The current coronavirus disease 2019 (COVID-19) pandemic has been described as a novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a disease condition at the beginning characterized by a massive number of cases, leading to unplanned intensive care unit (ICU) admissions, pneumonia with the multiorgan disease, and related mortality (particularly before mass vaccination programs) (1).

At the peak of the pandemic, a drop in notifications to emergency call centers for life-threatening diseases was observed, with a significant decrease in incidents attended by the emergency medical services (EMS) and the emergency department (ED) (2). A marked decrease in cases of acute myocardial infarction, stroke, or traffic accidents has also been reported (3–5), prioritizing COVID-19 (2). EMS were called upon to respond to biohazard medical emergencies, monopolizing patients with COVID-19 and virtually all ambulance transfers. Pre-hospital care was initially provided under unfavorable circumstances, e.g., the use of personal protective equipment, excessive evacuation delays, and, above all, a general unawareness concerning the transmission of the virus (6, 7).

With rapid tests, vaccinations, and effective therapies, the current pandemic has been kept under control, and health systems have managed to deal with COVID-19. We hypothesize that COVID-19 has likely been one of the factors, but not the unique one, of the exacerbation of chronic pathologies and of the observed over-mortality compared to the historical time series. This excess of mortality may result from the lack of appropriate and timely attention to life-threatening diseases, excess mortality due to COVID-19, or a combination of both circumstances (8).

Over the course of the outbreak, health systems have changed from assisting patients with COVID-19 and focusing all efforts on controlling the virus to assisting patients with diseases associated with COVID-19. In other words, COVID-19 has changed from being the primary disease to being treated for a patient in need of urgent care to being part of the full set of pathologies that may negatively affect the prognosis of the patient as a whole (9).

The objective of the present study was to compare long-term mortality (1-year mortality by all-cause and in- and out-of-hospital) in cases managed by EMS and subsequently transferred with high priority to ED in the following two contrasting prospective cohorts: cases with the acute disease without past COVID-19 vs. cases with the acute disease after COVID-19.

A total of 3,107 patients with acute disease managed by pre-hospital care and referred to the ED were included in the final evaluation: 2,594 in cohort #01 (non-COVID-19) and 513 in cohort #02 (COVID-19). We excluded 308 confirmed active COVID-19 cases in ED (Figure 1).

The median age was 67 years (IQR (interquartile range): 50–81 years), with 41.8% women (1,299 cases). Demographic characterization by COVID-19 cohort included older adults evacuated by ALSU mainly from urban areas to ED and derived to a large extent from nursing homes, with a significant number of comorbidities (especially congestive heart failure, myocardial infarction, dementia, connective disease, and severe chronic kidney disease). The non-COVID-19 cohort exhibited a similar median age, with more middle-aged cases, significantly reduced comorbidities, and a lower nursing home origin. Clinically, both groups reported the same qSOFA, and a similar percentage of pre-hospital advanced life support special procedures, with comparable hospitalization and ICU admission rates (Tables 1, 2).

The overall 1-year mortality was 20.3% (629 cases). Comparing both cohorts, the mortality rate in the COVID-19 group was 13.9 points higher than the one in the non-COVID-19 group (31.8 vs. 17.9%). Cumulative mortality by time points, respectively, 1, 2, 7, 30, 90, 180, and 365 days, in the COVID-19 cohort increased consistently over all time points, exhibiting about double the cumulative mortality vs. the non-COVID-19 cohort for all the time points (Table 3). This result was corroborated by the KM curve (Figure 2); as can be observed, both groups remained parallel throughout the follow-up.

When considering the whole cohort (Table 4A) or only those with 1-year mortality (Table 4B), the logistic multivariate analysis of COVID-19 vs. non-COVID-19 showed that the main characteristics of patients with COVID-19 were being in a nursing home and suffering from dementia or congestive heart failure. Additionally, when considering the whole cohort (Table 4A), patients with COVID-19 suffered from connective disease, presented a higher number of breaths per minute, and had higher 1-year mortality.

