Objective The utilisation of medical scribes in the USA has enabled productivity gains for emergency consultants, though their personal experiences have not been widely documented. We aimed to evaluate the consultant experience of working with scribes in an Australian ED.
Methods Emergency consultants working with scribes and those who declined to work with scribes were invited to participate in individual interviews (structured and semistructured questions) about scribes, scribe work and the scribe program in October 2016.
Results Of 16 consultants, 13 participated in interviews, that is, 11 worked with scribes and 2 did not and 3 left Cabrini prior to the interviews. Consultants working with scribes found them most useful for capturing initial patient encounters, for finding information and completing discharge tasks. Scribes captured more details than consultants usually did. Editing was required for omissions, misunderstandings and rearranging information order, but this improved with increasing scribe experience. Consultants described changing their style to give more information to the patient in the room. Consultants felt more productive and able to meet demands. They also described enjoyment, less stress, less cognitive loading, improved ability to multitask, see complex patients and less fatigue.
In interviews with the two consultants declining scribes, theme saturation was not achieved. Consultants declining scribes preferred to work independently. They did not like templated notes and felt that consultation nuances were lost. They valued their notes write-up time as time for cognitive processing of the presentation. They thought the scribe and computer impacted negatively on communication with the patient.
Conclusion Medical scribes were seen to improve physician productivity, enjoyment at work, ability to multitask and to lower stress levels. Those who declined scribes were concerned about losing important nuances and cognitive processing time for the case.
- emergency care systems
- emergency departments
- extended roles
- qualitative research
Statistics from Altmetric.com
- emergency care systems
- emergency departments
- extended roles
- qualitative research
What is already known on this subject
Particularly with the rise in the use of digital records, scribes have been increasingly seen as a solution to improving physician productivity. Only one survey study has looked at ED physician preferences, and this was in an academic centre in the USA.
What this study adds
We instituted a scribe programme in an Australian ED where consultants deliver care directly and interviewed those who used and did not use a scribe. Most physicians liked working with a scribe, finding it improved productivity, improved communications with patients and decreased stress. Those consultants who did not wish to use scribes felt that scribe notes missed nuance and readability, preferred writing their own notes as a way of organising their thinking and plan and felt that scribes negatively impacted on the patient–physician relationship.
Emergency consultants perform a significant number of clerical tasks in the provision of direct patient care, taking up to 43%1 2 of the doctors’ time. This is increasing with the transition to digital data records. Medical scribes may or may not provide cost-effective assistance with this.3–11 While productivity gains are worthy of pursuit, the major day-to-day impact of the scribe will be on the doctor.
The experience of working with medical scribes was previously described by primary care physicians in the USA, highlighting concerns about the accuracy and detail of scribed notes as well as the importance of the relationship between scribe and physician.12 In comparison to primary care physicians, however, emergency consultants often manage multiple and more acute patients simultaneously in longer, but one-off encounters that are often interrupted and necessitate shifting in and out of different patient’s charts.
Emergency consultants are continually required to adapt their work styles to optimise resource utilisation for their patients.13–16 Their experience adapting to work with scribes is not yet broadly described; however, Hess,10 in a single centre USA survey, notes that 60% of ED providers liked or really liked having a scribe, 74% had an overall positive attitude towards the scribe intervention, whereas 9% would be happier without one.
Scribe implementation should not be undertaken without an attempt to understand the impact on consultants who work with them and the reasons some consultants decline to work with a scribe.17 This might guide policies on scribe programme development, scribe recruitment and the allocation of a scribe to a consultant. The current qualitative study aimed to describe the experience of Australian emergency consultants working with scribes and the reasons some consultants elected to not work with a scribe.
This was a prospective, qualitative study conducted as part of a scribe economics study (ACTRN12615000607572), investigating the experiences of consultants during implementation of a scribe programme. The programme was initiated in January 2016. Individual, semistructured interviews were conducted in October 2016.
The study was undertaken at Cabrini Emergency Department. It is a tertiary, not-for-profit, Catholic private hospital in the southeast of Melbourne, Australia. The network (five campuses and one ED) has 832 beds; the ED has 24 000 annual visits (adult and paediatric). The average patient age is 56 years; the admission rate is 50%.
Consultants provide direct patient care with limited supervisory or flow responsibilities. The consultants take histories and examine patients, arrange all tests, obtain previous test results from external facilities, liaise with health workers and families, locate specialists, complete medication charts, book beds and appointments and complete free-text electronic medical record (EMR) documentation. Until the scribe programme, the consultants had no assistants. The average patients per-hour per-consultant is 0.83,11 which is typical of EDs serving complex, older patients in Australia with high admission rates.
