The demand for emergency care worldwide has increased in recent years according to the data of patient volumes at emergency departments (EDs) 1–61–61–61–61–61–6. Previous studies have analysed this phenomenon and investigated what events bring patients to the ED. It has been suggested that the increased attendance of patients at the ED has been due to a lack of primary care physicians 7–97–97–9. These suggestions have, however, not been supported by a number of studies 5,10–125,10–125,10–125,10–12. Instead, several studies indicate that patients come to the ED in the belief that they have a need for knowledge, investigation or treatment that they think cannot be provided elsewhere 5,10,11,135,10,11,135,10,11,135,10,11,13. The availability of the ED with access to care round the clock has also been suggested as a possible explanation to the increasing ED demand 10,1110,11.
In Sweden, the reasons for the increasing ED demand are fairly unknown, and despite several efforts by the Swedish national healthcare system to influence patient flows in the hope to relieve pressure on the emergency care system 14, the flow of ED-seeking patients continues to increase each year 6.
Further research is needed to find a solution to the problem with overcrowded EDs. An as yet uninvestigated topic is that of patients who have been in contact with other caregivers before their ED visit. It is unclear to what extent other healthcare services, such as primary healthcare settings or telephone advice lines, refer patients to the ED. To clarify this issue, a thorough investigation is needed that focuses on patients' pathways to the ED. Because the majority of studies on this topic are from other countries, it is uncertain whether such results are valid for Swedish conditions.
The overall aim of this study was to investigate whether patients come directly to the ED or whether they have taken any other actions or activities within the healthcare system before attending the ED. An additional aim was to increase our understanding of the potential determinants between patients’ ED-seeking behaviour and patient-related data.
This was a prospective descriptive study with adult ambulatory and ambulance arriving patients to the ED. Because the study was considered a quality improvement project, the regional ethics committee in Stockholm waived the need for approval. Patient information about the project emphasized the voluntary nature of participation and that the patients’ care at the ED was not influenced by their decision to participate or not in the study. The patients gave written informed consent.
Study setting and population
The study was carried out at an ED at a level one trauma centre at a university hospital with 70 000 patient visits (2008). Patients were excluded if they were under 18 years of age, did not understand Swedish, were too ill (allocated a high acuity level by the triage nurse), left the ED without being seen or were incapable of filling out the questionnaire. Patients were included conveniently during the data collection period. Both Internet-based and telephone-based healthcare guides were available to the public in the region during the time of data collection.
A questionnaire was designed for the study by the research group, the content of which was influenced by previous research 15–1715–1715–17. The questionnaire, which contained 20 questions, was piloted three times at the ED before the final version was determined. Six of the 20 questions were designed with ensuing questions, that is, if the patient answered yes to the question, two ensuing questions were to be answered. Two questions were open-ended, eight had dichotomous answers and the remaining 10 had multiple answers. The following variables were gathered from the electronic patient records: mode of arrival (with or without ambulance), chief complaint, triage level, admitted to hospital ward, pharmaceutical therapy in the ED, diagnostic investigations in the ED and referral to the ED.
During 12 weeks in 2008, research assistants (nurses from the ED) collected data during day and evening shifts (from 9 a.m. to midnight). These nurses did not take part in clinical work in the ED during this period. The nurses included patients by reading the patients’ triage notes in order to assess whether the patients fulfilled the inclusion criteria, and those who did were then approached and evaluated for suitability regarding physical and emotional status. Next, patients considered suitable for participation were informed about the project and given the questionnaire. The nurses then collected the completed questionnaire and written informed consent. When needed, the nurses helped the patients fill out the questionnaire.
The primary outcome was patients’ pathway to the ED. Further, the potential determinants between patients’ ED-seeking behaviour and patient-related data were a secondary outcome.
Descriptive and comparative analyses were performed using the Mann–Whitney U-test, the χ2-test and logistic regression. P-values less than 0.05 were considered statistically significant. Because the data were not normally distributed, age was dichotomised (on the basis of median age) in the logistic regression analysis. Sex and symptom duration were included in the multivariate analysis as they were statistically significant in the univariate analysis. Data management and statistical analyses were conducted with SPSS software (version 15, SPSS, Inc., Chicago, Illinois, USA). Statistics Sweden (SCB) carried out comparative analyses for sociodemographic variables (ethnicity, income and educational level). Chief complaints were categorized according to the origin of symptom and affiliation to the International Statistical Classification of Diseases and related health problems (ICD-10) 18. Post-high school education is defined as education at college and university.
