Deciding whether to consult the GP or an emergency department: A qualitative study of patient reasoning in Switzerland

Abstract

Background: Non-urgent care is an important factor responsible for rising healthcare costs and general practitioners (GPs) are known to be more cost-effective than emergency departments (EDs).

Objectives: To understand the reasons why patients confronted with a medical problem perceived as urgent choose to consult either a GP or an ED.

Methods: We conducted a qualitative study in Switzerland, using data collected between 2014 and 2015 through semi-structured interviews of adults with non-vital medical problems. Half were recruited after an ambulatory consultation in an ED, and half were patients who consulted their GP. The audio-recorded interviews were transcribed, coded, and analysed according to the constant comparative method.

Results: The main reason given by patients who consulted their GP first was the quality of the relationship. The more meaningful the relationship, the more likely patients were to seek advice from their GP. One marker of a privileged relationship was GPs supplying their mobile phone number to the patient. The perceived nature of the complaint, for example, symptoms considered as life-threatening or severe pain, together with the expected waiting time in an ED were additional factors influencing the patients’ choice.

Conclusion: Our study showed that when patients are confronted with what they perceive as a medical emergency, the quality of the relationship with the GP, in particular the continuity of care provided, seem to be the major reasons why they consult their GP rather than an ED.

KEY MESSAGES

• A close relationship with their GP seems a major reason for patients consulting their GP rather than an ED.

• The perceived nature of the complaint, such as symptoms considered as life-threatening or provoking severe pain, were important factors for participants to consult in an ED.

• Waiting time acts as a regulatory factor between the different consultation places.

Introduction

Emergency departments (EDs) in industrialized countries are facing a substantial increase in the number of consultations [1–3]. This phenomenon was observed in Switzerland in the early 1990s and continues today [4,5]. Several factors account for this increase. One is the growing number of patients seeking non-urgent care in hospitals [6]. Greater difficulty in accessing primary care facilities may account for this developing phenomenon. In addition, numerous patients no longer have a GP and refer to an ED when seeking medical advice. A nationwide review in the UK in 2013 stated that access to primary care services should be reinforced to limit the number of ambulatory consultations in EDs [7].

In Switzerland during office hours (8 am to 6 pm), patients can choose whether to consult their GP, walk-in facilities or a hospital. Most GPs do not have evening or weekend working hours. Walk-in clinics, mostly situated in larger cities, close at about 9 pm, some of them at 11 pm. Therefore, the ED is the only healthcare facility open 24/7. This unrestricted access may contribute to the increasing number of consultations in EDs [8], thereby inflating costs for healthcare systems, since studies have shown that costs are higher if patients consult in an ED rather than seeing a GP [9,10].

To reverse this trend, we need to understand better why patients choose to consult an ED rather than a GP when seeking medical care. Few studies have focused on patients’ rationales for deciding where to consult [11]. This study aimed to identify the personal reasons why patients when faced with a medical problem they perceive as urgent choose to consult their GP or an ED.

Methods

General design

Given the exploratory nature of the study, we chose a qualitative design, based in grounded theory and conducted individual semi-structured interviews [12]. After transcription and coding following the constant comparative method, we organized a focus group with the recruiters to present our first results and to obtain their input for our final analysis.

GPs and ED recruitment

We recruited 12 GPs through the ForOm NV network. This programme aims to promote family medicine in the northern region of the canton of Vaud in Switzerland. It links medical students and medical interns with teaching GPs and with the regional hospital of Yverdon, which has a catchment area of around 50 000 inhabitants, covering a typically Swiss population mixture of rural and urban areas. For this study, we collaborated with its ED, which had approximately 20 000 admissions in 2015 with an increase of about 7% each year [13].

Patient recruitment

Patients were recruited through GP surgeries (n = 9) during the opening hours and the Yverdon regional hospital ED during 24/7 (n = 11). Invitations were issued by the collaborating GPs and by medical ED staff briefed on the study. All invited patients accepted to participate in the study.