Consultants eligible to receive scribes as part of the economic study were eligible to participate in interviews. They were permanent staff, working at least one shift per week at Cabrini. They were all offered training in how to use a scribe. The primary investigator of the economic study (KJW) was excluded from interviews.
Scribes are premedical or medical university students as described previously.11 Scribes are randomly assigned to follow one doctor per shift to stand near the doctor during consultations, capture details of the history, physical examination, investigations and management plans. They complete a templated document at the time of patient consultation via a computer-on-wheels (COW), which consultants edit prior to addition to EMR. They facilitate tests, gather information and document medical decisions, patient progress and diagnoses. They also facilitate patient disposition, appointments and call primary care physicians, family and inpatient consultants. At the time of this study, the programme had been in place for 10 months, and scribes had worked a minimum of 45 shifts each.10
Methods and measurements
Consultants were approached face to face and informed that the ED scribe research group was investigating consultant views on the scribe programme. They were invited to participate in a 20 min voluntary interview containing a combination of structured and semistructured questions and were informed that responses would be anonymous. Interview questions(see online supplementary A) were designed by the author group to explore consultant perspectives on how working with scribes might affect stress, efficiency and safety. The interview was not piloted. A senior medical student (AT) who is independent of the scribe programme conducted interviews with consultants who participated in the scribe programme. These interviews were recorded and transcribed by a secure, independent transcription service. The interviewer for consultants who declined to work with a scribe was a medical student (TLC) and Cabrini scribe, who has previously undertaken qualitative research. Responses were recorded and transcribed within 24 hours by TLC. Responses were not reviewed by the participants for completeness. There were no repeat interviews, and participants were not informed of the findings from the interview phases.
Supplementary file 1
Interviews were analysed using a combination of deductive and inductive methodology. Responses to each question provided deductive results for those specific questions. In addition, inductive methods were used to derive subthemes that incorporated answers from each interview. Subthemes were derived independently from the transcripts using thematic analysis by TLC, KJW, WD and AT. TLC and WD are Cabrini scribes, KJW is the scribe programme director and AT has no relationship with the scribe programme. These researchers all had training in qualitative research, all but AT has prior experience. The researchers used open and axial coding of interview transcripts to extract themes. All phrases were sorted into themes. Quotations were selected that best identified positive and negative responses to each theme. No coding software was used for data analysis.
The study was approved by Cabrini Human Research Ethics Committee (06-21-03-16), prospectively registered (ACTRN12616000202460), and written informed consent was obtained from participants.
Characteristics of study subjects
Sixteen consultants were eligible for inclusion in the economic scribe trial (figure 1). Two declined to work with scribes but agreed to be interviewed. Three consultants who worked with scribes were unable to be interviewed because they had ceased employment with Cabrini. Interviewed consultants were predominantly men (71%), and ages ranged from 39 to 55 years. Self-reported participant characteristics are recorded in table 1.
For consultants who worked with scribes, theme saturation was achieved by the sixth consultant interview. Seven themes were identified, and interview responses for major identified themes have been summarised into quotation tables with interviewees recorded as ‘C.x’ for consultants who worked with scribes, and ‘D.x’ for consultants who declined to work with scribes (tables 2–4).
Themes (and subthemes) identified were the following:
Role of scribes (medical notes, other tasks and when they were most useful);
Quality of work (errors and safety and anticipation of next move);
Impact on consultant work ability (productivity and cognitive processing);
Impact on consultant work satisfaction and stress;
Impact on patient–physician relationship and communication;
Future scribe programme suggestions.
Both consultants who did not work with scribes agreed to be interviewed. One of the interviewed consultants who declined participation in the main scribe trial had previously used a medical scribe in the pilot study; the other had extensively reviewed ED scribed notes but had not worked with a scribe. Both were experienced consultants. There were not enough participants to ensure theme saturation. Their comments contributed to themes 1 (role of scribes), 3 (impact on consultant work ability) and 5 (impact on patient–physician relationship).
Theme 1: Role of scribes
Consultants who chose to work with scribes discussed the difference between physician and scribe capture and interpretation of consultations. In some areas, more detail was captured, in others less. The main areas for consultant editing were where medical synthesis was required. They valued capture of details when seeing complex patients (table 2).
Neither consultant who declined to work with a scribe was concerned about the accuracy of scribed medical notes; however, they were concerned about the nuance and ‘readability’, especially in the setting of sending correspondence to other medical practitioners.
