An integrative review of adult patient-reported reasons for non-urgent use of the emergency department

Abstract

Objective

To conduct an integrative review of the scientific literature to explore adult patient-reported reasons for using the emergency department (ED) non-urgently.

Method

A literature search of CINAHL, Cochrane, Embase, PsycINFO, and MEDLINE was conducted with filters for humans, published January 1, 1990-September 1, 2021, and English language.

Methodological quality was assessed using Critical Appraisal Skills Programme Qualitative Checklist for qualitative and National Institutes Health (NIH) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies for quantitative studies. Data was abstracted on study and sample characteristics, and themes/reasons for ED use. Cited reasons were coded using thematic analysis.

Results

Ninety-three studies met inclusion criteria. Seven themes were found: need to be risk averse with respect to the health issue; knowledge and awareness of alternative sources of care; dissatisfaction with primary care provider; satisfaction with ED; ED accessibility and convenience resulting in low access burden; referred to the ED by others; and relationships between patients and health care providers.

Discussion

This integrative review examined patient-reported reasons for attending the ED on a non-urgent basis. The results suggest that ED patients are heterogenous and many factors influence their decision-making. Considering the complexity with which patients live, treating them as a single entity may be problematic. Limiting excessive non-urgent visits likely requires a multi-pronged approach.

Conclusion

For many ED patients, they have a very clear problem which needed to be addressed. Future studies should explore psychosocial factors driving decision-making (e.g., health literacy, health-related personal beliefs, stress and coping ability).

Introduction

Internationally, there is increasing patient demand for health care services at accident and emergency departments (ED) [1]. According to Morgans et al. [2] a health emergency is defined as "a sudden or unexpected threat to physical health or wellbeing which requires an urgent assessment and alleviation of symptoms” (p. 288). There is little agreement between clinicians and patients as to what constitutes an emergency situation requiring urgent or emergency health care services. In clinical practice, health care providers tend to conceptualize emergencies as those which are structured around physiological metrics that suggest a critical threat to life or a limb (i.e., death or serious injury) [2]. Conversely, researchers have shown that patients commonly minimize, or fail to recognize, medically significant symptoms, and focus instead on the nature of their symptoms [2]. Symptoms which present with severe, sudden, or rapid onset tend to be interpreted as an emergency whereas those which are slow and intermittent are considered less urgent [2].

Despite the original intent of the ED, visits for low acuity reasons are common and have been described as non-urgent, inappropriate, preventable, avoidable, and/or misuse in the scientific literature. In an extensive review, the prevalence of non-urgent ED use has been reported to range from 10% to 90% with approximately half of included studies having a non-urgent ED use prevalence rate of 24% to 40% [3]. This may be problematic as research has shown that non-urgent users complicate the provision of medical services, impair treatment for patients with emergent health needs, and make it difficult to properly assess medical acuity [4]. With the goal of fully understanding and addressing this problem, an evaluation of patient-reported reasons for non-urgent ED is required.

Background

Numerous studies have specifically examined factors influencing use of the ED on a non-urgent basis. Since 2009, five reviews with different methods and foci have captured varying aspects of this literature [3, 5,6,7,8]. Carret et al. [3] and Uscher-Pines et al. [8] both conducted a systematic literature review of quantitative research studies (retrospective and prospective) and examined variables (i.e., sociodemographic and clinical factors) associated with non-urgent use. The reviews by Kraaijvanger et al., [7] Coster et al., [6] and O’Cathain et al. [5] included both qualitative and quantitative studies of both adult and pediatric populations; they all specifically explored patient-reported reasons for ED use. Kraaijvanger et al. [7] performed a systematic review and meta-analysis and Coster et al. [6] performed a rapid (non-systematic) review; O’Cathain et al. [5] performed a realist (non-systematic) review, expanding on these two reviews.

