Editorial note: this blog is co-authored by Chris Gray and Katie Walker. Katie was lead author on the study. Chris Gray is well known to you as a St Emlyn’s author and editorial board member. I have indicated where comments are limited to Chris (as critical appraisal lead) or Katie (as author). Both authors have reviewed the manuscript and are happy with the content.
Help a patient, write your notes, help a patient, write your notes. Rinse and repeat ad infinitum.
Unwittingly, we might have just condensed the entirety of medicine into just two actions. Hmm. Anyway, thinking about it further, with the increase in allied health professionals such as advanced nurse and clinical practitioners, emergency practitioners and physician associates there is more and more support available for the “help a patient” side of the profession. Emergency medicine however remains one of those specialties where clinicians of all grades still have to write their own notes for each patient they see. Medical or surgical consultants generally have junior doctors to scribe on the ward round, secretaries to type up dictated clinic letters, assistants to write the op note. Down in the ED, even the professors have to put their own (usually more fancy) pen to paper, or finger(s) to keyboard depending on how much your department has evolved.
There may be a solution – scribes.
What’s a scribe?
A scribe is a clerical assistant whose primary role is to complete the medical record. They accompany the clinician, documenting history, examination, recording procedures, results of investigations, further consultations or re-evaluations. They also have access to computer systems and patient records and are able to order investigations, book follow up appointments or complete referral letters. Whilst scribes are usually also health workers in training, such as medical students, they are not healthcare providers, and so are unable to carry out procedures such as giving medication, or tending to hygiene needs.
In 2014, Nat May and Damian Roland wrote a journal club review1,2 here at St Emlyn’s of a before and after study looking at the introduction of scribes into an American emergency department. They also discussed they ways that they take notes in their own practice and some valuable opinions on the pros and cons of scribes in the ED. It’s a great read and provides a lot of background information too, which I won’t repeat again here, so make sure you take a look!
The paper we’re going to be taking a look at today was published in the BMJ only a couple of weeks ago at the end of January 20193, looking at whether ED scribes could improve productivity and patient throughput. As ever we’d advise you to read the evidence for yourself to form your own opinion.
Tell me about the paper.
This is a prospective, multicentre, randomised controlled trial performed over five hospitals in Victoria, Australia over a little more than a two year period. They randomised emergency physicians to shifts with or without a scribe, then compared productivity between these shifts.
Of course, it would be difficult to blind such a study, and likely lead to an unfortunate situation in court a few months down the line where a lack of notes from the encounter would finally make you realise that that patient was in the ‘placebo’ arm. Not ideal. However, what is ideal is performing an RCT to assess the impact of an intervention here.
What’s the background?
Before undertaking this RCT, Katie Walker (who co-authored this blog) and her team undertook a feasibility study to assess the costs of training medical scribes, at the lead hospital in Melbourne. It was published in the EMJ and I wrote a brief commentary on this paper over at the EMJ blog in 2016. It is this training programme that nurtured the scribes who took part in today’s paper.
If you want to read more about the cost analysis, and information on the training process for the scribes, have a look at that paper.
Who were the patients?
Well, this study is slightly different, as it wasn’t patients that were randomised to the treatment, but rather the clinicians/scribes. Trained scribes were allocated what appears to be two weeks in advance to participating clinicians (88 in total) by random number generator. Full details are in the paper, but this seems a fair way to decide this whilst still ensuring rotas were distributed in a timely fashion.
Clinicians signed up voluntarily, and only included consultants, though an exception was made to include very senior trainees at the two smaller hospitals. There were no scribes allocated on night shifts or bank holidays.
Patients were not randomised, and the clinicians had no restrictions on who they could see, though of course consultants were free to adjust their own workload and select patients to see based on workflow and clinical priority needs.
What did they do?
Scribes accompanied the clinician for the full shift, documented the encounter, and performed other appropriate tasks. The medical record was checked by the clinician before sign-off. Emergency consultants (or occasionally advanced trainees) were recruited and trained scribes were randomised to their clinical shifts (or not). Junior doctors and night shifts were excluded.
The primary outcome measure was productivity, defined as patients per doctor per hour.
Secondary outcomes were primary patients per doctor per hour (i.e. patients the doctor had seen from start to end rather than handovers), productivity in various parts of the ED, door-to-doctor times, and emergency department length of stay.
They also looked at patient safety events related to the scribe, and completed another cost evaluation.
Was the study big enough?
The study was powered to detect a 15% increase in productivity using scribes which seems a lot, but actually (assuming an 8-10 hour shift and using their previous work which found on average doctors see 0.83 patients per hour) only equates to one extra patient per shift. Their power calculations are given in the paper, but look reasonable and you can read more about how they work on Simon’s intro to power calculations blog.
They met the required numbers of patients for their calculations.
What were the results?
Overall, 589 scribed and 3263 unscribed shifts were analysed. Each site needed to provide 100 scribed and non-scribed shifts
The team found a 15.9% increase in the number of patients seen overall per doctor per hour when a scribe was present. In real terms this is an increase from 1.13 to 1.31 patients per hour or around 1-2 patients per shift. Primary patient productivity increased from 0.83 to 1.04 patients per hour (25.6%). The most gains were obtained when a scribe was placed with a senior doctor in triage, however there were a low number of encounters here.
There was no change in door-to-doctor time, though ED length of stay decreased by an average of 19 minutes per patient.
Chris Gray on unresolved questions and concerns?
Chris: This is a well powered study and has set out its objectives well. Overall, it’s a good study. However, there may be a few issues, mainly relating to selection bias.