Inclusion criteria demanded that participants be older than 18 years of age, live in Switzerland and speak French fluently. Purposive sampling was applied to obtain equal gender representation, a wide age range (19–82 years) and a balanced proportion of Swiss and foreign origins with different socioeconomic backgrounds. Detailed information on participants is supplied in Table 1.

Table 1.

Participants’ characteristics.

Participant	Gender	Age	Nationality	Profession	GP	Recruitment source
1	Male	64	Swiss	Business owner	A	GP
2	Male	82	Swiss	Farmer	B	GP
3	Female	74	Swiss	Retired	C	ED
4	Female	52	Swiss	Teacher	A	GP
5	Male	52	Swiss	Representative	D	GP
6	Male	68	Swiss	Retired	D	GP
7	female	27	Swiss	Cook	E	ED
8	Female	44	Swiss	Waitress	E	ED
9	Female	44	Portuguese	Hairdresser	D	GP
10	Female	35	Kosovan	Saleswoman	D	GP
11	Male	41	Swiss	Business owner	E	ED
12	Female	20	Swiss	Student	F	ED
13	Male	22	Swiss	Student	G	ED
14	Male	36	Portuguese	Driver	H	ED
15	Male	25	Eritrean	Student	B	ED
16	Female	19	Swiss	Student	I	ED
17	Female	64	Swiss	Farmer	J	GP
18	Male	34	Tunisian	Engineer	H	ED
19	Female	48	Swiss	Secretary	J	GP
20	Female	33	Swiss	Teacher	K	ED

This Table summarizes the main characteristics of our participants including their nationality. In comparison, 34% of the residents in the canton of Vaud are non-Swiss nationals http://www.scris.vd.ch/Default.aspx?DomId=1956; 2017). GPs are identified in an anonymous way by an alphabetic letter.

GP-attending patients consulted for an urgent but ambulatory problem. In the hospital, patients were only eligible if scoring the lowest degree of emergency on the Swiss Emergency Triage Scale [14] and discharged after the consultation. All vital emergencies or severe pathologies needing hospitalization were excluded.

Interviews

Individual semi-structured interviews were conducted by our interviewers who were all clinicians aged 35–44 years with an MD degree. To pre-test the topic guide based on existing literature (Table 2), we arranged videotaped training interviews that were subsequently analysed collectively with the contribution of the second author who teaches qualitative research. This step allowed us to validate the topic guide and ensure correct use.

Table 2.

Semi-structured interview guide

Key question	Sub-question	Optional questions
Why did you consult recently?	What were your emotions? What did you most worry about?	Did you feel stressed or anxious before the consultation? Did you feel pain?
How was the consultation?	What were your feelings during the consultation?	Did you feel comfortable?
What kind of relationship do you have with the physician?	With your GP? In the ED?	Explain the differences between them.
What were your expectations before this consultation?	What did you expect from the doctor?	Did you need to be reassured? Did you need some medicine? Did you need a medical certificate? Did you want a specific exam (e.g., X-ray)?
Before your consultation, which illness did you think you had?	Had you already had this type of problem? What did the symptoms mean to you?
When you decided to consult, where did you think of going?	Did you consider consulting elsewhere? If so, where?	Did you consider going to your GP? Did you consider contacting a doctor on-call? Did you consider going to a walk-in facility?
For what reasons did you consult in this place?	When would you have consulted: A doctor on-call An ED	(For those who consulted in an ED.) Do you think that your problem could have been handled by a GP? (For those who consulted a GP.) Do you think that your problem could have been handled in an ED?
Which practical aspects are important for you when you have to consult urgently?	How important are the means of access from your point of view? How important are opening hours for you? How important is the availability of the doctor for you?	Did the ease of access, like parking facilities or public transport, influence your choice? Did opening hours influence your choice? Did the expected waiting time influence your choice? Did the availability influence your choice?
What is your impression of the ED?	Have you already consulted in an ED before? How many times have you consulted urgently in the last two years? To what extent do facilities such as CT-scan or laboratory tests influence your choice?
What image do you have of the ED?	Personal experience or movies, TV-series
According to you, how could urgent consultations be improved?