While the academic literature contains various syntheses related to the subject of non-urgent ED use, limitations exist in terms of the 1) heterogeneity of the included population, 2) type of methods employed (systematic versus non-systematic literature searching), 3) assessment of quality and completeness of data culling and abstraction, and 4) subjective reporting of reasons for non-urgent ED use (i.e., patient reported versus inferred). For example, inclusion criteria varied significantly between reviews; the pediatric population was included in three of the five reviews, [5,6,7] patients arriving at the ED via ambulance were included in one review, [5] and specific disease categories were included for two reviews [3, 7]. Combining these different populations is problematic as there are clear differences in the decision-making process for medical care of children by parents and caregivers, as well as for taking an ambulance ride (versus walking in the front door). Non-systematic methods were used in two reviews [5, 6] and another review [8] included only American articles. It was not always clear in the reviews how patients were triaged “non-urgent.” Assessments of methodological quality were only reported by Coster et al. [6]. In some reviews, tabled data and reference lists were incomplete. Given the heterogeneity of methods, design, quality appraisal of resources, and synthesis approach, it is difficult to draw meaningful conclusions on patient-reported reasons for ED use. A comprehensive review with defined criteria may better inform practice and policy moving forward.

The review

Objective

The objective of this review was to conduct an integrative review of the scientific literature to explore adult patient-reported reasons for using the ED non-urgently.

Design

A study protocol was not previously registered for this review.

An integrative review of the evidence was performed using the methodology described by Whittemore and Knafl [9]. An integrative review is a review method which summarizes empirical or theoretical literature in an effort to comprehensively understand phenomena or a health care problem [10]. They are often used in nursing science where a review of the state of science may directly inform research, practice and policy [9]. The integrative review method allows for the inclusion of a number of different methodologies (i.e., experimental, non-experimental, qualitative, and quantitative) and therefore may be applicable to problems of importance in nursing [9]. Given the nature of the topic, as well as the heterogeneity of design types, this method was deemed most appropriate. The five-stage review approach by Whittemore and Knafl [9] was undertaken: (1) problem identification (i.e., introduction, background, aim), (2) literature search, (3) data evaluation, (4) data analysis, and (5) presentation.

Search strategy

A literature search of multiple databases (i.e., CINAHL, Cochrane, Embase, PsycINFO, MEDLINE, and Scopus) was conducted by applying a systematic approach consistent with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). The search strategy was developed in an iterative form in consultation with a health sciences librarian. It involved three overarching constructs related to the ED, non-urgent care, and decision-making, with MeSH terms, key words, and subject headings used for each database, as appropriate. Filters were applied for the following restrictions: studies involving humans, published between January 1, 1990 and September 1, 2021, and in the English language. Supplementary search techniques consisted of scanning the reference lists of retrieved articles and reviews on the topic for missed citations.

Articles retrieved from each database search were downloaded to EndNote (Version 9.0). After removing duplicates, each article title was assessed for relevance by two screeners (AMc, SJ). Relevant abstracts and subsequent full-text articles were then screened according to the following four [4] a priori inclusion criteria:

• 1. Patients were adults (mean age = 18 years of age or older);

• 2. Patients were recruited prospectively from an ED, also commonly referred to as the emergency room, accident and emergency care, or accident and emergency department;

• 3. Patients were specifically asked for their reason for seeking emergency care services;

• 4. Using any method, patients were triaged on the basis of the severity and urgency of their presenting condition or reason for visit.

The following types of studies were excluded: general, non-systematic reviews, expository/textbook chapters, conference proceedings, program reviews/descriptions (without a study sample), continuous learning/education modules, and clinical practice guidelines. If a study included a sample with only one specific medical condition or disease (e.g., asthma or congestive heart failure or epilepsy), or the study examined reasons for taking an ambulance to the ED, it was excluded. Studies assessing frequent, repeat or high-use ED users only were excluded. There was no minimum sample size required for inclusion.