Scribes notes were more literal and this nuance was lost. (D.1)
I think that I wasn’t as happy with my notes with a scribe. The formulaic approach that bounded the scribed notes has weaknesses, but it is good at bringing all doctors up to the same standard. (D.1)
Generally, I think the notes are factually accurate but read very poorly if you haven’t seen the patient. The flow and intent and communication style needs review by the doctor. I would need to edit what they’d written because it wouldn’t be worded in my style. (D.2)
To address this concern, one respondent suggested
doctors put a paragraph in at the end that ties it all together.
Consultants described real-time capture of patient history and examination. They also used scribes to seek collateral information about previous patient care. They valued in-room ordering of beds and investigations, allowing them to move between patients without needing to find a computer desk to type notes and order investigations. They found scribes most useful at the beginning and end of patient processing and less useful during results review and specialist consultations (table 2).
Theme 2: Quality of work
Errors and safety
Those using scribes found the quality of the scribes’ work generally improved with time. One consultant who reported an inaccuracy in the medical record (identified in retrospect) could not recall the nature of the error that was noticed and noted improvement since the implementation phase of the scribe programme.
A couple of potentially harmful errors in the beginning (of the scribe program), not anymore. (C.1)
Scribe limitations in information synthesis and inability to prescribe were mentioned.
Anticipation of needs
Consultants valued scribes who were proactive and anticipated their next tasks.
(a good scribe) anticipates where I am going and what I need to do (C.5)
a good sense of timing (C.2)
Theme 3: Impact on consultant productivity and cognitive processing
Consultants using scribes reported that they allowed the physicians to feel more productive and spend less time using computers. They valued the in-room capture of information by the scribe and felt that this allowed them to focus on patient assessments and their clinical role, rather than documentation (table 3).
Doctors using scribes reported that they were less stressed and less tired at the end of shifts. They felt better able to manage several patients and to see several complex patients. At times, however, scribes pressured them to take on more than they had capacity to manage (table 3).
Notably, both doctors using scribes and those declining them felt there was something to be gained in writing in their own notes. Doctors using scribes reported that they felt that writing their own notes better allowed them to synthesise information.
Like the consultants using scribes, consultants without scribes reported concerns about the loss of the thinking time to process the information in the case that they usually have while writing notes.
I think writing my notes is a way of organising them and my plan. I would prefer if I wrote them myself. (D.2)
Theme 4: Impact on consultant work satisfaction and stress
Interviews with consultants using scribes
Many consultants working with scribes reported a reduction in stress levels and exhaustion. Shifts were more enjoyable, and doctors left work on time (table 4).
Theme 5: Impact on patient–physician relationship and communication
Both consultants using scribes and those who declined scribes agreed that the use of the scribe changed the patient–physician relationship and communication style. Those using scribes had to learn to tell patients in the room of the exam findings and their plans. They reported going on to use this style even without a scribe present.
the first sessions with a scribe can be confronting because it’s a different style of working (C.8)
you have to change the way you communicate information with patients (C.8)
within a couple of weeks it wasn’t an issue anymore—I even adopted [the stylistic changes] as a normal part of my practice (C.8)
One consultant noted that they spent less time answering patient questions when with a scribe. This consultant suggested that their clinical impression and management plan were articulated more when a scribe was in the room so the patient had fewer questions.
Patients are getting more information from me, understanding what I’m doing and why. (C.8)
Consultants declining scribes voiced concerns about how a scribe, as an extra staff member in the room, would affect the interaction between consultant and patient, particularly in regard to the patient revealing private or potentially embarrassing information,
I think the third person in the room negatively impacts on patient communication. (D.1)
They were also concerned about the logistics of scribes in terms of space on the department floor and in small patient cubicles.
Having more than one scribe on at once is physical nightmare due to space. (D.1)
Theme 6: The relationship between consultant and scribe
Consultants reported maximal productivity and enjoyment when working with a scribe with whom they had a good relationship owing to the ease of communication.
The ones that I gel with better, we understand each other better and communicate more clearly which makes things easier. (C.5)
The personalities of the consultants and scribes appeared important.
There are one or two scribes that irritate me. Just like there are one or two scribes that are fantastic… It’s about different personalities. (C.2)
Several consultants noted scribe–doctor working relationships improved with time together.
The more I worked with certain scribes, the more rapport we formed and the better I think we performed. (C.5)
The main reason reported by the two consultants who opted to not work with a medical scribe was a preference to work independently.
I think I don’t like the style, I prefer to work independently. (D.2)
Theme 7: Future scribe programme
All consultants receiving scribes stated that they would prefer to work with a scribe in the future. Most did not recommend any changes to the current programme. The most common suggestion was to work with
more scribes in the future (C.6)
followed by the consultant’s use of the same scribe to foster an optimal and efficient working relationship or to select which scribe they worked with.