After being contacted face-to-face, participants received detailed information about our study, signed a consent form and completed a short questionnaire about their characteristics. Interviews lasted approximately 45 min, were audio-recorded, anonymized and then transcribed verbatim (in French). The interviews took place in our offices in Yverdon, Switzerland, between October 2014 and December 2015. Interviewers had no previous knowledge of the patients participating in the study.

Data analysis

The first and third authors (SH, OP) coded each meaning within transcripts using NVivo software (NVivo qualitative data analysis software; QSR International Pty Ltd, Version 10, 2012, Melbourne, Australia). We ceased recruitment after 20 interviews because no new codes were identified, indicating we had reached saturation point. After careful comparison of the codes (open coding) between the two principal investigators (SH, OP), we organized them into categories to extract the major items (focused coding) relating to our research question.

We then presented the first results of our analysis to a focus group consisting of the head of Yverdon hospital’s ED and 8 GPs who had participated in the recruitment. This audio-recorded focus group aimed to synchronize our framework with their input, which we used for the final sequence of our analysis.

Results

Three key areas emerged as determined in a patient’s decision-making process regarding where to consult: the quality of their relationship with their GP, the perceived nature of the complaint, and the anticipated amount of time that they would need to wait before being seen. These are described below. Table 3 compares the two groups (patients recruited in the ED or by the GPs) and summarizes the perceived strong and weak points of consulting a GP or an ED.

Table 3.

Strong and weak points mentioned by participants recruited either by a GP or in the ED.

Participants’ recruitment place	Strong points	Weak points
GP	Trusting relationship Access to patient’s medical records Possibility of having an appointment	Limited opening hours Not always available on the same day Limited technical equipment
ED	Technical equipment (e.g. radiology) Opening hours 24/7 Access to specialists	Many inappropriate consultations Stressful environment Waiting time

Quality of patients’ relationship with their GP

The key result of this study was the importance of the participants’ relationship with their GP. When asked ‘What kind of relationship do you have with the physician?’ many of them cited their family doctor as someone they trust, a person of confidence.

My GP is a person whom I have known since I was nineteen years old. I’m now fifty-two, so I’ve known him for a long time. My father, my brothers, my wife and even my children are his patients. (P4)

To the question: ‘What kind of relationship do you have with your GP?’ some patients, especially the older ones, described their relationship as a friendship.

My GP is like a friend, a confidant. (P19)

The fact that GPs gave their private mobile phone number to their patients had a strong symbolic value in their relationship with patients. Patients appreciated it very much because they perceived it as a sign of confidence and virtually unlimited availability.

He even gave me his mobile phone number in case something serious happened, so I was able to reach him round the clock. I appreciated this gesture. [Interviewer: And did you use it?] … It didn’t prove necessary. (P10)

We often are in touch by texting when I have questions. You see how available she is, and I find that great. (P19)

We asked the GPs who participated in our focus group about providing their mobile phone number. All of them would give their number to patients under particular circumstances such as palliative care, severe chronic illness, psychiatric emergencies, etc.

Some patients’ trust in their GP was so strong that, when confronted with severe symptoms, they risked making an inappropriate choice. For instance, one patient with severe night dyspnoea explained how she would rather wait until her GP’s practice opened the next day than go directly to the ED, although it was nearer to her home.

My husband keeps telling me: there are doctors just nearby, you know, and yet you continue going to Dr X’s. And I tell him: well yes I do … but I have something special with her, a special relationship which I’m not sure I would find elsewhere. (P17)

After presenting this result to the GP focus group, such an extreme situation appeared to be rare. In their experience, this is particularly true concerning older people. Younger patients consult more easily in an ED or a walk-in facility because they are more likely not to already have a GP and have had fewer medical problems.