Quality appraisal

Two independent reviewers (AMc, SJ) assessed each study for methodological quality using two commonly used quality assessment tools. Qualitative studies were assessed using the Critical Appraisal Skills Programme Qualitative Checklist (CASP) [11]. The CASP is a 10-item questionnaire which allows one to evaluate qualitative studies among three broad areas: 1) Are the results of the study valid (Section A); 2) What are the results (Section B); and 3) Will the results help locally? (Section C). Items were rated either ‘yes’ or ‘not reported’ (i.e., not reported or could not tell). The CASP authors do not suggest scoring the items. Quantitative studies were assessed using the National Institutes Health (NIH) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies [12]. This tool includes 14 items evaluating a wide range of quality measures; however, some items are more relevant to cohort studies (items 6–10, 12, 13). We defined quantitative, cross-sectional studies as those using methods such as in-person or postal surveys and/or structured interviews where statistical analysis was performed. The CASP and NIH tools have been previously used to evaluate studies on non-urgent use of the ED in a previous rapid review by Coster et al. [6].

Data abstraction

Two reviewers (AMc, SJ) abstracted the following data from each study included for review: author(s), year of publication, country of first author’s origin, study design (i.e., quantitative, qualitative, mixed methods, or review), sample size, method of triage (i.e., triage system, definition or list of explicit criteria), patient characteristics (e.g., age, gender), study aim/objective, data collection method, and themes/reasons for ED use. The reasons for ED use were abstracted in the manner and language in which they were reported by the original authors. Data were abstracted and summarized in tabular form.

Synthesis

Whittemore and Knafl [9] describe the importance of identifying themes in the data abstraction and synthesis process. As such, the two reviewers coded and identified themes (data analysis stage) from each of the study’s key results. A structured, six-phase thematic analysis was applied using the approach by Braun and Clark [13] (Table 1). This method can be adapted for different types of data (including reviews) [14]. This approach has recently been used for assessing both qualitative and quantitative studies in an integrative review [15]. Specific data synthesis actions are outlined in Table 1. For qualitative studies, the original themes identified by individual studies (with supporting quotes and examples) were reviewed, coded and iteratively compared until large overarching themes between studies were uncovered. For quantitative studies, authors primarily reported results in tabular or list format whereby patient reported reasons were given (usually as statements), along with the proportion of the sample reporting this reason. Similar to qualitative articles, the list of reasons reported by quantitative studies were reviewed iteratively, coded and compared between studies until overarching themes emerged.

Table 1 Braun and Clarke’s [13] phases of thematic analysis, as adapted by Cooper et al. [15], and the research teams’ data synthesis actions

Results

Included studies

The literature review returned 3,268 studies; 401 abstracts and 171 full-length texts were reviewed, from which a total of 93 studies met the previously stated inclusion criteria (Fig. 1).

Fig. 1

PRISMA flow diagram for the systematic literature search

Among the 93 articles included, 69 were quantitative studies [16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84], 21 were qualitative studies [85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105] and 3 were reviews [5,6,7] (Tables 2 and 3). CASP and NIH quality criteria were applied to included articles to consider their rigour. Articles were not excluded based on rigour, rather, areas where there were concerns about rigour within an article were then discussed amongst the research team and themes did not rely solely on such articles. Overall, the included articles were shown to be of very high quality.

Table 2 Study and sample characteristics, quality appraisal score, and key themes/identified reasons for emergency department use among qualitative studies

Table 3 Study and sample characteristics, NIH quality appraisal score, and key themes/identified reasons for emergency department use among quantitative studies

In examining five-year intervals, 11 studies were published in 1995–1999, 8 studies in 2000–2004, 18 studies in 2005–2009, 23 studies in 2010–2014, 24 studies in 2015–2019 and 9 in 2020–2021. Studies were published across 16 different countries, the majority of which originated from just five countries (64.5%): United States (N = 20), United Kingdom (N = 18), Canada (N = 8), Australia/New Zealand (N = 7), and the Netherlands (N = 6). Among the remaining 33 studies, 18 originated from Europe, 3 from South America, 5 from Asia, 6 from the Middle East, and 2 from Africa. Some studies evaluated reasons for using the ED among patients with all types of medical severity. However, for this review, only reasons for attending were collected on non-urgent patients. The total number of patients included in the studies used for the review (excluding review papers) was 49,238. Approximately one quarter of studies had samples with either less than 100 patients (28.8%), or more than 500 patients (26.6%); the bulk of studies (44.6%) recruited between 100 and 500 patients. Fifteen studies (16.7%) did not provide information on the sex of patients (N = 7730 total patients). Among the 75 studies that reported sex ratios, there were 21,044 males (50.2%) and 20,864 females (49.8%) in total. A wide variety of formal triage classification systems were used in 39 studies to assign a severity and urgency of patients’ presenting complaints (Tables 2 and 3). A total of 44 studies did not specify which triage system was used and instead reported that “non-urgent” patients were recruited for participation. In seven studies, a 3-, 4-, or 5-level triage system was described but it was not formally named. The most commonly used triage system was the Canadian Triage and Acuity Scale (CTAS; N = 12) and this was used within and outside of Canada.