Similar to the comparison between patients with COVID-19 and patients without COVID-19, the longitudinal analysis of mortality for the whole cohort (Table 5) showed that factors associated with mortality included (results from Cox regression) age, respiratory support both invasive and non-invasive, noradrenaline administration, hospital admission, and hospital stay duration. The diagnosis groups that stood out as risk factors were respiratory, digestive, infection, and trauma and injury. Pathologies associated with mortality were a metastatic solid tumor, leukemia, and congestive heart failure. Those patients with COVID-19 presented a higher risk of mortality, a variable that remains statistically significant despite the high number of confounding factors. Finally, the mSOFA score was higher in those patients with a higher risk of mortality, suggesting its reliability in predicting clinical worsening even at long-term follow-ups. Further details of the results from this analysis can be found in Supplementary Table S1.

To determine the factors critical for mortality for patients with COVID-19, the same procedure applied in the previous analysis was used for the cohort of patients with COVID-19 (Table 6); further details of these results can be found in Supplementary Table S2. Again, age, mSOFA, respiratory disease, metastatic solid tumor, leukemia, and congestive heart failure were risk factors for mortality. This more detailed analysis showed that hemiplegia, high aCCI, diastolic blood pressure, and FiO₂ were critical factors for mortality within the COVID-19 group.

The massive caseload caused by SARS-CoV-2 has consequently led to an increase in mortality rates, associated both with the pandemic and with the suboptimal support provided to non-COVID-19 disease at the start of the outbreak.

Patients treated by pre-hospital care without COVID-19 (cases with an acute disease that did not present the previous COVID-19) showed a 1-year mortality rate close to 18%. According to our results, 1-year mortality for those from the COVID-19 group (cases formerly infected by COVID-19) was 13.9 points higher. A longitudinal analysis showed that presenting COVID-19 as an antecedent is a risk factor for long-term mortality.

Chronic preexisting health conditions are well-documented to play a key role in long-term survival; the greater the number of pathologies, the lower the likelihood of survival and the higher the likelihood of in-patient hospitalization, rehospitalization, and ICU admission rates (15, 16). The number of pathologies was observed as a key factor for short-, medium-, and long-term related mortality since the beginning of the pandemic (17, 18). Different studies examined long-term mortality in post-COVID-19 patients (19–21), but to the best of our knowledge, no research has analyzed the impact of COVID-19 as a previous condition among acute disease patients managed in pre-hospital care.

This over-mortality, according to our study, appears to have a multi-causal explanation. The cases included were multi-pathological patients, such as cardiovascular and neurologic diseases or trauma and injury. Pre-hospital care was homogeneous among both cohorts in terms of assessment using the mSOFA (13) (pulse saturation/inspired oxygen fraction ratio, mean arterial pressure, Glasgow Coma Scale, creatinine, and lactate), although some advanced life support techniques were preferred in COVID-19 cohort, e.g., non-invasive mechanical ventilation and noradrenaline use (22). The rates of hospital-inpatient, hospitalization-day, and ICU admission were statistically equivalent.

The above results reinforce the argument that over-mortality could be caused by a combination of variables. Chronological age is an unquestionable biological factor. In addition, chronological age plays a pivotal role in chronic diseases, so the older the age, the increased the comorbidities (23, 24). Despite age showing significant differences between groups, we believe that comorbidity burden was the most decisive factor since age was not statistically significant in the multivariate logistic regression. The COVID-19 cohort exhibited a median aCCI of 3 points vs. 1 point in the non-COVID-19 cohort. A detailed analysis highlighted an increase observed in cardiovascular pathology (congestive heart failure and myocardial infarction) and dementia in the COVID-19 cohort, with data in line with similar studies (25–27), since those conditions are associated with common exacerbations, hospital-inpatient, and ultimately poor long-term outcomes. Other pulmonary diseases, such as chronic obstructive pulmonary disease, were not related to a significant increase in 1-year mortality with similar results in both cohorts (28).

Nursing home affiliation was a critical factor directly involved in the mortality of the COVID-19 cohort. Remarkably, at the start of the outbreak, unacceptable mortality rates associated with nursing homes were observed. Admittedly, patients are multi-pathological, with multiple comorbidities, and generally of elderly age, but the over-mortality described in nursing homes should give us a wake-up call to reconsider this fact as a healthcare system (29, 30). Nursing home mortality was two times as high as in the COVID-19 cohort compared to patients managed by EMS due to acute disease without COVID-19; this irrefutable observation flags nursing home affiliation as a critical factor underlying poor long-term outcomes (31).