A retrospective analysis of nonparticipating patients (n=8176) was performed. The medical records of patients for whom no questionnaire was completed were reviewed as to age, sex, arrival mode, triage level and admission to hospital ward.
Of the 10 190 patients eligible for participation, 2014 (20%) were included. The 8176 (80%) nonparticipating patients differed statistically significantly from the participants in several variables (Table 1).
Patients seeking direct ED care (n=822) were more often men, who arrived at the ED using the prehospital emergency care services and with shorter symptom duration than patients with previous actions or activities before the ED visit (n=1192) (Table 1). In all, 1192 (59%) patients reported taking healthcare-related actions or activities before going to the ED. Although the direct ED group underwent diagnostic procedures and received pharmaceutical treatment to a higher extent than the previous action group, patients in the latter group were more often admitted to the hospital ward. Further, the groups differed in ethnicity and educational level. Table 2 illustrates patient data available for each variable.
Potential determinants between patients’ emergency department-seeking behaviour and patient-related data
The association between patients’ ED-seeking behaviour and potential determinants is outlined in Table 3. The variables that were significantly related to ED-seeking behaviour were sex and symptom duration. Multivariate logistic regression was performed to assess the contribution of the significant independent variables. In this model, both sex and symptom duration were found to be significantly related, meaning that female patients with long symptom duration (>24 h) more often carried out previous actions or activities before seeking ED care.
Figure 1 illustrates the patients’ chief complaints that drove them to seek ED care. Most common symptoms were musculoskeletal (ICD-10 code M), digestive (K), circulatory (I) and respiratory (J). A large proportion (18%) of patients had symptoms that were difficult to define based only on origin (R). Patients within this group presented with a variety of symptoms, including poor general health, fever, dizziness, fainting, convulsions, fatigue, achiness and headaches. There was a statistically significant difference with regard to patient symptoms between patients who had sought direct ED care and those who had been in contact with other caregivers before visiting the ED (P<0.001).
The actions or activities of the previous action group (n=1192) are presented in Table 4. The most common action/activity was to contact a general practitioner, carried out by 623 (29.1%) patients, whereas a private caregiver was most seldom (7.8%) contacted. On average, patients contacted 1.8 health caregivers before seeking ED care.
As can be seen in Table 5, patients to a high extent rated accessibility to the health caregivers previously contacted as easy or very easy. About half (48–58%) of the patients stated that their satisfaction with the care was insufficient/inadequate. The caregivers’ referral rate to the ED was 60.1–87.9%.
The main result in this study is that more than half (59%) of the patients reported having been in contact with other health caregivers before seeking ED care. Further, female patients and patients with longer (>24 h) symptom duration to a higher extent sought other health caregivers before the ED visit. About half of the patients with previous actions or activities rated their satisfaction with the care as insufficient/inadequate.
A common notion expressed by the ED staff is that patients seek care at the ED because of other limited accessibility of other healthcare providers. However, our study does not support this notion in that more than half of the patients had contacted another caregiver before seeking ED care. A recent study from the US was also unable to support this opinion 5. Instead, the authors found that the patients sought ED care because they desired something from the ED that they believed could not be delivered by other health professionals. Because many of the patients in our study who had been in contact with other caregivers before the ED visit rated other healthcare providers’ accessibility as easy or very easy, we believe that limited accessibility is not the main reason for seeking ED care without consulting other caregivers. One reason could be that patients seek something from ED care that other health professionals are unable to provide and hence the patients follow up the non-ED visit with an ED visit. Another possible reason is that these patients are more concerned about their symptoms than patients seeking primary care 19.
Several studies investigating nonurgent patients’ reasons for ED visits underscore other reasons for seeking ED care 10,11,1910,11,1910,11,19. These studies found that it was common for nonurgent patients to seek ED care because they believed they were suffering from a serious condition or that they were worried about their symptoms or they were referred to the ED. In our study, it was common (60.1–87.9%) for patients to be referred to the ED by other caregivers. It is noteworthy that the 250 (11.7%) patients who had been in contact with other hospital healthcare providers could not be cared for by those healthcare providers. We also find it surprising that, on average, the patients with previous actions or activities had contacted 1.8 healthcare providers. We believe that this finding supports the notion that patients actively seek other care alternatives before seeking ED care.