Themes

After comparing and contrasting major reasons for non-urgent ED use among studies, a total of seven major themes were identified:

• 1. Need to be Risk Averse with Respect to the Health Issue

• 2. Knowledge and Awareness of Alternative Sources of Care

• 3. Dissatisfaction with Primary Care Provider (PCP) (Subthemes: availability, competence, preference);

• 4. Satisfaction with ED (Subthemes: quality care, access to ED-specific services);

• 5. ED Accessibility and Convenience Resulting in Low Access Burden

• 6. Referred to the ED by Others (Subthemes: health care professionals, non-health care professionals); and

• 7. Relationships between Patients and Health Care Providers

Each theme and sub-theme will be described. Some patients reported that they had no specific reason for attending the ED [32, 35, 36, 63, 79, 81]. Several studies stated that there were “other, unspecified reasons” reported by patients; however, there were no further details provided [19, 28, 32,33,34,35,36, 38, 42, 47, 51, 60, 61, 64, 65, 69, 76, 77, 79, 81, 91, 103].

Theme 1: Need to be risk averse with respect to the health issue

One of the primary reasons reported in the literature for presentation at an ED was the tendency of patients to be risk averse in terms of their health issue. There was a self-perceived sense of severity or urgency to their medical matters, despite that their presenting complaints were deemed non-urgent [16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54, 85,86,87,88,89,90]. Many patients described having feelings of anxiety, uncertainty or significant concern about their health problem [22, 45, 65, 70, 88, 92,93,94,95]. Often patients had experienced pain or other discomforts which impacted their function and they desired immediate relief [22, 42, 45, 70,71,72,73, 88, 93, 94, 96, 97]. In some cases, they had attempted self-treatment at home, without good effect [73, 86, 96, 98], or had sought out primary care without resolve [21, 42, 53, 66, 74, 75, 88, 92]. Even when patients knew their condition was non-urgent, they still wanted reassurance, advice, or a second opinion [26, 33, 37, 43, 44, 48, 53, 55, 56, 64, 69, 72, 88, 94, 97]. One study found that patients had a self-perceived inability to cope [88]. The ability to leave the ED with a confirmed diagnosis or answer to their health problem (attestation) was particularly helpful in mitigating their fears of a real emergency [54, 55, 100, 102, 103].

Theme 2: Knowledge and awareness of alternative sources of care

Studies reported that some ED patients had limited knowledge and awareness of alternative sources of medical care. They were unaware or unsure of the differences between services [16, 27, 33, 35, 43, 51, 52, 55, 62, 66, 76, 78, 87, 89, 95] or simply had not considered going to their PCP [64, 90]. Patients reported that they believed the ED was the only and most appropriate option [22, 24, 31, 34, 40, 45, 54, 61, 63,64,65,66,67,68,69, 86, 90, 91]. Some people did not want to bother their PCP and did not feel it was necessary to seek primary care first [35, 43, 44, 53, 55, 77, 79].

Theme 3: Dissatisfaction with primary care provider

One of the most prevalent themes was related to patients’ dissatisfaction with primary care services. Within this theme there were three sub-themes: availability, competency, and preference.

