chest pain or dyspnea. We report the 18 most common codes. We also pre-specified critically ill subgroups: OHCA, stroke, STEMI, and major trauma. (Appendix 1).

Outcomes

We compared EHS encounter incidence rates per 100,000 person-months overall and for each of the most common 18 diagnostic codes between March 15 and May 15, 2020 to the same period in 2019, and calculated incidence rate differences with 95% confidence intervals. To ascertain whether patient decisions to call BCEHS were associated with local incidence of COVID-19, we analyzed the incidence rate difference in each of BC’s five health regions and compared this to the COVID-19 incidence in each of the health regions.

Analysis

Data were directly imported into an Excel spreadsheet and analyzed with R version 3.6.3. (Foundation for Statistical Computing, Vienna) We used descriptive methods for assessment.

Results

By May 15, 2020, the province had a cumulative total of 2425 laboratory-confirmed COVID-19-positive cases from 116,802 tests, for a 2.1% positive rate [7]. Comparing 2019 to 2020, total EMS encounters between March 15 and May 15 decreased from 83,925 (incidence rate 834 per 100,000 person-months) to 71,611 (incidence rate 701 per 100,000 person-months) for a decrease of 133 per 100,000 person-months (95% CI 126–141). Of the 18 prespecified codes, sixteen had a decrease in calls, most notably chest pain, abdominal pain, and syncope/collapse (Fig. 1). Two codes increased significantly: respiratory distress/symptoms from 31 to 35 per 100,000 (incidence rate difference 3.8 per 100,000 person-months, 95% CI 1.2–4.3) and anxiety from 17 to 21 per 100,000 (difference 3.4 per 100,000 person-months, 95% CI 2.2–4.6). Encounters for critically ill patients decreased significantly overall from 3019 to 2753 (incidence rate difference 3.1 per 100,000 person-months, 95% CI 1.6–4.5), including stroke, trauma, and STEMI. The incidence of OHCA was not statistically different.

Fig. 1
figure 1

March–May 2020 versus March–May 2019 change in British Columbia ambulance attendance per chief complaints

Full size image

While EMS encounters declined in all regions, the absolute reduction in incidence rate ranged from 70 per 100 000 person-months (95% CI 39–101) in the NHA to 179 per 100 000 person-months (95% CI 160–199) in VIHA. Dividing absolute regional reduction (in 100000 person-months) by the number of regional COVID-19 cases, results ranged from 81/1154 cases [0.070 per case] in FHA to (179/121 cases [1.5 per case]) in VIHA. (Appendix Fig. 1).

Discussion

In a province with a relatively low incidence of COVID-19, we demonstrate that all-cause EMS utilization decreased by 15% when compared with a historical control; while trauma, stroke, and STEMI decreased by 9%, and cardiac arrests did not decrease. Importantly, EMS calls for respiratory complaints and anxiety increased significantly during this time, similar to Israeli data [9]. Our findings extend to all health regions, and patient decisions to contact the ambulance were not associated with local COVID-19 incidence.

Studies have shown an apparent short-term decrease in EHS encounters [1, 9], or hospital visits for chest pain or myocardial infarction [2], stroke, [4] but a potential increase in OHCA [3]. These data have arisen from regions with higher incidences of COVID-19. Despite BCs low incidence, our results indicate significantly decreased EHS encounters across a broad spectrum of complaints, including critical illness. Patients might be reluctant to seek help due to concerns about contacting COVID-19, or a desire to not “burden” the health care system, perhaps bolstered by reports from regions that have been dramatically affected.

The reasons for decreased EHS encounters is unclear, illustrated by the disconnect between regional COVID-19 incidence and changes in encounters. Different health regions may have a different population age and comorbidity profile, and older, sicker patients may contact EHS at different rates. Furthermore, patients may have contacted EHS due to lack of primary care, and this may have been different between regions.

These findings are important to EHS and ED planners and clinicians. While some might defer calling EHS or attending an ED without consequence, critical illness cannot be deferred, and such patients will need to eventually attend an ED. For example, delayed STEMI could lead to severe complications such as heart failure, arrhythmia, or valve dysfunction, as observed in Italy [10]. Furthermore, with the corresponding increase in anxiety-related complaints, EHS and ED planners should ensure that mental health capacity is increased.

This analysis has limitations common to similar recent studies [1,2,3,4,5, 9, 10] including a short time frame at the start of the pandemic and unclear future directions. Inclusion of years prior to 2019 may have provided a different baseline. Paramedics classified all codes and interobserver reliability is unknown, but there is no evidence to suspect systematically different coding during the pandemic. Data are aggregated and results, therefore, unadjusted for age, comorbidity, illness severity, availability of primary care, or repeat visits; it is possible certain groups of patients are over- or underrepresented in our population.

Conclusion

In a single province with a low incidence of COVID-19, there was a 15% reduction in overall EHS use and a 9% reduction in critical illness. EHS planners will need to match patient need with available resources. Furthermore, there may be increased complications from delayed presentations of critically ill patients.