Displaying emergency patient estimated wait times: a multi-centre, qualitative study of patient, community, paramedic and health administrator perspectives

Study design

This qualitative interview and focus group study was conducted between July and November 2019 to determine consumer, community, pre-hospital and hospital administrator perspectives on emergency department wait time visibility. We employed a purposeful sampling strategy and used a grounded theory approach, using semi-structured interviews. The study was undertaken in Melbourne, Australia.

The study received Cabrini ethics committee approval (10-13-05-19) and was prospectively registered on ANZCTR.org.au: ACTRN12619000665134.

Characteristics of interviewer

The female interviewer was an emergency trainee introduced as a university researcher, whilst wearing office attire, with no previous relationship with participants.

Settings and participant selection

We sought information from six groups: patients/support persons in EDs, paramedics, well community members, paramedic controllers, hospital administrators and general practitioners (GPs). EDs were invited to participate if they were part of an academic health science centre (regional to tertiary hospitals). Other hospitals were recruited via researcher networks. Recruitment methodology for each participant group is described in Figure 1. Paramedics and controllers were recruited from Ambulance Victoria. GPs were invited to participate in phone interviews during education sessions for GPs at Cabrini. Community members included (i) sporting club members, (ii) special/preschool/primary/secondary school parents and (iii) older people receiving nursing home services from a large district home nursing services provider. Face-to-face recruitment was by convenience sampling.

• Download figure
• Open in new tab

Figure 1.

Recruitment methodology and flow through the study

Sampling continued until thematic and information saturation was achieved, then three further interviews were obtained from each group.

Data collection

Interviews and the focus group were semi-structured with potential questions and prompts as needed (Appendix 1). There were no repeat interviews or transcription checking by participants. Interviews were recorded as audio files for professional transcription. Field notes were undertaken during the focus group. Interview duration ranged from 15 to 30 minutes.

Data analysis and findings

Four researchers coded interviews using NVivo software (QSR, Melbourne, Australia) or paper. They worked independently, coding transcripts using open coding techniques. An iterative process was undertaken, refining the codes until most content could be classified. Themes were developed using an inductive and iterative approach and content and agreement on themes was reached at a researcher meeting.

Flow through the study

One-hundred and three interviews were conducted and eight people participated in a single home nursing service focus group. Eight people in EDs declined interviews. Few community members responded to invitations and no GPs participated. One transcript was lost (demographics unknown). All transcribed interviews were included in the study. (Figure 1.)

Hospitals and individual participants

Participating EDs included two medium metropolitan, three large metropolitan, two major, one major paediatric and one private hospital. (Table 1.) One hospital displays a wait time online, to Ambulance Victoria control, and in its ED waiting room (Cabrini). Four hospitals declined to participate (major, medium and large metropolitan, private). Interview participants are described in Table 1. Interviewees had a median age of 49 (range 20,93, IQR 33,58) and 59 (55%) were female.

Themes

Thematic saturation was achieved for patients, their advocates/carers and for ambulance administrators and paramedics. Theme saturation wasn’t achieved for hospital administrators or community members (parents of school children, sports club members, home nursing service clients). Themes relating to ED wait time visibility are presented below. Themes related to predicted ED wait time visualisation are presented in another manuscript.

Perspectives on not having a visible ED wait time

Consumers in the ED wanted to know their estimated wait time 48/51 (94%). Three consumers (6%) were happy to wait for care regardless of wait duration and didn’t wish to know how long this might be.

Consumers report feeling a lack of control when the wait time isn’t visible. They arrive at an ED seeking immediate treatment and then face a void of information following triage. Patients reported wondering if they should go home, feeling that if they had a true emergency, they would have been treated immediately. Some find their anxiety increases the longer they have to wait and discussed how having a wait time visible in the ED waiting room would lower their anxiety and assist them with planning other commitments.

“People get cross and angry, three hours, that’s a long time to be ignored” (consumer)

Others worry about planning their next few hours, whilst having competing demands on their time.

“I’ve got three kids…(it’s)annoying sitting there and thinking do I need to get someone to pick them up?” (consumer)

Many consumers commented that if wait times were visible, they could manage their wait better (including toileting, food, family/work communications, psychological expectations).

“I could have got a coffee or gone to the bathroom” (consumer)

Consumers report that after an hour of waiting, they start asking the triage nurse if they are still in the queue.

Health administrators consider that consumers have the right to transparent information about wait times and many hoped to make times available.

“I think if publicising wait times helps patients make choices” (health administrator)

Patients similarly requested wait time transparency. Paramedics remarked that transparency would reduce patient anxiety whilst waiting. Paramedics also commented that knowing wait times would help with planning care whilst patients remained on stretchers (eg toileting needs, administering pain relief, changing oxygen bottles, paramedic shift changes). (Table 2.)

“I did ask at one stage because I needed to know how to manage my patient, if they were going to go straight to a bed, I wouldn’t need to give more pain relief. We often need to know to organise another crew to come and relieve us.” (paramedic)

Expectations about waiting in an ED

Thirty-one consumers voiced opinions about how long it was reasonable to wait to see a provider. Seven thought less than one hour, 13 thought 1-2 hours and 11 were comfortable waiting longer than 2 hours. Most consumers interviewed had already been waiting longer than two hours.