Sub-theme 3a: Availability

For a variety of reasons, patients reported extreme difficulty in finding an available PCP [18, 32,33,34,35, 38, 41, 42, 45, 46, 52,53,54,55, 61, 62, 72, 89, 92, 94]. They could not obtain a PCP appointment at all [17, 20, 21, 27, 28, 33, 35, 36, 66, 69, 79, 80, 92, 93, 98, 99], or they could not obtain an appointment that did not interfere with work/school [46, 54, 66, 73, 93, 96, 101], or childcare [96]. There were significant issues obtaining care after hours or as a result of limited hours provided by the PCP [16, 19, 22, 28, 33, 35,36,37, 41, 44, 47,48,49,50, 54, 56, 57, 66, 72, 74, 85, 87, 92, 94, 97,98,99, 101,102,103]. Some patients were not registered with a PCP [18, 25, 27, 29, 30, 35, 36, 43, 49, 54, 55, 58, 61, 62, 64, 66, 77] or there were no primary care options at all [26, 28, 45, 46, 54, 62, 66, 70, 90, 92, 96, 97, 102]. A large majority of patients felt that it took too long to wait for an appointment with the PCP, even if they were successful in scheduling one [20, 21, 24, 25, 36, 37, 48, 49, 53, 56, 58,59,60, 62, 74, 81, 82, 88, 91, 92, 99,100,101]. Finally, some studies reported general PCP inconvenience as a reason for non-urgent ED use, although it was not further described [53, 54, 69].

Sub-theme 3b: Competency

A large number of patients reported dissatisfaction with their PCP’s ability to handle their ED concern, which was related to their perceived inadequacy and incompetency. Patients reported feeling dissatisfied with their PCP/staff and even discussed mistrusting them [16, 19, 25, 27, 32, 43, 45, 47, 52, 53, 57, 66, 69, 76, 77, 87, 89, 90, 104]. Some patients thought that their PCP was not capable, could not help them or did not have the necessary resources required to handle their presenting complaint [17, 25, 37, 48, 53, 55,56,57,58,59,60,61,62].

Sub-theme 3c: Preference

In certain health systems, PCP’s operate within a ‘cooperative’ whereby a team of physicians care for a roster of patients. Some patients indicated that, as a result of this model, they had an inconsistent PCP each time they made an appointment and this was less desirable to them [43, 56]. Other barriers to primary care included to varying language, culture and communication practices [45, 48, 56, 90].

Theme 4: Satisfaction with ED

Satisfaction with the ED was a highly cited reason for attending non-urgently. This theme included two sub-themes related to benefits of the ED, namely quality of care and access to ED-specific services.

Sub-theme 4a: Quality care

A large number patients reported that the ED afforded them superior care, beyond what could be obtained in primary care. Patients believed ED care was of higher quality and as such, they had greater trust and confidence in the ED [18, 19, 21, 22, 27, 30,31,32,33, 37, 39, 41,42,43,44,45,46,47,48, 54, 56, 58, 61,62,63,64,65, 74,75,76,77,78,79, 83, 84, 87, 94, 97, 98, 103]. Investigations were perceived to be more thorough [21, 35, 77, 78, 92, 103], with all resources available in one location [18, 22, 26, 28, 33, 37, 40, 45, 48, 56, 66, 82, 86, 93,94,95, 97, 98, 100]. Many patients reported that this was their preferred medical setting, that they were familiar with it, and had previous positive experiences in the ED [16, 22, 26, 30, 36, 42, 43, 45, 53, 56, 63, 68, 86,87,88, 90, 92, 94, 97].

Sub-theme 4b: Access to ED-specific services

The ED is unique in that it provides patients access to a wide variety of resources necessary for assessing, monitoring, managing and treating conditions for most medical problems. Patients reported attending the ED non-urgently to gain access to these ED-specific services they could not otherwise access through a PCP either in a timely fashion, or all in one visit [29, 43, 47, 49, 54, 55, 66, 70,71,72,73, 84, 95]. These included access to diagnostic investigations (e.g., imaging, bloodwork) [18, 25, 34, 42, 51, 56, 59, 62, 71,72,73,74, 79, 80, 93,94,95, 99, 101, 103], access to medication [22, 52, 73, 81, 93, 103], access to specialists [55, 62, 66, 73, 76, 100, 102, 103], or a pathway to hospital admission which they perceived was necessary [25, 40, 55, 59, 65, 66, 72, 103].