The fact that their GP knows them (personal history, medical file) seems to be reassuring for patients, as explained by this participant:

My GP knows me very well and for me it’s very reassuring. It’s in any case more reassuring than consulting in an ED where I’m seen by a person whom I don’t know and who doesn’t know me. (P4)

By contrast, another participant recruited in the ED explained why it was impossible for him to build a satisfactory relationship with a new GP after his family doctor had died. This was mainly due to the numerous changes in the GP’s surgery where he consulted subsequently. Now he almost always consults in an ED.

At the beginning, it was Dr Y and then there was a change. And now it is yet another doctor. Things are changing all the time. You’re at a loss! […] Now, I go to the nearest place. (P11)

Participants of the focus group pointed out that if the quality of relationships in medical services continues to decline, patients are likely to adopt a more pragmatic attitude when making a choice about where and when to consult.

Perceived nature of the complaint

The circumstances that induced this study’s participants to consult were also crucial. Drawing on their medical knowledge and experience, participants had to decide whether it was urgent to consult or not, and whether their GP—when they had one—was the right person to see. Symptoms that were considered life-threatening such as chest pain or severe headaches were important reasons for consulting in an ED.

Of course, if somebody has a heart problem, it is a real emergency! (P3)

Many participants understood that the ED should be used for urgent cases only, although according to some participants, many consultations in the ED were inappropriate; to quote this hairdresser:

I have several clients who go to their GP or even to the ED for nothing serious. And I question them: but you didn’t go to the ED for that, did you? And they answer: oh yes, I had to go because I was such in pain! (P9)

Waiting time

Our participants referred to the waiting time mostly about consultation in the ED.

Actually, the picture I have of the ED is that you have to wait a long time for nothing. (P12)

The waiting time until a consultation with a GP could be scheduled varied greatly between participants. Some GPs were more available and could see their patients within the same day. Other participants had to wait for two days or more, in which case, many of them decided to consult an ED.

Compared with my GP, where I sometimes have to wait for 24 hours […], I know that I will have to wait less than four hours to be received at the hospital. (P7)

Waiting time, therefore, seemed to act as a regulatory factor between the different consultation places.

Discussion

Main findings

When faced with what is perceived as a medical emergency, the quality of the participants’ relationship with their family physician, in particular, the continuity of care provided, seemed to be the major reason for participants choosing the family practitioner’s surgery as an entry into the healthcare system. All participants who reported having a relationship of trust with their GP preferred to consult them first.

The GP’s mobile phone number appeared to carry a strong symbolic value. Several patients understood their GP’s giving their mobile telephone number as a sign of confidence and explained how that reassured them and reinforced their mutually trusting relationship.

Another point that our study highlights is the direct link between the nature of the symptom or its severity and the decision made by the patient. For example, symptoms considered life-threatening, or severe pain, were identified as important reasons for consulting the ED.

Participants also referred to the waiting time as another important factor. They explained how the expected ‘wasted’ time influenced their decision-making process. On the one hand, a GP gives an appointment and therefore the time spent in the waiting room is reduced; on the other hand, some participants needed a rapid answer to their problem and preferred consulting in an ED directly. Moreover, participants who needed a rapid consultation and were confronted with their symptoms during out-of-office hours were more inclined to consult an ED directly. This was also true for people who could not, or did not want to, consult during their work hours.

Strengths and limitations

The strength of our study lies in its qualitative design permitting us to explore the subjective reasoning of participants: to our knowledge, the first such study ever conducted in Switzerland. One limitation is that our recruitment area was limited to a predominantly rural part of the country, a parameter that we were unable to influence during the purposive sampling process to balance the characteristics of subjects in terms of age range, occupation and medical experience before reaching data saturation [15].

