In this study, we found an overall increase in the number of non-conveyed patients and a decrease in the number of patients brought to a hospital during the COVID-19 pandemic compared to a control group of prehospital patients sampled the year before the pandemic. Despite a reduction in the number of patients admitted to hospital for diseases of the respiratory system, the overall mortality in the prehospital patient population increased during the COVID-19 pandemic. We further found that when the COVID-19 restrictions on society were loosened (July–September 2020), the reduction in the number of patients contacting the emergency medical dispatch centre was replaced by an increase in patient volume. This increase in patient volume appearing as restrictions were loosened surpassed the number of missions carried out the year before the pandemic.

Our findings are not uniformly supported by other studies. The burden imposed on the EMS by the COVID-19 pandemic thus differed in different parts of the world. Where some countries or EMS reported a reduction in the number of ambulance missions during the COVID-19 pandemic [5,6,7, 33], other systems reported massive increases in demands on the EMS resources or increased response times [8,9,10,11]. Especially the initial surge of the COVID-19 pandemic developed differently worldwide. This may explain the differences in EMS demand among countries. Several explanations for the apparent reduction in the need for ambulance transportation to the hospital have been proposed including effects of lifestyle changes. The lockdown measures may have reduced the risk of traffic injuries or injuries during recreational activities as shown in studies from trauma centres [13,14,15,16,17, 34]. Our study supports this, as we found a decrease in distress calls (1–1-2 calls) caused by accidents and a decrease in the number of patients assigned diagnoses of injuries.

Moreover, we found decreases in distress calls concerning all the most frequent symptoms and diagnostic groups. A decrease in EMS use has been attributed to the public's perception of the workload within the hospitals, as Hammad and co-workers reported that patients had refrained from calling on the healthcare system for fear of "disturbing" the system [18]. It has also been suggested that fear of acquiring COVID-19 at the hospital could have reduced the incentive to call for emergency medical assistance [6].

Vuilleumier et al. reported that the severity of the ambulance missions increased. Life-threatening emergencies thus increased significantly during the pandemic, while the proportion of non-urgent primary missions decreased in 2020 [35]. This perception is to some extent supported by our study where we found an increased overall mortality despite the EMS being requested to care for fewer patients. Consistent with these findings, a study of all acute hospital contacts during the first COVID-19 phase in Denmark reported increased mortality rate ratios during the pandemic [36]. More patients were attended to by ambulances without being brought to the hospital in 2020, and this may be partly explained by prehospital measures enabling paramedics to treat and release patients prehospitally [12, 37] in an attempt to mitigate the spreading of the virus to the hospitals. The slightly increased mortality in patients transported to hospital and the increased ratio of non-conveyed patients could suggest that the prehospital triage criteria may have been altered towards reducing hospitalization during the COVID-19 pandemic.

The total number of patients hospitalized with respiratory diseases in Denmark had a huge and persistent decrease in 2020. A Danish study of all acute hospital contacts found significantly fewer COPD patients with acute exacerbations in 2020 compared with 2017–2019 [36]. This reduction in exacerbations of COPD may also be the case here. One may speculate that the reduced number of social contacts [38] may reduce the risk of infections leading to acute exacerbation of COPD, thus playing a role, and so may the patients´ fears of acquiring COVID-19 at the hospital [6].

In our study, the reduction in the use of the EMS during the first lockdown period was followed by an increase in the number of ambulance missions when the restrictions were loosened. However, in the fall of 2020, measures addressed towards society to mitigate the COVID-19 pandemic were reinstated. This renewed lockdown resulted in a substantially more extensive reduction in EMS missions. These characteristics have also been reported in studies from North America [39, 40].

The multiple surges of the COVID-19 pandemic may have influenced the patients´ contact pattern towards the emergency medical dispatch centre and the subsequent hospital patient load. Jarvis and co-workers reported that there might have been some evidence of saturation concerning the restrictions in the public [41]. However, the number of public inter-personnel contacts remained low immediately after the severe restrictions [38, 42, 43]. Other papers have reported that a saturation of the public's willingness to adhere to lock-down measures may have led to unexpected high deaths [44]. The Danish EMS experienced a significant increase in the number of missions during the temporary lifting of the restrictions over the summer of 2020, which was reduced again once renewed restrictions were imposed again in the fall of 2020. One may speculate, as Zaildo and co-workers did, that financial and social support and trust in political authorities, which are relatively high in Denmark, may have enhanced the adherence to prevention and control measures for COVID-19 [45]. This may also explain the only minor increase in mortality observed in the current study. It is possible that patients, who hoped to avoid hospitals and EMS, inadvertently postponed seeking help with adverse effects. Assessing the possible long-term effects of reduced EMS contact were however beyond the scope of this study.

The major strength of the study is that this is a nationwide study population based on Danish clinical registries, which are generally considered to be of a quality well suited for research [26, 46]. Being a nationwide study increases the external validation.

Another strength of the study is that access to Danish hospitals and the Danish EMS is free for the individual patient. No immediate costs of using the system are imposed as the system is tax-financed. Thus, there is probably no bias related to the threshold for contacting the EMS.

The unique Danish patient identifier, the CPR number, allows for a rather comprehensive follow-up of patients [25]. However, a limitation of the study is the number of patients lost to follow-up. In this study, 6.5% of the patients were not identified.

Research regarding non-conveyance in Denmark are lacking. As such the included data from 2019 are the only available comparison in number of non-conveyed patients. It is not possible to assess if this is representative of the usual number of non-conveyed patients in Denmark.

Patient background, such as co-morbidities and status, as well as dispatch urgency level, was not included in the current study. This limits the assessment of illness severity, as results could be skewed towards both higher and fewer number of patients. Likewise, similar background for a control group, i.e. general population, was not included as this was beyond the scope and resources of the current study. A further limitation is the inherent limitation of observational and, specifically, retrospective cohort studies where only associations and not causality can be established.

A final limitation of the study is the lack of external generalisability. The results are obtained from Denmark only, and therefore are thus not necessarily representative for other countries.