Theme 5: ED accessibility and convenience resulting in low access burden

Compared to other primary care options, the ease, accessibility and convenience offered in the ED provided patients with a low burden of access to medical care [17, 18, 21, 22, 26, 30, 31, 34,35,36,37,38, 43, 47, 51, 59, 61, 63, 65, 68, 70, 72, 76, 79, 85, 87, 90, 92, 105]. Patients reported that it saved them time and overall the wait was short in order to receive help [18, 19, 26, 27, 30,31,32,33, 39, 42,43,44, 55, 60, 61, 64, 66, 81, 82, 94, 95, 99, 100, 103, 105]. Some patients faced transportation barriers getting to their PCP so it was easier to access the ED [54, 55, 89, 96]. Similarly, a great number of patients reported geographical proximity to the ED as a motivating factor for attending non-urgently [16, 19, 22, 27, 30, 32, 35, 36, 41,42,43, 45,46,47,48, 53,54,55,56,57, 60,61,62, 64, 66, 72, 74, 75, 77, 81,82,83, 92, 94, 95]. Other convenience factors, such as not requiring an appointment [18, 22, 33, 37, 43, 45, 48, 53, 56, 58, 76, 95] and unrestricted availability (open day and night) [25, 40, 72] were cited as important indicators for ED use. For patients seeking care where medical insurance coverage may be problematic, EDs were often sought out for relief of any financial burden [19, 20, 22, 24, 30, 32, 34, 35, 39, 41, 45, 46, 48, 51, 53, 54, 57, 65, 68, 72, 75, 78, 82, 84, 87, 89]. In a small number of studies, the reason was circumstantial. For example, patients reported being on vacation or were from out of town [56, 58] whereas others just happened to be at the hospital for an unrelated reason [53, 81].

Theme 6: Referred to the ED by others

Patients were often referred by others to attend the ED for their problem; there were two types of referrals discussed, those made by health care professionals and those made by non-health care professionals.

Sub-theme 6a: Health care professionals

Patients reported being told, although this was not verified by most studies, to go to the ED by their PCP [17, 19, 20, 28, 31,32,33,34,35,36, 41, 42, 50,51,52, 54, 56, 58, 61, 63, 67, 69, 71, 72, 74, 78, 80, 89, 91, 98, 100, 104], or by non-PCP clinic staff (e.g., medical secretaries) [17, 19, 33, 80, 90, 99, 104]. In some cases, patients reported attending because they believed their PCP would send them anyway, even if they had not contacted them at all [34, 38, 43, 56, 91]. Patients stated they had attended on the suggestion of non-physician health care providers [17, 19, 31,32,33, 51, 69, 72, 78, 96, 104], a health line [51] or a pharmacist [82].

Sub-theme 6b: Non-health care professionals

Patients stated that non-health care professionals referred them to the ED [26, 27, 30, 43, 59, 61, 63, 73, 91]. For example, family, friends, and others in patients’ social network were influential in telling them they should go to the ED [22, 33, 35, 38, 41, 42, 45, 62, 64, 70, 72, 82, 83, 86, 88, 90, 96, 105]. In two studies, patients stated their reason for attending was based on influences by the media (i.e., advertisements) [53, 97]. For others, patients were specifically directed to the ED by their employer [72, 105] or by the police [71].

Theme 7: Relationships between patients and health care providers

There are often dynamic interactions or relationships between patients and health care providers. In certain groups and geographical regions, use of the ED was an automatic, habitual behavioural or cultural practice shared by many patients [26, 33, 42, 45, 47, 48, 51, 52, 56, 66, 76, 83, 87]. For hospital staff or members of their family, the ED was a logical place to attend given their proximity to place of employment; the relationships these patients had with the ED (and the health care system at large) facilitated its use [19, 32, 42, 46, 47]. Attending the ED, even non-urgently, also made ‘sense’ for those who were currently (or previously) receiving treatment from that hospital already [19, 22, 27, 32, 34, 36, 45, 51, 65, 72, 78, 79, 95]. Conversely, for others, the ED acted as a place of anonymity because no relationship existed. The possibility of obtaining medical care from a doctor they did not know [37, 48, 56] or from someone of the same or opposite sex [48] was appealing.