“I would think one hour would be a very long waiting time, realistically, it can be two hours” (consumer)

Once waits were longer than 2 hours, consumers either felt they would wait as long as it took or were considering an alternative to their local ED. Alternatives considered were primary care or representing at a quieter time. Consumers who had been waiting over 6 hours seemed accepting that there was a prolonged wait and felt safer sitting in the ED waiting room than at home or in their GPs. (Table 3.)

Impact and decision making if ED wait times are visible

All consumers in ED waiting rooms perceived that they had an urgent condition needing treatment at an ED. If prolonged wait times were made visible, consumers discussed: choosing an ED with a shorter wait time; presenting to their closest ED anyway as they felt they would be treated quickly (queues were for everyone else); considering alternatives to ED treatment; calling an ambulance (anticipating a shorter queue) or attending their usual hospital, regardless of wait, due to a long-term relationship with the facility.

“I would travel further to not wait as long, definitely” (consumer)

“if it was crazy wait times and your patient was low-acuity, you could (go) to another hospital” (paramedic)

“it would just be the closest hospital” (consumer)

Paramedics reported being more interested in which was the closest appropriate facility than in the length of the wait. They said that if their protocols required it, they would be happy to choose the facility with the shortest wait time provided the hospital could meet the patient’s needs and the consumer accepted the decision.

Health administrators felt that consumers had capacity to integrate wait time information into appropriate decision making and that it would assist them in choosing a provider.

“it’s paternalistic of the profession not to trust patients to make valid decisions on where they want to go” (administrator)

One administrator felt that all patients should stay in their local area to facilitate post-hospital integrated care and family support and hence wait times weren’t relevant to their population. (Table 4.)

Safety issues regarding visibility

One out of 52 consumers stated that they would go home and not seek further care if wait times were many hours. The rest stated that they would wait for care, return the next day or seek alternate care (another ED, GP or other provider).

“if I thought something was seriously wrong, I’d stay however long it took” (consumer)

“(long waits) wouldn’t stop me from going to see somebody but it might change where I go” (consumer)

Most consumers understand that patients with time critical illness don’t wait and wait times don’t apply in this scenario. There was a variety of knowledge presented by consumers regarding triage and different waits for different people. Health administrators and paramedics expressed concerns about long wait predictions potentially deterring patient from seeking help. (Table 5.)

Definition of an ED wait time?

One hundred and one participants expressed opinions on this. The majority of non-ambulance participants wanted triage to doctor times displayed. Consumer interviewees defined a start time as when triage commenced (33/52) or when they arrived at the front door (18/52). One thought the time started when they got to the car park. The finish time was when consumers saw their doctor/definitive clinician (30/51). The majority of hospital administrators and community members felt the same (triage to doctor). A quarter of consumers wanted the total ED time to be provided (triage to discharge) (11/52). Very few people from any group wanted to know when they would go to a cubicle or have management commenced (medical triage, nursing assessment, medications, imaging, pathology). (See Appendix 2. for ED journey stages)

Ambulance personnel wanted the waiting time to start when they arrived at the door (22/31), with (9/31) feeling the start time was the triage time. Most (30/31) felt the finish time was when the patient was off-loaded from their stretcher (ie door to off-stretcher duration).

Would people use visible estimated ED wait times?

Of those asked, 47/50 of consumers felt they would use a wait time estimator. Twenty-eight of fifty consumers said they would look up a wait time prior to attending an ED as this would inform their choice of facility. Another 19/50 said they would look at the wait time on arrival as they needed to get to an ED before doing anything else (including researching where to go). Almost all wanted the wait time to be available in waiting rooms. Three interviewees didn’t want to see wait times at any stage of their journey. Well community members supported wait time visibility (6/7) as did hospital administrators (8/9).

Thirty of thirty-one ambulance employees (paramedics and controllers) felt they would use a wait time estimator. Twenty-five would use times to assist with load-spreading (choosing hospitals with shorter waits), three would use the time once in an ED, one wouldn’t use the information. Importantly, paramedics will only use wait time predictions for load-spreading if all hospitals in the region are supplying information and they are available in one place electronically for comparison.

Limitations

Participants were from a single city in Australia. The data were from a mixture of interviews and focus groups, findings from one method of data collection may differ from another. The study didn’t achieve theme saturation for health administrators, well community members or GPs, so there may be views that haven’t been presented in this jurisdiction, or different views in other locations. In particular, health administrators may have barriers to transparency that haven’t been mentioned (such as concerns about highlighting long waits or receiving more patients due to shorter waits). Future work should consider the views of emergency healthcare workers. The study doesn’t report on the best visualisation of data or how to safely display data. A comprehensive safety evaluation would require post-implementation studies. Wait time predictions would depend on EDs collecting and analysing electronic metadata in real-time (not currently universally available). Interviews presumed that predicted wait times are accurate, how consumers respond to inaccurate predictions hasn’t been explored.

Conclusion

Consumers and paramedics want wait time information visibility. They would use the information in a variety of ways, both pre-hospital and whilst waiting for care. Wait time visibility is likely to improve the patient experience of emergency medicine.