The result that patients with shorter symptom duration sought direct ED care is supported by another Swedish study 19. This result, that patients with shorter symptom duration seek direct ED care, is in line with the nature of ED care, that is, acute illness and injury. Our results support the conclusion that further accessibility to primary care or Internet-guided and telephone-guided advice and recommendations probably will not change the patient’s behaviour in seeking emergency care and thus the overcrowding problems in the EDs need new approaches and solutions different from those that have been performed so far by the health administration.
The main reason why 80% of all eligible patients were not included in this study has to do with the data collection method used. More specifically, the research assistants who conducted the data collection did not have time to include these eligible patients. However, selection bias cannot be ruled out. Because the nonparticipants differed in several variables (e.g. age, sex and socioeconomic status) from the participants, this unforeseen nonparticipant rate is unfortunate because it limits the possibility to generalize the results to other patient samples. Another limitation is that data were collected in only one ED. However, the patients have access to the same amount of primary healthcare facilities (such as general practitioners and nonhospital EDs) as in other parts of the city.
Forty-five patients were included despite the fact that they fulfilled the exclusion criterion of having a high acuity rating. The reason for this is that these patients were overtriaged (i.e. assessed as more ill than they actually were) upon ED arrival. Following re-evaluation of the triage level, the patients were allocated less urgent triage levels and subsequently included in the study. The patients missing in the analysis carried out by SCB could be due to the patients not having a permanent Swedish identification number (they only had a temporary one), which makes them unavailable to SCB, or that the patients were included more than once.
Conflicts of interest
There are no conflicts of interest.
3. Backman A Emergeny care seeking behavior in relation to patients' and providers' perceptions and attitudes. 2010 Stockholm, Sweden Karolinska Institutet, PhD
4. Lowthian J, Curtis A. Systematic review of trends in emergency department
attendences: an Australian perspective. Emerg Med J. 2010;5:373–377
5. Tranquada K, Denninghoff K, King M, Davis S, Rosen P. Emergency department
workload increase: dependence on primary care
? J Emerg Med. 2010;3:279–285
7. Murphy A. ‘Inappropriate’ attenders at accident and emergency departments I: definition, incidence and reasons attendance. Fam pract. 1998;15:23–32
8. Canadian Association of Emergency Physicians, Affiliation NEN. . Joint Position Statement on emergency department
overcrowding. Can J Emerg Med. 2001;3:82–88
9. Northington W, Brice J, Zou B. Use of an emergency department
by nonurgent patients. Am J Emerg Med. 2005;23:131–137
10. Field S, Lantz A. Emergency department
use by CTAS levels IV and V patients. Can J Emerg Med. 2006;5:317–322
11. Steele S, Anstett D, Milne W. Rural emergency department
use by CTAS IV and V patients. Can J Emerg Med. 2008;3:209–214
12. Hayden S, Jouriles N, Rosen P. Requiem for "non-urgent" patients in the emergency department
. J Emerg Med. 2010;3:381–383
13. Coleman P, Irons R, Nicholl J. Will alternative immediate care services reduce demands for non-urgent treatment at accident and emergency? Emerg Med. 2001;18:482–487
15. Diesburg-Stanwood A, Scott J, Oman K, Whitehill C. Nonemergent ED patiens referred to community resources after medical screening examination: characteristics, medical condition after 72 hours, and use of follow-up services. J Emerg Nurs. 2004;4:312–317
16. Howard M, Davis B, Anderson C, Cherry D, Koller P, Shelton D. Patients' perspective on choosing the emergency department
for nonurgent medical care: a qualitative study exploring one reason for overcrowding. J Emerg Nurs. 2005;5:429–435
17. Olsson M, Hansagi H. Repeated use of the emergency department
: qualitative study of the patient's perspective. Emerg Med J. 2001;18:430–434
19. Backman A, Blomqvist P, Lagerlund M, Carlsson-Holm E, Adami J. Characteristics of non-urgent patients. J Prim Health Care. 2008;3:181–187