Work out which scribe you are most productive with and only work with them. (C.5)
Value in having scribes was found particularly on busy shifts and mainly during the initial consultation by one consultant, so it was suggested
a scribe could be shared between clinicians (C.1).
For those declining scribes, value was identified in the use of medical scribes in both Australia and internationally,
I think it’s interesting and has huge potential in other settings. (D.1)
Interviewed consultants who worked with scribes (11/16) described themes of contentment with scribed notes and value gained from the scribe role. They edited mainly for medical synthesis of information, which was the main concern regarding notes of consultants who declined to work with scribes (2/16). Consultants in both groups agreed that writing their own notes allowed them time to think about presentations and synthesise information. Consultants who declined to work with scribes preferred their own nuanced notes and this thinking time. Yan et al identified a similar concern held by consultants regarding having control over their notes.12 Consultants who worked with scribes, however, reported improved productivity, workplace satisfaction and ability to meet key (time) performance indicators. They reported less workplace stress and exhaustion and less need to stay behind after their shift to complete notes.
Consultants in both groups agreed that scribes change the communication between doctor and patient; however, consultants who worked with scribes reported value from this change. They delivered more information to patients and incorporated this new style into their non-scribed interactions with patients. They were also concerned about the consultant–patient relationship with an observer and the COW size. Dunlop et al, however, report that scribes and the COW did not impact on the patient experience.18
Suggestions for future scribe programme indicated a preference to work with one scribe for a consistent working relationship. All consultants currently working with scribes would prefer to have a scribe with them every shift.
Fit with current knowledge
This is one of the first studies to begin to explore the consultant experience of using scribes. The preference of 11/16 consultants (all those interviewed who worked with scribes) to continue to work with scribes is supported by Hess et al 10 who found that 73.8% of surveyed consultants had a positive experience, and 76% felt that they spent more time with patients when working with a scribe.10 Compared with primary care physicians, emergency consultants in this study described similar concerns regarding the detail and accuracy of scribed notes and the importance of the relationship between doctor and scribe.12 Unlike primary care physician colleagues, consultants also felt an enhanced ability to simultaneously manage several complex patients. The divergence of opinions about working with scribes falls roughly into the proportions you would expect with change.19 This study suggests that attention needs to be paid when editing scribed notes, and a summary paragraph dictated by the consultant may be useful to provide improve nuance in the notes. The quality of scribed notes in ED has currently not been thoroughly examined; however, there is some evidence to suggest that scribed notes contain less omissions than physician’s notes.20 21
This was a small single-centre pilot evaluation in Australia using small cohorts of scribes and consultants in a private tertiary ED. Scribes were relatively junior, and consultants had relatively little exposure to scribes. The study interviewed consultants who were relatively inexperienced in working with scribes. Owing to scribe availability being mainly limited to evenings and weekends, consultants who worked weekends gained more exposure to scribes. Further work should include multiple settings and greater numbers of consultants, exploring themes identified in this pilot.
The study was susceptible to sponsor, social desirability and confirmation biases (investigators worked at the study site and were involved in the scribe programme). Owing to logistical restrictions, the primary author conducted interviews with consultants who did not work with scribes while an independent author interviewed the participating consultants. This may have introduced author bias. These limitations could bias the study in favour of scribes.
All consultants who had used medical scribes valued them and preferred to work with a scribe in the future. Consultants felt less stressed and felt a reduction in cognitive loading when they worked with a scribe. They felt more able to multitask when a scribe was present. Consultants suggested that working with the same scribe every shift and the development of rapport between consultant and scribe would be valuable. Consultants who did not want to work with a scribe preferred to work independently, were concerned with the loss of nuance in the notes and the synthesising they undertook while writing their own notes.
We thank our ED consultants, staff and scribes.
Contributors TLC, EG-B and WAD are Cabrini scribes. KJW is the scribe program director. AT has no relationship with the scribe program. KJW and MB-M secured funding. TLC, KJW, WAD and MS designed the methods. TLC and AT collected the data. TLC, KJW, WAD and AT analysed the data. TLC, KJW, EG-B and WAD wrote the manuscript. All authors revised the manuscript. TLC is the guarantor for the manuscript.
Funding The equity trustees and Cabrini Foundation/Institute funded this study.
Competing interests KJW reports grants from Cabrini Foundation, Equity Trustees and the Phyllis Connor Memorial Fund and non-financial support from Cabrini Institute and Cabrini Health during the conduct of the study.
Ethics approval Cabrini Human Research Ethics committee.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.