Chris: The clinicians in the study all volunteered, which may mean a particular type of doctor. Potentially those already enthusiastic about scribes, maybe those who have already worked with or trained those scribes at the centres. Maybe consultants who are a bit slower with technology/typing (if they even exist…) and leapt at the chance for someone else to do the computer work. We don’t know. However, this is probably the population who would use scribes day-to-day anyway, but it’s all self-selective, so could bias the results. We also know that in another study (also by Katie) increases in productivity did not always increase financial gains4 (but that was in a smaller, single centre study).
Chris: I was surprised to see that only day shifts were taken into account, however there were a small number of scribes, so this seems reasonable. During the day though, there are often more people around, with less pressure on the system. Could this have affected the study? On nights and bank holidays, would scribes make even more of an impact?
Chris: Consultants followed their usual work pattern, so were able to select patients to see if needed based on their clinical judgement. It’s possible that the case mix they elected to see varied depending on whether they had a scribe or not (see also the Hawthorne effect), though of course it’s hard to determine this from these data.
Chris: Due to the randomisation of scribe-clinician pairings, there also wasn’t the opportunity for working relationships to form, which could improve productivity more.
What’s the take home?
This is the first study of its type to have been performed, and although it was done over in Australia, it does suggest that scribes could improve things in the ED. We believe that we can extend that to say that it suggests that scribes could improve the productivity of some clinicians.
Many clinicians, us included, may find that writing things down helps us to consolidate the patient story, ensuring we’ve asked all the questions we need to, and allowing ourselves to repeat it back to them so that we can both make sure we haven’t missed something important out.
In the UK there may be more systematic issues that could be addressed within our EDs that could allow us to see one extra patient a shift, before thinking about scribes.
Chris: I have worked mainly in dual paper/electronic-based departments though, so my opinion may change in the future as we see more electronic medical record software take the forefront in the emergency department.
Scribes haven’t yet been evaluated in the UK as far as we know. However, we do know that patients tolerate scribes well5 and that emergency doctors really like working with scribes6 (85% don’t want to work another shift without a scribe). We also suspect that the scribes learn a tremendous amount that stands them in good stead for their future careers and we know that they like being paid to undertake meaningful work at the bedside as an embedded team member.
Ed – So Katie is obviously an advocate for scribes and we can see that in the paper and commentary. As for Chris as an impartial reviewer, what’s your bottom line?
Chris: I would certainly give scribes a go, to enable to me to see if it would help my practice. It would be interesting to hear if anyone in the UK has any experience of medical scribes as well! Let us know.
Katie: Thanks Chris. As an Australian researcher it’s really interesting to her the UK view on our paper. If anyone is seriously interested in evaluating scribes in the UK, I would be interested in partnering with them and helping them get the study off-the-ground and delivered. I’d love to understand whether scribes should be implemented in the NHS. You can contact me by email at email@example.com
Chris (aka @graydoc) and Katie
Katie is an emergency physician at Cabrini Hospital, Melbourne; where she is the director of emergency medicine research. She is an adjunct clinical associate professor at Monash University. Her undergraduate degree was from Bristol University, UK.
You can read more about this paper on Salim Reazzaie’s REBELEM blog written by Melanie Stephenson 7
How you can support St Emlyn’s
- Subscribe to the blog (look top right for the link)
- Subscribe to our PODCAST on iTunes
- Follow us on twitter @stemlyns
- PLEASE Like us on Facebook
- Find out more about the St.Emlyn’s team
- Find out more about the MMU MSc in Emergency Medicine here.
- 1.May N. JC: Time to Scribe? Introducing Scribes into the ED. St.Emlyn’s. https://www.stemlynsblog.org/jc-scribes-ed/. Published May 23, 2014. Accessed February 21, 2019.
- 2.Bastani A, Shaqiri B, Palomba K, Bananno D, Anderson W. An ED scribe program is able to improve throughput time and patient satisfaction. The American Journal of Emergency Medicine. May 2014:399-402. doi:10.1016/j.ajem.2013.03.040
- 3.Walker K, Ben-Meir M, Dunlop W, et al. Impact of scribes on emergency medicine doctors’ productivity and patient throughput: multicentre randomised trial. BMJ. January 2019:l121. doi:10.1136/bmj.l121
- 4.Walker KJ, Ben-Meir M, Phillips D, Staples M. Medical scribes in emergency medicine produce financially significant productivity gains for some, but not all emergency physicians. Emergency Medicine Australasia. March 2016:262-267. doi:10.1111/1742-6723.12562
- 5.Dunlop W, Hegarty L, Staples M, Levinson M, Ben-Meir M, Walker K. Medical scribes have no impact on the patient experience of an emergency department. Emergency Medicine Australasia. June 2017:61-66. doi:10.1111/1742-6723.12818
- 6.Cowan TL, Dunlop WA, Ben-Meir M, et al. Emergency consultants value medical scribes and most prefer to work with them, a few would rather not: a qualitative Australian study. Emerg Med J. September 2017:12-17. doi:10.1136/emermed-2017-206637
- 7.Hate Using Electronic Hospital Records? An Evaluation of Medical Scribes in Emergency Departments. – REBEL EM – Emergency Medicine Blog. REBEL EM – Emergency Medicine Blog. http://rebelem.com/hate-using-electronic-hospital-records-an-evaluation-of-medical-scribes-in-emergency-departments/. Published February 21, 2019. Accessed March 1, 2019.