Discussion

Summary of results

The aim of this study was to conduct an integrative review of the scientific literature to explore patient-reported reasons for using the ED non-urgently. The studies included for review reported that attending the ED was an intentional decision based on several influential factors. Seven main themes were identified: 1) Need to be risk averse with respect to the health issue; 2) Knowledge and awareness of alternative sources of care; 3) Dissatisfaction with PCP (Subthemes: availability, competence, preference); 4) Satisfaction with ED (Subthemes: quality care, access to ED-specific services); 5) ED accessibility and convenience resulting in low access burden; 6) Referred to the ED by others (Subthemes: health care professionals, non-health care professionals); and 7) Relationships between patients and health care providers. For many patients, there was a very clear problem which needed to be addressed, whether it was physical, psychological, or social. After weighing several options, from their perspective their need was real and the ED as an option for care was rational and justified, not just their last resort.

Context of other research

The results reported here are well-aligned with other reviews [5,6,7], but also extend the current knowledge of the subject by providing a comprehensive synthesis of all extant literature of reasons for non-urgent ED use. Recently, O’Cathain et al. [5] examined non-urgent ED use using a ‘realist review.’ Building on earlier reviews [6, 7, 106], they performed an updated literature search to the end of January 2017. They compiled and compared results from 29 quantitative studies, existing health behaviour theories, and 32 qualitative studies. Our integrative review was able to validate and supplement the ten program theories and six mechanisms of decision-making as described by O’Cathain et al. [5] with a larger compilation of studies. With respect to program theories, we did not uncover the theme of ‘fear of consequences when responsible for others’ found by these authors. This theme potentially relates to individuals’ responsibility to care for children, and we did not include studies on the pediatric population. Five mechanisms of decision-making were described by O’Cathain et al. [5] and shared with our integrative review (i.e., the need to be risk averse with respect to a health issue, ED accessibility and convenience resulting in low access burden, satisfaction with ED, dissatisfaction with PCP, and referral to ED by others). However, they reported that there was either limited or no support at all from the quantitative literature with respect to experiences of past traumatic events, anxiety, stress, coping, and need for immediate pain relief. In contrast, we found significant support for these reasons within the quantitative literature included here for review. Further, two themes not emphasized by O’Cathain et al. [5] were found to be highly influential in this review (i.e., Knowledge and awareness of alternative sources of care and Relationships between patients and health care providers). The additional studies incorporated in this integrative review (N = 60), not previously captured in other reviews, serve to both validate and enhance our previous understanding of the context surrounding decision-making for non-urgent ED.

Clinical implications

While there is the wealth of knowledge on this topic, the majority of studies were published from highly resourced nations (i.e., USA, UK, Canada, Australia); as such, the results should be considered in light of this context. For example, Canada has a publicly funded healthcare system which contrasts with the private health care model utilized in the United States, and various two-tiered systems adopted in Europe and Australia. American studies have reported financial barriers to primary care as a common reason for attending the ED [107]. Financial barriers are not particularly relevant to individuals from nations with public or semi-public health systems since they, in part, have a reduced (direct) financial responsibility for medical care. Our understanding of reasons for non-urgent ED use in less resourced nations is currently limited.

The results from this review suggest that ED patients are heterogenous and that many factors influence their decision-making. Considering the complexity of patients that EDs care for, treating them as a single entity may be problematic. Thus, a multi-pronged approach may be required to limit excessive non-urgent visits. For example, simply redirecting non-urgent patients to other settings has been shown not to be wholly effective [108]. Instead, ensuring health care providers (at both PCP clinics and ED) understand how and why patients make decisions may help to provide insight and direct patient education. Health education should be explicitly and intentionally embedded in all ED health care provider roles [109]. This involves communicating, managing knowledge, mitigating errors, and supporting decision-making [109]. Research suggests that basic educational expertise, fundamental knowledge and reasoning, as well as emotional self-regulation are all critical components of health [110]. Thus, education is a social determinant of health which can potentially impede or enhance patients’ health [110]. Routinely educating patients on the role of the ED, as well as alternatives in the community, is a critical aspect of improving the public’s health.

