In this population-based study of nearly 80,000 acute ambulance patients, assigned non-specific R or Z diagnoses at the hospital, the vast majority, two-third had non-specific sub-diagnosis such as ‘obs suspected diseases’ (Z03). Among these, nearly half of all patients were labeled ‘unspecified disease’ (Z039). Mortality was highest for patients diagnosed as ‘unspecified disease’ (Z039), where 8 % died within 30 days, exceeding even the mortality among ‘suspected cardiovascular disease’ (Z035). Though some patients with ‘causes of mortality, UNS’ (R99) could have died from cardiovascular disease, this is a high figure. Moreover, the number of deaths due to the nonspecific sub-diagnoses constituted a very large part of the total number of deaths.
The population-based design and large study population are strengths of this study. Each patient might be included more than once since we examined the total number of patient contacts. The Danish population has free and equal access to prehospital care, minimizing the risk of financial resources influencing the inclusion of patients. The linkage of prehospital- and hospital data and the complete follow up is another strength, minimizing the risk of selection and information bias.
A weakness is that only patients with known CPR number were included to allow linking prehospital and hospital data sources. We were not able to determine the number of patients excluded due to this, and missing CPR numbers are a well-known weakness in prehospital emergency service studies. This may introduce bias in either direction, since patients without known CPR number may be more or less ill [7].
Some patients may have moved out of the region during the study, which may result in incomplete mortality reporting. However, with the large size of the cohort and the relatively short period, we find this to be of minor influence.
Patients hospitalized more than once might have received different non-specific diagnoses each time. Mortality was analyzed from the most recent hospital contact, since the patients are ‘immortal’ in the time between two hospital contacts, hereby excluding the first non-specific diagnosis.
The sub-diagnoses describing death constituted 11 % of the 30-day mortality. Overall, the diagnoses constituted less than 1 % of the total number of diagnoses. Since the sensitivity analysis showed no considerable differences in mortality, these diagnoses are assumed to have no influence on the results.
Our study only included patients transported to the hospital by ambulance, and are not directly generalizable to all prehospital patients.
Several studies found that non-specific diagnoses represent a considerable part among acute patients: among patients taken to hospital by ambulance after calling 112 [3], among acute medical admissions [9, 10] and among all patients with contact to emergency departments [11]. Recently also, a German study showed that R-diagnoses (ICD-10 Chapter 18) was the third frequent diagnosis constituting 11-12% among patients admitted to hospital with ambulance [12]. Another Danish study found that ‘unspecified disease’ (Z039) and ‘observation for disease’ (Z038) comprised 76 % of all Z03-codes [13]. These findings agree with our study finding that two-thirds of all non-specific diagnoses consisted of Z03-codes, with the most frequent diagnoses unspecified diseases (Z039 and Z038).
Compared to the result from our previous study, showing that deaths among 112 patients with non-specific diagnoses constituted 20 % of the overall number of deaths after 30 days [3], this study confirmed the high mortality, also on the non-specific sub-diagnostic level both in proportions and in the large number of persons. However, we did not find any good explanations for these deaths apart from ‘suspected cardiovascular disease’ (Z035) and ‘suspected nervous disease’ (Z033).
Thus, the large number of non-specific diagnoses among patients brought to the hospital after calling the national emergency number does not represent only minor illness, and we cannot confirm that they represent unnecessary ambulance transports, even though such cases might be present in this patient group.
Worldwide the need for emergency care is rising and many studies including data from Bavaria, U.K., the U.S., Australia, New Zeeland, and Canada [14, 15, 16] confirm increased demand for ambulances.