This historic population-based cohort study of patients calling for an emergency ambulance in the North Denmark Region showed that the top five of Danish Index criteria contributing most to the total number of deaths among 112-patients was “unconscious adult/possible cardiac arrest”, “breathing difficulties”, “unclear problem”, “possible stroke”, and “chest pain”. Equivalent results were demonstrated for age and sex adjusted IR estimates of short-term mortality, i.e. death on the same or the following day after the 112-call, on annual population level per 100,000 people. There were large differences between each of the five criteria, with IR mortality for “unconscious adult/possible cardiac arrest” more than fifty-fold as high as for chest pain. These five conditions corresponded almost to the five most frequent criteria for calling 112. The only exception was “unconscious adult/possible cardiac arrest” which was much more seldom, whereas “accidents” was the third most frequent among the top five most frequent criteria.
A strength of this study was the availability of data and the opportunity to link registers, which enabled the follow-up and minimised the risk of information bias. Furthermore, the North Denmark Region covers mixed rural and urban areas and the free access to healthcare in Denmark minimises the risk of selection bias. A weakness of the study was the missing civil registration number for 8.7% of the emergency calls and the missing Danish Index in 4.9% of the calls. The number of missing Danish Index criteria decreased in the study period from 6.7% in 2016 to 3.0% in 2018. The Danish Index was implemented in 2011 and staff training might explain the increase in registrations. However, it cannot be ruled out that these excluded calls may represent a certain group of patients and thereby bias the results. They could represent patients with similar symptoms or potentially patients who are more, or less severely ill or injured, and thereby affect the results in either direction. However, it is a low percentage of the overall large number of included patients and is likely to have a minor impact on the results. Furthermore, the proportion of missing data was similar or even lower than in previous studies, where missing data due to unknow civil registration number reached 17.8% (1) and missing data due to unknow civil registration and missing Danish Index reached 45% (20).
Patients with repeated emergency calls within 30 days only had the last contact included for the mortality analysis, which may have resulted in underestimation as well as overestimation of mortality for some of the Danish Index criteria, as the initial call might be due to another symptom or mechanism of injury than the last call during the 30-day period. However, as only 8% of the patients had repeated runs, we assess this to have little influence. The repeated users are however a special group, that would benefit from further investigation
This study included patient symptoms or mechanism of injury in relation to the Danish Index and thereby each patient is only registered with one criteria as the main reason for calling, although patients may present with more than one symptom (21). This problem is well known, and similar to other studies on presenting main symptom (22, 23).
Our study confirms that the First Hour Quintet, i.e. cardiac arrest, respiratory failure, trauma, acute coronary syndrome, and stroke are serious conditions associated with high mortality, and we found that four of these were among the top five contributing to total mortality among 112-patients. “Accidents” was the third most frequent reason for calling 112, and number six contributing to deaths among 112-callers, with 1- and 30-day mortality of 0.6 and 2.4%, very similar to number five on our list, “chest pain” with 0.7 and 2.1%. Our study revealed that “unclear problem” is a high-risk symptom, as it had the third highest cumulative number of deaths within 30 days and was the second most frequent Danish Index criterion, and altogether making this the third most deadly symptom in 112-calls as reflected in the mortality IR, both crude and adjusted.
“Unclear problem” is usually assigned when the healthcare professional does not know the exact medical cause, a phenomenon which is already well-described for emergency departments. Likewise, the high mortality for “unclear symptom” corresponds to what have been previously described for non-specific symptoms presented in emergency departments (22, 24). A Swedish study of EMS patients arriving at an emergency department, found that the risk of having an ambulance dispatched with low priority by the EMCC was almost doubled among patients with non-specific complaints compared to randomly selected patients matched for age and sex (25). A Danish register-based study from the capital region in 2011–2013 found that 18% of emergency calls were categorized as unclear problem, and in calls assessed with emergency level B (without light and sirens), they found that unclear problem had a higher mortality than specific symptoms or problems (20). Another study, covering 75% of the Danish population, elucidated potential preventable deaths due to the medical dispatch decisions by auditing medical records for the 152 EMS patients dying the same day as receiving an ambulance with urgency level B. They found only few preventable deaths, but among these, the most frequent criteria were “unclear problem” and “breathing difficulties” (7). Thus, unclear problem in EMS seems to be a similar challenge as non-specific symptoms in the emergency departments.
Breathing difficulties has previously been demonstrated to be among the most common reasons for contacting EMS (26). Likewise, the high mortality for breathing difficulties has previously been established, both in EMS and the emergency departments (21–23, 27). A previous study assessed how well nurses, physicians, and patients agreed on an 11-point rating scales of breathlessness, and found an underestimation of breathlessness and respiratory function by nurses and physicians in the intensitive care units (28). This may play a role in the challenging task for healthcare professionals at the EMCC to estimate the severity of this particular symptom. A recent Swedish study confirmed high mortality rates for patients with dyspnoea, and interestlingly found that 84% of the patients had previously suffered from dyspnoea and more than half showed more than two days delays from symptoms onset to EMS contact (29). This shows that many patients with dyspnoa/breathing difficulties call for help as the condition has become unmanageable, which emphazises the vulnerability of these patients.
In 2012, the most frequently used Danish Index criteria were (1) unclarified problem, (2) chest pain, (3) minor wounds and injuries (4) accidents and (5) breathing difficulties (10), similar to our study, except for “possible stroke” now on top five. Early recognition of stroke leads to faster response and improves time to hospital arrival (30, 31) which in turn improves the diagnosis and treatment. The increased awareness for symptoms of stroke, may explain the increase in healthcare professionals’ assessment of the symptom “possible stroke”. In Denmark there has been several campaigns to raise awareness of symptoms and risk factors of stroke.
Our study showed a four to six times lower risk of short-term mortality for the symptom “chest pain” when compared to the frequently recorded symptoms, i.e. “breathing difficulties” and “unclear problem”. Likewise, an Irish study found that patients having chest pain as presenting complaint when admitted to a hospital, was associated with a decreased risk of 30-day mortality with a odds ratio of 0.47. Whereas, the study demonstrated breathing difficulties to be associated with an increased risk of death, with an odds ratio of 1.8 (32).
Recognition of time-critical conditions is import for patient outcome and the presenting symptoms carry valuable information of the risk of short-term mortality. The magnitude of “unclear problem” indicates the need for more research into this group of patients: on how they are handled on scene and in hospital; to which extent these patients are acutely ill requiring immediate help; whether patients with “unclear symptom” encompass terminal or old dying patients and/or socioeconomic vulnerable patients. This may be an overlooked and neglected patient group with potential for future improvements in the entire patient care pathway.