The above-mentioned results suggest that COVID-19 plays an important role in this long-term mortality, and three main reasons could be argued for the importance of COVID-19 in long-term mortality: First, in the selection of patients, all the patients were selected based on an opportunity sample method, i.e., selecting all the patients who accomplished criteria during the study time. The difference between the COVID-19 and non-COVID-19 groups regarding age or comorbidities was due to chance rather than a consequence of having suffered from COVID-19. When using the above-mentioned confusion factors in the Cox regression (Table 5), none of them (and the other confusion factors) exclude COVID-19 as a risk factor for mortality. In addition, when all statistically significant factors (including age and aCCI) were adjusted in a regression model to determine the final model that described the difference between the COVID-19 and non-COVID-19 groups (Table 4), age and aCCI were automatically (by the regression algorithm) excluded from the final model, and only a few comorbidities alone were included. Epidemiological studies have shown an excess of mortality in patients with COVID-19 compared to analogous historical series (32, 33). Even though mortality also increased in patients without COVID-19 in the early stages of the pandemic, this trend has gradually normalized to previous levels as healthcare returned to pre-pandemic attention levels and due to the improvement of COVID-19 handling (34). Therefore, as the pandemic evolved, one should expect a reduction in mortality, which was not the case according to our results.

Our study is not free of limitations. First, a pure convenience sample was collected consecutively. To control for potential bias, data input was gathered 24/7 non-stop throughout the study period in ambulance stations located in urban, suburban, and rural areas, patients transferred to ED of different hospitals, and hospitals with different clinical qualifications, attempting to be a true cross-section of the analyzed population. Second, the data extractors were not blinded. To avoid cross-contamination, the EMS staff was unaware of the scores being estimated and interpreted, and as a double fail-safe, the research associates from each hospital were unaware of the pre-hospital parameters as well. Only the data manager and the principal investigator could access the master database. Third, confirmed cases of COVID-19 were taken as patients with a positive polymerase chain reaction and/or rapid antigen test, but an underestimation is possible. Currently, some people skip screening or do not report self-test results. At the onset of the outbreak, the availability of test kits was limited, even though the incidence rates should be treated with caution. In this sense, antibody tests for the non-COVID-19 group were not available, so it cannot be completely ruled out that they did not have COVID-19. Fourth, the study was carried out across different provinces, all of which comprise the same health system. To validate the findings, multicenter studies in different regions involving several institutions should be carried out. Fifth, in the present study, we did not consider all the patients who could present long-term mortality; this is because patients could reach the emergency department by their means without requiring assistance from EMS. However, this study aimed to focus on patients who required pre-hospital emergency care. Sixth, since this study has been developed in the pre-hospital scenario, critical factors related to the long-term consequences of COVID-19 have not been considered due to the impossibility to achieve them, for instance: the date of infection (hampering determining the time between infection and the EMS attendance), the severity of COVID-19, the treatment the COVID-19 (whether it required intensive therapy or invasive mechanical ventilation), and the treatment after COVID-19 hospitalization. Seventh, the duration or diagnostic time of comorbidities was not available; however, despite being important information regarding the status of the patients, it is not included in the commonly used comorbidity-based scores.

According to our results, the COVID-19 group presented a higher mortality rate than the non-COVID-19 group. The predictive model, when adjusted by confusion factors, showed COVID-19 as a relevant risk factor for mortality.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Institutional Research Review Board of each health area (reference: PI-049-19/PI-GR-19-1258). The patients/participants provided their written informed consent to participate in this study.

FM-R conceptualized the project, managed and coordinated the project, assisted with the design of methodology, analyzed data, and prepared the initial and final drafts of the manuscript. AS-G takes responsibility for the data and their analysis. LM-M, JB-J, RC-S, BP-L, CP, MC, and JM-C assisted with the management and coordination of the project, assisted with the design of methodology, and helped review the manuscript. RL-I conceptualized the project and helped review and comment on the initial and final drafts of the manuscript. All authors performed a critical review and approved the final manuscript for interpretation of the data and important intellectual input.