Comparison with other studies

Our main result confirmed findings from a previous study conducted in Scotland. Farmer et al. described how the strong relationship between GPs and their patients in rural districts was one of the reasons why patients consulted their GP before going to an ED [16]. Swiss GPs also cite this close relationship as the keystone of their therapeutic approach [17].

Little research had explored the consequences of the fact that patients know their GP’s mobile phone number, although a Scottish study confirmed that patients showed satisfaction when they had the opportunity to reach their GP on his or her mobile phone [18]. Our participants described such an opportunity as proof of mutual trust.

The choice of the place of consultation can be founded on a pragmatic basis according to the availability of care, as confirmed by European studies which establish that younger patients are more likely to consult an ED in comparison with older patients who tend to see their GP first [4,19]. Patients who no longer have a GP for various reasons, such as being retired or moving house are also more likely to consult in an ED. A publication of the King’s Fund confirmed that better continuity of care with the GP was associated with fewer admissions in an ED [19]. Consequently, inadequate follow-up leads to more consultations at the hospital [20]. The patient’s relationship with their GP would, therefore, seem to have an impact on the appropriateness of their choice of whom to consult.

Our study shows how symptoms perceived as urgent or severe pain are determining factors in choosing an ED visit rather than seeing the GP. This is confirmed in a recent study by Detollenaere et al. in Flanders [21]. Furthermore, the accuracy of this patients’ assessment is influenced by their medical knowledge, anxiety levels and comorbidities. This assessment is known to be a particularly difficult exercise for many patients [11,21].

In a study by Campbell [22], waiting time tended to affect the intention of whether to consult. Our results also showed that waiting time acts as a regulatory factor in access to the healthcare system by affecting people’s choice of service rather than leading them not to consult at all: the longer the perceived waiting time in ED, the more likely the choice to consult a GP or a walk-in clinic. The reverse is also true, as described in the results of Watson’s study [10].

Implications

Our results show how important the availability of GPs is to our participants. Proposing GP surgery hours during the evenings and weekends would facilitate access considerably. Increasing the number of GPs to make them more available is another proposition to be considered by our healthcare policymakers when reorganizing non-urgent care to control rising costs. Further research with larger populations should be conducted in Switzerland to examine the effectiveness of these different approaches.

Conclusion

Our results show that when faced with a medical problem perceived as urgent, the quality of the relationship between patients and their GP, in particular the continuity of care they provide, is a key factor in determining where patients decide to consult. The nature of the patient’s symptoms was also identified as an important decision-making factor when consulting an ED, particularly if the symptoms were considered life-threatening. Finally, waiting time seems to play an important regulatory role in the choice to consult in an ED or see a GP.

Acknowledgements

The authors thank the Lausanne University Institute of Family Medicine (IUMF) and the Institute of Social and Preventive Medicine (IUMSP), especially L. Herzig, D. Widmer and D. Koutaissoff, for their advice and I. Aubin-Auger of the University of Paris Diderot for her expertise in qualitative research. The authors would also like to thank our colleagues involved in the North Vaudois curriculum of family medicine (ForOm NV), especially N. Haftgoli and M. Jayet, for their contribution at the beginning of the study; L. Sewer Burdet, H. Monigel, M. Sauter and T. Tai for the recruitment of patients; I. Hottinger, J.-D. Lavanchy, C. Béguin, P.-O. Muller for their participation in the focus group; J. Ombelli, Head of the Emergency Department of Yverdon Hospital, (eHnv) for his support and J. M. Pasche-Gray and C. Eidenbenz for their help with English editing.

Preliminary results were presented to the European General Practice Research Network (EGPRN) at the annual meeting in Timisoara (Romania) in May 2015.

Ethical approval and consent to participate

The Ethics Committee of the University of Lausanne (no. 24/13) accepted this research protocol (Switzerland). Written and informed consent was obtained from all participants.

Disclosure statement

The authors report no conflicts of interests. The authors alone are responsible for the content and writing of the paper.