This review found that many patients were anxious, uncertain, or fearful of their health problem. They had decreased ability to manage their discomforts and some reported the inability to cope. Guidance and support should be provided to patients with respect to managing recurring symptoms which may be directly or indirectly (e.g., anxiety, stress) related the presenting condition. Discharge teaching could include problem solving techniques for decision-making (e.g., accessing information) as well as self-management strategies (e.g., pain relief). While these “common sense” strategies may be commonplace among health care providers, it should not be assumed they are shared with lay persons. A recent systematic review highlighted and confirmed the disparity in patients’ and clinicians’ mutual understanding [111]. The authors examined the effectiveness of different methods of providing discharge instructions in the ED and found that communicating discharge instructions verbally may be insufficient; greater success could be achieved with the addition of video or written information [111], or via social media.

Finally, this integrative review demonstrated that there are notable deficiencies in various design and functioning of health care systems, where the literature was drawn. Many patients reported significant issues with accessing primary care, and were dissatisfied as a result. Simultaneously, patients were satisfied with the ED due to increased accessibility and quality of care, thus driving their attendance. This result has been supported by Van den Borg et al. [112] who examined the relationship between attending the ED and accessibility and continuity of primary care among 34 countries (60,991 patients). They found that ED visits had a significant and negative relationship with better primary care accessibility [112]. Systematically improving deficiencies in primary care may reduce non-urgent ED visits. Policy makers and practitioners should reflect and consider the complexities of their given health care environments to adequately design systems which are responsive to patients needs.

Research implications

There has been a significant amount of inquiry generated on patient-reported reasons for non-urgent ED use. Regardless, there are a few areas that should be targeted for deeper inspection which would assist in filling gaps in the knowledge and addressing certain methodological considerations. Future research should aim to explore, in greater depth, specific themes identified in this review. For example, the role of health knowledge, emotions, beliefs, attitudes, and behaviour response patterns have been indicated as influencing the decision-making process, specifically with respect to perceived severity and urgency of presenting condition. New studies should explicitly evaluate ED users’ health literacy, health-related personal beliefs, stress and coping ability using validated outcome measures. This approach has received little to no attention in the literature. Psychosocial factors (e.g., stress, coping) have been explored in only a dearth of studies, largely as an afterthought to the primary objective [2]. Linking this subjective data to large, objective administrative health data could provide greater context than simple patient-reported reasons. Researchers should endeavour to use standardized criteria to evaluate triage acuity, when possible, and to fully describe their patient population and geographic region for accurate interpretation of results and comparisons with others.

Limitations

This integrative review is not without its own limitations. The strict inclusion and exclusion criteria may have limited some articles from being included (e.g., all non-English studies). Further, specific populations (e.g., ambulance riders, pediatrics, specific presenting complaints, frequent ED users) have been cited as using the ED non-urgently but studies focused specifically on such subgroups were excluded from this review. This was intentional in an attempt to create a more homogenous sample for review. These ED subgroups may contribute unique results which could be informative to this topic. Nevertheless, results were drawn from a very large pool of general ED population studies. Finally, integrative reviews have the potential to suffer from lack of rigor given the process of combining diverse, complex methodologies [9]. The methods described herein were conducted using an iterative coding process by two individuals following well-cited, formulated guidance [9, 13].

Conclusion

This integrative review summarized over 30 years of research evidence on patient-reported reasons for non-urgent ED use. It was conducted using a rigorous systematic methodology and data analysis in accordance with widely accepted reporting criteria. The inclusion of both qualitative and quantitative studies led to a comprehensive understanding of seven major themes associated with decision-making, namely: Need to be risk averse with respect to the health issue; Knowledge and awareness of alternative sources of care; Dissatisfaction with PCP; Satisfaction with ED; ED accessibility and convenience resulting in low access burden; Referred to the ED by others; and Relationships between patients and health care providers. Future studies should use validated outcome measures to specifically explore the role of complex psychosocial factors driving decision-making including health literacy, health-related personal beliefs, stress and coping ability.